Ardila A 2016 La Neuropsicologia Como Una Forma de Vida Neuropsychology As A Way of Living

!
La neuropsicologa como una forma de vida
Neuropsychology as a way of life
Alfredo Ardila
2016
A todos mis queridos colegas y To all my dear colleagues and
amigos con quienes he compartido friends, with whom I have shared this
esta extraordinaria e incomprensible extraordinary and incomprehensible
aventura que se llama vida adventure that is known as life.
Septiembre 4, 2016 September 4th, 2016
Alfredo Ardila
Florida International University
Miami, Florida, USA
ardilaa@fiu.edu;
ardilaalfredo@gmail.com
Contenido Content
0. Un viaje hacia lo desconocido 0. A trip to the unknown
1. Asimetra cerebral 1. Brain asymmetry
Bernal, B. & Ardila, A. (2014). Bilateral Bernal, B. & Ardila, A. (2014). Bilateral
representation of language: A critical review representation of language: A critical review
and analysis of some unusual cases Journal and analysis of some unusual cases Journal
of Neurolinguistics, 28, 63-80 of Neurolinguistics, 28, 63-80
2. Afasias 2. Aphasias
Ardila, A. (2010). A proposed re- Ardila, A. (2010). A proposed re-

interpretation and reclassification of aphasia interpretation and reclassification of aphasia
syndromes. Aphasiology. 24 (3), 363394 syndromes. Aphasiology. 24 (3), 363394
3. Alexias y agrafias 3. Alexias and agraphias
Ardila, A. (2004). There is not any specific Ardila, A. (2004). There is not any specific
brain area for writing: From cave painting to brain area for writing: From cave painting to
computers. International Journal of computers. International Journal of
Psychology, 39, 61-67 Psychology, 39, 61-67
4. Modelos cognitivos de las alexias 4. Cognitive models of alexias
Ardila, A. & Rosselli, M. (2014). Spanish and Ardila, A. & Rosselli, M. (2014). Spanish and
the characteristics of acquired disorders in the characteristics of acquired disorders in
reading and writing / El espaol y las reading and writing / El espaol y las
caractersticas de los trastornos adquiridos caractersticas de los trastornos adquiridos
de la lectura y la escritura. Estudios de de la lectura y la escritura. Estudios de
Psicologa / Studies in Psychology, 2014 Psicologa / Studies in Psychology, 2014
5. Acalculias 5. Acalculias
Ardila, A. & Rosselli, M. (2002). Acalculia Ardila, A. & Rosselli, M. (2002). Acalculia
and dyscalculia. Neuropsychology Review, and dyscalculia. Neuropsychology Review,
12, 179-232. 12, 179-232.
6. Desarrollo de pruebas de diagnstico 6. Development of neuropsychological

neuropsicolgico assessment instruments
Ardila, A. (2007). Towards the development Ardila, A. (2007). Towards the development
of a cross-linguistic naming test. Archives of of a cross-linguistic naming test. Archives of
Clinical Neuropsychology. 22, 297-33. Clinical Neuropsychology. 22, 297-33.
7. Envejecimiento y demencia 7. Aging and dementia
Ardila, A. (2007). Normal aging increases Ardila, A. (2007). Normal aging increases
cognitive heterogeneity: Analysis of cognitive heterogeneity: Analysis of
dispersion in WAIS-III Archives of Clinical dispersion in WAIS-III Archives of Clinical
Neuropsychology, 22(8):1003-11 Neuropsychology, 22(8):1003-11
2
8. Neuropsicologa infantil 8. Child neuropsychology
Ardila, A., Matute, E. & Rosselli, M. (2005). Ardila, A., Matute, E. & Rosselli, M. (2005).
Parents educational level and development Parents educational level and development
of executive functions in five to 15-year-old of executive functions in five to 15-year-old
children. Developmental Neuropsychology, children. Developmental Neuropsychology,
28(1), 539-60. 28(1), 539-60.
9. Problemas especficos de aprendizaje 9. Specific learning disabilities
Ardila, A. (1997). Specific learning dis- Ardila, A. (1997). Specific learning dis-
abilities: A neuropsychological perspective. abilities: A neuropsychological perspective.
International Journal of Neuroscience, 89, International Journal of Neuroscience, 89,
189-205. 189-205.
10.Tartamudez 10.Stuttering
Ardila, A., Ramos, E., & Barrocas, B. (2011). Ardila, A., Ramos, E., & Barrocas, B. (2011).
Patterns of stuttering in a Spanish/English Patterns of stuttering in a Spanish/English
bilingual: A case report. Clinical Linguistics bilingual: A case report. Clinical Linguistics
& Phonetics, 25(1), 2336 & Phonetics, 25(1), 2336
11. Intoxicacin por cocana 11. Cocaine intoxication
Ardila, A., Rosselli, M., & Strumbasser, S. Ardila, A., Rosselli, M., & Strumbasser, S.
(1991) Neuropsychological effects of (1991) Neuropsychological effects of
cocaine abuse. International Journal of cocaine abuse. International Journal of
Neuroscience, 57, 73-79. Neuroscience, 57, 73-79.
12. Estandarizacin de pruebas 12. Standardization of assessment tests

diagnsticas
Ardila, A., Ostrosky-Solis, F., & Bernal, B.
Ardila, A., Ostrosky-Solis, F., & Bernal, B. (2006). Cognitive test performance toward
(2006). Cognitive test performance toward the future: The example of semantic verbal
the future: The example of semantic verbal fluency (ANIMALS). International Journal of
fluency (ANIMALS). International Journal of Psychology, 41, 324-332
Psychology, 41, 324-332
13. Variables educacionales 13. Educational variables
Ardila, A.., Bertolucci, P. H. , Braga, L.W., Ardila, A.., Bertolucci, P. H. , Braga, L.W.,
Castro-Caldas, A., Judd, T., Kosmidis, M.H., Castro-Caldas, A., Judd, T., Kosmidis, M.H.,
Matute, E., Nitrini, R., Ostrosky-Solis, F., & Matute, E., Nitrini, R., Ostrosky-Solis, F., &
Rosselli, M. (2010). Illiteracy: Neuro- Rosselli, M. (2010). Illiteracy: Neuro-
psychology of cognition without reading. psychology of cognition without reading.
Archives of Clinical Neuropsychology, 25, Archives of Clinical Neuropsychology, 25,
689712 689712
14. Neuropsicologa transcultural 14. Cross-cultural neuropsychology
Ardila, A. (2005). Cultural values underlying Ardila, A. (2005). Cultural values underlying
cognitive psychometric test performance. cognitive psychometric test performance.
Neuropsychology Review, 15, 185-195. Neuropsychology Review, 15, 185-195.
3
15. Sndromes neuropsicolgicos 15. Paroxysmal neuropsychological

paroxsticos syndromes
Ardila, A., & Snchez, E. (1988). Neuro- Ardila, A., & Snchez, E. (1988). Neuro-
psychological symptoms in the migraine psychological symptoms in the migraine
syndrome. Cephalalgia, 8, 67-70. syndrome. Cephalalgia, 8, 67-70.
16. Intoxicacin por escopolamina 16. Scopolamine intoxication
Ardila, A., Moreno, C. & Ardila, S.E. (2006). Ardila, A., Moreno, C. & Ardila, S.E. (2006).
Intoxicacin por escopolamina Intoxicacin por escopolamina
('Burundanga'): Controlando la habilidad ('Burundanga'): Controlando la habilidad
para tomar decisiones [Scopolamine para tomar decisiones [Scopolamine
(Burundanga) intoxication: Impairing the (Burundanga) intoxication: Impairing the
decision-making ability] Revista de decision-making ability] Revista de
Neurologia, 42, 125-128 Neurologia, 42, 125-128
17. Inteligencia 17. Intelligence
Ardila, A. (1999). A neuropsychological Ardila, A. (1999). A neuropsychological

approach to intelligence. Neuropsychology approach to intelligence. Neuropsychology
Review, 9, 117-136 Review, 9, 117-136
18. Sndrome disejecutivo 18. Dysexecutive syndrome
Ardila, A. (2008). On the evolutionary origins Ardila, A. (2008). On the evolutionary origins
of executive functions. Brain and Cognition, of executive functions. Brain and Cognition,
68, 92-99 68, 92-99
19. Neuropsicologa de la criminalidad 19. Neuropsychology of criminality
Ardila, A., & Ostrosky-Solis, F. (2009). Ardila, A., & Ostrosky-Solis, F. (2009).
Neuropsicologa de los asesinos en serie. [Neuropsychology of serial killers]. Revista
Revista de Neurologia, 48(3):162-163. de Neurologia, 48(3):162-163.
20. Origen de la cognicin humana 20. Origins of human cognition
Ardila, A. (2010). On the evolution of Ardila, A. (2010). On the evolution of

calculation abilities. Frontier in Evolutionary calculation abilities. Frontier in Evolutionary
Neurosciences, 2, 1-8 Neurosciences, 2, 1-8
21. Evolucin del lenguaje 21. Language evolution
Ardila, A. (2015). A Proposed Neurological Ardila, A. (2015). A Proposed Neurological

Interpretation of Language Evolution. Interpretation of Language Evolution.
Behavioral Neurology. Volume 2015, Article Behavioral Neurology. Volume 2015, Article
ID 872487, ID 872487,
22. Rehabilitacin neuropsicolgica 22. Neuropsychological rehabilitation
Rosselli, M. Ardila, A. & Beltran, C. (2001). Rosselli, M. Ardila, A. & Beltran, C. (2001).
Rehabilitation of Balints Syndrome: A Single Rehabilitation of Balints Syndrome: A Single
Case Report. Applied Neuropsychology, 8, Case Report. Applied Neuropsychology, 8,
242-247. 242-247.
4
23. Bilingismo 23. Bilingualism

Ardila, A., Rosselli, M., Ostrosky-Solis, F., Ardila, A., Rosselli, M., Ostrosky-Solis, F.,
Marcos, J., Granda, G., & Soto, M. (2000). Marcos, J., Granda, G., & Soto, M. (2000).
Syntactic comprehension, verbal memory Syntactic comprehension, verbal memory
and calculation abilities in Spanish-English and calculation abilities in Spanish-English
bilinguals. Applied Neuropsychology, 7, 3- bilinguals. Applied Neuropsychology, 7, 3-
16. 16.
24. Un modelo de la cognicin humana 24. A model of human cognition
Ardila, A. & Bernal, B. (2007). What can be Ardila, A. & Bernal, B. (2007). What can be
localized in the brain? Towards a "factor" localized in the brain? Towards a "factor"
theory on brain organization of cognition. theory on brain organization of cognition.
International Journal of Neuroscience, 117, International Journal of Neuroscience, 117,
935-69. 935-69.
25. reas de Brodmann en el lenguaje 25. Brodmanns areas in language
Ardila, A., Bernal, B., Rosselli, M. (2016). Ardila, A., Bernal, B., Rosselli, M. . (2016).
How localized are language brain How localized are language brain
areas? A review of Brodmann areas areas? A review of Brodmann areas
involvement in language. Archives of involvement in language. Archives of
Clinical Neuropsychology, 31, 112-122. Clinical Neuropsychology, 31, 112-122.
5
0. A trip to the unknown
0. Un viaje hacia lo desconocido When I was approximately 15 or 16 years

old, I read once that the reason why the right
Cuando tena probablemente unos 15 o 16 hand was predominantly used to write was
aos le en algn libro que utilizamos la because languages location on the brains
mano derecha para escribir simplemente left hemisphere. Consequently, I concluded
porque el lenguaje se encuentra localizado that if I taught myself to write with my left
en el hemisferio izquierdo del cerebro; hand, then language would be located in my
pens entonces que si escriba con la mano right hemisphere as well; expanding and
izquierda el lenguaje tambin se localizara developing language in both hemispheres, I
en el hemisferio derecho y supuse que eso supposed, this could be convenient! In the
seria muy conveniente: tener el lenguaje en weeks that followed I patiently practiced
ambos hemisferios cerebrales. Durante las switching hands while writing, hoping that
semanas siguientes estuve pacientemente my left-handed writing skills would soon
practicando escribir con la mano izquierda, develop. Unfortunately, in time, I realized it
pero cuando me di cuenta que nunca lo was a much more complex skill than I
lograra, desist y me olvid del asunto. Este expected. I decided to give up this personal
fue mi primer encuentro con la project and, eventually, forgot about it; this
neuropsicologa. was my first encounter with
neuropsychology.
En 1971, siendo profesor de psicologa en la
Universidad Javeriana en Bogot, Much later, in 1971, as a psychology
organizamos con mi gran amigo Carlos professor in the Universidad Javeriana in
Moreno la primera ctedra de psicofisiologa Bogot, I organized with my great friend,
que se ense en Colombia. Tom a mi Carlos Moreno, the first course of
cargo entre otros temas el capitulo de psychophysiology ever developed in
funciones corticales. Este fue mi segundo Colombia. I was primarily responsible for
encuentro con la neuropsicologa. Creo que topics concerning cortical functions,
fue la primera vez que vi el nombre de Luria. amongst others; this was my second
Qued convencido que esto era lo que encounter with neuropsychology. It was also
quera hacer en mi vida: estudiar cmo the first time I saw Lurias name mentioned.
cambia el cerebro cuando aprendemos algo, I was convinced this was what I wanted to
cmo cuando hablamos estamos utilizando do with my life, study how the brain supports
las regiones frontales del hemisferio cognition. For instance, the use of the
izquierdo, y cuando reconocemos una occipital lobes when we recognize a figure,
figura, este reconocimiento se basa en los or the use of the frontal left hemisphere
lbulos occipitales del cerebro. Encontr when we speak. I found themes like these
estos temas simplemente fascinantes y simply fascinating, so I decided to pursue a
tom la decisin de hacer un postgrado en graduate degree in a topic related to
algo relacionado con la actividad cerebral y cerebral activity and psychological
los procesos psicolgicos. En este momento processes. This happened to be a time
Colombia y la Unin Sovitica acababan de when Colombia and the Soviet Union had
firmar un acuerdo de intercambio cultural, just signed and approved a student
que inclua que Colombia recibira exchange program, which made Colombian
estudiantes soviticos y la Unin Sovitica students accessible to study in the USSR
recibira estudiantes colombianos. Viajar a and vice versa. I decided to take this
Mosc en cierta forma fue muy fcil, pero opportunity. Although the program made it
fue un viaje a lo desconocido; era casi como very easy for me to travel to Moscow, in a
ir a otro planeta. El mundo se encontraba en way I was taking a trip to what seemed like,
medio de la guerra fra, pero los to a different planet. It was a world caught in
colombianos ramos bienvenidos (o mas the middle of the cold war; yet we,
exactamente, irrelevantes) en uno y en otro Colombians, were welcomed (or more
campo. precisely, irrelevants!) into one or the other
side.
6
Recuerdo la primera vez que me encontr Upon my arrival, I remember with clarity the
con Luria; l coordinaba a los estudiantes first time I met Luria; he was the
extranjeros en la Facultad de Psicologa de international students coordinator in the
la Universidad Estatal de Mosc. Me pidi a Psychology Department for Moscow State
travs de la persona con quien me contact University. He invited me to his apartment
al llegar a Mosc, que fuera a su aparta- through an academic advisor; the person
mento. Estuve largamente practicando whom I contacted when I finally arrived to
cmo lo saludara en ruso, pero obviamente Moscow. I recall long hours trying to
cuando abri la puerta qued en blanco y practice an appropriate greeting in Russian;
solo atin a saludarlo en ingls. Nos needless to say, as soon as Luria opened
reunimos creo que unos 15 o 20 minutos. his door my mind went blank and, to my
Me explic que yo tenia dos posibilidades: o disappointment, I greeted him in English.
trabajar en un laboratorio experimental con We chatted for about 15 to 20 minutes
Sokolov, o trabajar en un rea clnica con where he explained to me my two options
pacientes con patologas cerebrales; en during my stay: I could work in an
este ltimo caso trabajara directamente con experimental laboratory with Sokolov; or
l; le contest que prefera la segunda with Luria in a clinic setting with patients with
opcin, y entonces hizo rpidamente un cerebral pathologies. I told him that I
plan de mis actividades durante los preferred the second option where I would
siguientes cuatro aos. Este fue mi tercer have direct guidance from him. Luria quickly
encuentro con la neuropsicologa. Desde planned my activities for the next four years.
ese da ya han transcurrido 44 aos. This was my third encounter with
neuropsychology; 44 years have passed
Mi vida como estudiante y profesional de la since that day.
neuropsicologa se podra dividir en siete
etapas. My life as a student and as a professional in
neuropsychology could be divided in seven
1. 1972-1976: Universidad Estatal de phases:
Mosc
1. 1972-1976: Moscow State University
Fue una poca llena de novedades de
diverso tipo. Vivir en un pas tan diferente, This was a time of discovery and novelty.
donde se habla una lengua tan lejana, con Living in such a different country from many
un sistema poltico tan distinto, fue una perspectives, where such a different
experiencia nica. Tres veces a la semana language was spoken, whos political
tena un seminario con Luria en el Instituto system was so different, made it a very
Burdenko, junto con otros estudiantes y unique experience. I had a seminar with
asistentes de neuropsicologa, en el que se Luria and his neuropsychology assistants in
examinaba y analizaba un paciente; pero the Burdenko Institute three days a week.
que duraba entre 3 y 4 horas. Luria era una There, each patient was assessed and
persona incansable, de una sencillez discussed for three to four hours. Luria was
natural, y con unos conocimientos e tenacious, with a natural simpathy, and an
intereses asombrosos. Por las tardes incredible knowledge and interest that
frecuentemente tena clases en la facultad defined him. During the afternoons I often
de psicologa; durante algn tiempo tambin had class in the psychology department and,
asista a clases de ruso. Recuerdo a mis for a period of time, I also attended a
compaeros de la facultad y a mis amigos Russian course. I can recall with great
colombianos con quienes nos reunamos los fondness my colleagues, professors, and my
sbados por la tarde. Un da todo termin y Colombian friends with whom I met every
regres a Colombia Saturday afternoon. One day, everything
came to an end and I returned to Colombia.
7
2. 1976-1980: Universidad Nacional de 2. 1976-1980: National University of

Colombia Universidad Nacional Colombia National Autonomous
Autnoma de Mxico University of Mexico
En Colombia me vincul a la Universidad Returning to Colombia I contacted the

Nacional de Colombia, y durante algn National University of Colombia and became
tiempo fui Director del Departamento de the Chair of the Psychology Department. I
Psicologa. Trat de iniciar un programa de attempted to initiate a research program
investigacin que inclua varios temas: including the following themes: cerebral
asimetra cerebral, atencin y funciones asymmetry, cortical functions, and attention.
corticales. Durante un ao viaj a Caracas y I spent one year in Caracas, Venezuela, and
luego a Mxico, a la Universidad Nacional later, I moved to Mexico, to the National
Autnoma se Mxico. Contine con los Autonomous University of Mexico, where I
trabajos sobre asimetra cerebral, y continued my work on cerebral asymmetry
comenc con mi gran amiga Feggy with my great friend Feggy Ostrosky. We
Ostrosky a trabajar sobre diagnstico focused on neuropsychological diagnosis,
neuropsicolgico y factores educacionales y and educational, along with cultural,
culturales en el funcionamiento intelectual y variables in intellectual neuropsychological
la evaluacin neuropsicolgica. . evaluation.
3. 1981-1992: Neurological Institute of

3. 1981-1992: Fundacin Instituto Colombia
Neurolgico de Colombia
On November 1980 I returned to Colombia
En Noviembre de 1980 regres a Colombia and began to work at the Neurological
y me vincul al Instituto Neurolgico de Institute of Colombia. It was a world of the
Colombia. Fue el mundo de la clnica, los clinical activity, with studies of aphasia,
trabajos sobre afasias, sndrome prefrontal, prefrontal syndromes, memory, etc. On
memoria, etc. En agosto de 1981 August 1981, we organized an International
organizamos un Congreso Internacional de Neuropsychological Conference with nearly
Neuropsicologa con cerca de 700 asis- 700 attendees from 14 different countries.
tentes provenientes de 14 pases diferentes. From here emerged the Colombian
Se crea la Asociacin Colombiana de Association of Neuropsychology. Later, with
Neuropsicologa. Mas adelante intentamos my wife, Monica Rosselli, we created the
junto con mi esposa Mnica Rosselli de Colombian Institute of Neuropsychology;
manera poco exitosa- crear el Instituto with less success than we hoped.
Colombiano de Neuropsicologa.
4. 1992-1993: Universidad de San 4. 1992-1993: University of San

Buenaventura, Medelln Buenaventura, Medelln
Dada la crisis del Instituto Neurolgico de Given the Colombian Institute of Neurologys
Colombia, y el poco xito del Instituto crisis, and the poor success of the
Colombiano de Neuropsicologa, estuve Colombian Institute of Neuropsychology, I
explorando varias posibilidades. Los planes was exploring other options. I planned to
iniciales eran trasladarme a Medelln, crear move to Medellin to create a neuro-
un instituto de rehabilitacin neuropsychological rehabilitation institute, along
psicolgica, desarrollar un programa de with develop a program of Neurolinguistics
neurolingstica en la Universidad de in the University of Antioquia and even
Antioquia, e iniciar la maestra en neuro- initiate a Master Program in Neuro-
psicologa de la Universidad San psychology in the University of San Buena-
Buenaventura. Slo el ltimo de estos ventura. However, only the last of those
proyectos pudo realizarse. projects was accomplished.
8
5. 1994-1998: Miami Institute of 5. 1994-1998: Miami Institute of

Psychology Psychology
Inicialmente consider que mi vinculacin al Initially, I felt that my time in the Miami
Miami Institute of Psychology seria durante Institute of Psychology would be for
aproximadamente unos dos aos, en tanto approximately two years; I would later be
lograba un nombramiento formal en el formally appointed part of the UCLA
Neuropsychiatric Institute de la UCLA en Neuropsychiatric Institute in Los Angeles, to
Los ngeles, para trabajar junto con D.F. work together with D.F. Benson. Benson, my
Benson. Ya mi amigo Benson haba iniciado soon-to-be coworker and my dear friend,
todos los trmites, pero apareci algo finished the paperwork so everything could
completamente inesperado: Benson se go as planned. However, an unexpected
enferm con un cncer, que unos pocos event occurred: Benson was diagnosed with
aos mas tarde acabara con su vida. Y a cancer that would eventually end his life.
todos mis planes hacia el futuro cambiaron. My future plans were put on hold. If not for
De no haber sucedido este infortunado this unfortunate event, a great portion of my
evento, probablemente gran parte de mi life would probably have been different. I
vida hubiera sido diferente. Contine en el continued at the Miami Institute of
Miami Institute of Psychology, donde tuve Psychology, where I worked with great
unos excelente colegas y amigos. Pero en colleagues and friends. But at a certain
algn momento decid explorar otras moment, I decided to explore other
posibilidades. possibilities.
6. 1999-2007: Memorial Regional Hospital 6. 1999-2007: Memorial Regional Hospital
Mi vinculacin al Memoria Regional Hospital My time during the Memorial Regional

en Hollywood fue como un regreso al Hospital felt like I was returning to the
Instituto Neurolgico de Colombia. Pero el Neurological Institute in Colombia. However,
trabajo clnico en Estados Unidos tiene un I found that clinical work in the United States
elemento de incertidumbre. Me vincul has an element of uncertainty. I also worked
tambin parcialmente a la Universidad de at the University of Miami where I
Miami para participar en un proyecto de participated in a research project. Still, for
investigacin. Igualmente, estuve durante many years I worked with the Public
varios aos trabajando con la Defensora Defense of Miami, making neuro-
Publica de Miami, realizando evaluaciones psychological evaluations on criminals and
neuropsicolgicas de criminales. Tambin also maintained a private clinical practice.
mantuve un prctica clnica privada.
7. 2002-hasta el da de hoy: Florida 7. 2002- up-to-date: Florida International

International University University
En forma inesperada surgi la posibilidad de I was unexpectedly given the opportunity to

vincularme al nuevo Departamento de become a part of Department of
Ciencias y Trastornos de la Comunicacin Communication Sciences and Disorders at
de la Universidad Internacional de la Florida. the Florida International University. This is a
Este es un programa pequeo donde me small program focused on bilingualism
encontr con excelentes colegas, y donde where I met some excellent colleagues, and
logr las dos cosas mas importantes que found the two most essential elements I
necesitaba para hacer el trabajo que quera needed to accomplish the projects that I
hacer: tiempo y apoyo. Ha sido la poca de wanted: time and support. Here, I have
los trabajos sobre bilingismo, pero tambin done some work on bilingualism along with
la poca de la teorizacin sobre la some theoretic work on cerebral
organizacin cerebral de la actividad organization of cognitive activity.
cognoscitiva.
9
Durante mi vida profesional he trabajado en Throughout my career I have been involved

25 programas diferentes de investigacin. in 25 different research programs. These
Estos programas no han sido secuenciales, programs have not been necessarily
sino que se superponen en el tiempo; en sequenced, but rather overlap in time; in
algunos de ellos he trabajado durante algn some, I have worked for a short period, then
tiempo, luego los he suspendido para took them up again after many years (eg,
retomarlos muchos aos despus (por cerebral asymmetry). Many others, I have
ejemplo, asimetra cerebral). Otros los he continued to partake throughout my life
mantenido a travs de toda mi vida (por (such as aphasia); in some, I started due to
ejemplo, afasias); otros mas los he iniciado certain circumstances in my life that has
como resultado de las condiciones drawn interest in that topic, such as
especificas que he vivido (por ejemplo, bilingualism; finally, in many other cases,
bilingismo); y en algunos otros casos son topics have corresponded with very specific
programas que corresponden a momentos moments of my life (such as neuro-
muy especficos de mi vida (por ejemplo, psychological paroxyamal syndromes). The
sndromes neuropsicolgicos paroxsticos). result of these research pragrams are
Los resultados de estos programas de presented in about 350 journal papers, 100
investigacin aparecen en unos 330 book chapters, and 40 books.
artculos, 100 captulos de libros y 40 libros.
In addition to clinical practice and research,
Adems de la actividad clnica e a good portion of my life has been dedicated
investigativa, he dedicado una parte to a professional activities in neuro-
importante de mi vida a las actividades psychology. This includes: organizing the
profesionales de la neuropsicologa: first international neuropsychological
particip en la organizacin y realizacin del congress developed in Latin America
primer congreso internacional de neuro- (August 1981, Bogota, Colombia), creating
psicologa que se desarroll en Latino- the Colombian Neuropsychological
amrica (Bogot, Agosto de1981); en la Association (1982), participation in founding
creacin de la Asociacin Colombiana de two Latin American neuropsychological
Neuropsicologa (1982); en la fundacin de societies (Sociedad Latinoamericana de
dos sociedades latinoamericanas de Neuropsicologa, SLAN, 1989; Asociacin
neuropsicologa (Sociedad Latinoamericana Latinoamericana de Neuropsicologa, ALAN,
de Neuropsicologa, SLAN, 1989; 1997), and founding and developing the
Asociacin Latinoamericana de Neuro- journal, Neuropsicologa, Neuropsiquiatra y
psicologa, ALAN, 1997); en la fundacin y Neurociencias. I have been president of the
desarrollo de la revista Neuropsicologa, SLAN, the ALAN, and the Hispanic
Neuropsiquiatra y Neurociencias. He sido Neuropsychological Society. I have had the
Presidente tanto de la SLAN como de la honor of contributing as a board member of
ALAN, al igual que de la Hispanic the International Neuropsychological
Neuropsychological Society; miembro de la Society, as well as the first director of the
Junta Directiva de la Sociedad Internacional first Masters Degree program in
de Neuropsicologa. Fui el primer director neuropsychology ever developed in
del Programa de Maestra en Colombia. During my time as a
neuropsicologa que se desarroll en neuropsychology professor I have been
Colombia (Universidad de San invited to teach at about 20 different
Buenaventura, Medelln); he sido profesor universities worldwide and have participated
de neuropsicologa o conferencista invitado in over 200 conferences and congresses in
en aproximadamente 20 universidades neuropsychology and topics related. All
colombianas y extranjeras; y he participado these activities occurred in a period of forty
en mas de 200 congresos o conferencias de years.
neuropsicologa o reas relacionadas. Todo
estas actividades cubren un periodo de 40
aos.
10
A continuacin, cinco fotografas corres- The following includes five pictures

pondientes a diferentes momentos, y un corresponding to different moments in time,
articulo representativo de cada uno de los along with one journal paper representing
programas de investigacin en los cuales he each research program I have participated
participado. in.
Con A.R. Luria. Mosc, 1976. With A.R. Luria, Moscow, 1976
International Conference of Neuropsychology, Bogota, 1981

!
Fundacin de la Sociedad Latinoamericana de Neuropsicologa (SLAN). Foundation of the

Latin American Society of Neuropsychology, Buenos Aires, 1989
2nd International Conference Brain-Mind (Latin American Association of

Neuropsychology ALAN). Izquierda a derecha (left to right): Esmeralda Matute,
Monica Rosselli, Alfredo Ardila, Feggy Ostrosky, Byron Bernal. Cartagena, 2001.
!
th
8 International Conference Brain-Mind (Latin American Association of Neuropsychology
ALAN). Izquierda a derecha (left to right): Feggy Ostrosky, Alfredo Ardila, Monica
Rosselli, Francisco Lopera, David Pineda, Esmeralda Matute. Barranquilla, 2014.
Author's personal copy
Journal of Neurolinguistics 28 (2014) 6380
Contents lists available at ScienceDirect
Journal of Neurolinguistics
journal homepage: www.elsevier.com/locate/
jneuroling
Review
Bilateral representation of language: A critical

review and analysis of some unusual cases
Byron Bernal a, Alfredo Ardila b, *
a
Department of Radiology/Research Institute, Miami Childrens Hospital, Miami, FL, USA
b
Department of Communication Sciences and Disorders, Florida International University, Miami, FL, USA
a r t i c l e i n f o a b s t r a c t
Article history: It is well known that for right-handed individuals, language is

Received 19 August 2013 usually and mostly associated with the left hemisphere activity.
Received in revised form 26 September 2013 The question of the potential bilateral representation of language,
Accepted 14 October 2013
however, has been barely approached. The evidence regarding the
bilateral representation of language taken from Wada test, PET,
Keywords:
fMRI, tractography, and magneto-encephalography is examined.
Brain asymmetry
Departing from the modularity concept and data ow computing
Language
Laterality models, two classications topographic and functional of po-
Brocas area tential language lateralization patterns are proposed; it is pointed
Wernickes area out that language can be bilaterally represented in different pat-
fMRI terns, accordingly with the distribution of the main domains
(expressive vs. receptive) and their subfunctions; and with respect
to different modalities of data ow. Five illustrative cases of
bilateral representation of language are presented. It is concluded
that language dominance is mostly a matter of hemispheric
advantage for a specic cognitive function.
! 2013 Elsevier Ltd. All rights reserved.
1. Introduction
Over one century ago, it was well established that for right-handed individuals, language is usually
and mostly associated with the left hemisphere activity (Broca, 1861, 1865; Dejerine, 1914; Wernicke,
1874). These early ndings were based on clinical-pathological observations but have been replicated
* Corresponding author. Department of Communication Sciences and Disorders, Florida International University, 11200 SW
8th Street, AHC3-431B, Miami, FL 33199, USA. Tel.: 1 305 348 2750; fax: 1 305 348 2740.
E-mail address: ardilaa@u.edu (A. Ardila).
0911-6044/$ see front matter ! 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jneuroling.2013.10.002
64 B. Bernal, A. Ardila / Journal of Neurolinguistics 28 (2014) 6380
in extensive studies using the intracarotid amytal test (so-called Wada test) (e.g., Kurthen et al., 1994;
Loring et al., 1999; Mddel, Lineweaver, Schuele, Reinholz, & Loddenkemper, 2009; Woermann et al.,
2003). Modern modalities of non-invasive neuroimaging and electro-physiology procedures have
validated those initial ndings as well: fMRI (e.g., Binder, 2011; Holland et al., 2007; Price, 2010), near
infra-red spectroscopy (Bembich, Demarini, Clarici, Massaccesi, & Grasso, 2011; Bisconti, Di Sante,
Ferrari, & Quaresima, 2012; Kennan, Kim, Maki, Koizumi, & Constable, 2002), magneto-
encephalography (Kadis et al., 2011), and diffusion tensor imaging/tractography (Matsumoto et al.,
2008; Powell et al., 2006; Rodrigo et al., 2008). However the question of bilateral representation of
language has been barely approached.
The departure question in examining the bilateral representation of language is up to what extent
the right hemisphere can hold language functions. Knecht et al. (2000) measured language laterali-
zation in 326 healthy individuals with functional transcranial Doppler sonography (a non invasive
technique measures the velocity of blood ow through the brains blood vessels using pulsed Doppler
transducer - ultrasonic pulse probe that detects the reected sound from moving blood) utilizing a
word-generation task. The incidence of right hemisphere language dominance was found to increase
linearly with the degree of left-handedness, from 4% in strong right-handers (right handedness
score 100) to 15% in ambidextrous individuals and 27% in strong left-handers (handedness "100).
However, this technique may be inuenced by expertise or research bias because of the potential poor
insonation conditions (Lorenz et al., 2008). Indeed, in a study with 150 healthy subjects (75 left-
handers and 75 right-handers), left-handers exhibited right language dominance in 77.3% of cases
while bilateral representation was observed in 14.7%, and left dominance in 8% of the subjects; 93.3% of
right-handers showed left side dominance, and 6.7% showed bilateral language representation (Basic
et al., 2004).
Khedr, Hamed, Said, and Basahi (2002) assessed language lateralization in normal subjects (25
right-handed and 25 left handed) using transcranial magnetic stimulation. The authors further sub-
divided the groups into strongly right-handed, moderately right-handed, strongly left-handed,
moderately left-handed and ambidextrous. In the strong right handed subjects, 87% of subjects
showed only language disruption with left hemisphere stimulation, while 8.2% exhibited disruption
with stimulation in either hemisphere. 4.2% of subjects had disruption with stimulus in the right
hemisphere. In strongly left-handed subjects, 73.7% of subjects had left hemisphere dominance, 15.8%
had bilateral representation and 10.5% had right side dominance. In ambidextrous subjects bilateral
representation was observed in 57% of cases. The authors concluded that speech lateralized to the left-
side cerebral dominance in strongly right- and left-handed subjects, but bilateral cerebral represen-
tation was frequent in mixed handedness and right-sided cerebral dominance rarely occurred.
2. Normal and anomalous right-sided lateralization of language
In 1865, at a meeting of the Socit de Anthropologie de Paris Paul Broca explicitly stated: "Nous
parlons avec lhmisphre gauche"d"We speak with the left hemisphere. (Harris, 1999). Since then, it
has been accepted that the left hemisphere plays a central role in language. Furthermore, Broca (1865)
assumed that left-handers are the mirror-reverse of right-handers for cerebral control of speech, with
the right hemisphere being dominant in left-handers, and the left hemisphere dominant in right-
handers. This hypothesis has been referred as conjunction hypothesis (Harris, 1991) and was
considered valid during the late XIX century. Nonetheless, it was later observed that aphasia can also be
associated with right hemisphere damage in dextrals; this type of aphasia is known as crossed
aphasia and was initially described by Bramwell in 1899. Bramwell applied this term to two different
conditions: (a) aphasia in a left-hander with right hemiplegia; and, (b) aphasia in a right-handed in-
dividual with left hemiplegia; that is, an aphasia resulting from a cerebral lesion ipsilateral to the
dominant hand. Noteworthy, just a few crossed aphasics were right-handers, but most were left-
handers. He considered that crossed aphasia is relatively frequent as a transient disorder, but it is
extremely unusual as a permanent syndrome; in this latter case, it is only found in left-handers. Hcaen
& Albert (1978) suggested that the term crossed aphasia should be used only to refer to aphasia
following right hemisphere pathology in a right-handed person, and this is the way the term crossed
aphasia is currently used (Ishizaki et al., 2012; Marin, Paghera, De Deyn, & Vignolo, 2004).
B. Bernal, A. Ardila / Journal of Neurolinguistics 28 (2014) 6380 65
The incidence of crossed aphasia is very low (Coppens, Hungerford, Yamaguchi, & Yamadori, 2002).
Hcaen, Mazurs, Ramier, Goldblum, and Merianne (1971) estimated an incidence of 0.38%; while
Benson and Geschwind (1973) proposed a gure of approximately 1%. In large clinical samples, it has
been found to be around 4% in the acute stage and 1% in the chronic stage (Pedersen, Jargensen,
Nakayama, Raaschou, & Olsen, 1995). It is generally accepted that crossed aphasia represents no
more than 3% of all cases of aphasia (Ha, Pyun, Hwang, & Sim, 2012). Some authors, however, have
suggested that the incidence of crossed aphasia could be even lower (Ardila, 2006; Castro-Caldas &
Confraria, 1984).
Right-hemisphere lesions, however, have been found more frequently associated with aphasia in
left-handed individuals (Basso & Rusconi, 1998). Some authors have reported that up to 50% of left-
handers with right hemisphere lesions present aphasia, although currently the accepted percentage
is notoriously lower. Indeed, left hemisphere damage in left-handers may be associated to aphasia in
more than 50% of cases (Benson & Ardila, 1996). The aphasia prole in left-handers in general is similar
between right and left-handers; although it has been suggested that left-handed aphasics are less
impaired in comprehension and writing, but they have reading disorders more frequently than right-
handed aphasics (Hcaen & Sauguet, 1971). Comparing the aphasia due to right and left hemisphere
pathology in left-handed individuals, just minor differences are found. By the same token, comparing
aphasia recovery in right- and left-handed individuals, only small and non-signicant differences are
found (Basso & Rusconi, 1998), regardless that in the past it was accepted that aphasia recovery is better
in left-handers.
Left hemisphere structural pre- and peri-natal lesions (Staudt et al., 2002), developmental tumors
(Anderson et al., 2002), vascular malformations (Vikingstad et al., 2000), and focal pre- and post-natal
intractable epilepsy (Hadac, Brozov, Tintera, & Krsek, 2007; Ligeois et al., 2004) are also associated
with right hemisphere activation in language tasks. Transferring of language functions from left to right
hemisphere has been reported in numerous articles after stroke and tumors of the left hemisphere
(e.g., Holodny, Schulder, Ybasco, & Liu, 2002; Ko!sla et al., 2012; Tyler, Wright, Randall, Marslen-Wilson,
& Stamatakis, 2010; Weiller et al., 1995). All these studies show right hemisphere overtake of homo-
topic areas after injury of language eloquent areas in the left hemisphere.
Hickok and Poeppel (2004, 2007) have proposed a new framework for understanding aspects of the
functional anatomy of language. This framework assumes that early cortical stages of speech percep-
tion involve auditory elds located bilaterally although asymmetrically in the superior temporal
gyrus. Normal language acquisition sets up a degree of bilateral representation in the superior tem-
poral gyrus, contributing to the formation of more widely distributed conceptual representations. This
cortical processing system then diverges into two streams, (1) a ventral stream, which is involved in
mapping sound onto meaning, and (2) a dorsal stream, which is involved in mapping sound onto
articulatory-based representations. The ventral stream projects toward the posterior middle temporal
gyrus; while the dorsal stream projects dorso-posteriorly involving a region in the posterior Sylvian
ssure at the parietal-temporal boundary, and ultimately projecting to the frontal regions.
3. Right hemisphere involvement in language across life span
Neuroimaging studies have supported the assumption that since early in life, language is pre-
dominantly processed by the left hemisphere (Dehaene-Lambertz, Dehaene, & Hertz-Pannier, 2002).
However, it has also been found that the degree of lateralization increases in the rst years of life.
Holland et al. (2001) studied a group of healthy children between 7 and 18 years, nding increasing left
lateralization across these ages. In a detailed study, Szaarski, Holland, Schmithorst, and Byars (2006)
found a statistical signicant positive correlation between age in children (range 517 years) and
lateralization indexes for Brocas area. Similar ndings of progressive lateralization of language to the
left hemisphere mostly related to the expressive language system are reported by Lidzba, Schwilling,
Grodd, Krgeloh-Mann, and Wilke (2011). Language progressive lateralization in childhood may be
related to the maturation of the corpus callosum (Allen, Richey, Chai, & Gorski, 1991).
Groen, Whitehouse, Badcock, and Bishop (2012) used functional transcranial Doppler ultrasound for
assessing cerebral lateralization for language production and for visuospatial memory; they selected 60
typically developing children (ages six and 16 years). The expected pattern of left-lateralized activation
for language and right-lateralized activation for visuospatial abilities was found in 58% of the children.
No age-related change in direction or strength of lateralization was found for language production. The
authors also tested the hypothesis whether having language and visuospatial functions in the same
hemisphere was associated with poor cognitive performance; no evidence for this functional
crowding hypothesis was found. They, however, observed that children with left-lateralized language
production had higher vocabulary and nonword reading age-adjusted standard scores than other
children, regardless of the laterality of visuospatial memory. They concluded that there is a link be-
tween language function and left-hemisphere lateralization that cannot be explained in terms of
maturational changes.
Interestingly, lateralization of language seemingly presents some changes during senescence. More
activation of the right hemisphere during language comprehension and production tasks has been
reported among elderly participants. This observation suggests that the degree of language laterali-
zation decreases with age, and the required cognitive processes become more symmetrical over time
(Wingeld & Grossman, 2006).
Cabeza (2002) proposed the so-called HAROLD model (Hemispheric Asymmetry Reduction in Older
Adults). This model is based on functional neuroimaging and other evidence in the domains of episodic
memory, semantic memory, working memory, perception, and inhibitory control. It was further pro-
posed that age-related hemispheric asymmetry reduction may have a compensatory function or may
reect a dedifferentiation process, having a cognitive or neural origin.
From all the aforementioned studies it seems that the right hemisphere may also hold language
functions either as a functional variant (as in some left handers), as a developmental feature (extremes
of life span) or as an adaptative response to pathology.
4. Bilateral brain representation of language
Different procedures have used to analyze the brain representation of language. These procedures
in general have yielded relatively similar results.
4.1. Wada test
The history of language lateralization has been partially dominated by the Intracarotid Amobarbital
Test described in Japan by J.A. Wada in 1949, but most popular since the 1960s and 1970s. This
technique consists of the injection of an anesthetic (sodium amobarbital) in one of the internal carotids
with the aim to anesthetize a single hemisphere. The subject should experience contralateral hemi-
plegia, without loss of consciousness. At this point the subject is examined to detect language and
memory decits. After a break, the other carotid artery is also injected and the language and memory
tests are repeated.
Table 1 summarizes a selective review of ve major publications in bilateral Wada tests. It totalizes
1799 bilateral Wada studies in 1446 right-handers and 353 subjects either left-handed or ambidextrus.
Although there is inhomogeneity in the procedures, classication and methods to determine brain
lateralization, the review shows similar distribution of language lateralization categories. In average,
bilateral representation of language is found in 10% of right-handers and in 27% of not right- handers.
From these early studies, there has been attempts to formulate a classication for bilateral language
dominance, that may describe the sometimes perplexing ndings of the Wada test. Kurthen et al.
(1994) described ve categories, based on lateralization indexes: (1) left dominant, (2) right domi-
nant, (3) incomplete left, (4) incomplete right, and (5) strongly bilateral. In this classication, incom-
plete left, incomplete right and strongly bilateral are subtypes of bilateral representation of language.
Qualitatively, the bilateral representation of language, may be also divided in three subgroups,
accordingly with the patients performance in the Wada test: Positive, negative, and general. Bilateral
positive representation occurred when the subject has incomplete loss of language on one side but lack
of language impairment on the other; Bilateral negative representation occurred when incomplete loss
of language was present on one side and complete loss on the other; General bilateral representation
occurred when partial language loss was present on both sides. In the analysis of the paper, and with
the aim to take into account patterns of language dissociation between the hemsipheres, Kurthens
Table 1
Bilateral language representation by Wada studies (selective review).
Author No. Hand (R) : (L) Left-H Right H Bilat Suggested classication of Observations
Sjcts Dom % Dom % Dom % bilateral representation
Rasmussen and 262 140 : 122 96 (R) 4 (R) 0 None
Milner (1977) 70 (L) 15 (L) 15 (L)
Kurthen et al. (1994) 173 142 : 31 77 (R) 4 (R) 19 (R) Bilateral positive (12 subjects) Left dominant (67.6%)
B. Bernal, A. Ardila / Journal of Neurolinguistics 28 (2014) 6380

23 (L) 32 (L) 45 (L) Bilateral negative (32 subjects) Right dominant (8.7%)
General bilateral (19 subjects) Incomplete left (16.2%)
-Subpatterns: Incomplete right (0%)
(1)Interhemispheric dissociation Strongly bilateral (7.5%)
(2)Double representation
(3)Unilateral representation
of subfunctions
(4)Distributed representation
of subfunctions
Risse, Gates, and 368 304 : 64 87 (R) 4 (R) 9 (R) Duplicated automatic speech 11 cases were not classied. They had modality-specic
Fangman (1997) 62 (L) 18 (L) 20 (L) Duplicated auditory limited language decit plus automatic speech. No
comprehension subjects found with interhemispheric dissocition
Incomplete right dominance of Broca/Wernicke type.
No Wada decit in either
hemisphere
Loring et al. (1999) 551 469 : 82 86 (R) 5 (R) 9 (R) Adopting Benbadis, Dinner, Bilateral autonomous: little or any decit following
48 (L) 29 (L) 23 (L) Chelune, Piedmonte, and either left or right carotid injection. Bilateral dependent:
Luders (1995) classication language impairment with either carotid injection.
(based on speech arrest): Benbadis et al. acknowledged in an ulterior work that
Bilateral autonomous speech arrest is not a good sign for language
Bilateral dependent lateralization (1998)
Mddel et al. (2009) 445 391 : 54 82 (R) 4 (R) 14 (R) Bilateral independent 69 patients (16%) with bilateral representation
48 (L) 22 (L) 30 (L) Bilateral dependent of language:
Bilateral dependent (6.5%)
Bilateral independent (9%)
Total 1799 1446 : 353 86 (R) 4 (R) 10 (R)
50 (L) 23 (L) 27 (L)
Conventions: Hand, handedness. (L) stands for not right-handed, including ambidextrous. (R), right-handed; Dom, dominant.
67
group proposes four bilaterality sub-patterns: interhemispheric dissociation, double representation,

unilateral representation of subfunctions, and distributed representation of subfunctions between the
hemispheres; this last one is observed when the patient exhibited incomplete loss of language on both
Wada tests.
The study of Risse, Gates, and Fangman (1997), based on automatic speech and auditory compre-
hension only, divided the patterns of bilateral language representation in four sub-groups: (1) dupli-
cation of automatic speech in the right hemisphere, (2) duplication of auditory comprehension in the
right hemisphere, (3) right dominance for all functions with some impairment with left Wada and (4)
no decit in either hemisphere.
Loring et al. (1999) adopted the classication suggested by Benbadis et al. (1995) consisting of two
types of bilateral language representation: bilateral autonomous language representation, in whose
cases there is little or no language alteration in either side, and bilateral dependent language, charac-
terized by language impairment with both left and right hemisphere injections. Mddel et al.s (2009)
classication is similar although utilizing different terminology: they propose two subgroups: bilateral-
dependent consisting of those subjects presenting speech arrest after injection of any of the carotids,
and speech-independent consisting of those subjects who did not have speech arrest after either in-
jection. Unfortunately, speech arrest has been proven to be a non-valid method to determine language
lateralization with Wada test (Benbadis et al., 1998).
Noteworthy, all subjects of these reports are epilepsy patients, and the effects of the disease may
play an important role in language reorganization. At least it can be argued that the bilateral language
representation on some of these patients is a brain adaptation to a deviant trajectory of development;
or a result of some mechanism of brain plasticity with re-arrangement of the language circuitry. It
would explain that patients (usually intractable epilepsy patients) with bilateral language represen-
tation perform worse on neuropsychological test measures obtained both pre- and postoperatively
(Pataraia et al., 2005). Nevertheless, bilateral language representation in normal subjects has been
found not to be associated to any academic achievement problem or language decit (Knecht et al.,
2001).
4.2. Modern neuroimaging studies
Aside Wada studies, bilateral representation of language has been also found in other different
techniques, including PET, fMRI, tractography and magneto-encephalography. The main advantage of
these studies over Wada tests is that they may be performed on normal volunteers.
PET studies on normal volunteers have found that receptive language tasks elicit more bilateral
activation than expressive language tasks (Mller et al., 1997; Papathanassiou et al., 2000), a nding
conrmed by fMRI studies (e.g., Lidzba et al., 2011). Bilateral activation of expressive areas are much
less frequent but may be found in patients with brain gliomas examined with PET (Thiel et al., 1998).
Tumors and epilepsy are also related to more bilateral or right language representation in fMRI studies
(Adcock, Wise, Oxbury, Oxbury, & Matthews, 2003; Springer et al., 1999) and magneto-
encephalography (Tanaka et al., 2013).
An extensive meta-analysis of functional neuroimaging studies including fMRI, PET and SPECT
studies have been published by Vigneau et al. (2011) aimed to describe the involvement of the right
hemisphere in distinct language tasks; the authors found 218 different areas of activation (referred to
as peaks), compared to 728 of the left hemisphere found in 105 experiments. Although the majority of
the peaks during the performance on linguistics tasks were unilateral in the left hemisphere (79%),
more bilateral peaks were observed when the right hemisphere was involved; that means, if the right
hemisphere was activated, usually it was also activated the left hemisphere. The authors conjecture the
right hemisphere works in an inter-hemispheric manner, that is that it somehow requires the left
hemisphere participation. When the peaks were analyzed with respect to each modality of language,
an interesting nding emerged: in phonological tasks, the motor representation for the mouth and
phonological working memory were exclusively found in the left hemisphere. The ndings suggest
that the right hemisphere does not host any phonological representation. In addition, the right frontal
lobe was found to participate in working memory, attentional functions and context processing in
sentence/word processing tasks.
4.3. Insights from tractography
Intraoperative electrical stimulation of the left arcuate fasciculus has demonstrated its involve-
ment in language transferring of phonology traits (Duffau, Gatignol, Mandonnet, Capelle, &
Taillandier, 2008; Mandonnet, Nouet, Gatignol, Capelle, & Duffau, 2007). In addition, the left
arcuate fasciculus has been reported in several articles to be associated to lateralization of language
(Bernal & Ardila, 2009; Nucifora, Verma, Melhem, Gur, & Gur, 2005; Powell et al., 2006; Rodrigo et al.,
2008). However, it has been found that the arcuate fasciculus is also seen left side dominant in
subjects with proven right side language dominance (Vernooij et al., 2007), a puzzling nding sug-
gesting a bilateral representation of language processing not evident or understood by current
technology and state of the art comprehension of language processing. This nding would suggest
that some trans-callosal data-ow may take place assuming that posterior (receptive) to anterior
(expressive) transferring is only suitable through the dominant arcuate fasciculus. There is, however,
at least two alternatives to explain a non-arcuate intrahemispheric transferring: transferring by in-
termediate relay modules via U bers, and transferring by proxy tracts utilizing other interlobar
associative bundle (e.g., the inferior-occipital-frontal fasciculi). In addition to these signicant nd-
ings related to the left arcuate fasciculus, it has been described that electrical stimulation of the left
ventral pathway, related to the inferior occipitofrontal fascicule, produces semantic paraphasias
(Duffau et al., 2008; Mandonet et al., 2007).
5. Toward a synthesis of bilateral representation of language
From the information collected to date and, in particular, with the insights provided by the modern
neurofunctional studies, a framework emerges of facts that may allow proposing a new explanatory
classication of the bilaterality of language representation.
These are the major facts:
- Expressive language is dissociated from the receptive language areas and is located in frontal areas
(mostly the left inferior frontal gyrus)
- Receptive language is dissociated from the expressive language areas and is mostly located in the
posterior third of the temporal lobe (mostly the left)
- There is a spectrum between left lateralization to right lateralization
- Bilateral representation of language may be found both in a limited number of patients and normal
subjects.
- Some epilepsy and tumor patients show language decit when either hemisphere is temporarily
disrupted.
- Some few epilepsy and tumor patients show no signicant language decit after temporary
disruption of either hemisphere.
- Some epilepsy and tumor patients show just minimal decit in only one hemisphere when
functional disruption is applied to both.
- Some degree of bilateral representation exists for the receptive function.
- Phonology has a segregated pathway (arcuate fasciculus) and it is mostly lateralized to the left
hemisphere
- Semantics is also segregated in the ventral pathway (possibly through the inferior occipital-frontal
fasciculus), and has a weaker lateralization.
In order to integrate these facts within the notion of bilateral representation of language two
conceptual elements can be proposed: (a) Modularity (understood here as a cortical regional speci-
cation for a given function), and (b) Data Flow Computing Models (understood for the sake of this
explanation as the transferring of a message, basically as a dichotomy between parallel vs. serial
transferring). Based on modularity, seven different patterns of bilateral language representation could
potentially be distinguished: (1) Bilateral expressive and receptive functions; (2) Bilateral expressive
with left receptive; (3) Bilateral expressive with right side receptive; (4) Bilateral receptive with left
expressive; (5) Bilateral receptive with right side expressive, (6) Left expressive with right receptive,
and (7) Right expressive with left receptive. This classication will be expanded to more levels after
taking in consideration of the data-ow. These seven patterns could be expressed:
1. ER: ER
2. ER: E
3. E: ER
4. ER: R
5. R: RE
6. E: R
7. R: E
Where E expressive and R receptive. The left side of the colon is left hemisphere and the right
side refers to the right hemisphere.
The concept of data ow computing stands for two different models of information processing: (a)
serial, and (b) parallel. In serial processing, steps occur one after the other, in such a manner that the
input for a given processor waits for the output of the prior processor in the chain of ow. An example
of this would be that semantic processing cannot occur until the phonological and lexical analysis has
taken place in another module (or processor, to keep the analogy with computers). In parallel pro-
cessing, the output of any given operation is the input for two or more processing steps in the algo-
rithm. An example of this would be the parallel processing of semantics of a word in conjunction with
the prosody of the intonation.
A number of publications have dealt with this dichotomized model borrowed from computational
sciences (Inui, Okamoto, Miki, Gunji & Kakigi, 2006; Mesulam, 1990; Townsend, 1990). Nonetheless,
some disagreement about which cognitive processes are parallel and which are serial remains;
neurophysiological facts and clinical ndings support the view that auditory processing is an example
of serial processing, and visual processing is a paradigmatic parallel processing, at least in its earliest
stages of processing. The notion of progressive language processing from phonological decoding to
syllables to words to phrases seems to empirically demonstrate the serial component of auditory
language processing. This has support of event-related studies showing that semantics appear in a
window of about 500 ms after word recognition (Hopf, Bader, Meng, & Bayer, 2003). However, parallel
processing has been also shown in language, for example, in reading (words and sentences) utilizing
event-related potentials (Dien, Frishkoff, Cerbone, & Tucker, 2003). In addition, there is parallel pro-
cessing of prosody and familiar words that can be read as a whole and not in a letter-by-letter decoding
fashion.
Patients with pathology may not only redistribute (reorganize) the modules in a new fashion, but
also introduce changes in the data ow. For example, some modules may become bilateral represented
in a redundant manner while others may be dissociated between the hemispheres. Fig. 1 illustrate the
different potential subtypes: (a) Bilateral receptive language representation with canonical Broca; (b)
The bilateral representation of receptive areas is accompanied by right hemisphere transferring of
Brocas area; (c) Bilateral distributed receptive language; (d) Distributed receptive with non-canonical
Broca; (e) Duplicated expressive-receptive; (f) Bilateral language representation with interhemispheric
dissociation of expressive and receptive sub-functions; (g) True duplication of expressive functions
alone; (h) Same as prior category with receptive transferring to the right hemisphere; (i) Isolated
bilateral expressive representation with interhemispheric dissociation; (j) Similar to prior case but
with Wernickes area transferring; phonological aspects are more probable to remain in the left
hemisphere; (k) Interhemispheric dissociation of language. (l) Subtype of interhemispheric dissocia-
tion of language with non-canonical Broca..
5.1. Serial distribution
Serial processing distributed between the hemispheres is probably the most common of all bilateral
representation of language, given the nature of language processing. The patterns of serial distribution
Fig. 1. Possible bilateral language representation based on two by two factors: receptive vs. expressive and phonology vs. semantics.
Drawings have a radiological orientation with left hemisphere represented on the right side. Ovals are divided in two halves
indicating domain dissociation between phonology and semantics (see text for explanation). Left column reects all possibilities for
bilateral representation of Wernickes area, whereas the right column shows bilateral representations of Brocas (otherwise not
included in the rst column) and two cases of expressive-receptive interhemispheric dissociation (k and l). Subtypes: (a) a frequent
normal subtype of bilateral language representation (11 of 39 cases in Risse et al. (1997) series, group II); (b) The bilateral repre-
sentation of receptive areas is accompanied by right hemisphere transferring of Brocas area. This is an infrequent subtype; the right
lateralized Broca suggests brain reorganization; (c) Bilateral distributed receptive language; this pattern includes some subtypes,
accordingly with the subdomain transferred to the right hemisphere; (d) Distributed receptive with non-canonical Broca; A pattern
highly suggestive of language brain reorganization. Some subtypes may emerge accordingly with the receptive dissociation; (e) This
subtype represents a truly global bilateral representation of language; Wada test should fails to produce decit in either carotid, as it
was found in 2 cases of 39 in Risses et al. series; (f) Bilateral language representation with interhemispheric dissociation of
expressive and receptive sub-functions; four subtypes may be found here: one mirroring the example, and two swapping only one
domain; (g) True duplication of expressive functions alone; it is probably only theoretical as it has not yet been described; (h) Same
as prior category with receptive transferring to the right hemisphere (not described); (i) Isolated bilateral expressive representation
with interhemispheric dissociation; at least 2 subtypes are possible; comprehension is only affected in left Wada, but some aspects
of expression are affected in each side; Wernickes area remain in the left side; (j) Similar to prior case but with Wernickes area
transferring; phonological aspects are more probable to remain in the left hemisphere; (k) Interhemispheric dissociation of lan-
guage. Rare condition, described in 4 of 490 epilepsy patients (Dongwook et al., 2008); (l) Subtype of interhemispheric dissociation
of language; expressive functions are most likely to transfer away from the seizure focus (Dongwook et al., 2008).
should be interpreted as an interdependence of the modules in which information ow has a recog-

nizable starting point, and from that point on each output proceeds to only one module up-stream in
cognitive complexity or even reaches the nal motor pathway that generates speech. A simple
approach would represent that chain in the following way: phonological discrimination >> word
recognition >> semantic at word level >> syntactic analysis >> working memory >> sentence
processing >> semantic at sentence level >> grammatical analysis >> motor encoding >> motor
response. In such a sequence no matter where the impairment is located, a nal expressive (albeit
disparate) decit is warranted. Assuming that a distributed inter-hemispheric network may have
modules in both sides of the brain, the ow of information would need to travel between the hemi-
spheres across the corpus callosum or the anterior commissure, back and forth to link all the steps in
the chain of a serial ow. Having this type of distribution, a temporary disruption of either hemisphere
would produce language decits, although they should be different and perhaps partial. Some Wada
tests ndings previously reported by different authors may be explained in this way.
5.2. Parallel distribution
Patterns of parallel processing, on the other hand, should be interpreted in two different ways: (a)
redundancy processing, and (b) distributed processing. Parsing instructions in a redundant manner
have some analogy with algorithms of Resilient Parallel Computing, which is intended to protect
processes form failures by repeating the process just in case one branch of the algorithm fails and
crashes (Liu, Deters, & Zhang, 2010). Parsing instructions in a distributed processing assigns specic
processors to a given function that could be executed while other processor nishes a required process.
In our analogy, the parallel redundancy would trigger two homologous brain modules, in two
different hemispheres to perform the same process; whereas the parallel distributed processing
modules located in different hemispheres will simultaneously process different functions, like in the
example of prosody/semantics. True parallel redundant process is probably inexistent since it would be
a source of conict messing up the cognitive data ow; however, it is possible to imagine a redundancy
between the hemispheres for a given function that theoretically would explain ndings of bilateral
failure on Wada tests.
The combination of the subtypes provided by modularity, and those explained by data- ow may
theoretically explain several types of possible bilateral language representation, as illustrated in Fig. 1.
6. fMRI ndings suggesting distributed bilateral processing
Clinical and fMRI studies have demonstrated the different cortical specication segmenting the
expressive and receptive language functions. Further, fMRI has shown anatomical sub-specication for
isolated expressive and receptive functions. The Brocas area seems to contain two major sub-
components; (a) the pars opercularis, BA 44, and the anterior insula, involved in phonological pro-
cessing and direct speech production, and (b) the pars triangularis, BA 45, more involved in semantic
and lexical processing (Amunts et al., 2004; Fiebach, Friederici, Mller, & von Cramon, 2002; Heim
et al., 2005; McDermott, Petersen, Watson, & Ojemann, 2003). This functional segregation has vali-
dation in the proven distinct structural connectivity that BA 44 and BA 45 exhibit in recent diffusion
tensor imaging studies (Klein et al., 2007; Lemaire et al., 2012). These areas seem to have many other
divergent functions beyond purely language processing (Bornkessel-Schlesewsky, Grewe, &
Schlesewsky, 2012); but of signicant relevance is the dorso-ventral differentiation of the pars oper-
cularis seemingly related with a mirror neuron system (Molnar-Szakacs, Iacoboni, Koski, & Mazziotta,
2005).
Wernickes area sub-specialization has received less attention, in spite of encompassing a large
distribution of Brodmanns areas. Perhaps it is due to the poor anatomical landmarks delimiting the
receptive language cortex. However, it is now accepted that at least transferring of language from
posterior to anterior areas are carried by two different systems, (a) the dorsal system, involved in
phonological processing and (b) the ventral system involved in semantic processing (Duffau et al.,
2002; Glasser & Rilling, 2008; Leclercq et al., 2010; Mandonnet et al., 2007). This subdivision sug-
gests some receptive phonological processing toward BA 40, and a more ventral and posterior semantic
analysis (McDermott et al., 2003).
The subdivision of phonology/semantic domains is only an example, may be the most relevant, but
not the only one. Several other subsystems are intervening in language that may have sub-
specialization. The dorsal pars opercularis has been found to be involved more specically in
sequencing linguistic and non-linguistic events (Ardila & Bernal, 2007; Makuuchi, Bahlmann,
Anwander & Friederici, 2009; Willems, Ozyrek, & Hagoort, 2009), whereas the ventral part has
been found to be involved in verbal working memory (Koelsch et al, 2009). Synonyms generation vs.
Fig. 2. Axial T1 MRI anatomical images with functional map from an auditory description/comprehension task. Left side of the
image corresponds to the right hemisphere (Radiological orientation). The left image shows activation of the posterior right tem-
poral lobe corresponding to Wernickes area. There are only minute activations in the left hemisphere. The image on the center
shows a large activation in the left inferior frontal gyrus corresponding to Brocas area. The right images show a hypointense image
located in the anterior temporal lobe (within the oval) consistent with a developmental tumor (patient 14 years old, right-handed
girl). (Images courtesy of Miami Childrens Hospital, Department of Radiology).
antonyms generation engages left lateralization regions. However, antonyms activation extends more
anteriorly (Jeon, Lee, Kim, & Cho, 2009); verbal working memory (Chein, Fissell, Jacobs, & Fiez, 2002),
semantic content vs. grammatical structure (Ni et al., 2000), and likely many other functions could be
distinguished.
Departing from the previous information it can be inferred that serial distribution between the
hemispheres can be quite complex, since a sub-specialized module may be located in one hemisphere
while the rest may remain in the other. Wada tests results seem to back up this assertion.
The following images were taken from patients with intractable epilepsy, who underwent fMRI for
language mapping. Some of these patterns are only evident when at least two language paradigms
with different linguistic loads are given. The cases will serve to illustrate key points in bilateral lan-
guage representation.
Case A (Fig. 2): Bilateral representation of language: Broca/Wernicke dissociation type. It clearly
shows a bilateral representation of language with reorganization occurring only for receptive language
function, most likely in keeping with the malfunction produced by the developmental lesion found in
the left temporal region. A case like this will have a positive bilateral Wada test, but each side producing
a different clinical picture. It would be classied by Kurthen et al. (1994) as a General Bilateral pattern
Fig. 3. Bilateral Brocas and left Wernickes area Orientation and background image as in Fig. 2 fMRI task: auditory description/
comprehension task. Left side of the image corresponds to the right hemisphere (Radiological orientation). Left image shows
complete left lateralization for receptive language. Right image shows bilateral activation of inferior frontal gyrus slightly more
prominent on the right. Patient: 11-year-old right-handed girl. (Images courtesy of Miami Childrens Hospital, Department of
Radiology).
Fig. 4. Bilateral global representation of language. Redundancy. Orientation and background are the same as aforementioned. fMRI
with auditory description/comprehension task. There is bilateral secondary auditory areas of activation in the posterior temporal
lobes, and bilateral activation of the inferior frontal gyrus. The patient is a 16-year-old right-handed girl with epilepsy. (Images
courtesy of Miami Childrens Hospital, Department of Radiology).
with interhemispheric dissociation. For Loring et al. (1990) and Mddel et al. (2009) this would be
classied as a bilateral dependent representation of language. Of note is the fact that this patient was a
right-hander.
Case B (Fig. 3) shows bilateral representation of expressive language functions, with well-dened
Wernickes left lateralization. The frontal areas are homologous, but only the right insula is involved.
This subject might have a bilateral positive (double representation) of Kurthen et al. (1994), where an
incomplete loss of language may be expected with a left carotid injection and no impairment with a
right carotid injection, provided there is redundancy of the Brocas area. For Loring et al. (1990) and
Mddel et al. (2009) this case will also be a bilateral dependent type. More difcult to classify
accordingly with Risse et al. (1997) subgroups of bilateral language representation, since they based it
too much on impairment of automatic speech. Perhaps the case could be classied as plain duplication
of automatic speech.
Case C (Fig. 4) is an example of global bilateral representation of language. There are some minor
asymmetries, but all canonical language areas are activated in both sides. A pattern like this may
explain all types of Kurthen et al. (1994), either bilateral-positive, bilateral-negative or bilateral global,
or double representation, unilateral representation of subfunctions or distributed representation of
subfunctions, all possible depending upon the distribution of the modules. If the bilateral condition
represents redundancy of both expressive and receptive modules, a bilateral global representation of
Kurthen et al.s is obtained. If, in contrast, a modular distribution (partial or complete) between the
hemispheres has taken place, either a bilateral-positive or bilateral negative is possible. Take for
example, that some phonological functions remain completely lateralized in the left hemisphere while
the rest of sub-modules are bilaterally represented (duplicated). In this case the right Wada will
produce a partial decit (if the reminder modules are distributed) or no decit (if the reminder
modules are duplicated), but the left Wada will produce deep language decit as the language pro-
cessing has been affected at its root. Likewise, this pattern may explain subtypes 1 (duplication of
automatic speech), 2 (duplication of comprehension) and 4 (no decit in either Wada) of Risse et al.
(1997) but not 3 for only right sided dominant activations. Likewise, the pattern may explain Loring
et al.s types 1 or 2 depending if the organization is just redundancy of processing (as the pattern
suggests) or redistribution of sub-functions.
Case D (Fig. 5) shows a frequent nding in language mapping whenever different paradigms are
applied. In the case shown, the auditory description/comprehension task only yields left hemisphere
activation. In this task, the subject has to respond pressing a button when judging a sentence as true. As
control the subject presses the button when hearing a tone amidst pseudosentences created by playing
the sentences backwards. When the subject performs a task in which it is required to discriminate, if
pairs of words are synonyms or antonyms, contrasted to discrimination of similar or different tones, the
right inferior frontal gyrus is also involved. This bilateral representation is task-related and it is
Fig. 5. Redistribution of sub-functions. Task-specic-dissociated bilateral representation of language. Left side of the image corre-
sponds to the right hemisphere (Radiological orientation). Upper row: fMRI with auditory description/comprehension task. Lower
row: fMRI with a semantic decision task based on antonyms. The rst task only shows left side involvement for expressive and
receptive language. The antonyms task shows bilateral activation of Brocas area with same left sided Wernicke lateralization. Pa-
tient: 13-year-old right-handed boy. (Images courtesy of Miami Childrens Hospital, Department of Radiology).
reecting a different strategy of the brain working under specic semantic constrains. It seems that it is
much easier for the brain to judge true that What is on top of the house is the....roof than to decide if
distant far are synonyms, not antonyms. A case like this would be classied as distributed repre-
sentation of sub-functions in Kurthen et al. (1994), classication; bilateral autonomous by Loring et al.
(1990) and not suitable for classication according to the proposal presented by Risse et al. (1997)
Fig. 6. Bilateral receptive lateralized expressive. Language activation map overlaid on an FLAIR MRI axial series. Left side of the
image corresponds to the right hemisphere (Radiological orientation). The subject was performing a semantic decision task based on
antonyms/synonyms discrimination. Notice the bilateral symmetrical activation of secondary auditory areas in homologous func-
tional areas, and the lateralization of the expressive areas to the left hemisphere. Patient was a 21-year-old man with a develop-
mental tumor in the right frontal lobe. (Images courtesy of Miami Childrens Hospital, Department of Radiology).
Case E (Fig. 6) Bilateral Wernickes representation has at least two different variants: with left Broca
or with right Broca localization. The rst subtype is the most frequent of all bilateral representations,
and it is probably the only found in normal right- handed patients. Bilateral representation of receptive
functions is well recognized since early clinical studies described less comprehension than expressive
impairment in cases of global aphasia due to hemispheric stroke (Benson & Ardila, 1996; Taylor-Sarno &
Levita, 1981). Likewise, recovery is also faster and better for receptive language than for expressive
functions. In that sense, bilateral representation of expressive functions may be seen always as a
suggesting sign of language network reorganization. Still, the bilateral receptive language represen-
tation may have some variants as some sub-receptive functions could be re-distributed or duplicated
between the hemispheres. In fMRI studies, the homologous bilateral activation supposedly would
suggest redundancy of representation, while non-homologous bilateral temporo-parietal activation
would suggest a rather distributed bilateral representation. Notice at this point that Kurthen et al.s
subtypes of double representation, unilateral representation of sub-functions (distributed) and
distributed representation of sub-functions are good for expressive alone, receptive alone or both,
given an ample spectrum of different subtypes of functional bilateral representation for language.
7. Toward an integration
A simple classication, taking into account the different language categorical dissociations that has
been found and demonstrated in Wada and fMRI studies, could be proposed from a topographic
perspective (Table 2).
Any bilateral representation may be subdivided functionally in parallel (this is redundantly pro-
cessed) or serial. In the rst case, we would see those cases in which the patient does not lose the
redundant function with the Wada test in neither of the carotids, while in the second (serial) he/she
will exhibit partial loss of speech (for example cases mentioned by Kurthen et al. (1994), Loring et al.
(1990) and Risse et al. (1997) cited before). In functional MRI, the serial processing between Brocas
areas may be seen as a task-related dissociation of expressive language (see Fig. 4).
The same rational may apply for cases with bilateral Wernickes representation - a more frequent
situation seen in clinical practice. Parallel processing between Wernickes areas would be seen as
subjects not referring any comprehension impairment after injection of the Amytal in either carotid,
while partial or limited comprehension may appear in both of them. Notice again that parallel pro-
cessing in these examples is a sort of duplicated data ow, while the serial implies a distributed
modularity.
Table 2
Proposed classication of language lateralization according to a topographic perspective. All main types and subtypes are self
explanatory. The task specic dissociation subtype refers to distribution of data ow in which some language functions reside in
only one hemisphere. Functions dissociated may be explained by transferring of subfunctions (phonology or semantics) to the
right hemisphere. This dissociation is only evident when the patient is presented with two or more paradigms with different
linguistic loads. In this case, brain activations may appear non-congruent between tasks. We have preferred this name, to a more
descriptive but less practical name like "specic-linguistic-sub-domain dissociated language representation."
Isolated Bilateral Expressive Representation

Subtype I: with left Wernickes
Subtype II: with right Wernickes
Isolated Bilateral Receptive Representation
Subtype I: with left Brocas
Subtype II: with right Brocas
Bilateral Global Representation
Bilateral Dissociated Representation
Subtype I: Transhemispheric Broca/Wernicke dissociation
Subtype I.1: Brocas transferred
Subtype I.2: Wernickes transferred
Subtype II: Task-specic dissociation
Subtype I: Brocas dissociation
Subtype II: Wernickes dissociation
Subtype III: Combined dissociation
Table 3
Proposed classication of language lateralization according to a functional perspective. Language modules may be duplicated in
the other hemisphere or distributed between the hemispheres. In the rst option process may occur in parallel assuming
complete capability of each side, or one side should remain quiescent. Notice that, in theroy, serial ow may still happen
assuming unbalanced efciency between the duplicated modules. However, in distributed modularity (types I and II) the serial
processing is obligued as data may ow crossing the hemispheres in the cognitive up-stream trajectory.
I. Duplicated Bilateral Representation of Language (each side sufces to hold the function: parallel processing)
Expressive
Receptive
Both
II. Distributed Bilateral Representation of Language (each hemisphere handles a particular subset of functions: serial
processing)
Expressive (phonology/semantics)
Receptive (Word-level vs. sentence level)
Both
III. Transhemispherical dissociated
Broca/Wernicke
A much more complex scenario is posed by bilaterality of both Brocas and Wernickes areas, since in
this case the serial vs. parallel processing may be distributed in two orthogonal dimensions: anterior to
posterior (receptive to expressive) and side to side between the homologous areas in which dissoci-
ations of the sort phonology/semantics may occur, among many others. The multidimensionality of
these intertwined factors may prompt for a quite complex and large classication. A functional clas-
sication could consequently also be proposed, as presented in Table 3.
8. Conclusion
Bilateral language representation has been a very complex and intricate aspect of brain organization
of cognition. It is frequent understanding that language lateralization is a matter of all or nothing.
However, language dominance is mostly a matter of hemispheric advantage for a specic multi-
modular cognitive function: language. As such, language in a strict sense is up to a certain point a
bilateral brain function. Aside expressive/receptive and phonological/semantic dichotomies there are
other many sub-functions for which we are looking for their anatomical and functional correlates with
new neuroimaging techniques such as diffusion tensor imaging tractography and fMRI.
The understanding of the language network in terms of submodules and connectivity will provide
ground to better understanding brain reorganization after structural and functional brain lesions.
References
Adcock, J. E., Wise, R. G., Oxbury, J. M., Oxbury, S. M., & Matthews, P. M. (2003). Quantitative fMRI assessment of the differences
in lateralization of language related brain activation in patients with temporal lobe epilepsy. Neuroimage, 8(2), 423438.
Allen, L. S., Richey, M. F., Chai, Y. M., & Gorski, R. A. (1991). Sex differences in the corpus callosum of the living human being.
Journal of Neuroscience, 11, 933942.
Amunts, K., Weiss, P. H., Mohlberg, H., Pieperhoff, P., Eickhoff, S., Gurd, J. M., et al. (2004). Analysis of neural mechanisms
underlying verbal uency in cytoarchitectonically dened stereotaxic spacedthe roles of Brodmann areas 44 and 45.
Neuroimage, 22(1), 4256.
Anderson, D. P., Harvey, A. S., Saling, M. M., Anderson, V., Kean, M., Jacobs, R., et al. (2002). Differential functional magnetic
resonance imaging language activation in twins discordant for a left frontal tumor. Journal of Child Neuralogy, 17(10),
766769.
Ardila, A. (2006). Las afasias Accessed March 29, 2013 http://neuropsicolog.blogspot.com/2009/04/libros-de-las afasias-alfredo-
ardila.html.
Ardila, A., & Bernal, B. (2007). What can be localized in the brain? towards a "factor" theory on brain organization of cognition.
International Journal of Neuroscience, 117, 935969.
Basic, S., Hajnsek, S., Poljakovic, Z., Basic, M., Culic, V., & Zadro, I. (2004). Determination of cortical language dominance using
functional transcranial Doppler sonography in left-handers. Clinical Neurophysiology, 115(1), 154160.
Basso, A., & Rusconi, M. L. (1998). Aphasia in left-handers. In P. Coppens, Y. Lebrun, & A. Basso (Eds.), Aphasia in atypical
populations (pp. 134). Mahwah, NJ: Lawrence Erlbaum Associates.
Bembich, S., Demarini, S., Clarici, A., Massaccesi, S., & Grasso, D. L. (2011). Non invasive assessment of hemispheric language
dominance by optical topography during a brief passive listening test: a pilot study. Medical Science Monitor, 17(12), CR692
CR697.
Benbadis, S. R., Binder, J. R., Swanson, S. J., Fischer, M., Hammeke, T. A., Morris, G. L., et al. (1998). Is speech arrest during Wada
testing a valid method for determining hemispheric representation of language? Brain and Language, 65(3), 441446.
Benbadis, S. R., Dinner, S. D., Chelune, G. J., Piedmonte, M., & Luders, H. O. (1995). Autonomous versus dependent: a classication
of bilateral language representation by intracarotid amobarbital procedure. Journal of Epilepsy, 8, 255263.
Benson, D. F., & Ardila, A. (1996). Aphasia: A clinical perspective. New York: Oxford University Press.
Benson, D., & Geschwind, N. (1973). Aphasia and related disturbances. In A. Baker (Ed.), Clinical Neurology (pp. 126). New York:
Harper and Row.
Bernal, B., & Ardila, A. (2009). The role of the arcuate fasciculus in conduction aphasia. Brain, 132(Pt 9), 23092316.
Binder, J. R. (2011). Functional MRI is a valid noninvasive alternative to Wada testing. Epilepsy Behavior, 20(2), 214222.
Bisconti, S., Di Sante, G., Ferrari, M., & Quaresima, V. (2012). Functional near-infrared spectroscopy reveals heterogeneous
patterns of language lateralization over frontopolar cortex. Neuroscience Research, 73(4), 328332.
Bornkessel-Schlesewsky, I., Grewe, T., & Schlesewsky, M. (2012). Prominence vs. aboutness in sequencing: a functional
distinction within the left inferior frontal gyrus. Brain and Language, 120(2), 96107.
Bramwell, B. (1899). On crossed aphasia. Lancet, 3, 14731479.
Broca, P. (1861). Remarques sur le sige de la facult du langage articul; suivies dune observation daphmie. Bulletin de la
Socit dAnthropologie, 2, 330357.
Broca, P. (1865). Du sige de la facult du langage articul. Bulletin de la Socit dAnthropologie, 6, 337393.
Cabeza, R. (2002). Hemispheric asymmetry reduction in older adults: the HAROLD model. Psychology and Aging, 17(1), 85100.
Castro-Caldas, A., & Confraria, A. (1984). Age and type of crossed aphasia in dextral due to stroke. Brain and Language, 23, 126133.
Chein, J. M., Fissell, K., Jacobs, S., & Fiez, J. A. (2002). Functional heterogeneity within Brocas area during verbal working
memory. Physiology and Behavior, 77(45), 635639.
Coppens, P., Hungerford, S., Yamaguchi, S., & Yamadori, A. (2002). Crossed aphasia: an analysis of the symptoms, their frequency,
and a comparison with left hemisphere aphasia symptomatology. Brain and Language, 83(3), 425463.
Dehaene-Lambertz, N., Dehaene, S., & Hertz-Pannier, L. (2002). Functional neuroimaging of speech perception in infants. Sci-
ence, 298, 20132015.
Dejerine, J. (1914). Semiologie des affections du systeme nerveux. Paris: Masson.
Dien, J., Frishkoff, G. A., Cerbone, A., & Tucker, D. M. (2003). Parametric analysis of event-related potentials in semantic
comprehension: evidence for parallel brain mechanisms. Cognitive Brain Research, 15(2), 137153.
Dongwook, L., Lee, S. J., Swanson, D. S., Sabsevitz, T. A., Hammeke, F., Winstanley, S., et al. (2008). Fmri and Wada studies in
patients with interhemispheric dissociation of language functions. Epilepsy and Behavior, 13, 350356.
Duffau, H., Capelle, L., Sichez, N., Denvil, D., Lopes, M., Sichez, J. P., et al. (2002). Intraoperative mapping of the subcortical
language pathways using direct stimulations. An anatomo-functional study. Brain, 125(Pt 1), 199214.
Duffau, H., Gatignol, S. T., Mandonnet, E., Capelle, L., & Taillandier, L. (2008). Intraoperative subcortical stimulation mapping of
language pathways in a consecutive series of 115 patients with Grade II glioma in the left dominant hemisphere. Journal of
Neurosurgery, 109(3), 461471.
Fiebach, C. J., Friederici, A. D., Mller, K., & von Cramon, D. Y. (2002). fMRI evidence for dual routes to the mental lexicon in
visual word recognition. Journal of Cognitive Neurosciences, 14(1), 1123.
Glasser, M. F., & Rilling, J. K. (2008). DTI tractography of the human brains language pathways. Cerebral Cortex, 18(11), 2471
2482.
Groen, M. A., Whitehouse, A. J., Badcock, N. A., & Bishop, D. V. (2012). Does cerebral lateralization develop? A study using
functional transcranial Doppler ultrasound assessing lateralization for language production and visuospatial memory. Brain
and Behavior, 2(3), 256269.
Hadac, J., Brozov, K., Tintera, J., & Krsek, P. (2007). Language lateralization in children with pre- and postnatal epileptogenic
lesions of the left hemisphere: an fMRI study. Epileptic Disorders, 9(Suppl. 1), S19S27.
Ha, J. W., Pyun, S. B., Hwang, Y. M., & Sim, H. (2012). Lateralization of cognitive functions in aphasia after right brain damage.
Yonsei Medical Journal, 53(3), 486494.
Harris, L. J. (1991). Cerebral control for speech in right-handers and left-handers: an analysis of the views of Paul Broca, his
contemporaries, and his successors. Brain and Language, 40, 150.
Harris, L. J. (1999). Early theory and research on hemispheric specialization. Schizophrenia Bulletin, 25(1), 1139.
Hcaen, H., & Albert, M. L. (1978). Human neuropsychology. New York: Wiley.
Hcaen, H., Mazurs, G., Ramier, A., Goldblum, M., & Merianne, L. (1971). Aphasie croisee chez un sujet droiter bilingue. Revue
Neurologique, 1, 319323.
Hcaen, H., & Sauguet, J. (1971). Cerebral dominance in left-handed subjects. Cortex, 7, 1947.
Heim, S., Alter, K., Ischebeck, A. K., Amunts, K., Eickhoff, S. B., Mohlberg, H., et al. (2005). The role of the left Brodmanns areas
and 45 in reading words and pseudowords. Cognitive Brain Research, 25(3), 982993.
Hickok, G., & Poeppel, D. (2004). Dorsal and ventral streams: a framework for understanding aspects of the functional anatomy
of language. Cognition, 92, 6799.
Hickok, G., & Poeppel, D. (2007). The cortical organization of speech processing. Nature Reviews, 8, 393402.
Holland, S. K., Plante, E., Weber Byars, A., Strawsburg, R. H., Schmithorst, V. J., & Ball, W. S., Jr. (2001). Normal fMRI brain
activation patterns in children performing a verb generation task. Neuroimage, 14, 837843.
Holland, S. K., Vannest, J., Mecoli, M., Jacola, L. M., Tillema, J. M., Karunanayaka, P. R., et al. (2007). Functional MRI of language
lateralization during development in children. International Journal of Audiology, 46(9), 533551.
Holodny, A. I., Schulder, M., Ybasco, A., & Liu, W. C. (2002). Translocation of Brocas area to the contralateral hemisphere as the
result of the growth of a left inferior frontal glioma. Journal of Computer Assisted Tomography, 26(6), 941943.
Hopf, J. M., Bader, M., Meng, M., & Bayer, J. (2003). Is human sentence parsing serial or parallel? Evidence from event-related
brain potentials. Cognitive Brain Research, 15(2), 165177.
Inui, K., Okamoto, H., Miki, K., Gunji, A., & Kakigi, R. (2006). Serial and parallel processing in the human auditory cortex: a
magnetoencephalographic study. Cerebral Cortex, 16(1), 1830.
Ishizaki, M., Ueyama, H., Nishida, Y., Imamura, S., Hirano, T., & Uchino, M. (2012). Crossed aphasia following an infarction in the
right corpus callosum. Clinical Neurology and Neurosurgery, 114(2), 161165.
Jeon, H. A., Lee, K. M., Kim, Y. B., & Cho, Z. H. (2009). Neural substrates of semantic relationships: common and distinct left-
frontal activities for generation of synonyms vs. antonyms. Neuroimage, 48(2), 449457.
Kadis, D. S., Pang, E. W., Mills, T., Taylor, M. J., McAndrews, M. P., & Smith, M. L. (2011). Characterizing the normal developmental
trajectory of expressive language lateralization using magnetoencephalography. Journal of the International Neuropsycho-
logical Society, 17(5), 896904.
Kennan, R. P., Kim, D., Maki, A., Koizumi, H., & Constable, R. T. (2002). Non-invasive assessment of language lateralization by
transcranial near infrared optical topography and functional MRI. Human Brain Mapping, 16(3), 183189.
Khedr, E. M., Hamed, E., Said, A., & Basahi, J. (2002). Handedness and language cerebral lateralization. European Journal of
Applied Physiology, 87(45), 469473.
Klein, J. C., Behrens, T. E., Robson, M. D., Mackay, C. E., Higham, D. J., & Johansen-Berg, H. (2007). Connectivity-based parcellation
of human cortex using diffusion MRI: establishing reproducibility, validity and observer independence in BA 44/45 and
SMA/pre-SMA. Neuroimage, 34(1), 204211.
Knecht, S., Drger, B., Deppe, M., Bobe, L., Lohmann, H., Flel, A., et al. (2000). Handedness and hemispheric language domi-
nance in healthy humans. Brain, 123(12), 25122518.
Knecht, S., Drger, B., Flel, A., Lohmann, H., Breitenstein, C., Deppe, M., et al. (2001). Behavioural relevance of atypical language
lateralization in healthy subjects. Brain, 124(Pt 8), 16571665.
Koelsch, S., Schulze, K., Sammler, D., Fritz, T., Mller, K., & Gruber, O. (2009). Functional architecture of verbal and tonal working
memory: an FMRI study. Human Brain Mapping, 30(3), 859873.
Ko!sla, K., Pfajfer, L., Bryszewski, B., Jasklski, D., Stefan
! czyk, L., & Majos, A. (2012). Functional rearrangement of language areas in
patients with tumors of the central nervous system using functional magnetic resonance imaging. Polish Journal of Radi-
ology, 77(3), 3945.
Kurthen, M., Helmstaedter, C., Linke, D. B., Hufnagel, A., Elger, C. E., & Schramm, J. (1994). Quantitative and qualitative evaluation
of patterns of cerebral language dominance. An amobarbital study. Brain and Language, 46(4), 536564.
Leclercq, D., Duffau, H., Delmaire, C., Capelle, L., Gatignol, P., Ducros, M., et al. (2010). Comparison of diffusion tensor imaging
tractography of language tracts and intraoperative subcortical stimulations. Journal of Neurosurgery, 112(3), 503511.
Lemaire, J. J., Golby, A., Wells, W. M., Pujol, S., Tie, Y., & Rigolo, L. (2012). Extended Brocas area in the functional connectome of
language in adults: combined cortical and subcortical single-subject analysis using fMRI and DTI tractography. Brain
Topography, 26(3), 428441.
Lidzba, K., Schwilling, E., Grodd, W., Krgeloh-Mann, I., & Wilke, M. (2011). Language comprehension vs. language production:
age effects on fMRI activation. Brain and Language, 119(1), 615.
Ligeois, F., Connelly, A., Cross, J. H., Boyd, S. G., Gadian, D. G., Vargha-Khadem, F., et al. (2004). Language reorganization in
children with early-onset lesions of the left hemisphere: an fMRI study. Brain, 127(Pt 6), 12291236.
Liu, D., Deters, R., & Zhang, W. J. (2010). Architectural design for resilience. Enterprise Information Systems, 4, 137152.
Lorenz, M. W., Thoelen, N., Loesel, N., Lienerth, C., Gonzalez, M., Humpich, M., et al. (2008). Assessment of cerebral autor-
egulation with transcranial Doppler sonography in poor bone windows using constant infusion of an ultrasound contrast
agent. Ultrasound Medical Biology, 34(3), 345353.
Loring, D. W., Meador, K. J., Lee, G. P., Murro, A. M., Smith, J. R., Flanigin, H. F., et al. (1990). Cerebral language lateralization:
evidence from intracarotid amobarbital testing. Neuropsychologia, 28(8), 831838.
Loring, D. W., Strauss, E., Hermann, B. P., Perrine, K., Trenerry, M. R., Barr, W. B., et al. (1999). Effects of anomalous language
representation on neuropsychological performance in temporal lobe epilepsy. Neurology, 53(2), 260264.
Makuuchi, M., Bahlmann, J., Anwander, A., & Friederici, A. D. (2009). Segregating the core computational faculty of human
language from working memory. Proceedingof the National Academy of Sciences of U S A, 106(20), 83628367.
Mandonnet, E., Nouet, A., Gatignol, P., Capelle, L., & Duffau, H. (2007). Does the left inferior longitudinal fasciculus play a role in
language? A brain stimulation study. Brain., 130(Pt 3), 623629.
Marin, P., Paghera, B., De Deyn, P. P., & Vignolo, L. A. (2004). Adult crossed aphasia in dextrals revisited. Cortex, 40, 4174.
Matsumoto, R., Okada, T., Mikuni, N., Mitsueda-Ono, T., Taki, J., Sawamoto, N., et al. (2008). Hemispheric asymmetry of the
arcuate fasciculus: a preliminary diffusion tensor tractography study in patients with unilateral language dominance
dened by Wada test. Journal of Neurology, 255(11), 17031711.
McDermott, K. B., Petersen, S. E., Watson, J. M., & Ojemann, J. G. (2003). A procedure for identifying regions preferentially
activated by attention to semantic and phonological relations using functional magnetic resonance imaging. Neuro-
psychologia, 41(3), 293303.
Mesulam, M. M. (1990). Large-scale neurocognitive networks and distributed processing for attention, language, and memory.
Annals of Neurology, 28(5), 597613.
Mddel, G., Lineweaver, T., Schuele, S. U., Reinholz, J., & Loddenkemper, T. (2009). Atypical language lateralization in epilepsy
patients. Epilepsia, 50(6), 15051516.
Molnar-Szakacs, I., Iacoboni, M., Koski, L., & Mazziotta, J. C. (2005). Functional segregation within pars opercularis of the inferior
frontal gyrus: evidence from fMRI studies of imitation and action observation. Cerebral Cortex, 15(7), 986994.
Mller, R. A., Rothermel, R. D., Behen, M. E., Muzik, O., Mangner, T. J., & Chugani, H. T. (1997). Receptive and expressive language
activations for sentences: a PET study. Neuroreport, 8(17), 37673770.
Ni, W., Constable, R. T., Mencl, W. E., Pugh, K. R., Fulbright, R. K., Shaywitz, S. E., et al. (2000). An event-related neuroimaging
study distinguishing form and content in sentence processing. Journal of Cognitive Neurosciences, 12(1), 120133.
Nucifora, P. G., Verma, R., Melhem, E. R., Gur, R. E., & Gur, R. C. (2005). Leftward asymmetry in relative ber density of the arcuate
fasciculus. Neuroreport, 16(8), 791794.
Papathanassiou, D., Etard, O., Mellet, E., Zago, L., Mazoyer, B., & Tzourio-Mazoyer, N. A. (2000). Common language network
for comprehension and production: a contribution to the denition of language epicenters with PET. Neuroimage, 11(4),
347357.
Pataraia, E., Billingsley-Marshall, R. L., Castillo, E. M., Breier, J. I., Simos, P. G., Sarkari, S., et al. (2005). Organization of receptive
language-specic cortex before and after left temporal lobectomy. Neurology, 64(3), 481487.
Pedersen, P. M., Jargensen, H. S., Nakayama, H., Raaschou, H. O., & Olsen, T. S. (1995). Aphasia in acute stroke: incidence, de-
terminants, and recovery. Annals of Neurology, 38, 659666.
Powell, H. W., Parker, G. J., Alexander, D. C., Symms, M. R., Boulby, P. A., Wheeler Kingshott, C. A., et al. (2006). Hemispheric
asymmetries in language related pathways: a combined functional MRI and tractography study. Neuroimage, 32(1),
388399.
Price, C. J. (2010). The anatomy of language: a review of 100 fMRI studies published in 2009. Annals of the New York Academy of
Sciences, 1191, 6288.
Rasmussen, T., & Milner, B. (1977). The role of early brain injury in the lateralization of cerebral speech functions. Annals of the
New York Academy of Sciences, 299, 3569.
Risse, G. L., Gates, J. R., & Fangman, M. C. (1997). A reconsideration of bilateral language representation based on the intracarotid
amobarbital procedure. Brain and Cognition, 33(1), 118132.
Rodrigo, S., Oppenheim, C., Chassoux, F., Hodel, J., de Vanssay, A., Baudoin-Chial, S., et al. (2008a). Language lateralization in
temporal lobe epilepsy using functional MRI and probabilistic tractography. Epilepsia, 49(8), 13671376.
Springer, J. A., Binder, J. R., Hammeke, T. A., Swanson, S. J., Frost, J. A., Bellgowan, P. S., et al. (1999). Language dominance in
neurologically normal and epilepsy subjects: a functional MRI study. Brain, 122(Pt 11), 20332046.
Staudt, M., Lidzba, K., Grodd, W., Wildgruber, D., Erb, M., & Krgeloh, M. (2002). Right hemispheric organization of language
following early left-sided brain lesions: functional MRI topography. Neuroimage, 16(4), 954967.
Szaarski, J. P., Holland, S. K., Schmithorst, V. J., & Byars, A. W. (2006). fMRI study of language lateralization in children and
adults. Human Brain Mapping, 27, 202212.
Tanaka, N., Liu, H., Reinsberger, C., Madsen, J. R., Bourgeois, B. F., Dworetzky, B. A., et al. (2013). Language lateralization rep-
resented by spatiotemporal mapping of magnetoencephalography. American Journal of Neuroradiology, 34(3), 558563.
Taylor-Sarno, M., & Levita, E. (1981). Some observations on the nature of recovery in global aphasia after stroke. Brain and
Language, 13(1), 112.
Thiel, A., Herholz, K., von Stockhausen, H. M., van Leyen-Pilgram, K., Pietrzyk, U., Kessler, J., et al. (1998). Localization of
language-related cortex with 15O-labeled water PET in patients with gliomas. Neuroimage, 7(4 Pt 1), 284295.
Townsend, J. (1990). Serial vs. Parallel processing: sometimes they Look like Tweedledum and Tweedledee but they can (and
should) Be distinguished. Psychology Science, 1, 3653.
Tyler, L. K., Wright, P., Randall, B., Marslen-Wilson, W. D., & Stamatakis, E. A. (2010). Reorganization of syntactic processing
following left-hemisphere brain damage: does right-hemisphere activity preserve function? Brain, 133(11), 33963408.
Vernooij, M. W., Smits, M., Wielopolski, P. A., Houston, G. C., Krestin, G. P., & van der Lugt, A. (2007). Fiber density asymmetry of
the arcuate fasciculus in relation to functional hemispheric language lateralization in both right- and left-handed healthy
subjects: a combined fMRI and DTI study. Neuroimage, 35(3), 10641076.
Vigneau, M., Beaucousin, V., Herv, P. Y., Jobard, G., Petit, L., Crivello, F., et al. (2011). What is right-hemisphere contribution to
phonological, lexico-semantic, and sentence processing? Insights from a meta-analysis. Neuroimage, 54(1), 577593.
Vikingstad, E. M., Cao, Y., Thomas, A. J., Johnson, A. F., Malik, G. M., & Welch, K. M. (2000). Language hemispheric dominance in
patients with congenital lesions of eloquent brain. Neurosurgery, 47(3), 562570.
Weiller, C., Isensee, C., Rijntjes, M., Huber, W., Mller, S., Bier, D., et al. (1995). Recovery from Wernickes aphasia: a positron
emission tomographic study. Annals of Neurology, 37(6), 723732.
Wernicke, C. (1874). Der Aphasiche Symptomencomplex. Breslau: Cohn & Weigert.
Willems, R. M., Ozyrek, A., & Hagoort, P. (2009). Differential roles for left inferior frontal and superior temporal cortex in
multimodal integration of action and language. Neuroimage, 47(4), 19922004.
Wingeld, A., & Grossman, M. (2006). Language and the aging brain: patterns of neural compensation revealed by functional
brain imaging. Journal of Neurophysiology, 96, 28302839.
Woermann, F. G., Jokeit, H., Luerding, R., Freitag, H., Schulz, R., Guertler, S., et al. (2003). Language lateralization by Wada test
and fMRI in 100 patients with epilepsy. Neurology, 61(5), 699701.
APHASIOLOGY, 2010, 24 (3), 363394
Forum
A proposed reinterpretation and reclassification of aphasic syndromes
Alfredo Ardila
Florida International University, Miami, FL, USA
Background: Different classifications of aphasic disorders have been proposed over the
years. During recent decades new approaches to aphasia study have been developed,
suggesting that current aphasia classifications can and should be reconsidered.
Aims: The purpose of this paper is to attempt to integrate contemporary knowledge
about brain organisation of language and to propose a new aphasia classification.
Main Contribution: It is emphasised that there are two fundamental forms of aphasia,
which are linked to impairments in the lexical/semantic and grammatical systems of
Downloaded By: [Ardila, Alfredo] At: 14:43 9 April 2010
language (Wernicke-type aphasia and Broca-type aphasia, respectively). Grammar

correlates with the ability to represent actions (verbs) and depends on what is known as
Brocas area and its related brain circuits, but it is also related to the ability to quickly
carry out the sequencing of articulatory movements required for speaking (speech
praxis). Lexical/semantic and grammatical systems not only depend on different brain
circuitries, but also on different types of memory and learning (declarative and
procedural). Other aphasic syndromes do not really impair language knowledge per se,
but rather peripheral mechanisms required to produce language (conduction aphasia
and aphasia of the supplementary motor area), or the executive control of the language
(extra-Sylvian or transcortical motor aphasia).
Conclusions: A new classification of aphasic syndromes is proposed: primary (or
central) aphasias (Wernickes aphasiathree subtypesand Broca aphasia);
secondary (or peripheral) aphasias (conduction aphasia and supplementary motor
area aphasia); and dysexecutive aphasia (extra-Sylviantranscorticalmotor aphasia),
are distinguished.
Keywords: Aphasia classification; Wernickes aphasia; Broca aphasia; Central aphasias;
Peripheral aphasias; Dysexecutive aphasia.
APHASIA CLASSIFICATIONS
Over 20 different aphasia classifications have been proposed since Brocas first
report (Broca, 1863). Table 1 presents the main aphasia classification proposals
presented in recent decades. However, two points should be emphasised: (1) some of
these classifications have played a guiding role in clinical practice and language
research; and (2) differences are frequently observed in the names used, not in the
clinical characterisation of the aphasia syndromes.
Address correspondence to: Alfredo Ardila PhD, ABPN, Florida International University,
Department of Communication Sciences and Disorders, HLS139, Miami, Florida 33199, USA. E-mail:
ardilaa@fiu.edu
Commentaries on the Lead Paper follow below and the Reply to the Commentaries appears on p. 413.
# 2010 Psychology Press, an imprint of the Taylor & Francis Group, an Informa business
http://www.psypress.com/aphasiology DOI: 10.1080/02687030802553704
364 ARDILA
TABLE 1
Some recent classifications of aphasia syndromes
Benson & Hecaen & Kertesz, Benson, Lecours et al.,

Luria, 1966 Geschwind, 1971 Albert, 1978 1979 1979 1983
Efferent motor Brocas Agrammatic Brocas Brocas Brocas

Sensory Wernickes Sensory Wernickes Wernickes Wernickes type I
Afferent motor Conduction Conduction Conduction Conduction Conduction
Dynamic Transcortical Transcortical Transcortical Transcortical Aspontaneity
motor motor motor motor
Transcortical Transcortical Transcortical Transcortical Wernickes type II
Sensory Sensory Sensory Sensory
Isolation Isolation Transcortical
language area mixed
Semantic Amnesic Anomic Amnesic Anomic Anomic Amnesic
Global Global Global

Aphemia Pure motor Aphemia Pure anarthria
Note. There are two popular aphasia tests that also include an aphasia classification: the Minnesota
Test for the Differential Diagnosis of Aphasia (Schuell, 1953, 1973); and Aachen Aphasia Test (Huber,
Poeck, Willmes, 1984; Willmes & Ratajczak, 1962).
The two most influential aphasia classifications are the Boston Group
classification (by Geschwind, Benson, Alexander, Goodglass, Kaplan, and others);
and Lurias aphasia interpretation. These two aphasia classifications have
significantly guided the area during the last decades.
Boston Group classification developed from Wernickes ideas, and two basic
distinctions are included: (1) aphasias can be fluent or non-fluent; and (2) aphasias
can be cortical, subcortical, or transcortical (e.g., Albert, Goodglass, Helm, Rubers,
& Alexander, 1981; Alexander & Benson, 1991; Benson, 1979; Benson & Geschwind,
1971, 1985; Geschwind, 1965; Goodglass, 1993; Goodglass & Kaplan, 1972).
Conduction aphasia (initially proposed by Wernicke in 1874 and described by
Lichtheim in 1885) is introduced to account for the language repetition impairments
frequently found in left parietal (or insular) damage.
Luria (1966, 1970, 1974, 1976) proposed six (or seven) aphasia subtypes. Until the
1960s he clearly distinguished six aphasia variants (motor efferent or kinetic, motor
afferent or kinaesthetic, acoustic-agnosic, acoustic-amnesic, semantic, and dynamic)
and suggested a seventh one (amnesic) (Luria, 1966, 1970). Later he overtly referred
to seven aphasia subtypes (Luria, 1976, 1980). The reason for this ambiguity is that
Luria was not sure if amnesic aphasia should be regarded as an independent aphasia
syndrome; or rather, whether the naming impairments associated with semantic
paraphasias should be included within semantic and/or acoustic-amnesic aphasia. In
Lurias approach, aphasia subtypes and names refer to the specific level of language
that is impaired (Table 2).
Benson and Ardila (1996) attempted to integrate the Boston and Lurian aphasia
interpretations. They proposed a classification based on two different anatomical
criteria: (1) aphasia can be pre-Rolandic (anterior, non-fluent) or post-Rolandic
(posterior, fluent); and (2) aphasia can be associated with pathology in the peri-
Sylvian language area (peri-Sylvian aphasias); or aphasia with damage beyond this
FORUM: LEAD PAPER 365
TABLE 2
Factors underlying different aphasic syndromes (according to Luria, 1976)
Type of aphasia Impaired factor
Acoustic-agnosic Phoneme discrimination

Acoustic-amnesic Verbal memory
Amnesic Semantic structure of the words
Semantic Understanding logic-grammatical quasi-
spatial relations
Afferent motor Articuleme discrimination
Efferent motor Kinetic melody of speech, inner speech
Dynamic Verbal initiative
area (extra-Sylvian). Subtypes were introduced for some aphasia syndromes.

Aphasias were also regarded as anatomical syndromes (Table 3). This classification
is currently used by different authors (e.g., Basso, 2003).
Following contemporary cognitive and psycholinguistic models of acquired

dyslexia (alexia)which usually identify three main central dyslexia subtypes
(phonological, surface, and deep) and two reading routes (direct and indirect) (e.g.,
Coltheart, 1980; Friedman, 1988; Friedman & Albert, 1985; Greenwald, 2001;
Jefferies, Sage & Ralph, 2007; Marshall & Newcombe, 1966, 1973; Morton &
Patterson, 1980; Patterson, 1978; Shallice, & Warrington, 1980)the terms
phonological aphasia (dysphasia), surface aphasia, and deep aphasia are
sometimes found in contemporary aphasia literature (e.g., Butterworth &
Warrington, 1995, Wilshire & Fisher, 2004; McCarthy & Warrington, 1984, 2001;
Michel & Andreewsky, 1983). This cognitive approach supposes that there are two
routes to language repetition: a phonological route and a semantic route. The
phonological representation of words is assumed to be functionally independent of
the semantic system (McCarthy & Warrington, 2001). Phonological aphasia would
be characterised by poor performance on auditory tasks with a strong phonological
component, whereas in spoken word production tasks errors are mainly phonemic
TABLE 3
Classification of aphasic syndromes (according to Benson & Ardila, 1996)
Pre-Rolandic Post-Rolandic
Peri-Sylvian Brocas Type I Conduction

(triangular syndrome) (parietal-insular syndrome)
Brocas Type II Wernickes Type I
(triangular-opercular-syndrome) (posterior insular-temporal
isthmus syndrome)
Wernickes Type II
(superior and middle temporal
gyrus syndrome)
Extra-Sylvian Extra-Sylvian Motor Type I Extra-Sylvian Sensory Type I

(left prefrontal dorsolateral syndrome) (temporal-occipital syndrome)
Extra-Sylvian Motor Type II Extra-Sylvian Sensory Type II
(supplementary motor area syndrome) (parieto-occipital angular
syndrome)
366 ARDILA
and formal paraphasias; single word repetition is poor but semantic errors are not
found (Fisher & Wilshire, 2004). In surface aphasia repetition performance is
affected by phonological similarity, word frequency, and stimulus lexicality, but is
not affected by linguistic complexity, word length, or any semantic variable
(McCarthy & Warrington, 2001). Deep aphasia is characterised by semantic errors
when repeating speech (Butterworth & Warrington, 1995). This interpretation of
language repetition errors assumes that the phonological and semantic spoken
language systems have different brain representation and can become dissociated in
cases of aphasia. Indeed, this interpretation approaches a specific language ability
language repetitionand, no question, it significantly contributes to further
understanding its normal and abnormal brain organisation. But it is not clear if it
can be regarded as a comprehensive aphasia classification.
During the last decades, in addition to the clinical descriptions and their
anatomical correlates, new approaches to aphasia research have developed:
Contemporary neuroimaging techniques, such as fMRI (e.g., Zahn et al., 2000),
PET (Cao, George, Ewing, Vikingstad & Johnson, 1998) and tractography (Yamada
et al., 2007), have been introduced, significantly extending our understanding of the
organisation of language in the brain under normal and abnormal conditions (Lee,
Kannan & Hillis, 2006; Small & Burton, 2002); an increased interest in identifying
the disrupted cognitive processes underlying different impaired functions has been
observed (Hillis, 2007); a significantly better understanding of the brain circuitries
supporting language has been developed (e.g., Ullman, 2004); a re-analysis of the
classical language areas (Brocas and Wernickes) has been developed (e.g.,
Grodzinky & Amunts, 2006); genetic advances have posited new perspectives to
the understanding of language organisation and language evolution (e.g., Vargha-
Khadem, Watkins, Alcock, Fletcher & Passingham, 1995); and new scientific
discoveries, such as the mirror neurons, have changed our understanding of the
functioning of the human brain, including language organisation (e.g., Rizzolati &
Arbib, 1998).
In this paper an attempt will be made to integrate these advances to propose a re-
classification of the aphasia syndromes.
TWO MAJOR APHASIC SYNDROMES

Since the 19th century it has been well established that there are two major and
fundamental aphasic syndromes, named in different ways, but roughly correspond-
ing to Wernicke-type aphasia and Broca-type aphasia (e.g., Albert et al., 1981;
Alexander & Benson, 1991; Bastian, 1898; Benson & Ardila, 1996; Freud, 1891/1973;
Goldstein, 1948; Head, 1926; Hecaen, 1972; Kertesz, 1979; Lichtheim, 1885; Luria,
1976; Pick, 1931; Schuell, Jenkins, & Jimenez-Pabon, 1964; Taylor-Sarno, 1998;
Wilson, 1926; see Tesak & Code, 2008, for review). This is a most basic departure
point in aphasia: Aphasia is not a single and unified clinical syndrome, but two rather
different (even opposed) clinical syndromes.
These two major aphasic syndromes have been related to the two basic linguistic
operations: selecting (language as paradigm) and sequencing (language as syntagm)
(Jakobson, 1971; Jakobson & Halle, 1956; Luria, 1972/1983). Jakobson (1964)
proposed that aphasia tends to involve one of two types of linguistic deficiency. A
patient may lose the ability to use language in two rather different ways: the
language impairment can be situated on the paradigmatic axis (similarity disorder)

or the syntagmatic axis (contiguity disorder).
The similarity disorder restricts the patients ability to select words on the
paradigmatic axis. These patients (Wernicke-type aphasia) cannot find words that
exist as parts of the system (vocabulary). These aphasic patients have severely limited
access to this language repertoire system. Specific nouns tend to be inaccessible.
These patients cannot select among alternative names (apple, pear, banana, etc) and
may instead fill out their discourse with circumlocutions (a clock may be referred to
as to know the time). Words no longer have a generic (paradigmatic) meaning for
these patients, and speech becomes empty. A dog can be referred to as fox, it
barks, etc.
Aphasic individuals presenting with what Jakobson referred to as contiguity
disorder (Broca-type aphasia) lose the ability to combine linguistic elements. Their
grammar is restricted or absent, and they can produce and understand only isolated
meaningful words. Words with purely grammatical functions (such as articles and
prepositions) tend to be omitted. Affixes may be substituted one for another but
more likely they are simply not produced. These patients thus tend to use only very
short sentences containing mostly meaningful words (nouns). In severe cases,
sentences can be as short as a single word (dog) and in general, there is a reduction
in resources available for syntactic processing (Caplan, 2006). In the Spanish
language with a significant flexibility in the word order, Broca aphasia is associated
with difficulties in understanding sentences with a non-canonical word order
(Ostrosky-Solis, Marcos-Ortega, Ardila, Rosselli, & Palacios, 1999). Luria further
developed Jakobsons ideas in his paper On the two basic forms of aphasic
disturbances (1972/1983). Luria emphasised that the selection disorder can be
observed at different levels of language, corresponding to different aphasia subtypes:
phoneme selection (aphasia acoustic agnosic), word selection (aphasia acoustic
amnesic), and meaning selection (amnesic aphasia). By the same token, the
contiguity disorder can be observed at different levels: sequencing words (kinetic
motor aphasiaBroca aphasia) or sequencing sentences (dynamic aphasia
transcortical motor aphasia). Noteworthy, different subtypes of Wernickes aphasia
are frequently distinguished (e.g., Ardila, 2006a). Lurias acoustic agnosic, acoustic
amnesic, and amnesic aphasia are indeed subtypes of the language impairment
syndrome referred to as a whole as Wernickes (sensory) aphasia.
Wernicke-type aphasia
Wernicke-type aphasia represents the clinical syndrome characterised by impair-
ments in the selection process (paradigmatic axis defect). In Wernickes aphasia the
lexical repertoire tends to decrease and language-understanding difficulties are
evident. Wernickes aphasia patients may not fully discriminate the acoustic
information contained in speech. Lexical (words) and semantic (meanings)
associations become deficient. In Wernicke-type aphasia obviously the language
defect is situated at the level of meaningful words (nouns). Phoneme and word
selection are deficient, but language syntax (contiguity: sequencing elements) is well
preserved and even overused (paragrammatism in Wernickes aphasia). Some times
these patients present errors that could be interpreted as sequencing errors in
phonemic paraphasias (such as left-to-right or right-to-left transpositions)
(Buckingham, 1989), but the percentage of this type of error is very low (less than 1%
368 ARDILA
of the total number of phonemic errors, at least in Spanish language; Ardila,

Montanes, Caro, Delgado, & Buckingham, 1989). Patients have problems in
recalling the words (memory of the words) and also in associating the words with
specific meanings. This means at least three different deficits underlie Wernicke-type
aphasia: (1) phoneme discrimination impairments (auditory verbal agnosia); (2)
verbal memory impairments; and finally (3) lexical/semantic association deficits.
Figure 1 presents in a summarised form the model proposed by Ardila (1993) to
account for the language recognition impairments observed in cases of Wernicke-
type of aphasia.
Nouns seem to depend on an organised pattern of brain activity. Contemporary
clinical and neuroimaging studies have corroborated that different semantic categories
are differentially impaired in cases of brain pathology. For instance, in anomia it has
been traditionally recognised that naming body-parts, external objects, and colours
depend (and are altered depending) on the activity of different brain areas (e.g.,
Hecaen & Albert, 1978). It has also been found that finer distinctions can be made with
regard to naming impairments, which can be limited to a rather specific semantic
category (e.g., peoples names, living things, tools, geographical names, etc.) (e.g.,
Goodglass, Wingfield, Hyde, & Theurkauf, 1986; Harris & Kay, 1995; Lyons, Hanley,
& Kay, 2002; Warrington & Shallice, 1984), and even as specific as medical terms
(Crosson, Moberg, Boone, Rothi & Raymer, 1997). A brain mapping of the
memory organisation of different semantic categories can be supposed.
Broca-type aphasia
Broca-type aphasia represents the clinical syndrome characterised by impairments in
the sequencing process (syntagmatic axis defect). It is usually recognised that Brocas
aphasia has two different distinguishing characteristics: (a) a motor component (lack
of fluency, disintegration of the speech kinetic melodies, verbal-articulatory
Figure 1. Three levels of language recognition potentially impaired in Wernicke-type aphasia can be
distinguished: phonemic (categorical perception level I), lexical (categorical perception level II), and
semantic (categorical perception level III). Three different sub-syndromes can be found: phonemic
discrimination defects (acoustic-agnosic or Wernickes aphasia type I), verbal-acoustic memory defects
(acoustic-amnesic or Wernickes aphasia type II), and semantic association defects (amnesic, nominal or
extra-Sylvian sensory aphasia). (Adapted from Ardila, 1993.)
impairments, etc., that is usually referred as apraxia of speech); and (b) agrammatism
(e.g., Benson & Ardila, 1996; Berndt & Caramazza, 1980; Goodglass, 1993; Kertesz,
1985; Luria, 1976). A large part of the fronto-parieto-temporal cortex has been
observed to be involved with syntactic-morphological functions (Bhatnagar,
Mandybur, Buckingham, & Andy, 2000). Apraxia of speech has been specifically
associated with damage in the left precentral gyrus of the insula (Dronkers, 1996; but
see Hillis et al., 2004b) It should be noted that not all of apraxia of speech is indeed a
contiguity disorder; there are many phonetic-level errors in apraxia of speech that have
more to do with segmental distortions, including lengthening of fricatives, intercala-
tion of schwas to break up difficult clusters, vocalic elongation due to an evening of
syllable stress, and a host of other phonetic aberrations that are not necessarily the
result of incorrect sequential ordering (Buckingham, 1998; Code & Ball, 1982).
If both impairments (apraxia of speech and agrammatism) are simultaneously
observed (i.e., they are very highly correlated), it simply means they are just two
different manifestations of a single underlying defect. It is not easy to understand
which one could be the single factor responsible for these two clinical manifestations,
but it may be an inability to sequence expressive elements (Figure 2). A single

common factor underlying both impairments may be assumed. Brocas area, most
likely, is not specialised in producing language, but in certain neural activity that can
support not only skilled movements required for speech, but also morphosyntax. It
has been observed that indeed language networks supporting grammar and fluency are
overlapped (Borovsky, Saygin, Bates, & Dronkers, 2007).
THE ISSUE OF BROCAS AREA

In the last decade there has been a significant interest in re-analysing the function of
Brocas area (e.g., Grodzinky & Amunts, 2006; Hagoort, 2005; Thompson-Schill,
2005). So-called Brocas area includes the pars triangularis (Brodmanns area
BA44) and probably the pars opercularis (BA 45) of the inferior frontal gyrus
(Foundas, Eure, Luevano, & Weinberger, 1998). BA 45 probably is more
cognitive than BA 44, which seems to be more motor, more phonetic. From the
traditional point of view, Brocas area corresponds to BA 44, but several
contemporary authors also include BA 45. In the traditional aphasia literature it
was assumed that damage in the Brocas area was responsible for the clinical
manifestations observed in Brocas aphasia. Only with the introduction of the CT
scan did it become evident that the damage restricted to the Brocas area was not
Figure 2. A single factor can account for the two major impairments observed in Broca aphasia. (Adapted
from Ardila & Bernal, 2007.)
370 ARDILA
enough to produce the classical Brocas aphasia; extension to the insula, lower
motor cortex, and subjacent subcortical and periventricular white matter is required
(Alexander, Naeser & Palumbo, 1990). Brocas area aphasia (minor Brocas
aphasia) is characterised by mildly non-fluent speech, relatively short sentences and
mild agrammatism; phonetic deviations and a few phonological paraphasias can be
observed (Mohr et al., 1978); some foreign accent can also be noticed (Ardila,
Rosselli, & Ardila, 1988).
Simultaneously including both BA 44 and BA 45 in Brocas area is
problematic. BA 44 is a premotor dysgranular area, whereas BA 45 has a
granular layer IV and belongs to the heteromodal prefrontal lobe (granular
cortex) (Mesulam, 2002). So, from a cytoarchitectonic point of view, BA 44 and
BA 45 are quite different. BA 44 is a premotor area whereas BA 45 corresponds to
the prefrontal cortex. From the aphasia perspective, some authors have referred
to different clinical manifestations associated with damage in BA 44 (Broca-type
aphasia) and BA 45 (transcortical motor/dynamic aphasia) (e.g., Luria, 1976).
Brocas area is, more than likely, involved in different language and language-
related functions (Fink et al., 2006). Some authors have pointed out that indeed
Brocas area is a collective term that can be fractionated in different sub-areas
(Lindenberg, Fangerau & Seitz, 2007).
Hagoort (2005, 2006) refers to the Brocas complex, including BA 44
(premotor), and also BA 45 and BA 47 (prefrontal cortex). He argues that Brocas
complex is not a language-specific area, and it becomes active during some non-
language activities, such as mental imagery of grasping movements (Decety et al.,
1994). Functional defined sub-regions could be distinguished in the Brocas complex:
BA 47 and BA 45 are involved in semantic processing, BA 44, BA 45, and BA 46
participate in syntactic processing, and BA and BA 44 are involved in phonological
processing. Hagoort (2005) proposes that the common denominator of the Brocas
complex is its role in selection and unification operations by which individual pieces
of lexical information are bound together into representational structures spanning
multiword utterances (p. 166). Its core function is, consequently, binding the
elements of the language.
Thompson-Schill (2005) analysed the different deficits observed in cases of
damage in the Brocas area: articulation, syntax, selection, and verbal working
memory, suggesting that there may be more than a single function of Brocas area.
Brocas area is related with several functions, and different disturbances are found in
differing cases of pathology. The author proposes a framework for describing the
deficits observed in different patients. The proposed framework suggests that
Brocas area may be involved in selecting information among competing sources.
Fadiga, Craighero, and Roy (2006) speculates that the original role played by
Brocas area relates to generating/extracting action meanings; that is, organising/
interpreting the sequence of individual meaningless movements. Ardila and Bernal
(2007) conjectured that the central role of Brocas area was related to sequencing
motor/expressive elements. Novick, Trueswell, and Thompson (2005) consider that
the role of Brocas area is related with a general cognitive control mechanism for the
syntactic processing of sentences.
Grodzinsky (2000, 2006) has presented an extensive analysis of the role of Brocas
area. He proposed that most syntax is not located in Brocas area and its vicinity
(operculum, insula, and subjacent white matter). This brain area does have a role in
syntactic processing, but a highly specific one: it is the neural home to receptive
mechanisms involved in the computation of the relation between transformationally

moved phrasal constituents and their extraction sites (syntactic movement). He further
assumes that Brocas area is also involved in the construction of higher parts of the
syntactic tree in speech production. Interestingly, blood flow in Brocas area
increases when participants process complex syntax (Caplan, Alpert, Waters, &
Olivieri, 2000). Santi and Grodzinsky (2007) also recognise its role in working
memory related with a specific syntactic role in processing fillergaps dependency
relations. Syntax is indeed neurologically segregated, and its components are housed
in several distinct cerebral locations, far beyond the traditional ones (Brocas and
Wernickes regions). A new brain map for syntax would also include portions of the
right cerebral hemisphere (Grodzinsky & Friederici, 2006).
Haverkort (2005) emphasises that a clear distinction should be established
between linguistic knowledge and linguistic use. Patients with Brocas aphasia have a
limitation in the use of grammar, but their grammatical knowledge is available.
Brocas aphasia patients present a simplified syntax and phrases are short. They
select simpler syntactic structures that are less complex because they impose less
burden on working memory. In consequence, one major factor in Brocas aphasia

relates to an impairment in verbal working memory.
In summary, regardless that expressive language disturbances have been
associated for over a century with damage in the left inferior frontal gyrus (later
known as Brocas area), currently there is incomplete agreement about its limits
and its specific functions in language. Different proposals have been presented to
explain language disturbances in so-called Brocas aphasia, including: binding the
elements of the language (Hagoort, 2005), selecting information among competing
sources (Thompson-Schill, 2005), generating/extracting action meanings (Fadiga et al.,
2006); sequencing motor/expressive elements (Ardila & Bernal, 2007); cognitive
control mechanism for the syntactic processing of sentences (Novick et al., 2005);
construction of higher parts of the syntactic tree in speech production (Grodzinsky,
2000, 2006); and verbal working memory (Haverkort, 2005).
WHAT ABOUT CONDUCTION APHASIA AND OTHER

APHASIA SYNDROMES?
Frequently it is assumed that three major aphasic (perisylvian) syndromes can be
distinguished: Brocas aphasia, Wernicke aphasia, and conduction aphasia (e.g.,
Benson, 1979; Goodglass, 1993). These are the three aphasia disorders associated
with damage in the so-called brain language area; a concept introduced by
Dejerine (1914), roughly corresponding to the perisylvian area of the left hemisphere.
Different aphasia classifications generally include a diversity of additional language
disturbances, such as transcortical (extra-Sylvian) aphasia and anomic aphasia (e.g.,
Alexander & Benson, 1991; Benson & Geschwind, 1971; Hecaen & Albert, 1978;
Kertesz, 1979; Lecours, Lhermitte, & Bryans, 1983; Luria, 1966).
Indeed, some aphasic syndromes can eventually be considered as variants of the
Brocas and Wernicke aphasias. For instance, amnesic or anomic or nominal
aphasia (usually due to damage in the vicinity of BA 37) (Head, 1926; Hecaen &
Albert, 1978; Luria, 1976) can be interpreted as a subtype of Wernicke aphasia in
which the semantic associations of the words are significantly impaired (see
Figure 1).
372 ARDILA
Transcortical (extra-Sylvian) sensory aphasia

Transcortical (extra-Sylvian) sensory aphasia (TSA) has been a polemic syndrome.
Seemingly, the polemic is related to the way TSA is defined and the elements
included in its definition. Some authors have even simply denied the existence of
such a syndrome.1 Currently, two integrative revisions of TSA are available
(Berthier, 1999; Boatman et al., 2000).
TSA is defined in two partially different ways; (1) according to its basic
definition, TSA is a fluent language disorder characterised by impaired auditory
comprehension, with preserved repetition (Albert et al., 1981; Berthier, 1999;
Goldstein, 1948; Lichtheim, 1885). Consequently, there are only three distinguishing
characteristics in TSA (normal fluency, impaired auditory comprehension, and
preserved repetition). In such a case, TSA presents similar deficits as in Wernickes
aphasia, but repetition ability is spared and phoneme discrimination impairments are
not found. (2) According to its extended definition, TSA also includes a semantic
jargon (Goodglass, 1993; Kertesz, 1982; Lecours, Osborn, Travies, Rouillon, &
Lavalle-Huyng, 1981). So, the question is: Should semantic jargon be included in the
definition of TSA? Kertesz (1985, p. 317) makes a comprehensive definition of TSA:
TSA is characterised by fluent and often irrelevant speech output, very poor
comprehension and well-preserved repetition. Spontaneous speech often consists of
semantic jargon that has no relationship to what is being asked of the patient
(italics are mine). This definition clearly recognises that there are three basic
characteristics, and sometimes jargon is found. But jargon is not a required symptom
for the diagnosis of TSA. By the same token, other language impairments can also be
found, such as poor naming, and preserved oral reading with impaired reading
comprehension, but their presence is not essential to establish the diagnosis of TSA
(Berthier, 1999).
According to Berthier (1999) the most common pattern of verbal expression is
represented by the so-called semantic or verbal jargon (e.g., Lecours &
Rouillon, 1976). There is an abundant language production, with reduction of
meaningful words conferring the impression of emptiness. The content of the
sentence is irrelevant. Furthermore, TSA patients appear unaware of their logorrhea
(Lebrun, 1987). A second pattern of spontaneous speech described by Berthier
(1999) is referred to as anomic and is associated with an impaired access to content
words. This second pattern corresponds to the TSA basic definition mentioned
above.
Because repetition is spared, phonological processing is assumed to be preserved,
at least partially, while lexical-semantic information included in the word meaning is
impaired (Boatman et al., 2000). Usually, it is accepted that TSA is associated with
relatively extensive posterior lesions including the temporo-parieto-occipital junction
of the left hemisphere but sparing the areas around the primary auditory cortex
(Berthier, 1999). Damasio (1991) observed that TSA is associated with lesions
involving the temporal-occipital area (BA 37), the angular gyrus (BA 39), or the
white matter underlying these regions, but sparing the primary auditory cortex (BA
41 and 42), and BA 22. Damasio suggested that the core area for TSA is the
1
Once somebody asked Luria: Why you do not ever refer to transcortical sensory aphasia? Luria
simply answered, because I never have seen a case of transcortical sensory aphasia (personal
observation).
temporal-occipital area (BA 37) with variable extension to the occipital lobe and the
angular gyrus. Crossed TSA has also been reported (e.g., Roebroek, Promes,
Korten, Lormans, & van der Laan, 1999). Some similarities between so-called
thalamic aphasia and TSA have been suggested (e.g., Berthier, 1999; Cappa &
Vignolo, 1979).
Kertesz (1982) analysed 15 patients with TSA and proposed two different
subgroups: one is more medial, inferior, and posterior and is clearly in the posterior
cerebral artery territory; and the other is relatively more lateral, superior, and
anterior and seems to be in a watershed area between middle cerebral and posterior
cerebral arteries. McCarthy and Warrington (1984) reported two patients with
impaired paraphasic performance in repetition tasks but relatively well-preserved
spontaneous speech (conduction aphasia); and one patient presenting impaired
paraphasic spontaneous speech but intact repetition (transcortical motor aphasia).
When active semantic processing was required the conduction aphasics were
facilitated and the transcortical motor aphasic impaired; in tasks requiring passive
repetition the opposite pattern of dissociation was observed. Departing from these
observations, the authors proposed a two-route model of the speech production

process. Coslett, Roeltgen, Gonzalez-Rothi, and Heilman (1987) proposed that
repetition may be mediated by at least two distinct processesa lexical process that
may involve the recognition and subsequent activation of discrete stored word
representations and a non-lexical process that involves phonologic decoding and
immediate phonologic encoding from immediate memory. Analysing spontaneous
speech, reading, and the tendency to recognise and spontaneously correct syntactic
errors in four patients with TSA, Coslett et al. (1987) concluded that there are two
subtypes of TSA. In one subtype both the lexical and direct repetition (or speech
production) mechanisms are preserved, but in the second subtype the lexical
mechanism is disrupted and repetition is mediated by the non-lexical mechanism.
Benson and Ardila (1996), considering this variability in TSA, also distinguished two
subtypes: the first one similar to Lurias amnesic aphasia (BA 37), and the second
one corresponding Lurias semantic aphasia (BA 39). This distinction is coincidental
with the neuroanatomical correlates of TSA found by Damasio (1991).
Recent reports support the assumption that TSA is usually found associated with
extensive lesions of the left hemisphere (e.g., Warabi, Bandoh, Kurisaki, Nishio, &
Hayashi, 2006), generally involving large portions of the temporal-parietal-occipital
areas. According to Alexander, Hiltbrunner, and Fischer (1989) the critical lesion for
transcortical sensory aphasia in these patients involved pathways in the posterior
periventricular white matter adjacent to the posterior temporal isthmus, pathways
that are most likely converging on the inferolateral temporo-occipital cortex.
However, frequently the lesions responsible for TSA account for the variability
observed in its clinical manifestations, suggesting that TSA does not necessarily
represent a single aphasic syndrome. When the lesions are restricted to BA 37 or BA
39, specific and well-described language impairments are observed (Benson & Ardila,
1996; Luria, 1976). With more extended lesions additional clinical manifestations,
such as jargon, can be found. These additional clinical manifestations are only
observed in the acute stage of the brain pathology, and progressively disappear
(Kertesz, 1979). Dronkers and Larsen (2001) state that transcortical sensory
aphasia always resolves into mild anomic aphasia (p. 29).
Thus, TSA represents a disorder in the semantic recognition of language that may
or may not be associated during the acute stage with other language impairments,
374 ARDILA
specially logorrhoea and jargon, depending on the extension of the lesion. But
logorrhoea and jargon are not required in the definition of TSA.
However, surprisingly TSA is also reported in cases of frontal damage (e.g.,
Berthier, 1999; Sethi, Burke, Torgovnick, & Arsura, 2007; Maeshima, Osawa,
Nakayama, & Miki, 2004; for an extensive review of these cases, see Berthier, 2001).
Berthier (2001), departing from the cases reported in the literature and his own
patients, concluded that syndrome of TSA associated with frontal lobe lesions is
phenomenologically similar to the classical variant, except for the presence in the
former of some deficits in speech articulation, prosody, phonology, and skilled
buccofacial movements indicative of frontal lobe involvement (p. 124). Otsuki et al.,
1998) compared 7 patients with TSA following the left frontal lesions (F) and 10
TSA patients following the left posterior lesions (P). The results indicated that the
disturbance of the P group was of semantic processing, per se. And the disturbance
of F was not only in semantic processing but also in the working memory process.
It sounds quite intriguing that an aphasia syndrome can have two clearly different
brain localisations. It seems reasonable to assume that both cortical areas (frontal
and posterior) should be included in a common brain circuitry. The similarity in the
language disturbances suggests that such a particular circuitry has been impaired at
different points. Zahn and colleagues (2000), using fMRI in normal participants,
observed that the areas specific to semantic processing were restricted to the left
hemisphere: the posterior middle frontal (BA 9) and posterior parietal (BA 7/40)
cortex, as well as an inferior temporal area (BA 20/21). They further analysed two
cases of TSA with lesions affecting either solely the prefrontal (Patient 1) or both the
prefrontal and posterior parietal part of the network activated in normal participants
(Patient 2) (Zahn et al., 2002). Both patients showed TSA on acute assessment. Their
recovery of language comprehension was associated with activation of a left
hemispheric network. Mainly activations of left perilesional prefrontal regions
(Patient 2), left Wernickes area (Patient 2 and Patient 1), or the left posterior middle
and inferior temporal cortex (Patient 1) were demonstrated in the TSA patients.
The observation that language can be impaired at the semantic level, with damage
located at rather different areas, reinforces the idea that semantics is represented in a
complex brain circuit including not only temporal, but also parietal and frontal areas
(e.g., Damasio, Tranel, Grabowski, Adolphs, & Damasio, 2004; Scott, 2005; Small,
Hart, Nguyen, & Gordon, 1995). Indeed, different brain areas participate in
language comprehension, including the posterior middle temporal gyrus and
underlying white matter, the anterior superior temporal gyrus, the superior temporal
sulcus and angular gyrus, mid-frontal cortex in BA 46, and BA 47 of the inferior
frontal gyrus. The middle temporal gyrus may be more important for comprehension
at the word level, while the other regions may play a greater role at the level of the
sentence (Dronkers, Wilkins, Van Valin, Redfern, & Jaeger, 2004).
Transcortical (extra-Sylvian) motor aphasia as a

dysexecutive aphasia
Some aphasic syndromes can be interpreted as language disturbances due to a more
general underlying disorder. For instance, extra-Sylvian (transcortical) motor
aphasia associated with left convexital prefrontal damage could be interpreted as
an executive function defect specifically affecting language use. The ability to
actively and appropriately generate language appears impaired while the phonology,
lexicon, semantics, and grammar are preserved. Simply speaking, the question is:
Should the ability to correctly generate language be regarded as a linguistic ability
(i.e., cognitive ability)? Or rather, should it be considered as an executive function
ability (i.e., metacognitive ability)? It does not seem difficult to argue that the ability
to correctly organise language sequences can be interpreted as an executive function
and as a metacognitive ability rather than a purely linguistic ability. Some rationales
to support this interpretation are: (1) It could be argued that in extra-Sylvian
(transcortical) motor aphasia there is a defect in verbal initiative rather than in
language knowledge (Kleist, 1934). (2) Some authors (Luria 1976, 1980) have
emphasised that this type of aphasia shares the general characteristics of prefrontal
(i.e., dysexecutive) syndrome but specifically with regard to verbal processes. This
means, it is the prefrontal (dysexecutive) syndrome affecting the verbal processes
(Gold et al., 1997). (3) Further, the impairment in extra-Sylvian (transcortical) motor
aphasia does not affect language understanding, and fundamental linguistic
processes are preserved (Berthier, 1999). And finally, (4) it could be argued that
the prefrontal cortex does not participate in basic cognition, but rather in
metacognition (e.g., Ardila & Surloff, 2006). In consequence, extra-Sylvian

(transcortical) motor aphasia does not necessarily have to be interpreted as a
primary aphasic syndrome, but rather as a language disturbance due to a more
general intellectual impairment (dysexecutive syndrome). In this regard, it is a
secondarynot primaryaphasia syndrome. Extra-Sylvian (transcortical) motor
aphasia could indeed be referred to as dysexecutive aphasia.
Some authors have previously interpreted extra-Sylvian motor aphasia in a
similar way (e.g., Luria 1976, 1980). Recently, Alexander (2006) suggested that
transcortical motor aphasia could be more accurately defined as an executive
function disorder rather than aphasia. He proposed that the progression of clinical
disorders from aphasia to discourse impairments can be interpreted as a sequence of
procedural impairments from basic morpho-syntax to elaborated grammar to
narrative language, correlated with a progression of the focus of the damage from
posterior frontal to polar and/or lateral frontal to medial frontal.
It is noteworthy that successful functional communication is significantly
associated with executive function in aphasia (Fridriksson., Nettles, Davis,
Morrow, & Montgomery, 2006). The ability for successful communication in
aphasic patients is apparently dependent on the integrity of executive function skills
(Purdy, 2002).
Conduction aphasia as a secondary (peripheral) aphasia

A crucial question is, how can conduction aphasiaa well recognised and
extensively studied aphasic syndrome (e.g., Benson & Ardila, 1994; Damasio &
Damasio, 1980; Goldstein, 1948; Kohn, 1992)be interpreted? The most frequent,
and classic, explanation of conduction aphasia is as a disconnection syndrome (e.g.,
Damasio & Damasio 1980; Geschwind 1965; Wernicke 1874), usually due to a lesion
affecting the arcuate fasciculus (Yamada et al., 2007) and sporadically in an indirect
pathway passing through the inferior parietal cortex (Catani, Jones, & Ffytche,
2005). Alternatively, conduction aphasia has also been interpreted as a segmentary
ideomotor apraxia (e.g., Ardila & Rosselli., 1990; Brown, 1972, 1975; Luria 1976,
1980). According to this second interpretation, conduction aphasia could be
regarded as a verbal apraxia, an ideomotor apraxia impairing the movements
376 ARDILA
required for speaking, or simply as a kinaesthetic apraxia of speech. Luria (1976)

suggested that paraphasias in conduction aphasia (Lurias kinaesthetic motor or
afferent motor aphasia) are articulatory-based (articulatory literal paraphasias).
These errors are due mainly to phoneme substitutions and phoneme deletions; they
result basically in switches in phoneme manner and place of articulation (Ardila,
1992). Similarities between errors in ideomotor apraxia and conduction aphasia
language deficits have been suggested.
According to Benson, Sheretaman, Bouchard, Segarra, Price, and Geschwind
(1973) conduction aphasia has three fundamental and five secondary characteristics;
so-called secondary characteristics are frequently but not necessarily found in
conduction aphasia. The three basic characteristics are: (1) fluent conversational
language; (2) comprehension almost normal; and (3) significant impairments in
repetition. Secondary characteristics include: (1) impairments in naming; (2) reading
impairments; (3) variable writing difficulties (apraxic agraphia); (4) ideomotor
apraxia; and (5) additional neurological impairments. Bartha and Benke (2003)
report that conduction aphasia patients present as relatively homogenic in their
aphasic manifestations: severe impairment of repetition and fluent expressive

language functions with frequent phonemic paraphasias, repetitive self-corrections,
word-finding difficulties, and paraphrasing. Repetitive self-corrections frequently
result in so-called conduit dapproche. Language comprehension (auditory and
reading) is only mildly impaired.
Benson et al.s. (1973) description of conduction aphasia clearly recognises that
spontaneous language production and language understanding are significantly
preserved. In consequence, some mechanisms required for correct language
repetition are impaired, but the knowledge of language itself (phonology, lexicon,
semantics, and grammar) is not impaired. Should conduction aphasia be interpreted
as a primary aphasic syndrome?
Indeed, language repetition impairments can be observed in different aphasia
syndromes and language repetition has also been interpreted as a right hemisphere
ability (Berthier et al., 1991). The distinction between aphasias with repetition
impairments vs aphasias without repetition impairments is a general and crude
distinction. Ardila and Rosselli (1992) analysed 38 Spanish-speaking aphasic
patients divided into six aphasia groups (transcortical motor, Brocas, conduction,
Wernicke, anomic, and global). Three repetition tests (words, high-probability and
low-probability sentences) taken from the Boston Diagnostic Aphasia Examination,
Spanish version (Goodglass & Kaplan, 1979) were administered. Repetition errors
were generally associated with perisylvian aphasias (Brocas, conduction, and
Wernickes). However, in all aphasic groups some repetition errors were observed
(Table 4). They were not only quantitative but also qualitatively different. It was
concluded that, depending on the specific repetition task, errors may be evident or
may be unnoticed in a particular aphasic group. Different mechanisms underlying
repetition deficits were proposed: limitation of auditory-verbal short-term memory,
difficulties at the level of phonological production, impairments in phoneme
recognition, and semantic and syntactic comprehension.
It is interesting to note that in the classification of reading and writing
disturbances, a frequent distinction is made between central and peripheral
alexias (dyslexias), and agraphias (dysgraphias) are introduced (e.g., Ellis & Young,
1988). Central alexias and agraphias impair the knowledge of written language
(recognition or production), whereas peripheral alexias and agraphias impair the
TABLE 4
Percentage of correct repetition for each patient group on the three repetition tasks from the
Boston Diagnostic Aphasia Examination.
Words High-Probability Low-Probability
Transc motor (n 5 6) 98.0 95.0 67.5

Broca (n 5 5) 46.0 50.0 45.0
Conduction (n 5 6) 63.0 53.7 21.2
Wernickes (n 5 13) 74.0 45.0 22.5
Anomic (n 5 4) 100.0 71.2 52.5
Global (n 5 4) 27.0 0.0 0.0
Adapted from Ardila and Rosselli, 1992.
normal access to written information (in cases of alexias) or the mechanisms required
for written language production (in cases of agraphias). Central agraphias are also
named linguistic or aphasic agraphias, whereas peripheral agraphias are
referred to as non-linguistic or non-aphasic agraphias (Benson & Ardila, 1996).

Could a similar distinction be introduced in cases of oral language disturbances?
Indeed, a similar rationale could be used in the case of conduction aphasia:
conduction aphasia is not a primary form of aphasia, but rather a secondary (or
peripheral) defect in language indirectly affecting specific language abilities (i.e.,
the ability to repeat). Language itself is not impaired, but rather it is an impaired
ability to reproduce aloud the auditory information that is heard. Of course, this is
an important skill used not only to develop language but also to use it correctly. The
correct visual-perceptual recognition of the written information is also an important
ability required not only in learning to read, but also in correctly accessing written
information. Interpreting conduction aphasia as a secondary (or peripheral) defect
in language (rather than a primary or central form of aphasia) does not in any way
decrease the importance of repetition in language.
In brief, I am proposing that conduction aphasia be interpreted as a secondary
(or peripheral) language disturbance, rather than a primary (or central) form of
aphasia. Language knowledge is well preserved in conduction aphasia, but there is a
limitation in a particular language function, i.e., repetition. Obviously, if parrots can
repeat, that means that language repetition cannot be considered as a primary
linguistic ability. Interestingly, Jakobson (1964) suggested a similar distinction when
proposing that in aphasia, language could be either disintegrated or limited
(disintegration vs limitation in aphasia). Obviously, in conduction aphasia language
is limited, not disintegrated.
Supplementary motor area (SMA) aphasia

In 1940 Brickner reported that electro-cortical stimulation of SMA (mesial aspect of
Brodmanns area 6) resulted in continuous perseveration. Penfield and Welch (1951)
observed arrest of speech associated with stimulation of this cortical region.
However, language disturbances associated with SMA pathology are reported
relatively late in the aphasia literature.
Clinical characteristics of this type of aphasia were described by Rubens (1975,
1976). Jonas (1981) later referred to the participation of the SMA in speech emission.
378 ARDILA
The occlusion of the left anterior cerebral artery is the most frequent aetiology, but it
has also been reported in cases of tumours and traumatic head injury (e.g., Ardila &
Lopez, 1984). Speech is characterised by (1) an initial mutism lasting about 210
days; (2) later, a virtually total inability to initiate speech, (3) a nearly normal speech
repetition, (4) a normal language understanding, and (5) absence of echolalia. A
right leg paresis and right leg sensory loss are observed; a mild right shoulder paresis
and Babinski sign are also found. Language recovery is outstanding and it is usually
observed during the following few weeks or months.
The SMA is a mainly mesial premotor area involved in the ability to sequence
multiple movements performed in a particular order (Tanji & Shima, 1994). SMA
participates in initiating, maintaining, coordinating, and planning complex
sequences of movements; it receives information from the posterior parietal and
frontal association areas, and projects to the primary motor cortex (Kandel,
Schwartz & Jessell, 1995). SMA damage is also associated with slow reaction time
(Alexander, Stuss, Picton, Shallice, & Gillingham, 2007). It has been observed that
activation of the SMA precedes voluntary movement (Erdler et al., 2000); a crucial
role in the motor expression of speech processing has also been postulated (Fried et al.,
1991). Nonetheless, the SMA is located some distanceand indeed far away
from the classic language area postulated by Dejerine (1914) and assumed in most
anatomical models of aphasia.
Recently it has been suggested that SMA has a close connectional relationship
with the prefrontal cortex and plays a critical role in the update of verbal
representations (Tanaka, Honda, & Sadato, 2005). Neuroimaging studies in humans
have demonstrated that SMA is active when performing various cognitive tasks,
such as spatial working memory (Jonides et al., 1993), verbal working memory
(Paulesu, Frith, & Frackowiak, 1993), arithmetic tasks (Dehaene et al., 1996;
Hanakawa et al., 2002), spatial mental imagery (Mellet et al., 1996), and spatial
attention (Simon et al., 2002).
Evidently, the SMA is a complex motor cortical area, not primarily a language-
related brain area. Its role in language seemingly refers to the motor ability to initiate
and maintain voluntary speech production.
The question of subcortical aphasia

Since Wernicke (1874) it has been assumed that aphasia can represent the
consequence of damage to neural networks including both cortical and subcortical
structures. Nonetheless, when Dejerine (1914) described the brains language area
no specific mention to subcortical structures was made. The idea of subcortical
aphasia was somehow forgotten during the following decades. Only with the
introduction of the CT scan was it observed that aphasia was frequently associated
with subcortical pathology, and the discussion and interpretation of subcortical
aphasia re-emerged (e.g., Basso, Della Sala, & Farabola, 1987; Cappa, Cavallotti,
Guidotti, Papagno, & Vignolo, 1983; Nasser et al., 1982). Contemporary
neuroimaging techniques have permitted far better understanding of subcortical
pathology involved in aphasia. Nonetheless, whether true aphasia results from
isolated subcortical brain damage, or whether it is due to a cortical extension or
cortical deactivation, remains unanswered (e.g., Craver & Small, 1997; Demonet,
1997). Subcortical pathology frequently includes altered speech (dysarthria), often
beginning with total mutism followed by hypophonic, slow, sparse output, and
poorly differentiated, amelodic speech (Alexander, Naeser, & Palumbo, 1987). In

addition to dysarthria, sometimes language impairments are also found. Two
neuroanatomical areas are most frequently discussed in subcortical aphasias: the
striatocapsular region and the thalamus. Patients with striatocapsular damage show
significant articulation impairments. Their language output appears truncated, but it
is not agrammatic. Speech mechanisms are generally impaired resulting in
impairments in articulation and prosody. Comprehension is intact for casual
conversation but breaks down when complex syntax is presented. Word-finding
problems may be noted (Alexander et al., 1987). Alexander and colleagues (1987)
have proposed six subtypes of verbal output impairment. These subtypes are
dependent on the specific neuroanatomical locus of striatocapsular damage,
demonstrating that considerable variation in speech and language impairment can
follow this type of pathology. Frequently, extension that involves the cortex is
present in these cases. Extensive subcortical damage is required to produce a pure
striatocapsular aphasia, but prognosis is worse when the posterior limb of the
internal capsule is involved (Liang et al., 2001). When no cortical abnormalities are
found on MRI, the severity of the left striatocapsular aphasia is associated with the
extent and severity of the left cerebral cortical hypoperfusion on brain perfusion
SPECT, particularly in the left temporal cortex (Choi et al., 2007).
Mega and Alexander (1994) evaluated 14 cases of striatocapsular aphasia. The
clinical profiles of the patients were quite similar, varying in severity in rough
proportion to lesion size and varying in quality in proportion to anterior
paraventricular extent. Large lesions were associated with impaired executive
and generative language functions. Similar aphasia profiles in patients with deep
frontal and paraventricular white matter lesions suggest that damage to a frontal-
caudate functional system underlies a core aphasia profile in these patients.
DEsposito and Alexander (1995) analysed 13 patients with left putaminal
haemorrhage. There was no apparent correlation between lesion site and acute
language profiles. During the post-acute epoch there were several distinct
correlations between lesion site and specific aphasia dimensions. Nadeau and
Crosson (1997), after a critical review of the literature, suggested that linguistic
impairments associated with striatocapsular pathology are predominantly related to
sustained cortical hypoperfusion and infarction not visible on structural imaging
studies.
It has been speculated that this heterogeneity in the aphasia pattern in cases of
basal ganglia pathology may reflect variations in cortical hypoperfusion resulting
from large vessel stenosis. To test this hypothesis, Hillis et al. (2004a) analysed a
consecutive series of 24 patients with left caudate infarct identified with diffusion-
weighted imaging who underwent language testing and perfusion-weighted imaging
less than 24 hours from onset of symptoms. Specific regions in perisylvian cortex
were rated for the percentage of the region that was hypoperfused. Aphasia type was
determined on the basis of speech fluency, comprehension, and repetition
performance on the language tests. Results demonstrated that in patients with acute
left caudate infarct the presence and type of aphasia reflected regions of
hypoperfusion and generally followed predictions based on chronic lesion studies
regarding anatomical lesions associated with classic aphasia types.
Thalamic pathology associated with aphasia usually produces an acute,
catastrophic clinical picture with hemiplegia, hemisensory loss, and alterations in
the level of consciousness (Benabdeljlil et al., 2001). The initial language abnormality
380 ARDILA
is mutism, which typically improves to a verbose, paraphasic, but hypophonic jargon

output. Anomia is often severe. Although thalamic aphasia resembles other fluent
paraphasic aphasias, patients with thalamic aphasia show decreased comprehension.
When they attempt to repeat a word or phrase, their verbal output is far better than
their conversational speech. A similarity to extra-Sylvian sensory aphasia has been
noted, even though syntactic impairments have also been reported (De Witte,
Wilssens, Engelborghs, De Deyn, & Marien, 2006). Language disorder syndromes
noted following thalamic damage can be categorised into three subtypes: (1) medial
(left paramedial thalamic area, involving dorsomedial and centromedian nuclei), (2)
anterior (left anterolateral nucleus), and (3) lateral (left lateral thalamus). It has been
suggested that thalamic nuclei and systems are involved in multiple processes that
directly or indirectly support cortical language functions: lexical-semantic functions,
working memory, visual processing in reading, and category-specific naming
(Crosson, 1999). It has been further proposed that the left thalamus seems to bring
online the cortical network involved in language processing (Metz-Lutz et al., 2000).
In brief, aphasia is sometimes associated with subcortical lesions, particularly left
striatocapsular and thalamic pathology. To account for subcortical aphasia it has

been proposed that aphasia may result from a cortical extension or cortical
deactivation. The idea of a cortical deactivation seems to prevail (e.g., Choi et al.,
2007; Hillis et al., 2004a). The question of subcortical aphasia suggests the existence
of cortical-subcortical circuits in language, as observed in other forms of cognition
(e.g., Cummings, 1993; Lichter & Cummings, 2001). The analysis of subcortical
aphasia can significantly advance the understanding of the language representation
in the brain, but indeed it does not affect the issue of aphasia classification.
TWO BRAIN LANGUAGE SYSTEMS: PRIMARY APHASIAS

It may be conjectured that language is supported by two different brain systems.
Lesions in each will result in Wernicke-type or Broca-type aphasia. These two brain
language systems correspond to the lexical/semantic (selecting; language as
paradigm) and morphosyntactic (sequencing; language as syntagm).
Ontogenetically, they appear at different ages during child language acquisition:
the ability to use single words for communication (holophrasic period) is observed
around age 1218 months; whereas the ability to combine several words in a single
sentence is found later, around age 2430 months (Hoff, 2003). Phylogenetically,
they probably developed at different stages in evolution: whereas language as a
lexical/semantic system may have appeared before the contemporary Homo sapiens,
language as a grammatical system may have appeared relatively recently in human
history (maybe some 10100 thousand years ago) and seems to be exclusive to Homo
sapiens (Ardila, 2006b).
Brain representation of nouns and verbs

It has been observed that verbs and nouns clearly depend on different areas of brain
activity, and naming objects and actions are disrupted in different types of brain
pathology. Increased activity is observed in the temporal lobe while speaking or
thinking in nouns, whereas speaking or thinking in verbs activates Brocas frontal
area (Raichle, 1994). By the same token, impairments in finding nouns are associated
with temporal lobe pathology, whereas impairments in finding verbs are associated
with left frontal damage and Brocas aphasia (Ardila & Rosselli, 1994; Damasio &
Tranel, 1993).
Naming actions activates the left frontal operculum roughly corresponding to
Brocas area (Damasio et al., 2001). The neural correlates of naming concrete entities
such as tools (with nouns) and naming actions (with verbs) are partially distinct: the
former are linked to the left inferotemporal region, whereas the latter are linked to
the left frontal opercular and left posterior middle temporal regions (Tranel, Martin,
Damasio, Grabowski, & Hichwa, 2005). Recently Luzzatti, Aggujaro, and Crepaldi
(2006) studied the lesions causing impairments related predominantly to verbs or
nouns. Results obtained on these studies indicated that lesions causing impairment
predominantly with nouns were located mostly in the middle and inferior left
temporal area. Three alternative lesion sites were associated with a verb deficit (left
posterior temporo-parietal lesions; large left fronto-temporal perisylvian lesions;
deep lesions of the insula and/or the basal ganglia). Arevalo et al. (2007) studied the
processing of words and pictures representing actions and objects in aphasic patients
and normal controls. Using a manipulability dimension (i.e., items that can/cannot
be manipulated and actions that do/do not involve fine hand movements) they found
that aphasic patients displayed relative difficulty with the manipulable items,
while controls displayed the opposite pattern.
Mirror neurons and language evolution

A mirror neuron is a neuron that fires both when an animal performs an action and
also when the animal observes the same action performed by another animal. In
monkeys, the rostral part of the ventral premotor cortex (area F5) contains neurons
that discharge both when the monkey grasps or manipulates objects and when it
observes the experimenter making similar actions. These neurons (mirror neurons)
appear to represent a system that matches observed events to similar, internally
generated actions. In humans, brain activity consistent with mirror neurons has been
found in the premotor cortex and the inferior parietal cortex (Rizzolatti &
Craighero, 2004; Rizzolatti, Fadiga, Gallese, & Fogassi, 1996).
Transcranial magnetic stimulation and positron emission tomography (PET)
experiments suggest that a mirror system for gesture recognition also exists in
humans and includes Brocas area (Rizzolati & Arbib, 1998). The discovery of
mirror neurons in Brocas area might have immense consequences for understanding
brain language organisation and language evolution (Arbib, 2006; Craighero, Metta,
Sandini & Fadiga, 2007). An obvious implication of mirror neurons is that they can
participate in the internal representation of actions.
Genetics of grammar
Recently a milestone observation was made that significantly enlightened our
understanding about the origin of language in general and grammar in particular. In
England it was found that about half the members of three generations of a family,
usually referred to as the KE family, had presented a significant disturbance in
language development. Speech was largely unintelligible; hence they were taught sign
language as children to supplement their speech. Affected members presented severe
disturbances in articulation and other linguistic skills, and broader intellectual and
physical problems. From the genetic point of view the disorder was associated with a
382 ARDILA
mutation in a single autosomal-dominant gene, FOXP2, located on chromosome 7

(Vargha-Khadem et al., 1995). The disorder was not restricted to speech and also
included the following characteristics: impairments in processing words according to
grammatical rules; understanding of more complex sentence structure such as
sentences with embedded relative clauses; inability to form intelligible speech;
impairments in the ability to move the mouth and face not associated with speaking
(relative immobility of the lower face and mouth, particularly the upper lip); and
significantly reduced IQ in the affected compared with the unaffected in both the
verbal and the non-verbal domain.
Further, affected family members presented a pronounced developmental
apraxia. The authors refer to the core deficit as one involving sequential articulation
and orofacial praxis (Vargha-Khadem et al., 1998; Vernes et al., 2006, 2007). PET
study revealed functional abnormalities in both cortical and subcortical motor-
related areas of the frontal lobe, while quantitative analyses of MRI revealed
structural abnormalities in several of these same areas, particularly the caudate
nucleus which was found to be abnormally small bilaterally. An abnormal gene
(SPCH1) in the chromosomal band 7q31 was localised. The genetic mutation or
deletion in this region was proposed to result in marked disruption of speech and
expressive language, including grammar (Fisher, Vargha-Khadem, Watkins,
Monaco, & Pembrey, 1998). Corballis (2004) suggests a possible link between
FOXP2 and the mirror-neuron system observed in the primate homologue of
Brocas area. He proposes that mutation of FOXP2 was the most recent event in the
incorporation of vocalisation into the mirror system, and thus the refinement of
vocal control to the point that it could carry the primary burden of language. Ardila
(2009) has suggested that language developed in two different stages: initially as a
lexical/semantic system, and only later as a grammatical system. Recent genetic
observations (Enard et al., 2002) have suggested that about 10,000100,000 years
ago there may have occurred a mutation in a single autosomal-dominant gene,
FOXP2, located on chromosome 7; this mutation has been proposed to be linked to
grammatical language evolution.
Memory systems for nouns and verbs

Different memory systems for nouns and verbs can be conjectured. As a matter of
fact, two major memory systems are frequently distinguished in contemporary
memory literature: Declarative memory (divided into semantic and episodic or
experiential) and procedural memory (Tulving, Fergus, & Craik, 2004). It has been
suggested that the lexical/semantic and grammar aspects of the language are
subserved by different neuroanatomic brain circuitries and depend on these two
different memory systems (Fabbro, 1999, 2001; Paradis, 2004; Ullman, 2001, 2004).
Whereas lexical/semantic aspects of language depend on declarative semantic
memory (knowledge about the meaning of the words), grammar depends on
procedural memory. Lexical/semantic aspects of the language (language as a
paradigm) are explicitly learned, and represent a type of knowledge we are aware of
(declarative memory). This depends on the retro-Rolandic cortical structures and the
hippocampus. Grammar (language sequences, contiguity) is acquired incidentally.
Procedural memory for grammar supposes implicit language knowledge. Procedural
grammatical learning is related to the execution of sequences of elements (skilled
articulatory acts and grammar) used for speaking and also for syntax. Procedural
memory is related with frontal/subcortical circuitries (Tulving et al., 2004).
Using verbs and using grammar as a single ability

Brocas area damage results in an impairment of grammar and also in an inability to
find verbs. In consequence, brain representation of actions and brain representation
of grammar is coincidental. Using verbs and using grammar depends on the very same
type of brain activity and both are simultaneously disrupted in cases of Brocas
aphasia. It can be conjectured that verbs and grammar appeared closely or
simultaneously in human language evolution; or rather, they are two sides of the
same coin. Furthermore, grammar is associated with oral praxis skills (i.e.,
agrammatism and apraxia of speech appear simultaneously in Brocas aphasia;
and language networks supporting both are overlapped; Borovsky et al., 2007), and
hence it seems reasonable to suppose that all three emerged simultaneously in the
evolution of human language: using verbs, using grammar, and rapidly sequencing
movements with the articulatory organs. The observation of the KE family seems to
support this inter-relationship.
However, there is a foundation condition for using verbsthe ability to
internally represent actions. That is, to internalise (to think) the actions. One can
wonder if mirror neurons may be involved in this internal representation of
actions. Neuroimaging data have shown that interactions involving Brocas area
and other cortical areas are weakest when listening to spoken language
accompanied by meaningful speech-associated gestures (hence, reducing semantic
ambiguity), and strongest when spoken language is accompanied by self-grooming
hand movements or by no hand movements at all, suggesting that Brocas area may
be involved in action recognition (Skipper, Goldin-Meadow, Nusbaum, & Small,
2007).
The obvious question at this point is: Can Brocas aphasic patients internally
represent actions? Or does Brocas aphasia also imply an inability to internally
represent (think) in actions? Even though specific research on this question is not
readily available, the answer seems to be no. Some observations point to a deficit in
internally representing actions in Brocas aphasia. For example Ardila and
Rossellis (1994) patient had to make the concrete action to retrieve the
corresponding verb. Hence, the internal representation of actions and the
understanding/using of verbs seem to be closely related abilities. Noteworthy,
Brocas area becomes active during the mental imagery of grasping movements
(Decety et al., 1994).
Interestingly, Luria (1976) supposed that Brocas aphasia (efferent or kinetic
motor aphasia) was the result of two associated impairments: impairment in the
kinetic melody of speech (kinetic apraxia), and disturbances in inner speech. Inner
speech is condensed and abbreviated internal speech, presumably responsible for
self-awareness, problem solving, and thinking. According to Vygotsky (1978, 1986)
inner speech follows a specific developmental sequence: first there is social speech
(communication with other people), then comes egocentric speech (that disappears
around the age of 7 or 8), and finally inner speech is formed. Inner speech represents
a major instrument of metacognition and self-control. PET studies have associated
the neural correlates of inner speech with activity of Brocas area (McGuire et al.,
1996). This is an indirect demonstration that patients with Brocas aphasia may
384 ARDILA
present with an impairment in inner speech and thinking in actions. Some

investigators (e.g., Levine, Calvanio, & Popovics, 1982) have overtly pointed out
that damage of the left inferior frontal area is associated with a loss of inner speech.
As an interesting clinical observation, a patient with a paroxysmal Brocas aphasia
reported that frequently she had the seizures when she was alone; on these occasions
she knew she was having an epileptic seizure because her mind became blank, she
had no idea in mind, and was sure that, if she had attempted to speak, she would be
unable to produce anything (Ardila & Lopez, 1988).
The question of variability in aphasic syndromes

For some time reports have been published describing unusual and unexpected
localisation of aphasia syndromes (e.g., Basso, Lecours, Moraschini, & Vanier, 1985;
Vignolo, Boccardi & Caverni, 1986). Vignolo et al. (1986), for instance, studied 37
stroke patients with global aphasia: it was found that while 22 patients presented the
expected large lesions including Brocas and Wernickes areas, 8 had anterior lesions
sparing Wernickes area, 3 had posterior lesions sparing Brocas area, and 4 had deep
lesions centred on the insula and lenticular nucleus. It is noteworthy that all the
patients were evaluated within 2 months after the aphasia onset.
In other aphasic syndromes different localisation of the pathology has been
reported. For instance, in TSA two rather different localisations of brain pathology
have been described (e.g., Berthier, 2001). The brain lesion responsible for
conduction aphasia can be not only left parietal but also insular (Damasio &
Damasio, 1980).
These unusual cases are obviously intriguing and deserve special attention.
Variability in the clinical manifestations of aphasia is nonetheless limited, and
aphasic syndromes are usually observed in cases of relatively specific brain
pathology: damage in BA 37 results in anomia associated with semantic
paraphasias; damage in or around BA 41 and 42 results in phoneme discrimination
impairments, etc. Lesion location is the main determinant of aphasic disorders even
at the acute stage. Kreisler et al. (2000) examined 107 stroke patients with aphasia
using a standardised aphasia battery and MRI, and concluded that most clinical
anatomical correlations support the classic anatomy of aphasia. How can these
unusual cases be explained? We do not have a clear and direct answer to this
question, but it can be speculated that at least three different factors might account
for this variability:
1. Aphasic symptomatology during the acute stage is more extended and complex
due to diaschisis (Von Monakow, 1914). Unusual aphasic syndromes are
generally observed during the acute stage of brain pathology (e.g., Vignolo et al.,
1986), Jargon and logorrhoea in TSA, for example, are only observed during the
acute stage (Dronkers & Larsen, 2001; Kertesz, 1979), suggesting that other
brain areas are also involved. The unusual aphasia manifestations may be the
result not only of focal brain pathology but also of more extended brain
dysfunction due to diaschisis.
2. The brain is a dynamic system and different areas are heavily interconnected.
Recent neuroimaging techniques, particularly fMRI, have shown that there are
some brain functions (i.e., simple) that can be localised into single brain areas
whereas there are others (i.e., complex) that cannot (e.g., Ardila & Bernal, 2007;
Cabeza & Nyberg, 2000). Complex functions depend on the simultaneous

activity of different brain areas, or more exactly, brain circuits. For example,
TSA is observed in cases of parietal-temporal-occipital damage, but also in
cases of frontal damage (Berthier, 2001), suggesting that both areas participate
in a single circuit or brain system involved in naming. Another example:
agrammatism and decreased verbal fluency are typically found in cases of
Brocas aphasia, but they can also be observed in cerebellar pathology
(Akshoomoff, Courchesne, Press, & Iragui, 1992; Daum & Ackermann, 1995;
Silveri, Leggio, & Molinari, 1994). A neural cerebellar-frontal loop affecting
cognitive, especially linguistic, functions has been suggested. It has been
hypothesised that the cerebellum contributes to cognitive processing, particu-
larly the processing of linguistic information, because the cerebellum has
anatomical connections to the cerebral cortex, through which it can affect
language function (Ackermann, Mathiak, & Riecker, 2007; Hernandez-Muela,
Mulas, & Mattos, 2005; Leiner, Leiner, & Dow, 1993). The neodentate
cerebellar area projects via the thalamus to the frontal lobe, especially the
prefrontal areas and Brocas area.

3. Individual differences in brain organisation of cognition in general, and
language in particular, have barely been examined (Telzrow, 1985). It seems to
make sense that individual linguistic experiences may affect the brain
organisation of cognition in general and language in particular. We know
well, for example, that developmental and acquired reading disturbances
depend on the idiosyncrasies of the reading system (Karanth, 2003; Paulesu et
al., 2001). We also know that over-practising a specific ability may result in
changing (enlarging) its cortical representation (e.g., Elbert, Pantev,
Wienbruch, Rockstroh, & Taub, 1995). We also know that literacy can
change the clinical manifestations of aphasia (e.g., Goldblum, & Matute, 1986;
Lecours et al., 1987). Linguistic experiences can be very different in an illiterate
person and in a writer; in a monolingual and in a polyglot; in an Inuit and in a
New Yorker. There are also very specific individual linguistic experiences,
depending on the professional activity, the individuals personal history, etc.
We simply do not know well how these differences in linguistic experiences can
affect the brain organisation of language (see, for example, Burton, Noll, &
Small, 2001).
CONCLUSION
From the above analysis it is evident that the term aphasia has been used to
refer both to primary language disturbances, affecting the language system itself
(phonology, lexicon, semantics, grammar), and to other impairments not affecting
the language system itself, but affecting some abilities required for using language.
A major distinction in aphasia can be established between primary language
disturbances (central aphasias), and secondary language disturbances resulting
from peripheral impairments (secondary or peripheral aphasias). Sometimes
language is not impaired, but the patient cannot use it appropriately because of
executive control impairments (dysexecutive aphasia). Table 5 presents a proposed
interpretation and classification of aphasia syndromes.
386 ARDILA
TABLE 5
Proposed interpretation and classification of aphasia syndromes
Type Impairment
Primary (central) aphasias Language system impaired

Wernicke-type aphasia (fluent aphasia) Phonological level
Lexical level
Semantic level
Broca-type aphasia(non-fluent aphasia) Sequencing expressive elements at syntactic and
phonetic level
Secondary (peripheral) aphasias Mechanisms of production impaired

Conduction aphasia Disconnection (or segmentary ideomotora verbal
apraxia)
SMA aphasia To initiate and maintain voluntary speech production
Dysexecutive aphasia Language executive control impaired

Extra-Sylvian (transcortical) motor aphasia Executive control of language
ACKNOWLEDGEMENTS
My sincere gratitude to Dr Hugh W. Buckingham for his most valuable suggestions
on a previous version of this paper. Many thanks to Dr Jean Mead for her invaluable
editorial support and many suggestions for this paper. Thanks to Dr Byron Bernal
for his observations. My sincere recognition to Dr Chris Code for his most
important comments on this paper.
REFERENCES
Ackermann, H., Mathiak, K., & Riecker, A. (2007). The contribution of the cerebellum to
speech production and speech perception: Clinical and functional imaging data. Cerebellum, 6,
202213.
Akshoomoff, N. A., Courchesne, E., Press, G. A., & Iragui, V. (1992). A new role of the cerebellum in
cognitive operations. Behavioral Neurosciences, 106, 731738.
Albert, M. L., Goodglass, H., Helm, N. A., Rubers, A. B., & Alexander, M. P. (1981). Clinical aspects of
dysphasia. New York: Springer-Verlag.
Alexander, M. P. (2006). Impairments of procedures for implementing complex language are due to
disruption of frontal attention processes. Journal of the International Neuropsychological Society, 12,
236247.
Alexander, M. P., & Benson, D. F. (1991). The aphasia and related disturbances. In R. J. Joynt (Ed.),
Clinical neurology (pp. 158). Philadelphia: Lippincott.
Alexander, M. P., Hiltbrunner, B., & Fischer, R. S. (1989). Distributed anatomy of transcortical sensory
aphasia. Archives of Neurology, 46, 885892.
Alexander, M. P., Naeser, M. A., & Palumbo, C. L. (1987). Correlations of subcortical CT lesion sites and
aplasia profiles. Brain, 110, 961991.
Alexander, M. P., Naeser, M. A., & Palumbo, C. (1990). Brocas area aphasias: Aphasia after lesions
including the frontal operculum. Neurology, 40, 353362.
Alexander, M. P., Stuss, D. T., Picton, T., Shallice, T., & Gillingham, S. (2007). Regional frontal injuries
cause distinct impairments in cognitive control. Neurology, 68, 15151523.
Arbib, M. A. (2006). Aphasia, apraxia and the evolution of the language-ready brain. Aphasiology, 20,
11251155.
Ardila, A. (1992). Phonological transformations in conduction aphasia. Journal of Psycholinguistic

Research, 21, 473484.
Ardila, A. (1993). Toward a model of phoneme perception. International Journal of Neuroscience, 70, 112.
Ardila, A. (2006a). Las afasias [The aphasias]. Retrieved 11 April 2007 from: http://www.aphasia.org/
libroespanol.php.
Ardila, A. (2006b). Orgenes del lenguaje: Un analisis desde las afasias [The origins of language: An
analysis from the aphasia perspective]. Revista de Neurologia, 43, 690698.
Ardila, A. (2009). Origins of the language: Correlation between brain evolution and language
development. In S. M. Platek & T. K. Shackelford (Eds.), Foundations of evolutionary cognitive
neuroscience (pp. 153174). New York: Cambridge University Press.
Ardila, A., & Bernal, B. (2007). What can be localised in the brain? Towards a factor theory on brain
organisation of cognition. International Journal of Neurosciences, 117, 93569.
Ardila, A., & Lopez, M. V. (1984). Transcortical motor aphasia: One or two aphasias? Brain and
Language, 22, 350353.
Ardila, A., & Lopez, M. V. (1988). Paroxysmal aphasias. Epilepsia, 29, 630634.
Ardila, A., Montanes, P., Caro, C., Delgado, R., & Buckingham, H. W. (1989). Phonological
transformations in Spanish-speaking aphasics. Journal of Psycholinguistic Research, 18, 163180.
Ardila, A., & Rosselli, M. (1990). Conduction aphasia and verbal apraxia. Journal of Neurolinguistics, 5,
114.
Ardila, A., & Rosselli, M. (1992). Repetition in aphasia. Journal of Neurolinguistics, 7, 103133.
Ardila, A., & Rosselli, M. (1994). Averbia as a selective naming disorder: A single case report. Jounral of
Psycholinguistic Research, 23, 139148.
Ardila, A., Rosselli, M., & Ardila, O. (1988). Foreign accent: An aphasic epiphenomenon? Aphasiology, 2,
493499.
Ardila, A., & Surloff, C. (2006). Executive dysfunction. In S. Gilman (Ed.) Medlink Neurology. San Diego,
CA: Arbor Publishing.
Arevalo, A., Perani, D., Cappa, S. F., Butler, A., Bates, E., & Dronkers, N. (2007). Action and object
processing in aphasia: from nouns and verbs to the effect of manipulability. Brain and Language, 100,
7994.
Bartha, L., & Benke, T. (2003). Acute conduction aphasia: An analysis of 20 cases. Brain and Language,
85, 93108.
Basso, A. (2003). Aphasia and its therapy. New York: Oxford University Press.
Basso, A., Della Sala, S., & Farabola, M. (1987). Aphasia arising from purely deep lesions. Cortex, 23,
2944.
Basso, A., Lecours, A. R., Moraschini, S., & Vanier, M. (1985). Anatomoclinical correlations of the
aphasias as defined through computerized tomography: Exceptions. Brain and Language, 26, 201229.
Bastian, D. C. (1898). Aphasia and other speech defects. London: H. K. Lewis.
Benabdeljlil, M., El Alaoui Farias, M., Kissani, N., Adi, S., Laaouina, Z., & Jiddane, M. et al. (2001).
Neuropsychological disorders after bithalamic infarct due to deep venous thrombosis. Revue de
Neurologie, 157, 6267.
Benson, D. F. (1979). Aphasia, alexia and agraphia. New York: Churchill Livingstone.
Benson, D. F., & Ardila, A. (1994). Conduction aphasia: A syndrome of language network disruption. In
H. Kirshner (Ed.), Handbook of speech and language disorders. New York: Mercel Dekker Inc.
Benson, D. F., & Ardila, A. (1996). Aphasia: A clinical perspective. New York: Oxford University Press.
Benson, D. F., & Geschwind, N. (1971). Aphasia and related cortical disturbances. In A. B. Baker & L. H.
Baker (Eds.), Clinical neurology. Philadelphia: Harper & Row.
Benson, D. F., & Geschwind, N. (1985). The aphasia and related disturbances. In A. B. Baker & R. J.
Joynt (Eds.), Clinical neurology. Philadelphia: Harper & Row.
Benson, D. F., Sheretaman, W. A., Bouchard, R., Segarra, J. M., Price, D., & Geschwind, N. (1973).
Conduction aphasia: A clinicopathological study. Archives of Neurology, 28, 339346.
Berndt, R. S., & Caramazza, A. (1980). A redefinition of the syndrome of Brocas aphasia: Implications
for a neuropsychological model of language. Applied Psycholinguistics, 1, 225278.
Berthier, M. (1999). Transcortical aphasias. Hove, UK: Psychology Press.
Berthier, M. L. (2001). Unexpected brainlanguage relationships in aphasia: Evidence from transcortical
sensory aphasia associated with frontal lobe lesions Aphasiology, 15, 99130.
Berthier, M. L., Starkstein, S. E., Leiguarda, R., Ruiz, A., Mayberg, H. S., & Wagner, H. et al. (1991).
Transcortical aphasia. Importance of the nonspeech dominant hemisphere in language repetition.
Brain, 114, 14091427.
388 ARDILA
Bhatnagar, S. C., Mandybur, G. T., Buckingham, H. W., & Andy, O. J. (2000). Language representation
in the human brain: Evidence from cortical mapping. Brain and Language, 74, 238259.
Boatman, D., Gordon, B., Hart, J., Selnes, O., Miglioretti, D., & Lenz, F. (2000). Transcortical sensory
aphasia: Revisited and revised. Brain, 123, 16341642.
Borovsky, A., Saygin, A. P., Bates, E., & Dronkers, N. (2007). Lesion correlates of conversational speech
production deficits. Neuropsychologia, 45, 25252533.
Brickner, R. M. (1940). A human cortical area producing repetitive phenomena when stimulated. Journal
of Neurophysiology, 3, 128130.
Broca, P. (1863). Localisation des fonctions cerebrales: Siege du langage articule. Bulletin de la Societe
dAnthropologie, 4, 200203.
Brown, J. M. (1975). The problem of repetition: A case study of conduction aphasia and the isolation
syndrome. Cortex, 11, 3752.
Brown, J. W. (1972). Aphasia, agnosia and apraxia. Springfield, IL: Thomas.
Brown, R. (1973). A first language: The early stages. Boston: Harvard University Press.
Buckingham, H. W. (1989). Mechanisms underlying aphasic transformations. In A. Ardila & F. Ostrosky
(Eds.), Brain organisation of language and cognitive processes. New York: Plenum.
Buckingham, H. W. (1998). Explanations for the concept of apraxia of speech. In M. T. Sarno (Ed.),
Acquired aphasia. (3rd ed., pp. 269307). San Diego, CA: Academic Press.
Burton, M. W., Noll, D. C., & Small, S. L. (2001). The anatomy of auditory word processing: individual
variability. Brain and Language, 77, 119131.

Butterworth, B., & Warrington, E. (1995). Two routes to repetition: Evidence from a case of deep
dysphasia Neurocase, 1, 5566.
Cabeza, R., & Nyberg, L. (2000). Imaging cognition II: An empirical review of 275 PET and fMRI studies.
Journal of Cognitive Neurosciences, 12, 147.
Cao, Y., George, K. P., Ewing, J. R., Vikingstad, E. M., & Johnson, A. F. (1998). Neuroimaging of
language and aphasia after stroke. Journal of Cerebrovascular Diseases, 7, 230233.
Caplan, D. (2006). Aphasic deficits in syntactic processing. Cortex, 42, 797804.
Caplan, D., Alpert, A., Waters, G., & Olivieri, A. (2000). Activation of Brocas area by syntactic
processing under conditions of concurrent articulation. Human Brain Mapping, 9, 6571.
Cappa, S. F., Cavallotti, G., Guidotti, M., Papagno, C., & Vignolo, L. A. (1983). Subcortical aphasia:
Two clinical-CT scan correlation studies. Cortex, 19, 227241.
Cappa, S. E., & Vignolo, L. A. (1979). Transcortical features of aphasia following left thalamic
hemorrhage. Cortex, 15, 121130.
Carthy, R. A., & Warrington, E. K. (1984). A two-route model of speech production: Evidence from
aphasia. Brain, 107, 463485.
Catani, M., Jones, D. K., & Ffytche, D. H. (2005). Perisylvian language networks of the human brain.
Annals of Neurology, 57, 816.
Choi, J. Y., Lee, K. H., Na, D. L., Byun, H. S., Lee, S. J., & Kim, H. et al. (2007). Subcortical aphasia
after striatocapsular infarction: Quantitative analysis of brain perfusion SPECT using statistical
parametric mapping and a statistical probabilistic anatomic map. Journal of Nuclear Medicine, 48,
194200.
Code, C., & Ball, M. (1982). Fricative production in Brocas aphasia. Journal of Phonetics, 10, 325331.
Coltheart, M. (1980). Deep dyslexia: A review of the syndrome. In M. Coltheart, K. Patterson, & J.
Marshall (Eds.), Deep dyslexia. London: Routledge & Kegan Paul.
Corballis, M. C. (2004). FOXP2 and the mirror system. Trends in Cognitive Sciences, 8, 9596.
Coslett, H. B., Roeltgen, D. P., Gonzales Rothi, L., & Heilman, K. M. (1987). Transcortical sensory
aphasia: evidence for subtypes. Brain and Language, 32, 362378.
Craighero, L., Metta, G., Sandini, G., & Fadiga, L. (2007). The mirror-neurons system: data and models.
Progress in Brain Research, 164, 3959.
Craver, C. F., & Small, S. L. (1997). Subcortical aphasia and the problem of attributing functional
responsibility to parts of distributed brain processes. Brain and Language, 58, 427435.
Crosson, B. (1999). Subcortical mechanisms in language: Lexical-semantic mechanisms and the thalamus.
Brain and Cognition, 40, 414438.
Crosson, B., Moberg, P. J., Boone, J. R., Rothi, L. J., & Raymer, A. (1997). Category-specific naming
deficit for medical terms after dominant thalamic/capsular hemorrhage. Brain and Language, 60,
407442.
Cummings, J. L. (1993). Frontal-subcortical circuits and human behavior. Archives of Neurology, 50,
87380.
Damasio, A. R., & Tranel, D. (1993). Nouns and verbs are retrieved with differently distributed neural
systems. Proceedings of the National Academy of Sciences, 90, 49574960.
Damasio, H. (1991). Neuroanatomical correlates of the aphasias. In M. Taylor Sarno (Ed.), Acquired
aphasia (pp. 4571). New York: Academic Press.
Damasio, H., & Damasio, A. (1980). The anatomical basis of conduction aphasia. Brain, 103, 337350.
Damasio, H., Grabowski, T. J., Tranel, D., Ponto, L. L., Hichwa, R. D., & Damasio, A. R. (2001). Neural
correlates of naming actions and of naming spatial relations. Neuroimage, 13, 10531064.
Damasio, H., Tranel, D., Grabowski, T., Adolphs, R., & Damasio, A. (2004). Neural systems behind word
and concept retrieval. Cognition, 92, 179222.
Daum, I., & Ackermann, N. (1995). Cerebellar contributions to cognition. Behavior and Brain Research,
67, 201210.
Decety, J., Perani, D., Jeannerod, M., Bettinard, V., Tadardy, B., & Woods, R. et al. (1994). Mapping
motor representations with positron emission tomography. Nature, 371, 600602.
Dehaene, S., Tzourio, N., Frak, V., Raynaud, L., Cohen, L., & Mehler, J. et al. (1996). Cerebral
activations during number multiplication and comparison: A PET study. Neuropsychologia, 34,
10971106.
Demonet, J. F. (1997). Subcortical aphasia(s): A controversial and promising topic. Brain and Language,
58, 410417.
DEsposito, M., & Alexander, M. P. (1995). Subcortical aphasia: Distinct profiles following left putaminal
hemorrhage. Neurology, 45, 3841.
De Witte, L., Wilssens, I., Engelborghs, S., De Deyn, P. P., & Marien, P. (2006). Impairment of syntax and
lexical semantics in a patient with bilateral paramedian thalamic infarction. Brain and Language, 96,
6977.
Dronkers, N. F. (1996). A new brain region for coordinating speech articulation. Nature, 384, 159161.
Dronkers, N. F., & Larsen, J. (2001). Neuroanatomy of the classical aphasia syndromes. In R. D. Berndt
(Ed.), Handbook of neuropsychology, vol 3: Language and aphasia (2nd ed., pp. 1930). Amsterdam:
Elsevier.
Dronkers, N. F., Wilkins, D. P., Van Valin, R. D., Redfern, B. B., & Jaeger, J. J. (2004). Lesion analysis of
the brain areas involved in language comprehension. Cognition, 92, 14577.
Elbert, T., Pantev, C., Wienbruch, C., Rockstroh, B., & Taub, E. (1995). Increased cortical representation
of the fingers of the left hand in string players. Science, 270, 305307.
Ellis, A. W., & Young, A. W. (1988). Cognitive neuropsychology. Hove, UK: Lawrence Erlbaum
Associates Ltd.
Enard, W., Przeworski, M., Fisher, S. E., Lai, C. S., Wiebe, V., & Kitano, T. et al. (2002). Molecular
evolution of FOXP2, a gene involved in speech and language. Nature, 418, 86972.
Erdler, M., Beisteiner, R., Mayer, D., Kaindl, T., Edward, V., & Windischberger, C. et al. (2000).
Supplementary motor area activation preceding voluntary movement is detectable with a whole-scalp
magnetoencephalography system. Neuroimage, 11, 697670.
Fabbro, F. (1999). The neurolinguistics of bilingualism: An introduction. Hove, UK: Psychology Press.
Fabbro, F. (2001). The bilingual brain, cerebral representation of languages. Brain and Language, 79,
211222.
Fadiga, L., Craighero, L., & Roy, A. (2006). Brocas region: A speech area? In Y. Grodzinky & K.
Amunts (Eds.), Brocas region (pp. 137152). New York: Oxford University Press.
Fink, G. R., Manjaly, Z. M., Stephan, K. E., Gurd, J. M., Zilles, K., & Amunts, K. et al. (2006). A role for
Brocas area beyond language processing: Evidence from neuropsychology and fMRI. In Y.
Grodzinky & K. Amunts (Eds.), Brocas region (pp. 254268). New York: Oxford University Press.
Fisher, S. E., Vargha-Khadem, F., Watkins, K. E., Monaco, A. P., & Pembrey, M. E. (1998). Localisation
of a gene implicated in a severe speech and language disorder. Nature Genetics, 18, 168170.
Foundas, A. L., Eure, K. F., Luevano, L. F., & Weinberger, D. R. (1998). MRI asymmetries of Brocas
area: The pars triangularis and pars opercularis. Brain and Language, 64, 282296.
Freud, S. (1891/1973). Las afasias. Buenos Aires: Ediciones Nueva Vision.
Fridriksson, J., Nettles, C., Davis, M., Morrow, L., & Montgomery, A. (2006). Functional
communication and executive function in aphasia. Clinical Linguistics & Phonetics, 20, 401410.
Fried, I., Katz, A., McCarthy, G., Sass, K. J., Williamson, P., & Spencer, S. S. et al. (1991). Functional
organisation of human supplementary motor cortex studied by electrical stimulation. Journal of
Neurosciences, 11, 36563666.
390 ARDILA
Friedman, R. B. (1988). Acquired alexia. In F. Boller, J. Grafman, G. Rizzolati, & H. Goodglass (Eds.),
Handbook of neuropsychology, vol 1. New York: Elsevier.
Friedman, R. B., & Albert, M. L. (1985). Alexia. In K. M. Heilman & E. Valenstein (Eds.), Clinical
neuropsychology (2nd ed.). New York: Oxford University Press.
Geschwind, N. (1965). Disconnection syndromes in animals and man. Brain, 88, 237294.
Gold, M., Nadeau, S. E., Jacobs, D. H., Adair, J. C., Rothi, L. J., & Heilman, K. M. (1997). Adynamic
aphasia: A transcortical motor aphasia with defective semantic strategy formation. Brain and
Language, 57, 374393.
Goldblum, M. C., & Matute, E. (1986). Are illiterate people deep dyslexics? Journal of Neurolinguistics, 2,
103114.
Goldstein, K. (1948). Language and language disturbances. New York: Grune & Stratton.
Goodglass, H. (1993). Understanding aphasia. New York: Academic Press.
Goodglass, H., & Kaplan, E. (1972). The assessment of aphasia and related disorders. Philadelphia: Lea &
Febiger.
Goodglass, H., & Kaplan, E. (1979). Evaluacion de la Afasia y de Transtornos Similares. Buenos Aires:
Editorial Medica Panamericana.
Goodglass, H., Wingfield, A., Hyde, M. R., & Theurkauf, J. C. (1986). Category specific dissociations in
naming and recognition by aphasic patients. Cortex, 22, 87102.
Greenwald, M. (2001). Acquired reading disorders. In F. Boller & J. Grafman (Eds.), Handbook of
neuropsychology. Vol 3, Language and aphasia (2nd ed.). Amsterdam: Elsevier.

Grodzinsky, Y. (2000). The neurology of syntax: Language use without Brocas area. Behavioral and Brain
Sciences, 23, 121.
Grodzinsky, Y. (2006). The language faculty, Brocas region, and the mirror system. Cortex, 42, 464468.
Grodzinky, Y., & Amunts, K. (Eds.). (2006). Brocas region. New York: Oxford University Press.
Grodzinsky, Y., & Friederici, A. D. (2006). Neuroimaging of syntax and syntactic processing. Current
Opinions in Neurobiology, 16, 240246.
Hagoort, P. (2005). Brocas complex as the unification of space for language. In A. Cutler (Ed.), Twenty-
first century psycholinguistics: Four cornerstones (pp. 157172). Mahwah, NJ: Lawrence Erlbaum
Associates Inc.
Hagoort, P. (2006). On Broca, brain, and binding. In Y. Grodzinky & K. Amunts (Eds.), Brocas region
(pp. 242253). New York: Oxford University Press.
Hanakawa, T., Hondam, M., Sawamoto, N., Okada, T., Yonekura, Y., & Fukuyama, H. et al. (2002). The
role of rostral Brodmann area 6 in mental-operation tasks: An integrative neuroimaging approach.
Cerebral Cortex, 12, 11571170.
Harris, D. M., & Kay, J. (1995). Selective impairment of the retrieval of peoples names: A case of category
specificity. Cortex, 31, 575582.
Haverkort, M. (2005). Linguistic representation and language use in aphasia. In A. Cutler (Ed.), Twenty-
first century psycholinguistics: Four cornerstones (pp. 5768). Mahwah, NJ: Lawrence Erlbaum
Associates Inc.
Head, H. (1926). Aphasia and kindred disorders of speech. London: Cambridge University Press.
Hecaen, H. (1972). Introduction a la neuropsychologie. Paris: Larousse.
Hecaen, H., & Albert, M. L. (1978). Human neuropsychology. New York: Wiley.
Hernandez-Muela, S., Mulas, F., & Mattos, L. (2005). [The contribution of the cerebellum to cognitive
processes]. Revista de Neurologia, 40, 5764.
Hillis, A. E. (2007). Aphasia: Progress in the last quarter of a century. Neurology, 69, 200213.
Hillis, A. E., Barker, P. B., Wityk, R. J., Aldrich, E. M., Restrepo, L., & Breese, E. L. et al. (2004a).
Variability in subcortical aphasia is due to variable sites of cortical hypoperfusion. Brain and
Language, 89, 524530.
Hillis, A. E., Work, M., Barker, P. B., Jacobs, M. A., Breese, E. L., & Maurer, K. (2004b). Re-examining
the brain regions crucial for orchestrating speech articulation. Brain, 127, 14791487.
Hoff, E. (2003). Language development (3rd ed.). New York: Wadsworth.
Huber, W., Poeck, K., & Willmes, K. (1984). The Aachen Aphasia Test. Advances in Neurology, 42,
291303.
Jakobson, R. (1964). Toward a linguistic typology of aphasic impairments. In A. V. S. DeReuck & M.
OConnor (Eds.), Disorders of language. New York: Little & Brown.
Jakobson, R. (1968). Child language, aphasia, and phonological universals. New York: Mouton.
Jakobson, R. (1971). Studies on child language and aphasia. New York: Mouton.
Jakobson, R., & Halle, M. (1956). Two aspects of language and two types of aphasic disturbances. New
York: Mouton.
Jefferies, E., Sage, K., & Ralph, M. A. (2007). Do deep dyslexia, dysphasia and dysgraphia share a
common phonological impairment? Neuropsychologia, 45, 15531570.
Jonas, S. (1981). The supplementary motor region and speech emission. Journal of Communication
Disorders, 14, 349373.
Jonides, J., Smith, E. E., Koeppe, R. A., Awh, E., Minoshima, S., & Mintun, M. A. (1993). Spatial
working memory in humans as revealed by PET. Nature, 363, 623625.
Kandel, E. R., Schwartz, J. H., & Jessell, T. M. (1995). Essentials of neural science and behavior. Norwalk,
CT: Appleton & Lange.
Karanth, P. (2003). Cross-linguistic study of acquired reading disorders. New York: Kluwer/Plenum.
Kreisler, A., Godefroy, O., Delmaire, C., Debachy, B., Leclercq, M., & Pruvo, J. P. et al. (2000). The
anatomy of aphasia revisited Neurology, 54, 11171123.
Kertesz, A. (1979). Aphasia and associated disorders. New York: Grune & Stratton.
Kertesz, A. (1982). The Western Aphasia Battery. New York: Grune & Stratton.
Kertesz, A. (1985). Aphasia. In J. A. M. Frederiks (Ed.), Handbook of clinical neurology, vol 45: Clinical
neuropsychology. Amsterdam: Elsevier.
Kertesz, A., Sheppard, A., & MacKenzie, R. (1982). Localisation in transcortical sensory aphasia.
Archives of Neurology, 39, 475478.
Kleist, K. (1934). Gehirnpathologie. Leipzig: Barth.

Kohn, S. (Ed.). (1992). Conduction aphasia. Mahwah, NJ: Lawrence Erlbaum Associates Inc.
Lebrun, Y. (1987). Anosognosia in aphasia. Cortex, 23, 251263.
Lecours, A. R., Lhermitte, F., & Bryans, B. (1983). Aphasiology. London: Baillere-Tindall.
Lecours, A. R., Osborn, E., Travies, L., Rouillon, F., & Lavalle-Huyng, G. (1981). Jargons. In J. Brown
(Ed.), Jargonaphasia (pp. 938). New York: Academic Press.
Lecours, A. R., & Rouillon, F. (1976). Neurolinguistic analysis of jargonaphasia and jargonagraphia. In
H. Whitaker & H. A. Whitaker (Eds.), Studies in neurolinguistics (vol. 2, pp. 95144). New York:
Academic Press.
Lecours, R. L., Mehler, J., Parente, M. A., Caldeira, A., Cary, L., & Castro, M. J. et al. (1987). Illiteracy
and brain damage I: Aphasia testing in culturally contrasted populations (control subjects).
Neuropsychologia, 25, 231245.
Lee, A., Kannan, V., & Hillis, A. E. (2006). The contribution of neuroimaging to the study of language
and aphasia. Neuropsychology Review, 16, 171183.
Lehmann, W. P. (1974). Proto Indoeuropean syntax. Houston, TX: University of Texas Press.
Leiner, H. C., Leiner, A. L., & Dow, R. S. (1993). Cognitive and language functions of the human
cerebellum. Trends in Neurosciences, 6, 444447.
Levine, D. N., Calvanio, R., & Popovics, A. (1982). Language in the absence of inner speech.
Neuropsychologia, 20, 391409.
Liang, C. L., Chang, H. W., Lu, K., Lee, T. C., Liliang, P. C., & Lu, C. H. et al. (2001). Early prediction of
aphasia outcome in left basal ganglia hemorrhage. Acta Neurologica Scandinavica, 103, 148152.
Lindenberg, R., Fangerau, H., & Seitz, R. J. (2007). Brocas area as a collective term? Brain and
Language, 102, 2229.
Lichter, D. G., & Cummings, J. L. (2001). Frontal-subcortical circuits in psychiatric and neurological
disorders. New York: Guilford Press.
Lichtheim, L. (1885). On aphasia. Brain, 7, 433484.
Luria, A. R. (1966). Human brain and psychological processes. New York: Harper & Row.
Luria, A. R. (1970). Traumatic aphasia: Its syndromes, psychology, and treatment. New York: Mouton.
Luria, A. R. (1972/1983). Sobre las dos formas basicas del alteraciones afasicas en el lenguaje [On the two
basic forms of aphasic disturbances]. In A. Ardila (Ed.), Psicobiologa del Lenguaje. Mexico: Trillas.
Luria, A. R. (1974). The working brain. New York: Basic Books.
Luria, A. R. (1976). Basic problems of neurolinguistics. New York: Mouton.
Luria, A. R. (1980). Higher cortical functions in man (2nd ed.). New York: Basic Books.
Luzzatti, C., Aggujaro, S., & Crepaldi, D. (2006). Verb-noun double dissociation in aphasia: theoretical
and neuroanatomical foundations. Cortex, 42, 875883.
Lyons, F., Hanley, J. R., & Kay, J. (2002). Anomia for common names and geographical names with
preserved retrieval of names of people: A semantic memory disorder. Cortex, 38, 2335.
Maeshima, S., Osawa, A., Nakayama, Y., & Miki, J. (2004). Transcortical sensory aphasia following
infarction in the left frontal lobe. European Neurology, 52, 125128.
392 ARDILA
Marshall, J. C., & Newcombe, F. (1966). Syntactic and semantic errors in paralexia. Neuropsychologia, 4,
169176.
Marshall, J. C., & Newcombe, F. (1973). Patterns of paralexias: A psycholinguistic approach. Journal of
McCarthy, R., & Warrington, E. (1984). A two-route model of speech production. Brain, 107, 463485.
McCarthy, R. A., & Warrington, E. (2001). Repeating without semantics: Surface dysphasia? Neurocase,
7, 7787.
McGuire, P. K., Silbersweig, D. A., Murray, R. M., David, A. S., Frackowiak, R. S. J., & Frith, C. D. (1996).
Functional anatomy of inner speech and auditory verbal imagery. Psychological Medicine, 26, 3938.
Mega, M. S., & Alexander, M. P. (1994). Subcortical aphasia: The core profile of capsulostriatal
infarction. Neurology, 44, 18241829.
Mellet, E., Tzourio, N., Crivello, F., Joliot, M., Denis, M., & Mazoyer, B. (1996). Functional anatomy of
spatial mental imagery generated from verbal instructions. Journal of Neuroscience, 16, 65046512.
Mesulam, M. M. (2002). The human frontal lobes: Transcending the default mode through contingent
encoding. In D. T. Stuss & R. T. Knight (Eds.), Principles of frontal lobe function (pp. 831). New
York: Oxford.
Metz-Lutz, M. N., Namer, I. J., Gounot, D., Kleitz, C., Armspach, J. P., & Kehrli, P. (2000). Language
functional neuroimaging changes following focal left thalamic infarction. Neuroreport, 11, 29072912.
Michel, F., & Andreewsky, E. (1983). Deep dysphasia: An auditory analog of deep dyslexia in the auditory
modality. Brain and Language, 18, 212223.

Mohr, J. P., Pessin, M. S., Finkelstein, S., Funkenstein, H. H., Duncan, G. W., & Davis, K. R. (1978).
Broca aphasia: Pathologic and clinical. Neurology, 28, 311324.
Morton, J., & Patterson, K. (1980). A new attempt of an interpretation, or, and attempt at a new
interpretation. In M. Coltheart, K. Patterson, & J. Marshall (Eds.), Deep dyslexia. London: Routledge
& Kegan Paul.
Nadeau, S. E., & Crosson, B. (1997). Subcortical aphasia. Brain and Language, 58, 355402.
Nasser, M. A., Alexander, M. P., Helm-Estabrooks, N., Levine, H. L., Laughlin, S. A., & Geschwind, N.
(1982). Aphasia with predominantly subcortical lesion sites: Description of three capsular/putaminal
aphasia syndromes. Archives of Neurology, 39, 214.
Novick, J. M., Trueswell, J. C., & Thompson, S. L. (2005). Cognitive control and parsing: Reexamining
the role of Brocas area in sentence comprehension. Cognitive, Affective, & Behavioral Neuroscience, 5,
263281.
Ostrosky-Solis, F., Marcos-Ortega, J., Ardila, A., Rosselli, M., & Palacios, S. (1999). Syntactic
comprehension in Brocas Spanish-speaking aphasics: Null effect of word-order. Aphasiology, 13,
553571.
Otsuki, M., Soma, Y., Aoki, K., Iizuka, O., Koyama, A., & Yoshimura, N. et al. (1998). [Difference
between transcortical sensory aphasia following the left frontal lesion and transcortical sensory aphasia
following the left posterior lesion]. No To Shinkei, 50, 9951002.
Paradis, M. (2004). A neurolinguistic theory of bilingualism. New York: John Benjamins.
Patterson, K. E. (1978). Phonemic dyslexia: Errors of meaning and the meaning of errors. Quarterly
Journal of Experimental Psychology, 30, 587601.
Paulesu, E., Demonet, J. F., Fazio, F., McCrory, E., Chanoine, V., & Brunswick, N. et al. (2001).
Dyslexia: Cultural diversity and biological unity. Science, 291, 21652167.
Paulesu, E., Frith, C. D., & Frackowiak, R. S. (1993). The neural correlates of the verbal component of
working memory. Nature, 362, 342345.
Penfield, W., & Welch, K. (1951). The supplementary motor area of the cerebral cortex: A clinical and
experimental study. AMA Archives of Neurology and Psychiatry, 66, 289317.
Pick, A. (1931). Aphasia. New York: Charles C. Thomas.
Purdy, M. (2002). Executive function ability in persons with aphasia. Aphasiology, 16, 549557.
Raichle, M. E. (1994). Visualizing the mind. Scientific American, 270, 5865.
Rizzolatti, G., & Arbib, M. A. (1998). Language within our grasp. Trends in Neurosciences, 21, 188194.
Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. Annual Review of Neuroscience, 27,
169192.
Rizzolatti, G., Fadiga, L., Gallese, V., & Fogassi, L. (1996). Premotor cortex and the recognition of motor
actions Cognitive Brain Research, 3, 131141.
Roebroek, R. M., Promes, M. M., Korten, J. J., Lormans, A. C., & van der Laan, R. T. (1999).
Transcortical sensory aphasia in a right-handed patient following watershed infarcts in the right
cerebral hemisphere: A 15-month evaluation of another case of crossed aphasia. Brain and Language,
70, 262267.
Rubens, A. B. (1975). Aphasia with infarction in the territory of the anterior cerebral artery. Cortex, 11,
239250.
Rubens, A. B. (1976). Transcortical motor aphasia. In H. Whitaker & H. A. Whitaker (Eds.), Studies in
neurolinguistics, vol 1. New York: Academic Press.
Santi, A., & Grodzinsky, Y. (2007). Working memory and syntax interact in Brocas area. Neuroimage, 37,
817.
Schuell, H., Jenkins, J. J., & Jimenez-Pabon, E. (1964). Aphasia in adults. New York: Harper & Row.
Scott, S. K. (2005). The neurobiology of speech perception. In A. Cutler (Ed.), Twenty-first century
psycholinguistics: Four cornerstones (pp. 141156). Mahwah, NJ: Lawrence Erlbaum Associates Inc.
Schuell, H. (1953). Minnesota Test for the Differential Diagnosis of Aphasia. Minneapolis, MN: University
of Minnesota Press.
Schuell, H. (1973). Differential diagnosis of aphasia with the Minnesota Test. Minneapolis, MN: University
of Minnesota Press.
Sethi, N. K., Burke, L., Torgovnick, J., & Arsura, E. (2007). Transcortical sensory aphasia as a result of
left frontal cortical-subcortical infarction. A case report. European Neurology, 57, 5253.
Shallice, T., & Warrington, E. K. (1980). Single and multiple components single dyslexic syndromes. In M.
Coltheart, K. Patterson, & J. Marshall (Eds.), Deep dyslexia. London: Routledge & Kegan Paul.
Silveri, M. C., Leggio, M. G., & Molinari, M. (1994). The cerebellum contributes to linguistic production:
A case of agrammatic speech following a right cerebellar lesion. Neurology, 44, 20472050.
Simon, S. R., Meunier, M., Piettre, L., Berardi, A. M., Segebarth, C. M., & Boussaoud, D. (2002). Spatial
attention and memory versus motor preparation: Premotor cortex involvement as revealed by fMRI.
Journal of Neurophysioloy, 88, 20472057.
Skipper, J. I., Goldin-Meadow, S., Nusbaum, H. C., & Small, S. L. (2007). Speech-associated gestures,
Brocas area, and the human mirror system. Brain and Language, 101, 260277.
Small, S. L., & Burton, M. W. (2002). Functional magnetic resonance imaging studies of language. Current
Neurology and Neuroscience Reports, 2, 505510.
Small, S. L., Hart, J., Nguyen, T., & Gordon, B. (1995). Distributed representations of semantic
knowledge in the brain. Brain, 118, 441453.
Tanaka, S., Honda, M., & Sadato, N. (2005). Modality-specific cognitive function of medial and lateral
human Brodmann area 6. The Journal of Neuroscience, 25(2), 496501.
Tanji, J., & Shima, K. (1994). Role for supplementary motor area cells in planning several movements
ahead. Nature, 371, 413416.
Taylor-Sarno, M. (1998). Acquired aphasia. New York: Academic Press.
Telzrow, C. F. (1985). The neuropsychology of individual differences: A developmental perspective. New
York: Plenum.
Tesak, J., & Code, C. (2008). Milestones in the history of aphasia: Theories and protagonists. London:
Psychology Press.
Thompson-Schill, S. L. (2005). Dissecting the language organ: A new look at the role of Brocas area in
language processing. In A. Cutler (Ed.), Twenty-first century psycholinguistics: Four cornerstones
(pp. 173190). Mahwah, NJ: Lawrence Erlbaum Associates Inc.
Tranel, D., Martin, C., Damasio, H., Grabowski, T. J., & Hichwa, R. (2005). Effects of noun-verb
homonymy on the neural correlates of naming concrete entities and actions. Brain and Language, 92,
288299.
Tulving, E., Fergus, I., & Craik, M. (2004). The Oxford handbook of memory. New York: Oxford
University Press.
Ullman, M. T. (2001). The declarative/procedural model of lexicon and grammar. Journal of
Ullman, M. T. (2004). Contributions of memory circuits to language: The declarative/ procedural model.
Cognition, 92, 231270.
Vargha-Khadem, F., Watkins, K., Alcock, F., Fletcher, P., & Passingham, R. (1995). Praxic and
nonverbal cognitive deficits in a large family with a genetically transmitted speech and language
disorder. Proceedings of the National Academy of Sciences, 92, 930933.
Vargha-Khadem, F., Watkins, K. E., Price, C. J., Ashburner, J., Alcock, K. J., & Connelly, A. et al.
(1998). Neural basis of an inherited speech and language disorder. Proceedings of the National
Academy of Sciences, 95, 1269512700.
394 ARDILA
Vernes, S. C., Nicod, J., Elahi, F. M., Coventry, J. A., Kenny, N., & Coupe, A. M. et al. (2006).
Functional genetic analysis of mutations implicated in a human speech and language disorder. Human
Molecular Genetics, 15, 31543167.
Vernes, S. C., Spiteri, E., Groszer, M., Taylor, J. M., Davies, K. E., & Geschwind, D. H. et al. (2007).
High-throughput analysis of promoter occupancy reveals direct neural targets of FOXP2, a gene
mutated in speech and language disorders. American Journal of Human Genetics, 81, 12321250.
Vignolo, L. A., Boccardi, E., & Caverni, L. (1986). Unexpected CT-scan findings in global aphasia.
Cortex, 22, 5569.
Vygotsky, L. S. (1934/1978). Mind in society. Cambridge, MA: Harvard University Press.
Vygotsky, L. S. (1986). Thought and language. Cambridge, MA: The MIT Press.
Von Monakow, C. (1914). Die Lokalisation im Grosshirn und der Abbau der Funktion durch kortikale
Herne. Wieswaden: Bergmann.
Warabi, Y., Bandoh, M., Kurisaki, H., Nishio, S., & Hayashi, H. (2006). [Transcortical sensory aphasia
due to extensive infarction of left cerebral hemisphere]. Rinsho Shinkeigaku, 46, 317321.
Warrington, E. K., & Shallice, T. (1984). Category specific semantic impairments. Brain, 107, 829854.
Wernicke, C. (1874). Der Aphasiche Symptomencomplex. Breslau: Cohn & Weigert.
Willmes, K., & Ratajczak, K. (1962). The design and application of a data and method-based system for
the Aachen Aphasia Test. Neuropsychologia, 25, 725733.
Wilshire, C., & Fisher, C. (2004). Phonological dysphasia: A cross-modal phonological impairment
affecting repetition, production, and comprehension. Cognitive Neuropsychology, 21, 187210.

Wilson, S. A. K. (1926). Aphasia. London: Kegan Paul.
Yamada, K., Nagakane, Y., Mizuno, T., Hosomi, A., Nakagawa, M., & Nishimura, T. (2007). MR
tractography depicting damage to the arcuate fasciculus in a patient with conduction aphasia.
Neurology, 68, 789790.
Zahn, R., Huber, W., Drews, E., Erberich, S., Krings, T., & Willmes, K. et al. (2000). Hemispheric
lateralization at different levels of human auditory word processing: A functional magnetic resonance
imaging study. Neuroscience Letters, 287, 195198.
Zahn, R., Huber, W., Drews, E., Specht, K., Kemeny, S., & Reith, W. et al. (2002). Recovery of semantic
word processing in transcortical sensory aphasia: A functional magnetic resonance imaging study.
Neurocase, 8, 376386.
INTERNATIONAL JOURNAL OF PSYCHOLOGY, 2004, 39 (1), 6167
There is not any specific brain area for writing:

From cave-paintings to computers
Alfredo Ardila
Florida International University, Miami, USA
I n this paper it is pointed out that human brain adaptation was accomplished to survive in certain living
conditions that existed long before classical civilizations did. It is argued that there is no brain area
specialized for writing, but rather that writing relies on some basic abilities that existed long before writing was
invented. Pre-writing was initially a visuoconstructive and ideomotor ability, and only later did it become the
language-related ability of writing. It is also emphasized that most of the neuropsychological syndromes,
Downloaded By: [Florida International University] At: 21:00 3 March 2011
including agraphia, were described during the late 19th and early 20th century, but living conditions have
changed dramatically during the last 100 years. Writing no longer means only using pencil and paper, but using
computer word processing programs. Writing using paper and pencil does not require the same cognitive,
motor, and spatial tasks as those required when using a computer keyboard. Although the conceptual
knowledge of written language can be the same, the motor activity and the spatial abilities that are used are
rather different. It can be anticipated that new neuropsychological syndromes resulting from these new living
conditions will be described in the future.
C et article attire lattention sur le fait que le cerveau humain sest adapte afin de survivre dans des
conditions de vie existant longtemps avant les civilisations classiques. Il est propose quil ny a aucune
zone cerebrale specialisee pour lecriture, mais plutot que lecriture repose sur certaines habiletes de base
existant bien avant que lecriture soit inventee. La pre-ecriture etait initialement une habilete visuoconstructive
et ideomotrice et ce nest que plus tard quelle est devenue une habilete reliee au langage (ecriture). Par ailleurs,
cet article met lemphase sur le fait que la plupart des syndromes neuropsychologiques, incluant lagraphie,
furent decrits au cours de la fin du 19eme siecle et le debut du 20eme siecle. Cependant, les conditions de vie ont
enormement evoluees durant les 100 dernieres annees. Ecrire ne represente plus uniquement le fait dutiliser un
papier et un crayon, mais egalement lutilisation dun programme de traitement de texte informatise. Lecriture
au moyen dun papier et dun crayon ne fait pas reference aux memes taches cognitives, motrices et spatiales
que lecriture a partir dun clavier dordinateur. Quoique la connaissance conceptuelle du langage ecrit puisse
etre la meme, lactivite motrice et les habiletes spatiales utilisees sont differentes. Il peut etre anticipe que, dans
le futur, les syndromes neuropsychologiques resultant de ces nouvelles conditions de vie seront decrits.
E n este artculo se senala que la adaptacion del cerebro humano fue lograda para sobrevivir bajo ciertas
condiciones de vida mucho antes de las civilizaciones clasicas. Se discute que no existe un area
especializada para la escritura, sino que, la escritura se basa en habilidades basicas que han existido desde antes
que la escritura fuera inventada. La pre-escritura fue inicialmente una habilidad visuoconstructiva e
ideomotora, y despues se convirtio en una habilidad relacionada con el lenguaje (escritura). Se enfatiza que la
mayora de los sndromes neuropsicologicos, incluyendo la agrafa, fueron descritas durante finales del siglo
19 y principios del siglo 20, sin embrago, las condiciones de vida han cambiado dramaticamente durante los
ultimos cien anos. La escritura ya no solo significa el uso de papel y lapiz sino tambien el uso de programas de
procesamiento de texto. La escritura con lapiz y papel no requiere las mismas tareas cognitivas, espaciales y
motoras que se necesitan cuando se utiliza un teclado de computadora. A pesar de que el concepto de lenguaje
escrito pueda ser el mismo, la actividad motora y las habilidades espaciales que son usadas son totalmente
diferentes. Se anticipa que en el futuro se puedan describir nuevos sndromes neuropsicologicos resultado de las
nuevas condiciones de vida.
Correspondence should be sent to Alfredo Ardila, PhD, 12230 NW 8 Street, Miami, Florida 33182, USA (E-mail:
alfredoardila@cs.com).
# 2004 International Union of Psychological Science

http://www.tandf.co.uk/journals/pp/00207594.html DOI: 10.1080/00207590344000295
62 ARDILA
Anthropology has striven to adequately under- abilities. Our understanding of cultural differ-
stand mans living conditions 10,000, 20,000, or ences is optimistic yet limited (Ardila, 1995;
100,000 years ago. The Stone Age (usually divided Fletcher-Janzen, Strickland, & Reynolds, 2000;
into the early Stone Age, or Palaeolithic, and the Uzzell, Ponton, & Ardila, in press). Matthews
later Stone Age, or Neolithic) extends to approxi- (1992), in his International Neuropsychological
mately 6,000 to 7,000 years ago (Hours, 1982). Society presidential address, accurately observed
Agriculture appeared some 10,000 years ago. The that ... a very limited kind of neuropsychology,
first cities appeared some 6000 years ago and the appropriate to only a fraction of the worlds
first civilizations appeared about 5000 years ago. population, is presented to the rest of the world as
Writing only has a 5000- or 6000-year history and if there could be no other kind of neuropsy-
arithmetical abilities have a history of about 6000 chology, and as if the education and cultural
years (Childe, 1936; Sampson, 1985). However, it assumptions on which ... neuropsychology is
has been thought that contemporary man (Homo based were obviously universals that applied
sapiens sapiens) has lived on earth for at least everywhere in the world (p. 421).
50,000 years but perhaps, according to current Furthermore, most of the neuropsychological
evidence, it may be since at least 100,000 years syndromes were described during the late 19th
ago. We can state with some certainty that, during and early 20th century: aphasia (Broca, 1863;
this time, the structural changes of the brain in Wernicke, 1874), alexia (Dejerine, 1891, 1892),
man have been minimal (Harris, 1983; Kochetkova, agraphia (Exner, 1881), acalculia (Henschen,
1973; Tomasello, 2000). Human brain adaptation 1925), apraxia (Liepmann, 1900), spatial orienta-
was for survival in Stone-Age life conditions tion disturbances (Jackson, 1874/1932), prosop-
(existing for about 98% of this life span) rather agnosia (Bodamer, 1947), visuoconstructive
than in those life conditions existing nowadays. disturbances (Henschen, 1925; Poppelreuter,
Only by departing from the analysis of these 1917), and executive functioning defects
original conditions can we understand the specific (Harlow, 1868), among others. Nonetheless,
characteristics and idiosyncrasies of the brains living conditions have changed dramatically
adaptation. It would seem reasonable for any during the last 100 years. Writing no longer
neuroscientist to put the question: What type means only the use of a pencil and a paper, but
of information did the human brain become also the use of a computer word processor
adapted to process? Consequently, What are program. Arithmetical abilities, too, have chan-
mans basic cognitive abilities? The search for ged; instead of writing numbers down on paper
universals has guided an important proportion and applying certain computational rules, we
of anthropological and linguistic activity during more often require the ability to use a pocket
the last decades. calculator. One major source of knowledge of
Anthropologists and linguists have attempted other peoples faces is through television, and a
to find out basic and universal ways of social major source of knowledge of other peoples
organization in different human groups (Van den voices is through the telephone. Intensive expo-
Berghe, 1979) and fundamental language charac- sure to these media has been observed only over
teristics (Greenberg, 1978; Hagege, 1982). Attempts the last few decades.
are made to infer the social organization of The need for a clear understanding of the
prehistoric man and languages existing before origins of current cognitive abilities is evident
living languages. An excellent example of this last (Ardila, 1993a, 1993b, 1993c, 1993d). The exam-
point is the reconstruction obtained for the Indo- ple of reading and writing may be illustrative of
European language (Anderson, 1973; Lehmann, the need for a historical/anthropological analysis
1974; Martinet, 1975), whose last speaker passed of neuropsychological syndromes. Varney (2002)
away several thousands of years ago. Efforts are points out that reading is a cultural, not an
currently being made to reconstruct even older evolutionary, development. He emphasizes that
proto-languages (Shevoroshkin, 1990). our capacity of reading did not evolve biologi-
Undoubtedly, neuropsychology has advanced cally; it evolved through cultural developments
tremendously in some specific areas. Significant that were only acquired as typical human
advancements have been in the assessment of the abilities within the last 200 years in Europe and
sequelae of brain pathology and in the establish- America, and only after World War II in the rest
ment of clinical/anatomical correlations. How- of the World (p. 3). The origins of reading can be
ever, we do not yet sufficiently understand or found in certain abilities that existed long before
know what can be considered as basic cognitive reading was developed.
THE BRAIN AND WRITING 63
Reading and writing were far from universal cave paintings dating from the Palaeolithic age
even at the beginning of the 21st century. have been found. Mainly animals, but also
According to the United Nations, a person people, instruments, and environmental condi-
who is literate can, with understanding, both read tions, are represented in these paintings. Further
and write a short simple statement on his or her evolution in pre-writing is represented by paint-
everyday life. A person is functionally literate ings becoming standardized for representing
who can engage in all of those activities in which specific elements (i.e., a standard bird means
literacy is required for effective function of his or bird). Lecours, Pena-Casanova, and Ardila
her group and community and also for enabling (1998) point out that writing begins with concrete
him or her to continue to use reading, writing, pictograms that reflect realities accessible to the
and calculation for his or her own and the senses, particularly to vision. These pictograms
communitys development (UNESCO, 2003). further evolved and became abstract, progres-
Surveys throughout the world have been con- sively separating from the concrete representation.
ducted to observe populations speaking various This situation was observed in Sumer (contem-
languages and their inability to read or write a porary Iraq) about 53 centuries ago, and it is
simple message. In the first survey (1950), at least usually regarded as the beginning of writing in
44% of the worlds population were found to be human history. Symbols (graphemes) referred
illiterate. A 1978 study showed the rate to have to the meaning of the words, so these original
dropped to 32.5%. In 1990 illiteracy worldwide writing systems are regarded as logographic.
dropped to about 27%, and by 1998 to 16%. Graphemes representing sounds (syllables)
However, a study by the United Nations Chil- appeared later, about 4000 years ago in Phoenicia
drens Fund (UNICEF) published in 1998 pre- (Sampson, 1985), and graphemes representing
dicted that the world illiteracy rate would increase phonemes appeared even later in Greece.
in the 21st century because only a quarter of the The sequence of the evolution of writing in
worlds children were in school by the end of the consequence was:
20th century. The highest illiteracy rates were
found in the less developed nations of Africa, Drawings p pictograms p logograms p syllabic
Asia, and South America. The lowest illiteracy graphemes p phonemic graphemes
rates were found in Australia, Japan, North
Writing systems can be divided in different
Korea, and the more technologically advanced
ways; a major distinction between logographic
nations of Europe and North America. Currently,
(representing meanings) and sonographic (repre-
there are an estimated 862 million illiterate adults
senting sounds) systems can be established
in the world, of whom about two thirds are
(OMNIGLOT, 2003; Sampson, 1985).
women (UNESCO, 2003). The mean educational
The fundamental difference between logographic
level of contemporary man is only about 34
writing systems and other scripts is that each
years of school!
logographic symbol means something. As a result,
Varney (2002) analysed the origins of reading
logographic writing systems generally contain a
ability. He suggested that the ancient skills of
large number of symbols: anything from several
gesture comprehension and animal tracking were
hundred to tens of thousands. In fact there is no
the underpinnings of brain organization that
theoretical upper limit to the number of symbols
permitted reading to occur. He demonstrated
in some logographic scripts, such as Chinese.
that alexia is significantly associated with
Logographic scripts may include the following
impaired pantomime and animal footprint recog-
types of symbols.
nition. Thus, these abilities, existing since early
human history, were prerequisites that led the way 1. Logogramssymbols that represent parts of
to the cultural development of reading. Gesture words or whole words. Some logograms resemble
recognition may have existed for several millions the things they represent and are sometimes
of years, but reading developed just a few known as pictograms or pictographs.
millennia ago. 2. Ideogramssymbols that graphically repre-
sent abstract ideas.
HOW DID WRITING APPEAR? 3. Semantic-phonetic compoundssymbols that
include a semantic element, which represents or
Wall paintings appeared during the Palaeolithic hints at the meaning of the symbol, and a
era, some 3035,000 years ago (Childe, 1936). phonetic element, which denotes or hints at the
Across Europe, particularly in France and Spain, pronunciation.
64 ARDILA
4. Sometimes symbols are used for their pho- read or write. Reading and writing is obviously
netic value alone, without regard for their meaning. far from being a primary or biologically based
cognitive ability. Clearly, writing represents a
In sonographic writing systems, syllables (syllabic cognitive ability that depends on the human
alphabets) or phonemes (phonemic alphabets) can cultural evolution (Vygotsky, 1962).
be used. Alphabetic writing systems come in two
varieties.
AGRAPHIA AS A
1. Abjads (consonant alphabets) represent con- NEUROPSYCHOLOGICAL SYNDROME
sonants only, or consonants plus some vowels.
Even though not common, full vowel indication Agraphia can be defined as the partial or total
(vocalization) can be added, usually by means of loss of the ability to produce written language,
diacritics. and is associated with brain pathology. The
2. Alphabets (phonemic alphabets) represent ability to write can be impaired as a result of
consonants and vowels. linguistic defects (aphasia), but other elements
not related to language (e.g., motor and spatial)
Thus, initial writing (or rather, pre-writing) was a also participate in the writing ability. It supposes
visuoconstructive ability (i.e., representing exter- at least a knowledge of the language codes
nal elements visually), and only later did it (phonemes, words), an ability to convert language
become an ideomotor praxis ability (i.e., making sounds in graphemes, a knowledge of the
certain learned and fixed sequences of movements graphemic system (alphabet), an ability to per-
with the hand to create a pictograma standar- form fine movements, and an appropriate use
dized representation of external elements). Still of the space for distributing, joining, and
later, after writing became an ideomotor praxis separating letters. It is evident that diverse types
ability, it became a linguistic ability (i.e., associat- of writing disturbances can be found in clinical
ing the pictogram with a word, and further practice.
analysing the word in its constituting sounds). It Different attempts to classify writing distur-
is not surprising that three major disorders in bances are found in the history of neuropsy-
writing can be observed as a result of brain chology. Goldstein (1948) distinguished two
pathology: visuoconstructive (spatial or visuospa- major types of agraphia: apractoamnesic and
tial agraphia), ideomotor (apraxic agraphia), and aphasic-amnesic. Luria (1976, 1980) referred to
linguistic (aphasic agraphia). In addition, of course, five different types of agraphia, three of them
writing requires visual and motor integrity. associated with aphasia (sensory agraphia, affer-
ent motor agraphia, and kinetic agraphia) and
HOW MANY PEOPLE CAN WRITE? two associated with visuospatial defects. Hecaen
and Albert (1978) distinguished four types of
Even though writing began several millennia ago, agraphia: pure, apraxic, spatial, and aphasic.
as recently as the 1950s about half of the worlds Regardless of the diversity of classifications of
population was illiterate. The percentage of agraphia, a basic distinction can be established
illiteracy dramatically increases as we go back between (1) agraphias due to a language impair-
in time, and up to only a couple of centuries ago, ment (linguistic or aphasic agraphias), (2) agra-
the overwhelming majority were illiterate. Until phias due to other types of impairments (most
the 15th century, when the printing press was often, motor or spatial) disturbing the normal
invented, writing may well have been limited to a ability to write (Benson & Ardila, 1996), or simply
few intellectual people and monks. Even though (3) central and peripheral agraphias (Ellis, 1988).
there are no statistics available, it may be In the first case, agraphia is just a secondary
conjectured that 99% or more of the population manifestation of the aphasic syndrome. In the
was illiterate. Furthermore, it has to be kept second, it can be interpreted as a result of a
in mind that the mean level of education of broader visuoconstructive/visuospatial impairment
contemporary man is about 3 years of school, (Ardila & Rosselli, 1993), or motor-apraxic
which may not be enough to develop automatic disturbance (Hecaen & Albert, 1978). Conse-
reading and writing. quently, writing can be interpreted as a particular
It is evident that writing represents an unusual type of cross-modal learning. Certain visuo-
ability in humans. The overwhelming majority of constructive and ideomotor abilities become
members of our species who have lived could not associated with language.
IS ANY AREA IN THE BRAIN Similarly, the use of the space is different. The
SPECIALIZED FOR WRITING? normal spatial distribution of the words on the
page is automatic on the computer and, hence,
Writing is a functional system (Luria, 1976) writing in this way cannot be spatially disorga-
that requires, and is based on, some more nized, as may be the case in handwriting. By the
fundamental abilities: praxis abilities (i.e., learning same token, letters are neatly written and easily
sequences of movements required to write the recognizable. When typing, we are not using a
letters), spatial abilities (distributing letters and space that is directly manipulated with the hands
words in the space, and understanding the value (constructional space), but only a visual
of space in writing), constructional abilities space. Furthermore, typing is not a construc-
(reproducing a model using certain movements), tional task (we do not have to construct the
and obviously, the knowledge of the language and letters) but rather a motor-spatial task.
the association between verbal auditory elements Many people type using a spatial memory for
and visual symbols. the position of the letters in the keyboard. This
Varney (2002) refers to the anthropological is a type of memory not required in handwriting,
concept of pre-adaptation, which states that the and it probably depends on right hippocampal
evolution of a structure for one purpose can and parietal activity (Moser, Hollup, & Moser,
enable that structure to perform another purpose. 2002). Other people have to look at the keys
This must be true for writing (as for any other to select the letters when typing. In this case,
abilities depending on cultural evolution). Visuo- literal reading is a prerequisite for writing. Letters
constructive and ideomotor abilities represent have to be recognized visually before they are
prerequisites for writing; they are probably related written. In handwriting, we use a mental repre-
to the ability to make tools and weapons and sentation of the visual form of the letters.
generally to use the hands in a skilled way. Interestingly, few peopleif any, regardless of
how well they can typeare able to reproduce
(i.e., describe verbally or by drawing) how the
FROM AGRAPHIA TO DYSTYPIA different letters are arranged on the keyboard.
Memory for their location seems to be a purely
Contemporary literate man is using handwriting spatial and motor memory of which we are
less and less, and relying on computers more and poorly aware.
more. In an informal survey to 40 people with a For typing some special symbols (e.g., inter-
college-level education background, they reported rogation marks) and letters (the Spanish N), some
using a computer about 90% of the time when relatively sophisticated motor manoeuvres are
writing and handwrote only 10% of the time. required, sometimes requiring the use of special
Obviously, this sample does not represent all of keys or sequences of movements. In handwriting,
humankind, and computers are not accessible to however, special symbols are written using the
a large percentage of the human population. But mental forms that we have learned. When typing,
this sample seems to illustrate the way in which if a letter needs to be lower or upper case, a
writing is evolving: from handwriting to typing key has to be pushed. No other change to the
on a computer. movement is made. We can also select different
Handwriting and using computers represent signi- writing styles and letter sizes using some special
ficantly different cognitive and motor abilities. commands and menus, all without changing the
During handwriting, fingers are maintained in a sequences of the hand movements.
relatively steady position while the hand moves. In cases of brain damage, how is typewriting
In typing, the opposite pattern is observed. When altered? To the best of my knowledge, only
typing, the right hand does not move from one one case of agraphia for typewriting has been
side to the other and back as in handwriting, but published (Otsuki, Soma, Arihiro, Watanabe,
the hands remain relatively stationary and only Moriwaki, & Naritomi, 2002). Nonetheless, it
the fingers are moved. Letters are not written but can be assumed that different types of brain
selected. Both hands have to be used in a similar pathology may affect the ability for typing on a
way when typing. Because of using both hands, computer word processor. The following can be
we have to assume that a major interhemispheric conjectured.
integration is required. It is obvious to assume
that right-hemisphere lesions located in the 1. An anterior callosal lesion would impair
frontal and parietal areas should significantly the ability to coordinate the movements between
impair the typewriting ability of the left hand. the hands. Furthermore, the left hand would be
66 ARDILA
isolated from the linguistic left hemisphere, and to understand the principles governing the
would be unable to write. Left-hand hemi- functioning of a computer.
agraphia in callosal lesions has been observed 2. Some visuoperceptual abilities to recognize
(Benson & Ardila, 1996). icons, windows, etc.
2. By the same token, it has been observed that 3. Some skilled movements to type on the
damage in the supplementary motor area results keyboard and manoeuvre the mouse correctly.
in disturbances in the coordinated movements 4. Some spatial abilities to handle the working
between both hands (Middleton & Strick, 2001). space (monitor screen).
We can anticipate supplementary motor area 5. Some memory abilities to learn programs,
typing agraphia. to use the spatial position of the keys, etc.
3. Spatial memory disturbances should result
in difficulties in recalling the positions of the Obviously, the ability to use computers can
letters on the keyboard. Typing would be slow, potentially be disrupted as a consequence of a
and would require a continual search for the failure in any one of these abilities (acumputuria
letters. syndrome). In the future, apart from dystypia,
more complex disturbances in the ability to use
Otsuki et al. (2002) reported on a 60-year-old computers will probably be established.
right-handed Japanese man who showed an

isolated persistent typing impairment without
aphasia, agraphia, apraxia, or any other neuro- CONCLUSION
psychological deficit. They proposed the term
dystypia for this peculiar neuropsychological The origins of writing can be traced back to cave
manifestation. The symptom was caused by an paintings. Writing (or pre-writing) was initially a
infarction in the left frontal lobe involving the visuoconstructive ability, later involving some
foot of the second frontal convolution and the stereotyped movements to represent pictograms,
frontal operculum. The patients typing impair- and finally involving spoken language. It makes
ment was not attributable to a disturbance of the sense, therefore, that the ability to write can be
linguistic process, since he had no aphasia or disturbed in three major forms: as a visuospatial/
agraphia. Nor was it attributable to an impair- visuoconstructive dexterity, as an ideomotor skill,
ment of the motor execution process, since he had and as a linguistic ability.
no apraxia. Thus, it was deduced that his typing Writing has followed a long evolution since
impairment was based on a disturbance of the cave painting during the Palaeolithic times.
intermediate process where the linguistic phono- Different strategies have been used to represent
logical information is converted into the corre- spoken language visually (ideograms, alphabets,
sponding performance. The authors hypothesized etc). Writing, however, has continued to evolve
that the foot of the left second frontal convolu- since its initial invention. The use of punctuation
tion and the operculum may play an important marks and the distinction between upper and
role in the manifestation of dystypia. lower case in writingto mention just two
Using a computer is somehow equivalent to examplesare relatively recent in history (Samp-
a new writing system. Obviously, there is no brain son, 1985). Evolution has continued with the
area related to typing on a computer, as there is development of different technical instruments for
no brain area related to reading and writing. These writing: the feather, the pencil, the typewriter, and
are cultural and technological elements recently the computer. Brain representation of written
developed through human evolution. Rather, language has necessarily changed in some way,
there are basic cognitive abilities (pre-adaptative too. Neuropsychological syndromes associated
abilities) that are required for the use of these new with brain pathology have evolved over time.
cultural elements: e.g., certain visuoperceptual We can assume that the consequences of brain
abilities and cross-modal associations for reading, pathology in a Palaeolithic man were not the
phonological awareness and some fine movements same as for a 19th-century individual (when
for writing, etc. Using computers is notoriously agraphia was first described), or for contemporary
more complex, yet we can assume a functional man or woman (some of whom frequently
system participating in their use. spending most of their working day in front of
It can be conjectured that using computers a computer screen). It can be anticipated that in
requires at least the following abilities. the future new neuropsychological syndromes
resulting from new living conditions will be
1. A conceptual ability (executive functioning) described.
REFERENCES Kochetkova, V. I. (1973). Paleoneurology [in Russian].

Moscow: Moscow State University Press.
Anderson, J. M. (1973). Structural aspects of language Lecours, A. R., Pena-Casanova, J., & Ardila, A. (1998).
change. London: Longman. Orgenes y evolucion de la escritura [Origins and
Ardila, A. (1993a). Introduction: Toward a historical/ evolution of writing]. In A. R. Lecours, J. Pena-
anthropological approach in neuropsychology. Casanova, & F. Dieguez-Vide (Eds.), Dislexia y
Behavioural Neurology, 6, 7174. disgrafia: Teora, formas clnicas y exploracion
Ardila, A. (1993b). Historical evolution of spatial (pp. 18). Barcelona: Masson.
abilities. Behavioral Neurology, 6, 8388. Lehmann, W. P. (1974). Proto-Indian-European syntax.
Ardila, A. (1993c). People recognition: A historical/ Austin, TX: University of Texas Press.
anthropological perspective. Behavioural Neurology, Liepmann, H. (1900). Das Krankheitsbild der Apraxie
6, 99106. (motorische Asymbolie) auf Grund aines Falles
Ardila, A. (1993d). On the origins of calculation von einseitiger Apraxie. Monatschrie Psychiatrie
abilities. Behavioural Neurology, 6, 8998. Neuroligie, 10, 214227.
Ardila, A. (1995). Directions of research in cross- Luria, A. R. (1976). Basic problems of neurolinguistics.
cultural neuropsychology. Journal of Clinical and The Hague: Mouton.
Experimental Neuropsychology, 17, 143150. Luria, A. R. (1980). Higher cortical functions in man.
Ardila, A., & Rosselli, M. (1993). Spatial agraphia. New York: Basic Books.
Brain and Cognition, 22, 7595. Martinet, A. (1975). Evolution de langues et recon-
Benson, D. F., & Ardila, A. (1996). Aphasia: A struction. Paris: Presses Universitaires de France.
clinical perspective. New York: Oxford University Matthews, C. G. (1992). Truth in labeling: Are we
Press. really an international society? Journal of Clinical
Bodamer, J. (1947). Die Prosopagnosie. Archiv fur and Experimental Neuropsychology, 14, 418426.
Psychiatrie und Nervenkrankheiten, 179, 654. Middleton, F. A., & Strick, P. L. (2001). A revised
Broca, P. (1863). Localization des fonctions cerebrales: neuroanatomy of frontal-subcortical circuits. In
Siege du langage articule. Bulletin de la Societe D. G. Lichter & J. L. Cummings (Eds.), Frontal-
dAnthropologie, 4, 200203. subcortical circuits in psychiatry and neurological
Childe, V. G. (1936). Man makes himself. London: disorders (pp. 4458). New York: Guilford Press.
Pitman Publishing. Moser, E. I., Hollup, S. A., & Moser, M. B. (2002).
Dejerine, J. (1891). Sur un cas de cecite verbale avec Representation of spatial information in dynamic
agraphie suivi dautopsie. Comptes rendus. Societe neuronal circuits in the hippocampus. In L. R.
de Biologie, 3, 197201. Squire & D. L. Schacter (Eds.), Neuropsychology of
Dejerine, J. (1892). Contribution a letude anatomo- memory (pp. 361376). New York: Guilford Press.
pathologique et clinique des differerents varietes de OMNIGLOT. (2003). Retrieved May 1, 2003, from
cecite verbale. Comptes Rendus, Societe de Biologie, www.omniglot.com.
4, 6190. Otsuki, M., Soma, Y., Arihiro, S., Watanabe, Y.,
Ellis, A. W. (1988). Normal writing processes and Moriwaki, H., & Naritomi, H. (2002). Dystypia:
peripheral acquired dysgraphias. Language and Isolated typing impairment without aphasia, apraxia
Cognitive Processes, 3, 99127. or visuospatial impairment. European Neurology, 47,
Exner, S. (1881). Unersuchungen uber die lokalisation 136140.
der Functionen in der Grosshimirinde des Menschen. Poppelreuter, W. (1917). Die psychischen Schadigungen
Wien: Braumuller. durch Kopfschuss im Kriege 19141916: Die
Fletcher-Janzen, E., Strickland, T. L., & Reynolds, Storungen der neideren und hoheren Schleitungen
C. R. (Eds.). (2000). Handbook of cross-cultural durch Verletzungen des Okzipitalhirns. Leipzig,
neuropsychology. New York: Kluwer Academic/ Germany: Voss.
Plenum Publishers. Sampson, G. (1985). Writing systems. Stanford, CA:
Goldstein, K. (1948). Language and language Stanford University Press.
disturbances. New York: Grune & Stratton. Shevoroshkin, V. (1990). The mother tongue: How
Greenberg, J. H. (1978). Universals of human language. linguistics have reconstructed the ancestor of all
Stanford, CA: Stanford University Press. living languages. The Sciences, May/June, 2027.
Hagege, C. (1982). La structure des langues. Paris: Tomasello, M. (2000). The cultural origins of human cog-
Presses Universitaires de France. nition. Cambridge, MA: Harvard University Press.
Harlow, J. M. (1868). Recovery from the passage of an UNESCO. (2003). Retrieved May 1, 2003, from www.
iron bar through the head. Massachusetts Medical portal.unesco.org.
Society Publication, 2, 327346. Uzzell, B., Ponton, M., & Ardila, A. (Eds.) (in press). Inter-
Harris, M. (1983). Cultural anthropology. New York: national handbook of cross-cultural neuropsychology.
Harper & Row. Mahwah, NJ: Lawrence Erlbaum Associates Inc.
Hecaen, H., & Albert, M. L. (1978). Human Van den Berghe, P. L. (1979). Human family systems.
neuropsychology. New York: Wiley. New York: Elsevier.
Henschen, S. E. (1925). Clinical and anatomical Varney, N. R. (2002). How reading works: Considera-
contributions on brain pathology. Archives of tions from prehistory to present. Applied Neuropsy-
Neurology and Psychiatry, 13, 226249. chology, 9, 312.
Hours, F. (1982). Les civilisations du Paleolithique. Vygotsky, L. S. (1962). Thought and language.
Paris: Presses Universitaires de la France. Cambridge, MA: MI T Press.
Jackson, J. H. (1932). Selected writings. London: Wernicke, C. (1874). Der aphasiche Symptomencomplex.
Hodder & Stoughton. (First published 1874). Breslau, Germany: Cohn & Weigert.
Estudios de Psicologa / Studies in Psychology, 2014
http://dx.doi.org/10.1080/02109395.2014.965453
Spanish and the characteristics of acquired disorders in reading

and writing / El espaol y las caractersticas de los trastornos
adquiridos de la lectura y la escritura
Alfredo Ardilaa and Mnica Rossellib
a
Florida International University; bFlorida Atlantic University
(Received 4 February 2014; accepted 15 May 2014)
Downloaded by [Alfredo Ardila] at 11:01 04 December 2014
Abstract: Spanish has unique phonological, lexical and grammatical features

and a unique literacy system; these characteristics may influence the specific
way acquired disorders in written language manifest themselves. The purpose
of this article is to firstly review the specific characteristics of Spanish and
subsequently analyse alterations in cases of alexia. A review of reading
models is presented as well. It is concluded that a direct application of reading
models developed for other languages that have noticeably irregular writing
systems does not seem appropriate to interpret Spanish errors after brain
damage.
Keywords: Spanish; alexia; written language; reading skills; writing skills
Resumen: Se enfatiza que el espaol presenta ciertas caractersticas propias

desde el punto de vista fonolgico, lxico y gramatical, al igual que en su
sistema de lectoescritura; estas caractersticas pueden incidir sobre la forma
especfica en que se manifiestan los trastornos adquiridos en el lenguaje
escrito (alexias y agrafias). Se revisan los estudios sobre alexias y agrafias
en espaol. Se concluye que no parece apropiada la aplicacin directa al
espaol de modelos de lectura desarrollados en otras lenguas con sistemas
de escritura notoriamente ms irregulares.
Palabras clave: espaol; alexia; agrafia; lenguaje escrito; habilidades de
lectura; habilidades de escritura
Spanish has certain specific phonological, lexical and grammatical features that
are also seen in its writing system. These characteristics can influence learning
how to read and the specific way acquired disorders are manifest in written
language (alexia) as a result of brain pathology. The purpose of this article is to
firstly review the specific characteristics of Spanish so as to then subsequently
analyse alterations in cases of alexia.
English version: pp. 18 / Versin en espaol: pp. 916

References / Referencias: pp. 1617
Translated from Spanish / Traduccin del espaol: Liza DArcy
Authors Address / Correspondencia con los autores: Alfredo Ardila, Department of
Communication Sciences and Disorders, Florida International University, Miami,
Florida 33199, USA. E-mail: ardilaa@fiu.edu
2014 Fundacion Infancia y Aprendizaje

2 A. Ardila and M. Rosselli
Some characteristics of the Spanish language

The Spanish phonological system is similar to other Indo-European languages
(Table 1). It has about 23 phonemes (there are two phonemes that are only found
in parts of Spain and Latin America. An example is the dental fricative Z or
C that can be heard in most regions of Spain (except Andalusia) and the
voiced palatal lateral // LL that can be heard in some Latin American
countries such as Colombia and Paraguay). In addition, some phonemes sound
different according to the region. The system has five phonemic vowels: /a/, /e/,
/i/, /o/, /u/, 1618 consonants and two glides [i] and [u] which can
function as glides in a postnuclear position of the syllable (Penny, 2002;
Quilis, 1963).
The most outstanding Spanish phonological characteristics are: (1) The
diphthongization. Spanish has a very extensive amount of diphthongs that can
alternate with vowels (e.g., in Spanish poder puedo) inasmuch as that it tends
to not use hiatuses (accents to dissolve the diphthong; e.g., in the Spanish word
for day da). (2) In the evolution from Latin to Spanish, voiceless occlusive
phonemes (/p/, /t/, /k/) that were positioned between vowels were voiced (/b/,
/d/, /g/), but only in vulgar words (higher frequency words) inasmuch as they
remained voiceless in derivative words (examples in Spanish, abeja, apicultura
[bee, beekeeping]); and voiced occlusive phonemes were fricativized (some-
times these fricativized voiced occlusive phonemes are described as approxi-
mants or spirants) (, , ). (3) Spanish has a very definite syllabic structure; for
Table 1. Spanish phonological system. Above: consonant system. Below: vowel system.
Dental
Bilabial Labiodental fricative Dental Alveolar Palatal Velar
Plosive Voiceless /p/ /t/ /k/
Voiced /b/ /d/ /g/
Affricate Voiceless /t/
Voiced
Fricative Voiceless /f/ // /s/ /x/
Voiced //
Nasal Voiced /m/ /n/ //
Lateral Voiced /l/ //
Trill Voiced /r/ //
simple
multiple
Location
Open Front Central Back
Closed /i/ /u/
Central /e/ /o/
Open /a/
Reading and writing disorders / Trastornos de la lectura y la escritura 3
each word within the language one of the syllables carries a distinctive, func-
tional accent.
From a language rhythm standpoint, Spanish belongs to the group of lan-
guages whose syllables take roughly the same amount of time (syllable-timed)
(the duration of every syllable is equal, even though each word has a prosodically
stressed syllable) (Berg, 1991), in contrast to languages such as English, which is
timed by accentual units (stress-timed) (syllables are sometimes different lengths,
but the temporal duration between two stressed syllables is equal), Spanish has a
very evident accentual structure.
From a grammatical standpoint, Spanish is an inflected language: it uses
inflections to indicate the relationship between elements. This involves significant
morphological changes in words. However, there are no declensions (inflectional
paradigm as according to the accusative, locative, genitive, etc., context in which
they are used) as there are in Latin. But they also use prepositions (type of
grammatical morpheme: a, opposite, under, etc.) to indicate those relationships
between the elements (Seco, 1988). In terms of syntax, like other romance
languages, the word order in the clause is flexible (e.g., the clause Juan ama a
Mara [John loves Mary], can be said or written another way without altering its
meaning: A Mara ama Juan; Juan a Mara ama; A Mara Juan ama, etc.),
although Spanish is clearly more flexible than other Latin languages.
The Spanish writing system has certain peculiar characteristics (Table 2). The
Spanish reading system is transparent (that is, there are specific rules for convert-
ing graphemes into phonemes), except in: (1) the reading of certain words
borrowed from other languages (e.g., in Spanish, souvenir, jeep, etc.); (2) some
irregularities in the transcription of words borrowed from Native American
languages, particularly evident in Mexico (e.g., in Spanish, Oaxaca); (3) certain
written forms that do not correspond to the actual pronunciation of the words
(e.g., in Spanish, pelear [fight]); and (4) certain archaisms in writing (e.g., in
Spanish, Mxico).
Ambiguity in the literacy Spanish system is found in writing, as many words
could potentially be written in different ways (spelling rules) (e.g., /muxer/ could
be written MUJER or MUGER), although there is only one possible way to read
any word or nonword (Table 3). Leal and Matute (2001) found that the phoneme
grapheme relationship that is used to read Spanish is much higher (.70) than the
phoneme-grapheme relationship (.30) used to write it. In Spanish it is thus
possible to find homophonic heterography (alternative ways to write the phone-
mic sequence of a word), but there is no holographic heterophony (there is only
one way to read a sequence of letters). In writing, it is possible to distinguish two
types of errors: homophones (usually referred to as spelling errors where the
pronunciation of the word does not change, e.g., MUGER; and non-homophones
(usually referred to as typing errors, e.g., NUJER for MUJER). Non-homophone
errors are due to changes, additions or substitutions of letters altering written
representation of the spoken word and therefore its meaning. The first type of
error is very common, particularly in people with low levels of education, but
Table 2. Writing phonemes to graphemes (adapted from Leal and Matute (2001)).
Number Phoneme Grapheme (s) Example
1 /p/ P /pensar/
2 /t/ T /toldo/
3 /d/ D /dar/
4 /f/ F /fama/
5 / t/ CH / tosa/
6 /l/ L /lima/
7 /r/ R /klaro/
8 /m/ M /malo/
9 / / /pia/
10 /e/ E HE /edad/, /ermoso/
11 /a/ A HA /ala/, /ada/
12 /o/ O HO /oliba/, /onor/
13 /b/ BV /base/, /baso/

14 /g/ G GU /gato/, /giso/
15 /n/ N NN /nube/, /perene/
16 /y/ Y LL /yema/. /caye/
17 /R/ R RR /radio/, /ko ida/
18 /i/ I Y HI /bida/, /ielo/
19 /u/ /kulpa/, /pinguino/, /ueso/
20 /k/ C K QU /kosa/, /kilo/, /keso/
21 /x/ JG /kaxa/, /xiro/
22 /s/ C S Z SC PS /siklo/, /pasado/, /sarpar/, /irasible/, /sikiko/
23 // CZ /ielo/, /ankudo/
24 LL
could also be found in people who have a university level education (Ardila,
Rosselli, & Ostrosky-Solis, 1996).
There is no agreement on what the reading unit in different languages is, but
the reading unit in Spanish is probably the syllable (Ardila, 1998). As mentioned
before, Spanish has a very well defined syllabic structure (syllables are built
around a vowel or a diphthong) and from an expressive standpoint, Spanish
syllables clearly take a similar amount of time (syllable-timed). It is easy to
assume that Spanish speakers have a strong syllabic awareness; distinguishing
syllables in a word is considered (unlike other languages such as English) a very
easy task, one that is almost self-evident given the syllabic pronunciation in the
language.
In Spanish, the longest unit that needs to be read as a whole, when only
considering the complete sequence of letters needed to work out its phonology, is
the syllable, more specifically certain syllables that consists of three letters, such
as GUI and GUA. It seems reasonable then to suggest that reading in Spanish is
performed using a sequence of syllables (syllable by syllable). Interestingly, in
written words recognition tasks (lexical decision), the time required to make the
decision of whether or not the stimulus is a word is affected by both word
frequency (frequency effect) and its length. Further, the correlation with the
number of syllables (phonological length) is greater than the correlation with
Table 3. Reading graphemes to phonemes (adapted from Leal and Matute (2001)).
Number Grapheme Phoneme Example
1 A /a/ AROMA
2 B /b/ BOTELLA
3 D /d/ DENTRO
4 E /e/ ESTELA
5 F /f/ FESTA
6 I /i/ PILA
7 J /x/ JUGO
8 K /k/ KIOSKO
9 L /l/ LUZ
10 M /m/ MALO
11 N /n/ NADA
12 // CAA
13 O /o/ OTRO
14 S /s/ SALTAR
15 T /t/ TODO
16 U /u/ UNO
17 V /b/ VACA
18 Z /s/ o // ZAPATO
19 LL // o // LLANTO
20 NN /n/ PERENNE
21 RR // PERRO
22 SC /s/ ASCTICO
23 CH / t/ CHISTE
24 GU /g/ /u/ AGERO
25 QU /k/ QUISO
26 HA /a/ HASTA
27 HE /e/ HELADO
28 HI /i/ HIJO
29 HO /o/ HONDO
30 HU /u/ HUELE
31 Y /y/, /i/ YEMA, LEY
32 G /g/, /x/ GAS, GELATINA
33 C /s/, //, /k/ CIEN, CAMA
34 R /r/, // ARO, RATN
35 P /p/, PILA, PSIQUE
36 PS /s/, /ps/ PSICOLOGA, CLEPSIDRA
37 GU /g/, /gu/ GUITARRA, AGUANTAR
38 X /ks/ AXIOMA
the number of letters in the word (Table 4). This observation further supports the
assumption that the fundamental unit of reading in Spanish is the syllable.
Spelling errors are found in those syllabic sets where alternate spellings are
possible. Different complex script (e.g., representation of the phoneme /k/,
representation of the phoneme /g/, etc.) can be distinguished (Ardila, 1998). In
the case of brain pathology, spelling errors in Spanish are not necessarily asso-
ciated with aphasia and therefore damage to the left cerebral hemisphere; in the
case of injury to the right hemispheric side, an increase in spelling errors in
Table 4. Lexical decision (word recognition) in Spanish. The correlations between

reaction time and word frequency, number of letters and number of syllables in each
word are presented (adapted from Ardila, Rosselli, and Lecours (1993)).
Frequency Number of letters Number of syllables
Correlation .319 .247 .281
P< .0004 .007 .002
Table 5. Errors when writing 95 words that included all possible Spanish spelling
decisions (35 children from 3rd grade, average age 8.94 years, and 57 university students,
average age 20.14 years). A. Non-homophone errors; B. Homophone errors (adapted from
Ardila (1998)).
Children Adults
Non-homophones (average per subject) Letter substitutions 1.06 0.05

Letter omissions 0.20 0.00
Letter exchanges 0.00 0.00
Morpheme substitutions 0.03 0.00
Homophone errors (% of subjects) 0 errors 0.0% 8.8%
15 errors 2.9% 56.1%
610 errors 5.7% 29.8%
1115 errors 14.3% 1.7%
1620 errors 17.1% 3.5%
2125 errors 28.6% 0.0%
More than 25 errors 31.4% 0.0%
writing is often found, which may suggest that the use of spelling in some way
represents a visual perceptual capacity for Spanish subjects (how the word is
seen written one way or another). (Table 5)
Transparency or opacity of a languages writing system is the most relevant
variable when determining the difficulty of learning how to read in that language
(Ziegler & Goswami, 2005). A transparent language is one where there is a
correspondence between phonemes and graphemes; that is why these languages
are known as consistent. In other words, a grapheme is always read in the same
way. The prototypes of transparent languages include Italian, Spanish, Russian
and Finnish. By contrast, an opaque language such as English presents numerous
variations in how phonemes are read and that is why it is also called an incon-
sistent language. It is estimated that learning how to read is much faster in Spanish
than in English because of the differences in the spelling system between these
two languages (Goswami, 2002; Ziegler & Goswami, 2005). How fast reading is
learnt in a language is also influenced by that languages spelling system (Rosselli
& Matute, 2012). For example, Finnish children that begin to read at the age of
seven can reach a reading precision level (reading without any errors) of 90%
after 10 weeks of training while English children who start learning how to read at
the age of five reach this level of precision at the age of nine or 10 (Goswami,
Gombert, & de Barrera, 1998).
The spelling system also determines the prevalence of learning problems in

reading (Helmuth, 2001). This is why prevalence of dyslexia in Italian children
who are 10 years old is estimated to be 12%, while in the US the prevalence is
5% (Paulesu et al., 2001), and rarely seen in Japanese children (Wydell, 2003).
We are not aware of any studies done on the prevalence of dyslexia in a Spanish-
speaking country but we could assume that it is probably similar to that found in
Italy.
Despite the variability of dyslexia found in various languages, the difficulty
behind it is similar in all languages. Jimnez and colleagues found that Spanish-
speaking children who had dyslexia showed the same difficulties with phonemic
segmentation that is described in English speaking dyslexic people (Jimnez
Gonzlez, 1997). Jimnez et al. (2005) examined the effects of the linguistic
complexity in Spanish-speaking children and found that those who had problems
learning how to read had a defect in phonological awareness, similar to that

described in children who had dyslexia and were speakers of other languages.
Reading models and acquired reading disorders

During the last decades of the twentieth century a new approach was proposed in
the study of acquired reading disorders (alexia or acquired dyslexia). Researchers
began to wonder about the nature of cognitive and linguistic defects responsible
for reading difficulties (Beauvois & Drouesen, 1981; Coltheart, 1980; Marshall
& Newcombe, 1973). Interest slightly shifted from the anatomical correlates to the
functional mechanisms responsible for alexia (Ellis, 1993). This new cognitive
and linguistic approach involved the development of reading models in indivi-
duals with normal cognitive functioning. During the following years several
cognitive models partial but not entirely coinciding for the normal reading
process in different languages were proposed. In general, it was suggested that
reading, subsequent to the initial identification of letters, can be done through two
different routes: (1) the direct route: the written word is visualized whole and is
associated with a visual representation of the word that is held in the memory; (2)
the indirect route: the written word becomes a spoken word through a set of
graphophonemic rules and the meaning of the word is reached through phonolo-
gical mediation, in the same way that oral language is understood.
These cognitive models of reading were developed on the basis of the English
language, which has a very irregular reading system and falls within the languages
that are considered opaque. The English writing system is not directly based on
the phonological system of the language, it could more accurately be described as
an intermediate system between early phonological and logographic reading
(Sampson, 1985). Reading words is done partly following the principles of
grapheme/phoneme correspondence, and partly following a global recognition,
not necessarily separating individual letters. Words like dog, man and pistol are
regular words (that is, the grapheme/ phoneme correspondence is regular) and
their pronunciation can be inferred from their spelling, while words like island,
women, yacht and knight are irregular (that is, their pronunciation cannot be
inferred from their spelling). Words like through (as opposed to thought or
though) can be read only by considering the whole word. It is thus comprehen-
sible that two different reading systems have been proposed for reading in English
(direct and indirect or simply phonographic and logographic) (Shallice &
Warrington, 1980).
In attempting to apply these models to Spanish (and other languages with
transparent reading systems) two standpoints emerged: (1) these models can be
applied directly as the reading strategies (routes) are the same in English and
Spanish (Davies & Cuetos, 2005; Ferreres & Miravalles, 1995; Ferreres, Cuitio,
& Olmedo, 2005; Iribarren, Jarema, & Lecours, 2001; Ruiz, Ansaldo, & Lecours,
1994); (2) different languages emphasize different reading strategies, and conse-
quently, these models are not directly applicable to Spanish (Ardila, 1998; Ardila
et al., 1996; Ardila, Rosselli, & Pinzn, 1989). A comparison of the character-
istics of alexias observed in different languages seems to favour the second view
(Karanth, 2003).
An important debate has developed regarding the existence of semantic errors
in the reading of words in Spanish (semantic paralexias) after brain damage.
Ardila (1998) suggested that although semantic paralexias (when one word is
read for another one which often has a similar meaning) are a common phenom-
enon in English-speaking aphasic patients (Landis, Regard, Graves, & Goodglass,
1983), in Spanish it is not a frequent error. Most likely, and based on the reported
cases, semantic paralexias are found in a very specific subgroup of aphasic
patients: chronic patients who are highly educated, often multilingual and who
suffer from a type of motor aphasia. In English, there are many patients who
suffer from semantic paralexias and who can be found in various groups of
aphasic patients.
Conclusions
Spanish has certain distinctive characteristics in its phonology, lexicon, morpho-
syntax and writing system that affect how and the speed with which Spanish
speakers learn how to read and how acquired disorders are manifested in the
written language. Given that Spanish has a transparent system, a direct application
of reading models developed from languages that have a more opaque and
inconsistent writing system does not seem appropriate. A similar conclusion
was reached by studying reading disorders observed in other languages with
transparent systems scriptures such as Hindi (Karanth, 2003) and Serbo-
Croatian (Lukatela & Turvey, 1990).
El espaol y las caractersticas de los trastornos adquiridos de la

lectura y la escritura
El espaol presenta ciertas caractersticas especficas desde el punto de vista

fonolgico, lxico y gramatical que se observan tambin en su sistema de
escritura. Tales particularidades pueden incidir sobre el aprendizaje lector y la
forma especfica en la que se manifiestan los trastornos adquiridos en el lenguaje
escrito (alexias y agrafias) como consecuencia de una patologa cerebral. El
objetivo de este artculo es inicialmente revisar las caractersticas especficas del

espaol y posteriormente analizar las alteraciones en casos de alexias y agrafias
Algunas caractersticas del espaol

El sistema fonolgico del espaol es similar al de otras lenguas indoeuropeas
(Tabla 1). Posee aproximadamente 23 fonemas (hay dos fonemas que slo se
encuentran en algunas regiones de Espaa y Latinoamrica. Por ejemplo la
fricativa interdental Z o C se observa en la mayora de las regiones de
Espaa (excepto Andaluca) y la lateral sonora palatal // LL se observa en
Tabla 1. Sistema Fonolgico del Espaol. Arriba: Sistema Consonntico. Abajo: Sistema
Voclico.
Bilabial Labiodental Interdental Dental Alveolar Palatal Velar
Oclusiva Sorda /p/ /t/ /k/
Sonora /b/ /d/ /g/
Africada Sorda /t/
Sonora
Fricativa Sorda /f/ // /s/ /x/
Sonora //
Nasal Sonora /m/ /n/ //
Lateral Sonora /l/ //
Vibrante Sonora /r/ //
simple
multiple
Localizacin
Apertura Anterior Media Posterior
Cerrada /i/ /u/
Media /e/ /o/
Abierta /a/
algunos pases latinoamericanos como Colombia y Paraguay). Adems, algunos

fonemas tienen realizacin diferente segn la regin geogrfica. El sistema
incluye cinco vocales fonolgicas: /a/, /e/, /i/, /o/, /u/, 1618 consonantes y dos
semivocales: la [i] y la [u], que pueden funcionar como semivocales en una
posicin posnuclear de la slaba (Penny, 2002; Quilis, 1963).
Dentro de las caractersticas fonolgicas ms sobresalientes del espaol
estaran: (1) La diptongacin. El espaol posee una cantidad muy extensa de
diptongos que pueden alternar con las vocales (p. ej., poder puedo), en tanto
que tiende a no utilizar hiatos (acentos para disolver el diptongo; p. ej., da). (2)
En la evolucin del latn al espaol los fonemas oclusivos sordos (/p/, /t/, /k/) en
posicin intervoclica se hicieron sonoros (/b/, /d/, /g/), pero slo en los vocablos
vulgares (de ms alta frecuencia), en tanto que continuaron siendo sordos en
palabras derivadas (p. ej., abeja, apicultura); y los fonemas oclusivos sonoros se
fricativizaron (algunas veces estos fonemas oclusivos sonoros fricativizados reci-

ben el calificativo de aproximantes o espirantes) (, , ). (3) El espaol posee una
estructura silbica muy definida; en cada palabra, una de las silabas lleva un
acento distintivo, funcional dentro de la lengua.
Desde el punto de vista del ritmo de la lengua, el espaol pertenece al grupo de
las lenguas cronometradas por slabas (syllable-timed) (cada slaba se produce con
una duracin aproximadamente igual, a pesar de que cada palabra posee una
slaba acentuada prosdicamente) (Berg, 1991), en contraste con las lenguas como
el ingls, cronometradas por unidades acentuales (stress-timed) (las slabas pueden
tener diferente duracin, pero el tiempo entre dos conjuntos acentuales es similar),
el espaol posee una estructura acentual muy evidente.
Desde el punto de vista gramatical, el espaol es una lengua flexiva: utiliza
flexiones para indicar la relacin existente entre los elementos. Esto implica
cambios morfolgicos importantes en las palabras. Sin embargo, no existen
declinaciones (paradigmas de flexin segn el caso acusativo, locativo, geni-
tivo, etc. en que se usa), como as suceda en el latn. Pero tambin utiliza
preposiciones (tipo de morfema gramatical: a, ante, bajo, etc.) para indicar estas
relaciones entre los elementos (Seco,1988). En cuanto a su sintaxis, al igual que
otras lenguas romances, el orden de las palabras dentro de la oracin es flexible
(p. ej., la oracin, Juan ama a Mara, se puede expresar con mltiples rdenes sin
alterar su significado: a Mara ama Juan; Juan a Mara ama; a Mara Juan ama,
etc.), aunque claramente ms flexible que otras lenguas latinas.
El sistema de escritura del espaol presenta ciertas peculiaridades (Tabla 2). El
espaol posee un sistema de lectura transparente (es decir, existen reglas precisas
para la conversin de grafemas a fonemas), exceptuando: (1) la lectura de ciertas
palabras tomadas de otras lenguas (p. ej., souvenir, jeep, etc.), (2) algunas
irregularidades en la transcripcin de palabras tomadas de lenguas nativas amer-
icanas, evidente particularmente en Mxico (p. ej., Oaxaca); (3) ciertas formas
escritas que no corresponden a la pronunciacin actual de las palabras (p. ej.,
pelear); y (4) ciertos arcasmos en la escritura (p. ej., Mxico).
La ambigedad en el sistema de lectoescritura va en el sentido de la escritura,
ya que muchas palabras pueden potencialmente escribirse de diferentes formas
Tabla 2. Escritura: Fonemas a Grafemas (adaptado de Leal y Matute (2001)).

Numero Fonema Grafema (s) Ejemplo
1 /p/ P /pensar/
2 /t/ T /toldo/
3 /d/ D /dar/
4 /f/ F /fama/
5 / t/ CH / tosa/
6 /l/ L /lima/
7 /r/ R /klaro/
8 /m/ M /malo/
9 / / /pia/
10 /e/ E HE /edad/, /ermoso/
11 /a/ A HA /ala/, /ada/
12 /o/ O HO /oliba/, /onor/
13 /b/ BV /base/, /baso/

14 /g/ G GU /gato/, /giso/
15 /n/ N NN /nube/, /perene/
16 /y/ Y LL /yema/. /caye/
17 /R/ R RR /radio/, /ko ida/
18 /i/ I Y HI /bida/, /ielo/
19 /u/ /kulpa/, /pinguino/, /ueso/
20 /k/ C K QU /kosa/, /kilo/, /keso/
21 /x/ JG /kaxa/, /xiro/
22 /s/ C S Z SC PS /siklo/, /pasado/, /sarpar/, /irasible/, /sikiko/
23 // CZ /ielo/, /ankudo/
24 LL /ka e/
(reglas ortogrficas) (p. ej., /muxer/ podra escribirse MUJER o MUGER), aunque
slo hay una lectura posible en cualquier palabra o pseudopalabra (Tabla 3). Leal
y Matute (2001) encontraron que la relacin grafema fonema que se usa para leer
en espaol es mucho ms alta (.70) que la relacin fonema-grafema (.30) que usa
para escribirlo. En espaol es posible, entonces, encontrar heterografa
homofnica (pueden haber formas alternas para escribir la secuencia de fonemas
de una palabra), pero no existe heterofonia hologrfica (slo hay una forma
posible de leer una secuencia de letras). En la escritura es posible distinguir dos
tipos de errores: homfonos (referidos usualmente como errores de ortografa en
los que la pronunciacin de la palabra no cambia; p. ej., MUGER); y no
homfonos (referidos usualmente como errores de escritura; p. ej., NUJER). Los
errores no homfonos son debidos a cambios, adiciones o substituciones de letras
que alteran la representacin escrita de la palabra hablada y por lo tanto su
significado. El primer tipo de errores es muy frecuente, particularmente en
personas con niveles limitados de escolaridad, pero tambin se encuentran en
personas con niveles educativos universitarios (Ardila, Rosselli y Ostrosky-Solis,
1996).
No hay acuerdo sobre cul es la unidad de lectura en diferentes lenguas, pero
probablemente la unidad de lectura en espaol sea la slaba (Ardila, 1998).
Como se mencion antes el espaol posee una estructura silbica muy bien
Tabla 3. Lectura de Grafemas a Fonemas (adaptado de Leal y Matute (2001)).

Numero Grafema Fonema Ejemplo
1 A /a/ AROMA
2 B /b/ BOTELLA
3 D /d/ DENTRO
4 E /e/ ESTELA
5 F /f/ FESTA
6 I /i/ PILA
7 J /x/ JUGO
8 K /k/ KIOSKO
9 L /l/ LUZ
10 M /m/ MALO
11 N /n/ NADA
12 // CAA
13 O /o/ OTRO
14 S /s/ SALTAR
15 T /t/ TODO
16 U /u/ UNO
17 V /b/ VACA
18 Z /s/ o // ZAPATO
19 LL // o // LLANTO
20 NN /n/ PERENNE
21 RR // PERRO
22 SC /s/ ASCTICO
23 CH / t/ CHISTE
24 GU /g/ /u/ AGERO
25 QU /k/ QUISO
26 HA /a/ HASTA
27 HE /e/ HELADO
28 HI /i/ HIJO
29 HO /o/ HONDO
30 HU /u/ HUELE
31 Y /y/, /i/ YEMA, LEY
32 G /g/, /x/ GAS, GELATINA
33 C /s/, //, /k/ CIEN, CAMA
34 R /r/, // ARO, RATN
35 P /p/, PILA, PSIQUE
36 PS /s/, /ps/ PSICOLOGA, CLEPSIDRA
37 GU /g/, /gu/ GUITARRA, AGUANTAR
38 X /ks/ AXIOMA
definida (las slabas se construyen alrededor de una vocal o un diptongo) y desde

el punto de vista expresivo, el espaol es claramente un lenguaje cronometrado
por slabas (syllable-timed). Es fcil suponer que los hispanohablantes poseen una
conciencia silbica muy fuerte; distinguir las silabas en una palabra es conside-
rado (a diferencia de otras lenguas como el ingls) como una tarea con un nivel de
dificultad muy bajo, casi autoevidente, dada la produccin silbica de la lengua.
Tabla 4. Decisin lxica (reconocimiento de palabras) en espaol. Se presentan las

correlaciones entre tiempo de reaccin y frecuencia de la palabra, nmero de letras y
nmero de silabas en cada palabra (adaptado de Ardila et al. (1993)).
Frecuencia Nmero de letras Nmero de slabas
Correlacin .319 .247 .281
P< .0004 .007 .002
En espaol, la unidad ms larga que se requiere leer globalmente, solo con-

siderando la secuencia completa de letras para deducir la fonologa es la slaba, en
particular ciertas slabas compuestas por tres letras, tales como GUI y GUA.
Parecera entonces razonable sugerir que la lectura en espaol se lleva a cabo
utilizando una secuencia silbica (slaba por slaba). Es interesante que en tareas
de reconocimiento de palabras escritas (decisin lxica) el tiempo requerido para

tomar la decisin de si el estmulo es o no una palabra, est afectado tanto por la
frecuencia de las palabras (efecto de la frecuencia) como por su longitud. Ms
an, la correlacin con el nmero de slabas (longitud fonolgica) es mayor que la
correlacin con el nmero de letras en la palabra (Tabla 4). Esta observacin
apoya an ms el supuesto de que la unidad fundamental de lectura en espaol es
la slaba.
Los errores ortogrficos se encuentran en aquellos conjuntos silbicos en los
cuales son posibles formas alternas de escritura. Se pueden distinguir diferentes
complejos de escritura (p. ej., la representacin del fonema /k/, la representacin
del fonema /g/, etc.) (Ardila, 1998). En caso de patologa cerebral los errores
ortogrficos en espaol no necesariamente se asocian con afasia; en caso de
lesiones hemisfricas derechas es frecuente encontrar un incremento de los errores
ortogrficos en la escritura, lo cual podra sugerir que el uso de la ortografa
representa en alguna forma una aptitud visoperceptual para los sujetos hispanoha-
blantes (cmo se ve la palabra, escrita en una u otra forma) (Tabla 5).
La transparencia u opacidad del Sistema de escritura de una lengua es la
variable ms relevante en la determinacin de la dificultad del aprendizaje lector
en esa lengua (Ziegler y Goswami, 2005). Una lengua transparente es aquella en
la que existe una correspondencia entre los fonemas y los grafemas y por eso estas
lenguas son conocidas como consistentes. Es decir un grafema siempre se lee de
la misma forma Entre los prototipos de lenguas transparentes se incluyen el
italiano, el espaol, el ruso, y el finlands. Una lengua opaca como es el ingls
por el contrario presenta numerosas variaciones en la forma como se leen los
fonemas y por eso recibe tambin el nombre de lengua inconsistente. Se estima
que el aprendizaje de las habilidades de lectura es mucho ms rpido en espaol
que en ingles debido a las diferencias que existen en el sistema ortogrfico entre
estas dos lenguas (Goswami, 2002; Ziegler y Goswami, 2005). La rapidez en el
progreso lector en una lengua esta tambin marcado por el sistema ortogrfico de
esta (Rosselli y Matute, 2012). As por ejemplo los nios finlandeses que empie-
zan a leer a la edad de siete aos, pueden alcanzar un nivel de precisin en la
Tabla 5. Errores en la escritura de 95 palabras que incluan todas las decisiones

ortogrficas posibles (35 nios del tercer grado de educacin primaria, edad promedio
8.94 aos y 57 estudiantes universitarios, edad promedio 20.14 aos). A. Errores no
homfonos. B. Errores homfonos (adaptado de Ardila (1998)).
Nios Adultos
No homfonos (promedio por sujeto) Sustitucin de letras 1.06 0.05
Omisiones de letras 0.20 0.00
Intercambio de letras 0.00 0.00
Sustitucin de morfemas 0.03 0.00
Errores homfonos (% de sujetos) 0 errores 0.0% 8.8%
15 errores 2.9% 56.1%
610 errores 5.7% 29.8%
1115 errores 14.3% 1.7%
1620 errores 17.1% 3.5%
2125 errores 28.6% 0.0%

Ms de 25 errores 31.4 0.0
mecnica de la lectura (leer sin errores) de 90% despus de 10 semanas de

entrenamiento mientras que los nios ingleses que inician el aprendizaje de la
lectura a los cinco aos nicamente alcanza este nivel de precisin a los nueve o
10 aos (Goswami, Gombert y de Barrera, 1998).
El sistema ortogrfico tambin determina la prevalencia de los problemas de
aprendizaje de la lectura (Helmuth, 2001). Es as como la prevalencia de dislexia
en nios italianos de 10 aos de edad se estima entre un 1 a 2% mientras que en
EEUU esa prevalencia es de 5% (Paulesu et al., 2001) y rara vez se observa en
nios japoneses (Wydell, 2003). No conocemos un estudio de prevalencia de la
dislexia en un pas hispanohablante pero asumimos que puede ser semejante a la
encontrada en Italia.
A pesar de la variabilidad en la incidencia de dislexia hallada en diversas
lenguas, la dificultad que subyace a ella es semejante en todas las lenguas.
Jimnez y colaboradores han encontrado que los nios con dislexia en espaol
exhiben la misma dificultad en la segmentacin fonmica descrita en ingls
(Jimnez Gonzlez, 1997). Jimnez et al. (2005) examinaron los efectos de la
complejidad lingstica en nios hablantes del espaol y encontraron que aquellos
con problemas de aprendizaje de la lectura tenan un defecto en la conciencia
fonolgica semejante al que se ha descrito en nios con dislexia hablantes de otras
lenguas.
Trastornos adquiridos en la lectura y la escritura

Durante las ltimas dcadas del siglo XX se propuso un nuevo enfoque en el
estudio de los trastornos adquiridos en la lectura (alexias o dislexias adquiridas).
Los investigadores comenzaron a preguntarse sobre la naturaleza de los defectos
cognitivos y lingsticos responsables de las dificultades en la lectura (Beauvois y
Drouesen, 1981; Coltheart, 1980; Marshall y Newcombe, 1973). El inters un
poco cambi de los correlatos anatmicos a los mecanismos funcionales respon-

sables de las alexias (Ellis, 1993). Este nuevo enfoque cognitivo y lingstico
supuso el desarrollo de modelos de la lectura en individuos con un funciona-
miento cognoscitivo normal. Durante los aos siguientes se propusieron varios
modelos cognitivos parcial pero no totalmente coincidentes del proceso
normal de lectura en diferentes lenguas. En general, se propuso que la lectura,
luego de la identificacin inicial de la letras, puede realizarse a travs de dos rutas
diferentes: (1) la ruta directa: la palabra escrita se visualiza de manera global y
se asocia con representacin visual en el lxico que se tiene en la memoria; (2) la
ruta indirecta: la palabra escrita se transforma en una palabra hablada por medio
de un conjunto de reglas grafofonmicas y el significado de la palabra se logra a
travs de su mediacin fonolgica, en la misma forma en que se entiendo el
lenguaje oral.
Estos modelos cognitivos de la lectura se desarrollaron partiendo del ingls,

que tiene un sistema de lectura muy irregular y entrara dentro de las lenguas que
se consideran opacas. El sistema de escritura del ingls no se basa directamente en
el sistema fonolgico de la lengua; ms exactamente se podra describir como un
sistema intermedio entre los principios de lectura fonolgica y logogrfica
(Sampson, 1985). La lectura de palabras se realiza siguiendo algunos principios
de correspondencia grafema/fonema, y parcialmente siguiendo un reconocimiento
global, no necesariamente separable en letras individuales. Palabras como dog,
man, y pistol son palabras regulares (es decir, la correspondencia grafema/fonema
es regular) y su pronunciacin se puede deducir de su escritura. En tanto que
palabras como island, women, yacht, y knight son irregulares (es decir, su
pronunciacin no se puede deducir de su escritura). Palabras como through, (en
oposicin a thought o though) pueden leerse nicamente teniendo en cuenta el
conjunto de la palabra. Es entonces comprensible que para leer en ingls se hayan
propuesto dos sistemas diferentes de lectura (directo e indirecto o simplemente
fonogrfico y logogrfico) (Shallice y Warrington, 1980).
Al intentar aplicar estos modelos al espaol (y otras lenguas con sistemas
transparentes de lectura) surgieron dos puntos de vista: (1) se puede hacer una
aplicacin directa de estos modelos ya que las estrategias (vas) de la lectura son
las mismas en ingls y en espaol (Davies y Cuetos, 2005; Ferreres y Miravalles,
1995; Ferreres, Cuitio, y Olmedo, 2005; Iribarren, Jarema, y Lecours, 2001; Ruiz,
Ansaldo, y Lecours, 1994); (2) diferentes lenguas enfatizan diferentes estrategias de
lecturas, y en consecuencia, estos modelos no son directamente aplicables al
espaol (Ardila, 1998; Ardilla et al., 1996; Ardila, Rosselli, y Pinzn, 1989). La
comparacin de las caractersticas de alexias observadas a travs de diferentes
lenguas parece favorecer el segundo punto de vista (Karanth, 2003).
Una discusin importante se ha desarrollado alrededor de la existencia de errores
semnticos en la lectura de palabras en espaol (paralexias semnticas). Ardila
(1998) propuso que a pesar de que las paralexias semnticas (sustitucin al leer de
una palabra por otra frecuentemente con significado similar) son un fenmeno
comn en pacientes afsicos hablantes del ingls (Landis, Regard, Graves, y
Goodglass, 1983), en espaol representan un tipo de error infrecuente. Muy
probablemente, y partiendo de los casos reportados, las paralexias semnticas se

encuentran en un subgrupo muy especfico de pacientes afsicos: pacientes crnicos
con niveles educaciones altos, frecuentemente multilinges, y con un tipo de afasia
motora. En ingls, las paralexias semnticas son abundantes y se pueden encontrar
en diversos grupos de pacientes afsicos.
Conclusiones
El espaol posee ciertas caractersticas distintivas en su fonologa, lxico, morfo-
sintaxis y sistema de escritura que afectan la forma y rapidez con la que se
aprende a leer y la manera como se manifiestan los trastornos adquiridos en el
lenguaje escrito. Dado el sistema transparente que tiene el espaol, no parece
apropiada la aplicacin directa de modelos de lectura desarrollados en lenguas con
sistemas de escritura ms opacos e inconsistente . A una conclusin similar se ha

llegado al estudiar los trastornos en la lectura observado en otras lenguas con
sistemas de escrituras trasparentes, como el hindi (Karanth, 2003) y el serbo-
croata (Lukatela y Turvey, 1990).
References / Referencias
Ardila, A. (1998). Semantic paralexias in the Spanish Language. Aphasiology, 12, 885900.
doi:10.1080/02687039808249457
Ardila, A., Rosselli, M., & Lecours, A. R. (1993). Decisin lxica en sujetos hispano-
parlantes: Efecto de la frecuencia y la longitud. Montevideo, Uruguay: III Congreso
Latinoamericano de Neuropsicologa.
Ardila, A., Rosselli, M., & Ostrosky-Solis, F. (1996). Agraphia in the Spanish language.
Aphasiology, 10, 723739. doi:10.1080/02687039608248446
Ardila, A., Rosselli, M., & Pinzn, O. (1989). Alexia and agraphia in Spanish-speakers. In
A. Ardila & F. Ostrosky-Solis (Eds.), Brain organization of language and cognitive
processes (pp. 147175). New York, NY: Plenum Press.
Beauvois, M-F., & Drouesen, J. (1981). Lexical or orthographic agraphia. Brain, 104, 21
49. doi:10.1093/brain/104.1.21
Berg, T. (1991). Phonological processing in a syllable-timed language with pre-final stress:
Evidence from Spanish speech. Language and Cognitive Processes, 6, 265301.
doi:10.1080/01690969108406945
Coltheart, M. (1980). Deep dyslexia: A review of the syndrome. In M. Coltheart, K.
Patterson, & J. Marshall (Eds.), Deep dyslexia (pp. 326380). London: Routledge and
Kegan Paul.
Davies, R., & Cuetos, F. (2005). Acquired dyslexia in Spanish: A review and some
observations on a new case of deep dyslexia. Behavioural Neurology, 16, 85101.
doi:10.1155/2005/872181
Ellis, A. W. (1993). Reading, writing and dyslexia: A cognitive analysis. Hillsdale, IL:
Lawrence Erlbaum Associates.
Ferreres, A. R., Cuitio, M. M., & Olmedo, A. (2005). Acquired surface alexia in
Spanish: A case report. Behavioural Neurology, 16, 7184. doi:10.1155/2005/473407
Ferreres, A. R., & Miravalles, G. (1995). The production of semantic paralexias in a
Spanish-speaking aphasic. Brain and Language, 49, 153172. doi:10.1006/
brln.1995.1026
Goswami, U. (2002). Phonology, Reading development, and dyslexia: A cross-linguistic
perspective. Annals of Dyslexia, 52, 139163. doi:10.1007/s11881-002-0010-0
Goswami, U., Gombert, J. E., & de Barrera, L. F. (1998). Childrens orthographic

representations and linguistic transparency: Nonsense word reading in English,
French and Spanish. Applied Psycholinguistics, 19, 1952. doi:10.1017/
S0142716400010560
Helmuth, L. (2001). Dyslexia: Same brains, different languages. Science, 16 March 2001,
20642065.
Iribarren, I. C., Jarema, G., & Lecours, A. R. (2001). Two different dysgraphic syndromes
in a regular orthography, Spanish. Brain and Language, 77, 166175. doi:10.1006/
brln.2000.2418
Jimnez, J. E., Garca, E., Ortiz, R., HernndezValle, I., Guzmn, R., Rodrigo, M.,
Hernndez, S. (2005). Is the deficit in phonological awareness better explained in
terms of task differences or effects of syllable structure? Applied Psycholinguistics, 26,
267283. doi:10.1017/S0142716405050174
Jimnez Gonzlez, J. E. (1997). A reading-level match study of phonemic processes
underlying reading disabilities in a transparent orthography. Reading and Writing:
An Interdisciplinary Journal, 9, 2340. doi:10.1023/A:1007925424563
Karanth, P. (2003). Cross-linguistic study of acquired reading disorders. New York, NY:
Kluwer/ Plenum.
Landis, T., Regard, M., Graves, R., & Goodglass, H. (1983). Semantic paralexia: A release of
right hemispheric function from left hemispheric control? Neuropsychologia, 21, 359
364. doi:10.1016/0028-3932(83)90022-2
Leal, F., & Matute, E. (2001). La transparencia de los sistemas ortogrficos y la idea de
estrategias diferenciales de procesamiento de la lengua escrita. In G. Lpez Cruz & M.
C. Mora Leyva, Encuentro Internacional de lingstica en el Noroeste. Tomo III.
Hermosillo, Mxico: Editorial UniSon.
Lukatela, G., & Turvey, M. T. (1990). Automatic and pre-lexical computation of phonol-
ogy in visual word identification. European Journal of Cognitive Psychology, 2(4),
325343. doi:10.1080/09541449008406211
Marshall, J. C., & Newcombe, F. (1973). Patterns of paralexia: A psycholinguistic approach.
Journal of Psycholinguistic Research, 2, 175199. doi:10.1007/BF01067101
Paulesu, E., Demonet, J. F., Fazio, F., McCrory, E., Chanoine, V., Brunswick, N., Frith,
U. (2001). Dyslexia: Cultural diversity and biological unity. Science, 291, 21652167.
doi:10.1126/science.1057179
Penny, R. (2002). A history of Spanish language. Boston: Cambridge University Press.
Quilis, A. (1963). Fontica y fonologa del espaol. Madrid: Consejo Superior de
Investigaciones Cientficas.
Rosselli, M., & Matute, E. (2012). Importancia de los factores lingsticos y ambientales
en el diagnstico de dislexia. In E. Matute & S. Guajardo (Eds.), Dislexia: Definicin
e intervencin en hispanohablantes. Mxico: Editorial Manual Moderno.
Ruiz, A., Ansaldo, A. I., & Lecours, A. R. (1994). Two cases of deep dyslexia in
unilingual Hispanophone aphasics. Brain and Language, 46, 245256. doi:10.1006/
brln.1994.1015
Sampson, G. (1985). Writing systems. Stanford, CA: Stanford University Press.
Seco, R. (1988). Manual de gramtica espaola. Buenos Aires: Aguilar.
Shallice, T., & Warrington, E. K. (1980). Single and multiple components single dyslexic
syndromes. In M. Coltheart, K. Patterson, & J. Marshall (Eds.), Deep dyslexia (pp.
109145). London: Routledge and Kegan Paul.
Wydell, T. N. (2003). Dyslexia in Japanese and the hypothesis of granularity and
transparency. In G. Nata (Ed.), Dyslexia in different languages: Cross-linguistic
comparisons (pp. 255276). London: Whurr Publishers, Ltd.
Ziegler, J. C., & Goswami, U. (2005). Reading acquisition, developmental dyslexia and
skilled reading across languages: A psycholinguistic grain size theory. Psychological
Bulletin, 131, 329. doi:10.1037/0033-2909.131.1.3
Neuropsychology Review pp628-nerv-452511 November 23, 2002 8:23 Style file version sep 03, 2002
Neuropsychology Review, Vol. 12, No. 4, December 2002 (

C 2002)
Acalculia and Dyscalculia
Alfredo Ardila1,3 and Monica Rosselli2
Even though it is generally recognized that calculation ability represents a most important type of
cognition, there is a significant paucity in the study of acalculia. In this paper the historical evolution
of calculation abilities in humankind and the appearance of numerical concepts in child development
are reviewed. Developmental calculation disturbances (developmental dyscalculia) are analyzed. It is
proposed that calculation ability represents a multifactor skill, including verbal, spatial, memory, body
knowledge, and executive function abilities. A general distinction between primary and secondary
acalculias is presented, and different types of acquired calculation disturbances are analyzed. The
association between acalculia and aphasia, apraxia and dementia is further considered, and special
mention to the so-called Gerstmann syndrome is made. A model for the neuropsychological assessment
of numerical abilities is proposed, and some general guidelines for the rehabilitation of calculation
disturbances are presented.
KEY WORDS: acalculia; dyscalculia; numerical knowledge; calculation ability.
INTRODUCTION Acalculia is frequently mentioned in neurological

and neuropsychological clinical reports, but research di-
Calculation ability represents an extremely complex rected specifically to the analysis of acalculia is rather
cognitive process. It has been understood to represent a limited. During 199099, there were 83 Medline and 56
multifactor skill, including verbal, spatial, memory, and PsychInfo entries for acalculia/dyscalculia as key word.
executive function abilities (Ardila et al., 1998). Cal- The word acalculia or dyscalculias was included in the
culation ability is quite frequently impaired in cases of journal paper or general publication titles 31 and 13 times
focal brain pathology (Ardila and Rosselli, 1992; respectively according to these two databases during this
Grafman, 1988; Harvey et al., 1993; Hecaen 10-year period (Table 1).
et al., 1961; Rosselli and Ardila, 1989) and dementia The same applies to testing for calculation abili-
(Deloche et al., 1995; Grafman et al., 1989; Parlatto ties. Tests for calculation abilities are always included
et al., 1992; Rosselli and Ardila 1998). The loss of in the psychological or neuropsychological evaluation of
the ability to perform calculation tasks resulting from cognition. Calculation abilities are included when test-
a cerebral pathology is known as acalculia or ac- ing for general intelligence (e.g., WAIS-III; Wechsler,
quired dyscalculia. Acalculia has been defined as an 1997) and most neuropsychological assessment proce-
acquired disturbance in computational ability (Loring, dures worldwide include the assessment of calculation
1999). The developmental defect in the acquisition abilities. Moreover, they are frequently included in the
of numerical abilities, on the other hand, is usually Mini-Mental State Examination (MMSE) (Folstein et al.,
referred to as developmental dyscalculia (DD) or 1975) and other brief neuropsychological assessment pro-
dyscalculia. cedures (e.g., Neuropsi; Ostrosky et al., 1997). But, a spe-
cific standardized test battery with norms for acalculia is
hardly found (Deloche et al., 1994).
1 Department of Communication Sciences and Disorders, Florida Inter- Thus, acalculia is in a somewhat peculiar position
national University, Miami, Florida.
2 Department of Psychology, Florida Atlantic University, Davie, Florida. amid the cognitive disturbances encountered in cases of
3 To whom correspondence should be addressed at 12230 NW 8 Street, brain pathology. Although there is a general consensus
Miami, Florida 33182. E-mail: alfredoardila@cs.com that calculation ability represents an extremely important
179
1040-7308/02/1200-0179
C 2002 Plenum Publishing Corporation
180 Ardila and Rosselli
Table 1. Number of Entries of Some Neuropsychological Syndromes Berger (1926) introduced the distinction between pri-
Found in Medline and Psychinfo During 199099 mary and secondary acalculia. Primary or pure acalcu-
Medline PsychInfo lia corresponds to the loss of numerical concepts and to
the inability to understand or execute basic arithmetical
As key In the As key In the
word title word title
operations. Secondary acalculia refers to the defect in cal-
culation derived from other cognitive deficits (e.g., mem-
Dementia 18,180 5,580 4,269 1,978 ory, language, etc.). This distinction became particularly
Amnesia 2,528 713 965 318 influential, and then was increasing recognition that cal-
Aphasia 2,344 632 972 417 culation disturbances could be associated with and depen-
Alexia/dyslexia 1,278 409 601 317
Apraxia 665 240 222 91
dent upon other cognitive defects, such as aphasia, alexia,
Agnosia 505 104 146 61 and agraphia. The polemic, however, has revolved around
Agraphia/dysgraphia 266 84 123 39 the possible existence of a primary acalculia, because
Acalculia/dyscalculia 88 31 54 13 some authors have questioned the existence of acalculia as
an independent cognitive deficit (Collington et al., 1977;
Goldstein, 1948).
type of cognition, and calculation abilities are tested in vir- Gerstmann (1940) proposed that primary acalculia
tually any psychological and neuropsychological assess- is associated with agraphia, rightleft disorientation, and
ment procedure, research on acalculia is quite limited. digital agnosia, conforming a single brain syndrome that
since then has been known as Gerstmann syndrome.
Historical Note Neuroimaging methods have shown a correlation between
the Gerstmann syndrome and left posterior parietal in-
Henschen (1925) proposed the term acalculia juries (Mazzoni et al., 1990).
(akalkulia). He then defined acalculia as the impairment in Lindquist (1936) distinguished different types of
computational skills resulting from brain injury. Nonethe- acalculia associated with lesions in different brain areas.
less, before Henschen some mention of calculation dis- He conjectured that calculation disturbances are not ho-
turbances associated with brain damage is found in the mogeneous, and in consequence, acalculia subtypes may
literature. In most cases, those calculation disturbances be distinguished. As a result, several classifications of
were interpreted as sequelae of the language impairment acalculias have been presented (e.g., Ardila and Rosselli,
(aphasia). 1990; Grafman, 1988; Grafman et al., 1982; Hecaen et al.,
Lewandowsky and Stadelmann published the first de- 1961; Lindquist, 1936; Luria, 1973), and different patterns
tailed report of a patient suffering from calculation distur- of errors have been described in patients with right and
bances in 1908. Their patient presented with focal brain left hemispheric injuries (Levin et al., 1993; Rosselli and
damage associated with right homonymous hemianopia, Ardila, 1989).
and had significant difficulties in written and mental cal- Boller and Grafman (1983) claim that calculation
culations. The authors further reported that the patient had abilities can be impaired as a result of different types
impairments in reading arithmetical signs, even though the of defects. They believe that calculation skills can be al-
patient could carry out arithmetical operations. This paper tered as a result of (1) inability to appreciate the signifi-
represents a landmark in the development of the concept cance of the names of numbers, (2) visualspatial defects
of acalculia, because it considers that calculation disorders that interfere with the spatial organization of the numbers
are different and dissociated from language disturbances. and the mechanical aspects of the operations, (3) inabil-
Henschen (1925) reviewed 305 cases in the litera- ity to remember mathematical facts and to appropriately
ture that reported calculation disturbances associated with use them, and (4) defects in mathematical thought and in
brain damage, and 67 of his own patients. He identified the comprehension of the underlying operations. Interest-
those patients presenting calculation disturbances without ingly, occasionally brain damage may result in a relatively
evident language impairments, and proposed an anatomi- restricted disorder in performing arithmetical operations,
cal substrate for arithmetical operations, different from but for instance, a limited alexia for arithmetical signs (Ferro
close to the anatomical substrate for language and musi- and Botelho, 1980), or a specific deficit for arithmetical
cal ability. The third frontal convolution was suggested procedures (Semenza, 1988), without additional calcula-
to represent the center for the pronunciation of numbers. tion disturbances.
The angular gyrus and the fissure interparietalis were pro- Calculation ability implies the use of numerical con-
posed as the brain areas participating in number reading, cepts. The concept of numbers can be associated with the
and the angular gyrus was regarded as the brain structure presence of at least four factors: (1) immediate representa-
responsible for writing numbers. tion of quantity, implicit in the number; (2) understanding
Acalculia and Dyscalculia 181
the numerical position within the system of other numer- the neuropsychological domain, however, very little re-
ical symbols (i.e., its position in the series of digits and search indeed has been carried out to explore the rela-
its place in class); (3) understanding the relationships be- tionship between mathematical test performance and per-
tween a number and other numbers; and (d) understand- formance on other cognitive tests. Ardila et al. (1998)
ing the relationships between numerical symbols and their administered a comprehensive neuropsychological test
verbal representations (Tsvetkova, 1996). battery to a 300-participant sample, aged 1725 years. All
McCloskey et al. (1985, 1986, 1991a,b) and were right-handed male university students. The battery
McCloskey and Caramazza (1987) proposed a cognitive included some basic psychological and neuropsycholog-
model regarding the processing of numbers and the rela- ical tests directed to assess language, calculation abili-
tionship of arithmetical operations. This model includes ties, spatial cognition, praxis abilities, memory, percep-
a distinction between the processing system of numbers tual abilities, and executive functions. The following tests
(an understanding mechanism and the production of num- were included: (1) Auditory Recognition (recognition of
bers), in addition to the numerical calculation system that songs and the Seashore Rhythm). (2) Verbal Fluency
includes the necessary processing components to accom- (phonologic and semantic). (3) Nonverbal Fluency. (4) Se-
plish mathematical operations. In the event of brain in- rial Verbal Learning (5) Finger Tapping Test (FTT) (left
jury, these components can be disassociated (Dagenbach hand and right hand). (6) The ReyOsterrieth Complex
and McCloskey, 1992; Pesenti et al., 1994). The princi- Figure (ROCF) (copy and immediate recall). (7) Ratcliffs
ples (multiplication tables), rules (N 0 = 0), and pro- Mental Rotation Test. (8) Arithmetical Abilities. Two dif-
cedures (multiplication proceeds from right to left) form ferent tests were used: (a) Mental Arithmetical Opera-
part of the numerical calculation system. Errors of calcu- tions (two additions, two subtractions, two multiplica-
lation observed in patients with brain injury and in normal tions, and two divisions) and (b) arithmetical problems.
participants can result from inappropriate recall of princi- Sixteen arithmetical problems were orally presented. The
ples, inadequate use of rules, and/or errors in procedures. participants were allowed to use pencil and paper if so
Cognitive modeling has helped to establish similarities desired. (9). Localization of cities on a map. (10) Orthog-
between acquired acalculias and DDs (Temple, 1991). raphy Test. (11) Perceptual Recognition (similarities be-
Clark and Campbell (1991) have presented a spe- tween two figures, differences between two figures and
cific integrated theory of calculation abilities. This theory hidden figures). (12) Reading Speed. (13) Wechsler Adult
presumes that visuospatial, verbal, and other modality- Intelligence Scale (WAIS). (14) Wechsler Memory Scale
specific number codes are associatively connected as an (WMS). And (15). Wisconsin Card Sorting Test (WSCT).
encoding complex and that different facets of number pro- Forty-one different scores were calculated.
cessing generally involve common rather than indepen- Four different tests were considered in the analysis of
dent, processes. This point of view emphasizes the par- the calculation abilities (Mental arithmetical operations,
ticipation of multiple components in calculation. It seems Arithmetical problems, WAIS Arithmetic, and WAIS Dig-
evident that normal calculation ability requires verbal, vi- its subtests). Numerical ability tests turned out to present
suospatial, and other fundamental cognitive skills. Calcu- a notably complex correlation system. Mental arithmetic
lation impairments, as a matter of fact, can be observed significantly correlated with 25 (out of 41) test scores, and
in cases of a wide diversity of brain disturbances: Left or Arithmetical problems with 17 test scores. WAIS Arith-
right hemisphere, frontal, parietal, temporal, and occipital. metic subtest correlated with 15 test scores, and WAIS
Moreover, subcortical lesions can also result in some cal- Digits correlated with 7 test scores. Some correlations
culation defects (Dehaene and Cohen, 1997). Almost any were quite understandable (e.g., different mathematical
type of brain pathology may produce difficulties on calcu- test scores are highly intercorrelated), whereas other cor-
lation tests, even though the specific pattern of difficulties relations were rather unexpected (e.g., arithmetical ability
can be different (Rosselli and Ardila, 1989). tests highly correlate with the Orthography test). Main cor-
relations were observed with verbal memory, visuospatial,
visuoperceptual, language, and visuoconstructive ability
Calculation Ability: One or Several Abilities? tests (Table 2). It may therefore be assumed that arith-
The Question of Modularity metical ability is associated with and depends upon some
verbal, visuoperceptual, visuospatial, and memory abili-
Calculating has been identified as a concept and goal ties. Consequently, it is not surprising that there is such a
formation cognitive skill (Mandell et al., 1994). Calcu- wide variety of calculation disturbances observed in brain
lation ability under normal circumstances requires not pathology. Calculation ability disturbances are even as-
only the comprehension of numerical concepts, but also sociated with body knowledge disturbances (autotopag-
that of conceptual abilities and other cognitive skills. In nosia, finger agnosia) (Gerstmann, 1940).
Table 2. Correlations over.15 (p < .01) Between Different Testsa Table 2. (Continued )
Test Correlates with r Phonologic Verbal Fluency .23

WAIS: Arithmetic .22
Mental arithmetic Arithmetical problems .49 Arithmetical problems .20
Orthography test .37 Orthography test .17
WAIS: Arithmetic .35 FTT: Left hand .16
WMS: Associative Learning .29 WAIS: Similarities .16
Ratcliffs test .25
WMS: Logical Memory .26 a Mental Arithmetic, Arithmetical Problems, WAIS: Arithmetic, and Dig-
Perceptual speed: Similarities .25 its Subtests (adapted from Ardila et al., 1998a).
WAIS: Information .24
WAIS: Similarities .23
WAIS: Comprehension .23 A factor analysis with varimax rotation disclosed
WAIS: Digits .23 five factors accounting for 63.6% of the total variance.
Phonologic Verbal Fluency .22 Interestingly, none of these factors was a calculation or
WAIS: Vocabulary .21 numerical factor. The first factor, accounting for over
Localization of cities on a map .20
one fourth of the variance, was clearly verbal ability.
WAIS: Picture completion .20
WAIS: Picture arrangement .20 It was best tested with the Similarities, Information,
Perceptual speed: differences .19 Vocabulary, and Comprehension subtests from the WAIS.
Seashore Rhythm Test .18 This result reinforces the assumption that Arithmetic and
Perceptual speed: Hidden .18 Digits subtests are not purely verbal subtests. Factor II
WAIS: Digitsymbol .18
was a perceptual or nonverbal factor, whereas factor III
WAIS: Block design .17
WAIS: Object assembly .16 was simply a WCST factor, nonsignificantly correlated
Reading speed .16 with other test scores. Minor correlations were observed
WMS: Orientation .16 with the WMS Information and Orientation subtests.
WMS: Mental Control .15 Factor IV was a type of fine motor factor or (motor and
Solving arithmetical Mental arithmetic .49
verbal) fluency factor. Factor V was a memory, especially
problems WAIS: Arithmetic .46
Orthography test .32 verbal memory, factor (Table 3).
WAIS: Block design .31 General intelligence (Full Scale IQ) as understood
WAIS: Information .28 in the WAIS was best associated with the mathematical
Ratcliffs test .26 ability tests and the orthography test (Table 4). These cor-
WAIS: Digitsymbol .26
relations were particularly high and significant. The asso-
WAIS: Comprehension .24
WMS: Mental Control .24 ciation with some perceptual tests was somehow lower,
WMS: Logical Memory .23 yet strongly significant. All of these tests (mathemati-
WAIS: Similarities .22 cal, orthographic, and perceptual tests) presented a very
WAIS: Picture Arrangement .21 complex intercorrelation system with a wide variety of
WAIS: Digits .20
other psychological and neuropsychological tests. Con-
WAIS: Vocabulary .19
Perceptual speed: Similarities .18 sequently, they may be measuring a diversity of cogni-
Perceptual speed: differences .18 tive abilities. Still, mathematical abilities represented the
WAIS: Picture completion .18 best predictors of the general intelligence according to
WAIS: Arithmetic Arithmetical problems .46 the psychometric concept of intelligence. Interestingly, no
WAIS: Information .40
Mental Arithmetic .35
WAIS: Comprehension .32
WAIS: Vocabulary .31 Table 3. Correlations Between the Different Calculation Tests and the
WAIS: Picture completion .26 Five Factors (Adapted from Ardila et al., 1998a)
WAIS: Digits .22
Factors
WAIS: Block design .21
Phonologic Verbal Fluency .21 I II III IV V
Localization of cities on a map .21 Tests Verbal Perceptual WCST Fluency Memory
WAIS: Similarities .18
Semantic Verbal Fluency .17 Mental arithmetic .40 .35 .10 .03 .19
Orthography test .17 Arithmetical problems .37 .40 .03 .07 .23
Perceptual speed: Hidden .16 WAIS: Arithmetic .44 .28 .05 .15 .19
?> WAIS: Picture arrangement .15 WAIS: Digits .30 .02 .17 .18 .08
WAIS: Digits Mental arithmetic .23
Table 4. Correlations over.15 (p < .01) Between Verbal IQ, Performance nitive deficits in cases of dementia (Roudier et al., 1991).
IQ, and Full Scale IQ with Different Test Scores (Adapted from Ardila Noteworthy, using functional magnetic resonance imaging
et al., 1998a)
(fMRI) and positron emission tomography (PET) while
Correlates with r performing arithmetical operations, a complex pattern of
activity has been demonstrated. Most active areas include
Full Scale IQ Arithmetical problems .40 the left prefrontal areas and the posterior superior temporal
Orthography test .36
Mental arithmetic .35
gyrus (Burbaud et al., 1995; Sakurai et al., 1996).
Perceptual speed: Similarities .32
Seashore Rhythm Test .28 HISTORICAL DEVELOPMENT
Perceptual speed: Hidden .27
Phonologic verbal fluency .25
OF CALCULATION ABILITIES
ROCF: Memory .24
Localization of cities .24 Calculation abilities have followed a long process,
Reading speed .20 since initial quantification of events and elements, to mod-
Semantic verbal fluency .19 ern algebra, geometry, and physics. Some rudimentary nu-
Ratcliffs test .16
Verbal IQ WAIS: Vocabulary .81
merical concepts are observed in animals and no question
WAIS: Comprehension .68 prehistorical man used some quantification. However, the
WAIS: Information .67 ability to represent quantities, the development of a nu-
WAIS: Similarities .59 merical system, and the use of arithmetical operations, is
WAIS: Arithmetic .50 found only in old civilizations.
Arithmetical problems .32
Phonologic verbal fluency .32
Mental arithmetic .31 Numerical Concepts in Animals
Orthography test .31
WAIS: Digits .28 The origin of mathematical concepts can be traced
Localization of cities .25
Semantic verbal fluency .18
to subhuman species. Throughout recent history differ-
Performance IQ Arithmetical problems .35 ent reports have argued that animals (horses, rats, dogs,
Perceptual speed: Similarities .33 chimpanzees, dolphins, and even birds) can use numerical
Seashore Rhythm Test .28 concepts and perform arithmetical operations.
Mental arithmetic .28
There is general agreement that some rudimentary
Perceptual speed: Hidden .28
ROCF: Memory .24 numerical concepts are observed in animals. These basic
Orthography test .24 numerical skills can be considered as the real origin of
Reading speed .20 the calculation abilities found in contemporary man. For
Ratcliffs test .17 instance, pigeons can be trained to pick a specific num-
ber of times on a board, and rats can be trained to press
a lever a certain amount of times to obtain food (Calpadi
single WAIS verbal or performance subtests, according to and Miller, 1988; Koehler, 1951; Mechner, 1958). It could
current results, could be considered a good predictor of be conjectured that pigeons and rats can count, at least up
Full Scale IQ. to a certain quantity: They can recognize how many times
In conclusion, calculation abilities are associated a motor actto pick on a board or to press a leverhas
with a diversity of other cognitive abilities, including ver- been repeated. If this behavior can or cannot be really in-
bal, perceptual, spatial, memory, and executive function terpreted as counting is nonetheless questionable. But it
abilities. In this regard, calculation ability represents a is observed, at least, after a long and painstaking training.
multifactor ability. To a significant extent, it reflects the Interestingly, these animal responses (to pick or to press
ability to manipulate acquired knowledge (Mandell et al., the lever) are not accurate but just approximate. In other
1994). The ability to use numerical information is highly words, when the rat is required to press the lever seven
correlated with general intellectual level. It is not sur- times, the rat presses it about seven times (i.e., five, six,
prising that calculation disturbances are heterogeneous, seven, eight times) As Dehaene (1997) emphasizes, for
and can be observed in cases of brain pathology of quite an animal 5 + 5 does not make 10, but only about 10.
different locations. Virtually, any cognitive defect may re- According to him, such fuzziness in the internal repre-
sult in impairments in calculation abilities. By the same sentation of numbers prevents the emergence of exact nu-
token, mental calculation impairments have been demon- merical arithmetical knowledge in animals. Using highly
strated to represent an important factor in predicting cog- controlled and sophisticated designs, it has been pointed
out that chimpanzees can even use and add simple nu- The immediate recognition of certain small quantities
merical fractions (e.g., 1/2 + 1/4 = 3/4) (Woodruff and is found not only in animals but also in small children. Ani-
Premack, 1981). These observations support the assump- mals and children can readily distinguish one, two, or three
tion that some quantity concepts can be found in different objects (Fuson, 1988; Wynn, 1990, 1992). Beyond this
animals. point, however, errors are observed. Oneness, twoness,
Counting (or rather, approximately counting) motor and threeness seemingly are basic perceptual qualities that
responses is just a motor act as it is walking or running. our brain can distinguish and process without counting. It
Counting lever pressing is not so different to estimate can be conjectured that when the prehistoric humans be-
the effort (e.g., number of steps or general motor activity) gan to speak, they might have been able to name only the
required going from one point to another. Counting could numbers 1, 2, and perhaps 3, corresponding to specific
be linked to some motor and propioceptive information. perceptions. To name them was probably no more diffi-
Not only chimpanzees but also rats and many other cult than to name any other sensory attribute (Dehaene,
animals can distinguish numerosity (i.e., global quantifi- 1997). Noteworthy, all the world languages can count up
cation): they prefer the bowl containing the larger num- to three, even though three may represent many, sev-
ber of nutritive elements (chocolates, pellets, or whatever) eral, or a lot (Hurford, 1987). One is obviously the
when selecting between two bowls containing different unit, the individual (the speaker may also be one). Two
amounts (Davis and Perusse, 1988). It may be conjec- conveys the meaning of another (e.g., in English and
tured that global quantification (numerosity perception) also in Spanish, second is related with the verb to sec-
and counting (at least the approximate counting of mo- ond and the adjective secondary). Three may be a
tor responses) represent kind of basic calculation abilities residual form from a lot, beyond the others, many
found at the animal level. Rats prefer the bowel contain- (e.g., troppo that in Italian means too much is seem-
ing 10 pellets to the bowel containing 20 pellets; however, ingly related with the word three -tre-). In the original
they do not prefer the bowel containing 20 pellets to the European language, spoken perhaps some 15,00020,000
bowel containing 21 pellets. Obviously, numerosity per- years ago, apparently the only numbers were one, one
ception is related to size and shape of the visual image and another (two), and a lot, several, many (three)
projected to the retina. It can be assumed that 20 pellets in (Dehaene, 1997). In some languages, two different plurals
a bowel result in a larger and more complex retinal image are found: a plural for small quantities (usually 2, some-
than do 10 pellets. But the visual image corresponding to times 3 and 4) and a second plural for large quantities.
20 pellets is difficult to distinguish from the visual image Interestingly, in different world languages the word
corresponding to 21 pellets. one has not any apparent relationship with the word first,
and the word two is not related either with the word sec-
ond. Three may sometimes but not always holds some
Calculation Abilities in Prehistorical Man relationship with third. Beyond three, ordinals are clearly
associated with cardinal numbers (Table 5). The conclu-
Chimpanzees are capable of various forms of nu- sion is obvious: for small quantities, cardinals and ordinals
merical competence including some correspondence con- must have a different origin. For larger quantities, ordinal
structions for low quantities (Davis and Perusse, 1988; numbers are derived from cardinals. As a matter of fact,
Premack, 1976). Most likely, these numerical abilities also one/first and two/second correspond to different concep-
existed in prehistorical man. Homo sapiens antecessors tual categories.
may have been capable to use correspondence construc- It may be speculated that for the prehistorical man
tions in some social activities such as food sharing. It the first person and the second person in a line (or the first
has been proposed that Homo habilis (ancestor of Homo animal and the second animal during hunting or whatever)
erectus, living about 2.5 million years ago) required to do not seem to be related with numbers 1 and 2. For small
use correspondence constructions when butchering large children first has the meaning of initial (e.g., I go
animal carcasses (Parker and Gibson, 1979). Distribut- first) whereas second is related to later or after
ing pieces of a divided whole (e.g., a pray) into equal (you go second). They have a temporal and also spatial
parts required the ability to construct one-to-one corre- meaning, but not an evident cardinal meaning. The associ-
spondences. Probably, Paleolithic man was able to match ations between one and first, and between two and
the number of objects in different groups. And eventually, second seem a relatively advanced process in the de-
the number of objects in a collection with the number of velopment of numerical concepts. That is, the numerical
items in some external cue system, for example, fingers or meaning of first and second seems to appear after its
pebbles (incidentally, calculus means pebbles). temporal and spatial meaning. The association between
Table 5. Cardinal and Ordinal Numbers in Different Languages
English Spanish Russian Greek Persian Arabic Hindi Aymaraa Ibob
One Uno Odin Ena Yek Wahid Ek Maya Nbu

First Primero Pervie Proto Aval Awal Pahla Nairankiri Onye-nbu
Two Dos Dva Dio Dou Ethnaim Do Phaya Ibua
Second Segundo Vtoroi Deftero Douvoum Thani Dusra Payairi Onye-ibua
Three Tres Tri Tria Seh Thalatha Tin Kimsa Ito
Third Tercero Treti Trito Sevoum Thalith Tisra Kimsairi Nke-ito
Four Cuatro Cheterie Tesera Chahaar Arrbaa Char Pusi Ano
Fourth Cuarto Chetviorti Tetarto Chaharoum Rabiek Chautha Pusiiri Nke-ano
Five Cinco Piat Pente Pang Khamsa Panch Pheschka Ise
Fifth Quinto Piati Pemto Panjoum Khamis Panchvan Pheskairi Nke-ise
Six Seis Shest Exi Shash Sitta Chha Sojjta Isi
Sixth Sexto Shestoi Ekto Shashoom Sadis Chhatha Sojjtairi Nke-isi
Seven Siete Siem Epta Haft Sabaa Sat Pakallko Isaa
Seventh Septimo Sidmoi Evthomo Haftoom Sabieh Satvan Pakallkoiri Nke-isaa
Eight Ocho Vosiem Octo Hasht Thamania Ath Kimsakallko Asato
Eighth Octavo Vosmoi Ogdoo Hashtoom Thamin Athvan Kimsakallkiri Nke-asato
Nine Nueve Dievit Enea Nouh Tisaa Nau Llatunca Itonu
Ninth Noveno Diviati Enato Houhum Tasih Nauvan Llatuncairi Mke-Itonu
Ten Diez Diesit Deka Dah Ashra Das Tunca Iri
Tenth Decimo Disiati Dekato Dahoom Asher Dasvan Tuncairi Mke-iri
a Ameridian language spoken in Bolivia.

b Ibo: Eastern Nigeria.
ordinals and cardinals becomes evident only for larger misrepresentation on Draw-a-Person Test (Pontius, 1985,
quantities (more than three) and seems to represent a later 1989). This observation has been confirmed in different
acquisition in human evolution and complexization of nu- cultural groups.
merical concepts. In many contemporary languages (e.g., Taking a typical example as an illustration, the
Uitoto language, spoken in South America) there are not Colombian Sikuani Amazonian jungle Indians count in
ordinal numbers. For the first Uitoto language uses the the following way: the person (a child when learning to
beginning; to express second the word another is used. count or an adult when counting) places his/her left hand
Arithmetical abilities are clearly related with count- in supination to point the number 1, the right index points
ing. Counting, not simply recording the approximate to the left little finger, which is then bent (Queixalos, 1985,
amount of motor responses required to obtain a reinforce- 1989). The order followed in counting is always from the
ment, but to say a series of number words that correspond little finger to the index. To point to the number 5, the hand
to a collection of objects, is relatively recent in human his- is turned and the fingers opened; for 6, both thumbs are
tory. Counting is also relatively late in child development. joined, the left fingers are closed, and the right opened;
In human history as well as in child development (Hitch they are opened one after the other for 7, 8, 9, and 10.
et al., 1987) counting using number words begins with Between 11 and 20, the head points to the feet and the
sequencing the fingers (i.e., using a correspondence consequence is reinitiated. The lexicon used is as follows:
struction). The name of the finger and the corresponding
number can be represented using the very same word (that 1. kae (the unit, one)
means, the very same word is used for naming the thumb 2. aniha-behe (a pair, both)
and the number 1; the very same word is used to name the 3. akueyabi
index finger and the number 2, etc.). The fingers [and toes; 4. penayanatsi (accompanied; i.e., the fingers to-
as a matter of fact, many languages (e.g., Spanish), use a gether)
single word (dedo) to name the fingers and toes] are usu- 5. kae-kabe (one hand)
ally sequenced in a particular order. This strategy repre- Numbers from 6 to 9 are formed with one hand and
sents a basic procedure found in different ancient and con- (a certain number) of fingers. Ten becomes two hands.
temporary, cultures around the world (Cauty, 1984; Levy-
Bruhl, 1947). Interestingly, it has been demonstrated that 6. kae-kabe kae-kabesito-nua (one hand and one
children with low arithmetical skills also present a finger finger)
7. kae-kabe aniha-kabesito-behe (one hand and a Digit (from digitus, Latin) in English or Spanish
pair of finger) (degito) means not only number but also finger. The cor-
10. aniha-kabe-behe (two hands) respondence construction between numbers and fingers
is evident. Latin number notation was originally Etruscan
Two hands is maintained between 10 and 20. Toes (Turner, 1984), and referred (as everywhere) to the fingers.
(taxawusito) are added between 11 and 14, and one foot One, two, and three were written simply making vertical
(kae-taxu) is used in 15. Twenty is two hands together strokes. In four the Latin system recurs to a simplification.
with two feet. Originally, four was written IIII, but later on it became IV.
11. aniha-kabe-behe kae-taxuwusito (two hands and Five (V) represented the whole hand with the arm bent
one toe) (i.e., all the fingers of the hand), and 10 (X) the two arms
12. aniha-kabe-behe aniha-tuxuwusito-behe (two crossed.
hands and two toes) From a neuropsychological perspective, the strong
15. aniha-kabe-behe kae-taxu-behe (two hands and relationship existing between numerical knowledge, fin-
one foot) ger gnosis, and even lateral (rightleft) knowledge be-
16. aniha-kae-behe kae-taxu-behe kae-taxuwusito comes understandable. Finger agnosia (and probably
(two hands, one foot, and one toe) rightleft discrimination disturbances) could be inter-
20. aniha-kabe-behe aniha-taxu-behe (two hands preted as a restricted form of autotopagnosia (Ardila,
and two feet) 1993). It is not surprising to find that a decimal (or vigecial)
system has been most often developed. Simultaneously or
Fingers are named according to their order in count- very close in time, decimal systems appeared in different
ing (as mentioned previously, counting begins always with countries (Sumer, Egypt, India, and Creta). Different sym-
the little finger of the left hand). Sikuani language pos- bols were used to represent 1, 10, 100, and 1,000 (Childe,
sesses number words only up to three (kae, aniha-behe, 1936).
and akueyabi). Four (penayanatsi = accompanied, to- There is, however, an interesting and intriguing ex-
gether) represents a correspondence construction. Strictly ception: Sumerian and later Babylonians (about 2,000 BC)
speaking, Sikuani language counts only up to three. From developed not only a decimal system but also a sexagesi-
4 to 20, they use a correspondence construction, not really mal system: a symbol represented 60 or any 60-multiple,
counting; and for higher quantities, they recur to a global and other different symbol represented the number 10 and
quantification. any 10-multiple. Thus, for example, the number 173 was
Sometimes not only the fingers (and toes) but also then represented: 2 60 (the symbol for 60 repeated
other body segments may be used in counting: the wrist, twice) + 5 10 (the symbol for 10 repeated five times) +
the shoulders, the knees, and so on (Cauty, 1984; Levy- 3 (a symbol for units repeated three times). A base of 60
Bruhl, 1947). But sequencing the fingers (and toes) rep- has remained for some contemporary time measures (e.g.,
resents the most universal procedure in counting. Some hours, minutes, seconds). Twelve is also frequently main-
languages (e.g., some Mayan dialects and Greenland tained as a second-order unit (e.g., a dozen). Evidently,
Eskimo) use the same word to denote the number 20 (i.e., 60 results from five times 12. Five obviously is one
all the fingers and all the toes) and a person. hand, and the question becomes where 12 comes from.
In different Amerindian languages, for higher than Which are the two additional units? It might be speculated
10 or 20 figures, most often many is used (global quan- that 12 means the 10 fingers plus the 2 feet or even the
tification principle) (Cauty, 1984). Or, they can recur to 2 elbows or the 2 shoulders or the 2 knees (individual-
other peoples hands (correspondence construction) (e.g., ity of components is easier to appreciate in the hands
35 might be something like my two hands, my two feet, than in the feet). But this is only speculation, although
my fathers two hands, my fathers one foot). As men- feasible according to our knowledge about counting pro-
tioned, 20 sometimes becomes something like one per- cedures used in different cultural groups (Levy-Bruhl,
son, a sort of higher order numeral. It is interesting to 1947).
note that in some contemporary languages (like English It is interesting to note that the Maya Indians devel-
and Spanish) one means the unit, but it is also used as a oped a similar system, but having 20 as a base (Leon-
sort of indefinite personal pronoun. In English and Span- Portilla, 1986). They used different symbols to represent
ish we can also use one as synonymous of myself. 20, 400 (20 20), and 8,000 (20 20 20) (Cauty,
Twenty is found to be the base number in the Mayas nu- 1984).
merical system (Cauty, 1984; Swadesh, 1967). In many So, reviewing the history of numerical concepts, it is
contemporary languages, a 10 and/or 20 base can be found that world languages developed a base 10 (10 fin-
evident. gers) or 20 (10 fingers plus 10 toes) or even 5 (5 fingers)
to group quantities. In some contemporary languages, for originally horizontal lines that became tied together by be-
example in French, a residual 20-base can be found (e.g., ing handwritten. This observation may be related with the
in French 80 can be four twenties). In many contempo- inborn ability to perceptually recognize up to three ele-
rary languages, different words are used between 1 and ments. Beyond three, errors become progressively more
10. Between 10 and 20 the numerical systems usually be- likely. Perceptually distinguishing 8 and 9 is not so easy as
come irregular, unpredictable, and idiosyncratic. From 20 distinguishing between 2 and 3 strokes. The introduction
ahead, new numbers are formed simply with the words of a different representation for quantities over 3 was a
twenty plus one, twenty plus two, and so on. Some useful and practical simplification.
contemporary languages still use a 5-base in counting. For Not only the numerical system but also the measure
instance, in the Amerindian language Tanimuca in South units were developed departing from the body dimensions
America, they count up to five. Between 5 and 10, numbers (fingers, hands, arm, steps, etc.). This tendency to use the
are five one, five two, and so on. human body not only to count but also as measure units
is currently still observed in some contemporary measure
units (e.g., foot).
Further Developments of Arithmetical Abilities In neuropsychology, some common brain activity for
finger knowledge and calculation abilities can be sup-
Writing numbers appeared earlier in history than posed. Finger agnosia and acalculia appear as two simul-
writing language. Some cultures (e.g., Incas) devel- taneous signs of a single clinical syndrome (Gerstmann,
oped a number-representing system, but not a language- 1940), usually known as Gerstmann syndrome or an-
representing system (Swadesh, 1967). As mentioned, cal- gular gyrus syndrome. For prehistorical man, finger ag-
culus means pebble. Pebbles, or marks, or knots, or any nosia and acalculia could have represented just the same
other element were used as a correspondence construction defect.
to record the number of elements (people, cows, fishes, Adding, subtracting, multiplying, and dividing were
houses, etc.). In summer the first number writing system possible in the Egyptian system, but of course, following
has been found (about 3,000 BC) (Childe, 1936): Instead procedures quite different than those procedures we cur-
of using pebbles, fingers, or knots it was simpler just to rently use. They based multiplication and division in the
make a mark (a stroke or a point) on the floor, or on a tree duplication and halving method (Childe, 1936). Inter-
branch or a on board if you wanted to keep the record. In estingly, this very same procedure (duplicating and halv-
Egypt, India, and later in Crete, a similar system was de- ing quantities) is also observed in illiterate people when
veloped: units were represented by a conventional symbol performing arithmetical operations. So, in the Egyptian
(usually a stroke) repeated several times to mean a digit system to multiply 12 by 18, the following procedure was
between one and nine; a different symbol was used for 10 followed:
and 10-multiples.
Positional digit value is clearly disclosed in 1 18
Babylonians, and about 1,000 BC the zero was introduced. 2 36

Positional value and zero are also disclosed in Maya 4 72

Indians (Leon-Portilla, 1986). Egyptians and Babyloni- 8 144
ans commonly used fractions. Small fractions (1/2, 1/3, Total 216
and 1/4) are relatively simple numerical concepts, and
even chimpanzees can be trained to use small fractions
(The number 18 is duplicated one or several times, and
(Woodruff and Premack, 1981).
the amounts corresponding to 12 (4 + 8 in this example)
As mentioned previously, recognition of individual
are selected and summed up as follows: 72 + 144 = 216.
marks or elements up to 3 is easy: It represents an immedi-
To divide, the inverse procedure was used. So, to divide
ate perception readily recognizable. Beyond 3, the number
19 by 8 would be as follows:
of marks (strokes or dots) has to be counted and errors can
be observed. Furthermore, it is rather time-consuming and
cumbersome to be all the time counting marks. Notewor- 1 8

2 16
thy, the different digit notational systems always represent
2 4
one, two, and three with strokes (or points, or any specific
4 2
mark). It means, the numbers 1, 2, and 3 are written mak-
8 1
ing one, two, or three strokes. But beyond that figure, digit
writing may recur to other strategies. In our Arabic digit
notation system 1 is a vertical line; whereas 2 and 3 were That is, 2 + 4 + 8 (2 + 1/4 + 1/8), that is 2.375.
In brief, arithmetical abilities and number represen- Table 6. Different Levels of Numerical Knowledge (Adapted from
tation have been around only for some 5,0006,000 years. Klein and Starkey, 1987)
Most likely, during the Stone Age only simple counting Global quantification What collection is bigger and smaller
up to 3 was present and of course, bigger and smaller
(magnitude judgment) concepts. Global quantification al- Recognition small Differentiate one, two, and three elements
ready was existing at prehuman levels. Correspondence quantities
Enumeration Sequencing the elements in a collection
constructions allowed increasing the amount of numbers. Correspondence To compare collections
The most immediate correspondence construction is done construction
with the fingers. Finger knowledge and counting repre- Counting A unique number name is paired with each
sent in a certain extent the same cognitive ability, as it is object
still evident in some contemporary languages, as Sikuani Oneone principle Each object in a collection is to be paired
with one and only one number name
language. Stable order principle Each name is assigned to a permanent
Counting, finger gnosis, and even lateral spatial ordinal position in the list
knowledge may present a common historical origin. Cardinal principle The final number name used in a counting
Seemingly, calculation abilities were derived from finger sequence refers to the cardinal value of
sequencing. Number representation and arithmetical op- the sequence
Arithmetics Number permutability (e.g., adding,
erations are observed only for some 5,0006,000 years. subtracting)
Currently, calculation abilities are rapidly evolving be-
cause of the introduction of modern technology.
Rightleft discrimination (as well as the use of other
spatial concepts) most likely was present in prehistori- faster in the acquisition of numerical abilities. The differ-
cal man. Requirements of spatial abilities may have been ent stages appear in a sequential way, and the understand-
very high, even higher, than in contemporary man (Ardila, ing of more complex concepts requires the acquisition of
1993; Ardila and Ostrosky, 1984; Hours, 1982). Rightleft more basic levels. A percentage of otherwise normal chil-
discrimination and finger gnosis are strongly interdepen- dren can fail in using numerical concepts normally ex-
dent and even they can be interpreted as components of pected at their age. The term developmental dyscalculia
the autotopagnosia syndrome. It seems, in consequence, has been used to refer to this group of underperforming
that there is a rationale for finding a common brain activ- children.
ity for finger gnosis, calculation, and rightleft discrim-
ination (and, in general, spatial knowledge mediated by
language). Numerical Abilities in Preschool Children
Global quantification or numerosity perception rep-

DEVELOPMENT OF CALCULATION resents the most elementary quantification process. Global
ABILITIES IN CHILDREN quantification supposes the discrimination between col-
lections containing different number of objects (Davis
During child development, different stages in the ac- et al., 1985). Global quantification simply means what
quisition of numerical knowledge are observed (Klein set of elements is bigger and which one is smaller. Global
and Starkey, 1987). They include global quantification, quantification is observed at the animal level: Many ani-
recognition of small quantities, numeration, correspon- mals can select the larger collection of elements when they
dence construction, counting, and arithmetics (Table 6). have to choose. However, the ability to distinguish which
As mentioned, some fundamental numerical concepts can collection is larger depends upon the number of elements
be observed at the animal level and it is not surprising to in the collections. To distinguish 3 and 4 elements may be
find them in small children. The initial levels of numer- easy (4 is 25% larger than 3). To distinguish 10 and 11 ele-
ical knowledge are found in preschool children. The de- ments is obviously harder (11 is only 10% larger than 10).
velopment of complex numerical concepts requires long By the same token, to distinguish 10 from 20 elements is
school training. Complex arithmetical concepts depend easy (the double), but to distinguish 100 from 110 is hard
upon a painstaking learning process, and they are not usu- (one tenth). It means, what is important is the ratio ex-
ally found in illiterate people. The different stages in the isting between the two collections of elements (so-called
acquisition of numerical concepts are associates with the psychophysics Webers fraction).
language, perceptual, and general cognitive development. Global quantification is expressed in the language
Variability is normally observed, and some children can be with words such as many, a lot, and similar terms.
For small quantities, the words several, a few, and similar 1986). During the 2nd year the child also begins to use
quantity adverbs are included in the language. Quantity some number names, and usually develops the ability
adverbs used in everyday speech represent global quan- to correct counting up to 3. The child thus acquires the
tifiers. Quantity adverbs appear early in language his- knowledge of two basic principles in counting: (1) one-
tory and also in child language development than in the to-one principle (i.e., each object in a collection is to
numerical system. As mentioned previously, all known be paired with one and only one number name) and
world languages use global quantification and possess (2) the stable order principle (each name is assigned to
words to refer to many, a lot. All languages op- a permanent ordinal position in the list; the sequence
pose small quantities (one, two, a few) to many, a of numbers is always the very same: one, two, three,
lot. As a matter of fact, many, much, and simi- etc.). At this point, however, the child does not exhibit
lar global quantifiers represent early words in language yet a cardinal principle; that is, the final number name
development. used in a counting sequence refers to the cardinal value
Global quantification, however, does not represent of the sequence. If a collection is counted one, two,
yet a truly numerical process, because it does not suppose three, it means that in that collection there are three ob-
a one-to-one correspondence. Enumeration (sequencing jects (Klein and Starkey, 1987). Cardinal principle will
the elements in a collection of elements; this process sup- be observed in 3-year-old children (Gelman and Meck,
poses the individualization of each element) represents 1983).
the most elementary type truly of numerical knowledge At this point, the child can count small quantities,
(Klein and Starkey, 1987). Enumeration requires to dis- usually below 10. During this period the child is also learn-
tinguish the individual elements in the collection (this, ing how the numerical system works and memorizing the
this, and this, etc.). In child language development, the number words. Most often, the numerical system contains
most elementary distinction is between this and other. three different segments: (1) From 1 to 10 different words
This and other are also early words in child language are used. (2) From 10 to 20 counting becomes idiosyn-
development. cratic and quite frequently irregular. In English 11 has
Correspondence construction constitutes a type of not any apparent relationship with 1; 12 has an ev-
enumeration used to represent the number of objects in ident relation with 2 but it is a unique word number;
a collection and to compare collections. The amount of from 13 to 19 the ending teen is used. In Spanish, from
elements in a collection is matched with the amount of el- 11 (once) to 15 (quince) the ending ce is used. From 15
ements in an external aid (fingers, pebbles, knots, strokes, to 19 the word numbers are formed as ten and six, ten
marks, dots, etc.). It implies, in consequence, a one-to-one and seven and so on, 20 (veinte) has not any apparent
correspondence: Each one of the elements in the collection relation with 2 (dos). And (3) from 20 ahead the numer-
corresponds to one finger or pebble, knot, stroke, mark, ical system becomes regular. Word numbers are formed
dot, or whatever. An external device can be used for mak- as twenty and one, twenty and two, and so on. Learn-
ing the correspondence construction. The most immediate ing the whole numerical system usually is completed at
devices are the fingers. During enumeration usually the school.
fingers are used to point the objects. Computational strategies (e.g., adding; if a new item
Counting represents a sophisticated form of enumer- is included in a collection, the collection will become
ation: a unique number name is paired with each object in larger and the next cardinal number name will be given
a collection, and the final number name that is used stands to that collection) are found in 3- to 5-year-old children,
for the cardinal value of that collection. The initial object initially only for small quantities.
corresponds to one, the following to two, and so on.
Some times, the very same finger name is used as number
name (i.e., the very same word is used for one and thumb, Development of Numerical Abilities at School
two and index finger, etc.). The collection has the amount
of objects that corresponds to the last pointed object (car- Adding and subtracting numerical quantities and the
dinal principle). Arithmetics represents an advanced nu- use of computational principles are observed during in
merical system, which comprises number permutability firstsecond-grade children, but they only become able
(e.g., adding, subtracting). to manipulate the principles of multiplying and dividing
Human infants are able to recognize numerosity for after a long and painstaking training period, usually during
small quantities (usually up to 36 items) (Antell and thirdfifth school grade.
Keating, 1983), but the ability to construct correspon- Understanding that subtracting is the inverse oper-
dences emerges only during the childs 2nd year (Langer, ation of adding is usually acquired about 56 years. At
this age the child begins to use three different procedures DEVELOPMENTAL DYSCALCULIA
for performing additions and subtractions: (1) counting
using the fingers, (2) counting aloud not using the fin- The term developmental dyscalculia refers to a cog-
gers, and (3) memorizing additions and subtractions for nitive disorder of childhood, impairing the normal ac-
small quantities (one plus one is two, two plus two is quisition of arithmetical skills (American Psychiatric
four, two minus one is one, etc.). The last strategy be- Association, 1987). Frequently, dyscalculia has been used
comes progressively stronger when advancing age and as a general term encompassing all aspects of arithmeti-
schooling. Nonetheless, children continue using the fin- cal difficulty (Shalev et al., 1988). The term DD has been
ger for adding and subtracting larger quantities. From the changed to Mathematics Disorder in the DSM-IV (Amer-
age of 10 until about 13 years, counting using the fin- ican Psychiatric Association, 1994). However in the neu-
gers progressively disappears, but counting aloud, and ropsychology literature the term DD remains. It is esti-
performing arithmetical operations aloud, remains. Au- mated that approximately 6% of school-age children in the
tomatic memory not only for additions and subtractions United States suffer from this disorder (Grafman, 1988;
but also for multiplications (multiplication tables) and di- Gross-Tsur et al., 1996). The prevalence of the disorder is
visions becomes progressively more important (Grafman, difficult to establish because it is frequently found in com-
1988; Siegler, 1987). As a matter of fact, adding and sub- bination with other developmental disorders (American
tracting one digit quantities (e.g., 7 + 5 = 12; 4 + 5 = Psychiatric Association, 1994). Gross-Tsur et al. (1996)
9; 8 5 = 3; etc.) represents a type of numerical rote observed that 17% of the children with DD were diag-
learning, similar to the multiplication tables. Interesting nosed with dyslexia and 25% had ADHD-like symptoms.
to note, the performance of arithmetical operations aloud It is estimated that 1% of the school-age children have DD
may remain during adulthood, even in highly educated alone (American Psychiatric Association, 1994). DD is a
people. disorder frequently encountered in children with epilepsy
It should be emphasized that there is a significant (Seidenberg et al., 1986) and in girls with sex chromosome
variability in the specific strategies used by different chil- abnormalities such as Turners syndrome (Gross-Tsur
dren at the same age. Furthermore, the very same child can et al., 1996).
recur to different strategies when solving different arith- Gender distribution has been controversial. Some-
metical problems. In some situation, for instance, the child times it is assumed that it is more frequent in boys, whereas
can recur to the fingers, whereas in a different operation, some authors consider that DD tends to affect both sexes
he may not require using the fingers. Or, the child can be equally. Equal ratios between the sexes for arithmetical
able to use some multiplication tables whereas failing with difficulties have been reported by Lewis et al. (1994) and
others. Gross-Tsur et al. (1996). In Gross-Tsur et al., the ratio of
About the age 89 usually the children learn to multi- girls to boys was 11:10.
ply. This requires the memorization of the multiplication The origin of DD has not been established. Some re-
tables. The errors most frequently found when learning searchers considered DD to be a genetically determined
the multiplication tables are those answers that could be brain-based disorder (Rourke, 1989). However some have
correct for other number within the same series (e.g., 4 hypothesized that the childs environment and social con-
5 = 16). These errors may be the result of some inter- text are causal for mathematical disorders (Fergusson
ference. They can be observed in children at any age, and et al., 1990). In acalculia (acquired dyscalculia), contrary
even they are sometimes found in normal adults (Graham, to DD, the mathematical disorder is consequence of a well-
1987). known cerebral lesion. The cognitive processes dysfunc-
Development of abstract reasoning and increase in tion that underlie the two defects could, however, be the
working memory span contribute to the use of mathemat- same ones, because the types of errors that people present
ical algorithms (i.e., the set of rules used for solving arith- with either of the two-dyscalculia types, are similar.
metical problems following a minimal number of steps).
The development of algorithms begins when learning the
basic arithmetical operations. Progressively, they become Neuropsychological Characteristics
more automatic, representing basic strategies for solving
arithmetical problems. Development of abstract thinking Children with DD can fail in a whole array of nu-
allows the use of magnitudes applied to different systems merical tasks including performing arithmetical operation,
(use of the numerical system in measuring time, temper- solving arithmetical problems, and using numerical rea-
ature, etc.) and the understanding of quantities expressed soning. According to Strang and Rourke (1985), the er-
in a symbolic way. rors found in children with dyscalculia can be classified
Table 7. Errors Most Frequently Found in Children with DD (Adapted other appropriate strategies. Cohen (1971) has proposed
from Strang and Rourke, 1985) that short-memory difficulties explain the incompetence
Error Characteristics in arithmetics of children with DD. In fact, the inability
to carry and recall number tables may be the result of
Spatial Difficulties in placing numbers in columns, memory deficits (Shalev et al., 1988). Davis et al. (1992)
following appropriate directionality of the suggest a sequential processing deficit as the underlining
procedure, v. gr., to subtract the substrand from
the minuend
deficit of DD.
Visual Difficulties in reading arithmetic signs, forgetting In addition to memory deficits, children with DD
the points of the units of thousand, etc. present attentional difficulties. The association between
Procedural Omission or addition of a step of the arithmetical attentional problems and dyscalculia has been well doc-
procedure, and application of a learned rule for umented. Badian (1983) described attentionalsequential
a procedure to a different one, v. gr., 75 + 8 =
163, an operation in which the multiplication
problems in 42% of the children with DD. More recently,
rule is applied in the sum Shalev et al. (1995a,b) demonstrated the presence of at-
Graphomotor Difficulty in forming the appropriate numbers. tention deficit disorder symptoms in 32% of the dyscalcu-
Judgment Errors that imply impossible results, such as one lia sample studied. Equally children with attention deficit
in which the result of subtracting is bigger than syndrome with or without hyperactivity make mathemat-
the numbers being subtracted
Memory Problems in the recall of multiplication tables or
ical errors secondary to impulsiveness and inattention
arithmetical procedures (Sokol et al., 1994).
Perseveration Difficulty in changing from one task to another Rosenberger (1989) found that visualperceptual and
one, repetition of the same number attention disorders were evident in children who had spe-
cific difficulties in mathematics. Strang and Rourke (1985)
not only corroborated the presence of significant diffi-
into seven categories: (1) errors in spatial organization of culties in visualperceptual organization in children with
quantities, (2) errors in visual attention, (3) arithmetical dyscalculia, but also described difficulties in other neu-
procedural errors, (4) graphic motor errors when writing ropsychological tasks. These children manifested difficul-
quantities, (5) numerical judgment and reasoning errors, ties in the tactile analysis of objects, particularly with the
(6) memory errors for quantities, and (7) preserveration in left hand, as well as impairments in the interpretation of
solving arithmetical operations and numerical problems. facial and emotional expressions (Rourke, 1987). Chil-
Table 7 describes the characteristics of the most frequent dren with dyscalculia also present an inadequate prosody
types of errors found in children with DD. in verbal language (Rourke, 1988), and difficulties in the
Kosc (1970) described six types of difficulties ob- interpretation of nonverbal events (Loveland et al., 1990).
served in DD: (1) problems in verbal organization of num- These neuropsychological findings have suggested the
bers and mathematical procedures, (2) difficulties in the presence of a functional immaturity of the right hemi-
management of mathematical symbols or objects, (3) er- sphere as a structural fact underlying dyscalculia.
rors in reading numbers, (4) errors in writing numbers, (5) Hernadek and Rourke (1994) described the disorder
difficulties in the understanding of mathematical ideas, of nonverbal learning associated with a dysfunction of the
and (6) in the carrying over when performing arithmetical right hemisphere and characterized by visuospatial diffi-
operations. Students with dyscalculia may present a clus- culties, visuomotor coordination and reasoning problems,
ter of problems in their ability to perform mathematical defects in concept formation and in mathematical skills.
tasks. This disorder has also been known as developmental right-
It is unclear what arithmetical function is impaired in hemisphere syndrome (Gross-Tsur et al., 1995; Weintraub
children with DD. Shalev et al. (1988) studied the arith- and Mesulam, 1983). The children with nonverbal learn-
metical errors found in 11 children with dyscalculia and 10 ing disorder also present defects in the recognition of
matched control children. No differences were observed faces and of emotional expressions and poor adaptation to
in number comprehension (matching numbers to quanti- novel social situations. In accordance with Hernadek and
ties, appreciation of relative quantity, numerical rules, and Rourke, children with dyscalculia associated with the dis-
serial order) and number production (counting, reading, order of nonverbal learning can be distinguished clearly
and writing numbers) scores. But significant differences from children with dyscalculia associated with reading
emerged in the performance of fact retrieval, addition, sub- disorder. In the later group, problems are manifested in
traction, multiplication, and division scores. The group the performance of language recognition tasks whereas
with DD had difficulty in fact retrieval but could show that performance in tactile and visual perception tasks is well
they knew how to calculate by using finger counting and preserved. Gross-Tsur et al. (1995) describe the clinical
characteristics of 20 children with the nonverbal learn- cases of DGS in otherwise normal children, however, have
ing disability disorder. Dyscalculia was the most frequent been described (Benson and Geschwind, 1970; PeBenito,
encountered scholastic problem. 1987), and the syndrome may represent a delayed cerebral
Although an evident association exists between maturation (PeBenito et al., 1988).
dyscalculia and dyslexia, different underlying cognitive Besides dyslexic children, those children with emo-
explanations have been suggested for the two disorders. tional problems can also fail easily in tasks that demand
Rosenberger (1989) found that those children with dif- high attention levels such as in the case of mathematics.
ficulties in mathematics present more evident visuospa- Problems in the emotional domain are recurrently men-
tial and attentional dysfunctions. These children obtained tioned in the context of arithmetical disabilities (Shalev
lower scores in visuomotor tests, for example, in the et al., 1995a,b). Withdrawal and social problems have been
Bender Visuomotor Gestalt Test, and in the digitsymbol noted in children with DD (Rourke, 1989).
subtest of Wechsler Intelligence Scale, than did a group of In summary, children with DD can fail in a whole
children with specific difficulties in reading. The authors array of numerical and arithmetical tasks. Errors found in
postulate that visuomotor and visuospatial execution and these children are of spatial organization, visual attention,
organization defects can distinguish children with dyscal- procedural, motor, judgment, reasoning, and memory. DD
culia from children with dyslexia. Not all researchers, is commonly associated with reading dysfunctions, atten-
however, have supported this distinction between children tion disorders, and emotional difficulties. The involvement
with DD and children with dyslexia. Rasanen and Ahonen of one or both cerebral hemispheres in dyscalculia is still
(1995) have suggested functional communality between controversial.
dyslexia and dyscalculia. They found that reading accu-
racy and reading speed correlate with the number of errors
in arithmetical operations, particularly in multiplication. Subtypes
The authors concluded that difficulties in visuoverbal rep-
resentations might explain both reading and mathematical DD is not a uniformed disorder. Children with DD
disorders. can manifest an array of different numerical errors, and
Some authors consider that DD does not appear there are variations in terms of the type of dyscalculia and
like an isolated manifestation of cerebral dysfunction, the severity of the disorder (Grafman, 1988). Different
but as part of a syndrome called Gerstmann syndrome. subtypes of DD have been proposed.
This syndrome is composed of the tetrad of dyscalcu- Kosc (1970) describes six dyscalculia subtypes char-
lia, digital agnosia, dysgraphia, and leftright disorien- acterized by difficulties in (1) the verbalization of terms
tation. This syndrome is found in skillful adults as a and mathematical relationships, (2) the handling of sym-
result of lesions in the left parietal lobe, but has been bols/mathematical objects, (3) the reading of numbers, (4)
described in children with specific learning disabili- the writing of numbers, (5) the comprehension of mathe-
ties, and has been given the name of developmental matical ideas, and (6) the ability for carry in arithmetical
Gerstmann syndrome (DGS). Children with DGS present operations.
intact language skills, and reading is usually at the normal Badian (1983) finds in children with DD a high
grade level (PeBenito et al., 1988). However, Croxen and frequency of spatial numerical difficulties, developmen-
Lytton (1971) found that more children who are slow read- tal anarithmetia (primary difficulties in calculating), and
ers had difficulties with finger recognition and rightleft attentionalsequential defects, but very few cases of
discrimination. dyslexia and dysgraphia for numbers. Geary (1993) in-
Gerstmann syndrome may occur in children with tended to classify dyscalculia in three groups based in
brain damage or in children who are apparently normal. three types of errors: (1) visuospatial, (2) semantic mem-
Gerstmann syndrome in brain damage children is usually ory, and (3) procedural. Two types of developmental
associated with multiple symptoms, such as hyperactiv- dysacalculias have been recognized by Rourke (1993):
ity, short attention span, lower intellectual performance, (1) dyscalculia associated with language problems
and poor reading. This group accounts for most of the re- (dyslexia), defects in the understanding of instructions
ported Gerstmann syndrome in children (Kinsbourne and and verbal problems, and reduction in the capacity of ver-
Warrington, 1963; PeBenito et al., 1988; Pirozzolo and bal memory; and (2) dyscalculia associated to spatial
Payner, 1978). The nature of the neurological disorder in temporal difficulties, with sequence problems and rever-
these cases is usually diffuse and bilateral. DGS, on the sion of numbers (Spreen et al., 1995). The dichotomy
other hand, seems to be relatively common among individ- between fundamental visuospatial dyscalculia indicative
uals with fragile X syndrome (Grigsby et al., 1987). Few of right hemispheric dysfunction and another fundamental
dyslexic dyscalculia, suggestive of left hemispheric dyscal operation demands visuospatial discrimination ability
function, has not been corroborated by other authors (e.g., to organize the numbers in columns, to arrange the appro-
Shalev et al., 1995a,b; Sokol et al., 1994). It is frequent priate spaces among numbers, and to begin the operation
to find that those children with difficulties in reading and from right to left. The working memory, or operative mem-
writing also present defects in learning arithmetics. For ory, associated with sustained attention, obviously plays
the child with learning difficulties, the achievement of a central role in the performance of any arithmetical op-
a mathematical problem becomes a task more difficult eration. The algorithmic plan of action that unchains the
than reading, because the change of a single digit alters numerical symbol in a particular arithmetical operation
the result of the operation completely. Additionally, the is evoked, or recovered from previous learnings whose
achievement of any mathematical problem, even the sim- engrams are found in long-term memory (Boller and
plest, demands the pursuit of certain systematic steps that Grafman, 1985). The stages of the necessary cognitive
can be a high-level difficult task for a child with dyslexia. process for the achievement of an arithmetical operation
The prevalence of some right hemisphere dysfunc- can be observed in a simple example: 34 + 26 = 60.
tion in DD has been recently reanalyzed. Shalev et al. When this operation is presented to a person, in the
(1995a,b) studied a group of children with DD and neu- first place they have to perceive the spatial organization of
ropsychological profiles suggestive of right or left hemi- the quantities, the relationship amongst themselves, un-
spheric dysfunction. They analyzed the types of errors in derstand the meaning of the plus (+) symbol, recognize
each group, and correlated the scores of hemispheric later- the numerical symbols, and to know the steps that should
alization with the results in tests of mathematics. Contrary be followed to carry out adding appropriately. The sum of
to that proposed by Rourke et al., the children with suppos- the numbers 6 and 4 is automated, and the only answer that
edly dysfunction of the left hemisphere presented a sig- needs attention is the number 10. If the operation is not
nificantly larger number of visuospatial errors and more automated, the individual can use the controlled process
notorious difficulty for the achievement of arithmetical of count. The number 10 should be maintained in mem-
operations that the group with right neuropsychological ory whereas the zero is placed in the right column, and the
dysfunction. The authors did not find differences in the unit 1 is taken and conserved in the short-term memory
profile of errors of each group, neither a correlation be- (mentally), or written on the following column. The same
tween the hemispheric dysfunction and the type of errors process is continued for the following column. The infor-
in mathematics tests. Despite the fact that Shalev et al.s mation that is stored in the long-term memory seems to
data indicates that arithmetical impairments are more se- correspond to two types: (1) syntactic information, that is,
vere in cases of DD associated with left hemisphere dys- the knowledge of the rules of the numerical procedures,
function, their conclusion is that there is a participation and (2) semantic information (i.e., the comprehension of
of both cerebral hemispheres the arithmetical processes the meaning of the procedures implied in the solution of
(Shalve et al., 1995b). particular problems). When the problems are solved using
automatic codes, semantic reasoning is not required.
PROCESSES INVOLVED IN ARITHMETICAL In performing an arithmetical operation, two differ-
OPERATIONS ent cognitive systems should be distinguished: the proces-
sor of numbers and the operating system of calculation.
Numerical handling represents a language that in- Additionally, during the processing of numbers it is nec-
volves a system of symbols. This system of symbols im- essary to distinguish between the comprehension and the
plicated in calculations can be divided into two groups: (1) production of numbers, and in each one of these subpro-
a logographic system including Arabic numbers from 0 to cesses, in turn, it is necessary to consider the presence
9, and (2) a phonographic system that provides the verbal of a double code (verbal and numerical) and of a dou-
name to numbers; for example, one, nine. The perfor- ble analysis (lexical and syntactic). To process quantities
mance of a given arithmetical operation begins with the in Arabic numbers, for example, 823, or numbers in
numerical recognition, which depends on a verbal process verbal codes, eight hundred twenty-three, demands a
and perceptual recognition: numbersymbol or symbol lexical analysis (comprehension and production of the in-
number. Each number provides two types of information. dividual elements) and a syntactic analysis (processing of
On the one hand the base group to which the number the relationships among the elements).
belongs (units, dozens, hundreds), and on the other, the McCloskey et al. (1985) and McCloskey and
ordinal position of the number inside the base. Thus, the Caramazza (1987) focused on a detailed analysis of the
number 5 belongs to the units and occupies the fifth place calculation system. They proposed a cognitive model
inside them. The successful performance of an arithmeti- of the numerical processing on the basis of three basic
Fig. 1. Schematic representation of the number comprehension, number production, and calculation system (adapted fromCaramazza and
McCloskey, 1987; Levin et al., 1993).
abilities: understanding of a number, production of a principle of communality). They demonstrated, in a

number, and processing of the mathematical procedures. case study design, that conceptual knowledge can be
The category of understanding of the number (input) dissociated from arithmetical fact and procedures. They
includes the understanding of quantities, of the symbolic found impaired fact retrieval with intact processing of
character of those quantities (lexical processing), and of arithmetical principles and problems. This selective
the order of the digits (syntactic processing). Inside the vulnerability of arithmetical facts had been demon-
numerical production (output) count is found, and the strated before by Warrington (1982) and McCloskey
reading and the writing of numbers. The two subsystems, et al. (1985). Arithmetical facts represent a separate subset
number comprehension and number production, include of the semantic memory and may also dissociate from
the verbal system (written or spoken words, e.g., thirty) other mathematical and numerical knowledge (Hittmair-
and the Arabic system (element of a number, e.g., 30). Delazer et al., 1995). Dissociations, according to the
These subsystems are illustrated in Fig. 1. The verbal specific arithmetical operations, have also been reported.
system is the lexical processing mechanism. Three are the For example, patients with severe addition difficulties and
mechanisms included inside the calculation system. The selective preservation of subtraction have been reported
first one is the recognition of the arithmetical symbols, (Dagenbach and McCloskey, 1992; McNeil and
the second and the third ones are the understanding, Warrington, 1994).
memorization, and execution of the arithmetical facts Using a cognitive neuropsychology perspective,
(i.e., table facts, such as 2 3), and arithmetical proce- McCloskey et al. (1985, 1986) have described different
dures (i.e., when performing an addition, start at the right forms of acquired acalculia. They distinguished between
most column, sum the right most digits, write down the disorders of number processing (reading, writing, pro-
ones, carry the tens, etc.) (Hittmair-Delazer et al., 1995; duction, comprehension, or repetition of numbers) and
Sokol et al., 1994). Arithmetical facts are retrieved from a disorders of calculation (number facts, knowledge of
semantic network system independently from calculation procedures).
procedures. The calculation procedures are the sequence Caramazza and McCloskeys model has been applied
of steps necessary to perform multidigit operations. to the acquisition of mathematical abilities in preschool
Although both components are learned, their functional children and to the study of DD (Sokol et al., 1994). An
independence has been demonstrated by several authors specific impairment in number processing has been de-
(McCloskey et al., 1985, 1991a,b; Sokol et al., 1994; Sokol scribed in a child with DD (Temple, 1989). This child had
et al., 1991; Warrington, 1982). Hittmair-Delazer a selective impairment in the lexical processes of reading
et al. (1995) have pointed out that conceptual knowledge Arabic numbers with intact syntactic processes. Temple
plays a crucial role in arithmetical processing, frequently (1991) also reports two cases of DD in which an accu-
neglected in the neuropsychology models of calculation. rate number processing skill had been developed. How-
Hittmair et al. refer to conceptual knowledge as the ever, one of the cases showed a selective difficulty with
understanding and use of arithmetical principles (i.e., the procedures of calculation, and the other one showed a
selective disorder in mastery of those arithmetical facts metical problems. Hittmair-Delazer et al. (1995) describe
that comprised the multiplication tables. Sokol et al. a patient that presented great difficulty in carrying out
(1994) studied the range of functional dissociations pre- arithmetical operations despite having the appropriate
dicted by McCloskey et al.s cognitive model in 20 stu- knowledge of arithmetical principles. These dissociations
dents with DD. Their results supported the usefulness of among diverse mathematical components suggest a func-
the model to understand DD. tional independence of each one of them.
The modules delineated by McCloskey et al. (1985, The functional independence of the diverse math-
1986) may be also separable during the normal develop- ematical components has been corroborated for the
ment of calculation abilities. Children are considered to analysis of the errors of numerical count in patients
have considerable conceptual knowledge before they ac- with cerebral damage. Lesions in diverse parts of the
quire automaticity in number processing and calculation cerebral cortex originate different errors (Rosselli and
processes (Hittmair-Delazer et al., 1995). For example, Ardila, 1989). Some patients can present difficulties
children know counting principles before they count cor- in carrying quantities, others in the placement of the
rectly (Gelman and Meck, 1983), and they can judge arith- appropriate numbers, and in other patients the difficulty
metical transformations correctly before they can perform to execute arithmetical operations one can observe in the
them (Starkey, 1992). combination of the two procedures.
The studies of the alterations of number count in pa-
Cerebral Mechanisms tient with cerebral damage have demonstrated that lesions
in either of the two cerebral hemispheres can produce acal-
The appropriate solution of a numerical problem de- culia, although of different characteristics. The lesions in
mands verbal, spatial, and conceptual abilities that very the areas of language in the left cerebral hemisphere pro-
probably require of the active participation of numerous duce alterations in the comprehension and in the produc-
cerebral structures. It has been suggested, however, that tion of numbers, and therefore in carrying out the arith-
the central neural mechanisms implied in the recogni- metical operations. On the contrary, the lesions in the right
tion of numbers seem to be different to those that par- cerebral hemisphere cause alterations in the spatial orga-
ticipate in the solution of arithmetical problems as such nization quantities and in the comprehension and achieve-
mechanisms can be altered differentially in cases of cere- ment of abstract problems (Ardila and Rosselli, 1990;
bral focal damage. A patient can present difficulties in Rosselli and Ardila, 1989). The experimental studies with
the recognition of numbers with appropriate conserva- dichotic audition and the tachistoscopic presentation of
tion of the ability to carry out arithmetical operations. visual information support the participation of the two
Additionally, the handling of numbers can be dissociated cerebral hemispheres in carrying out arithmetical prob-
alternating independently from the numerical production lems (Grafman, 1988; Holender and Peereman, 1987).
and conserving their understanding (Benson and Denckla,
1969). McCloskey and Caramazza (1987) describe, as a
consequence of localized cerebral damage, a dissociation CALCULATION ABILITIES
between the capacity to understand and the capacity to IN NORMAL POPULATIONS
produce numbers. Evidence of selective alterations exists
for the processing of Arabic numbers and the processing Few neuropsychological studies have approached the
of lexical numbers (Ardila and Rosselli, 1990; Deloche question of calculation abilities in the general population.
and Seron, 1987; Rosselli and Ardila, 1989). Similar dis- Intuitive observation points to a significant dispersion of
sociations have been described between the lexical and arithmetical abilities in normal people. Usually, however,
syntactic processing of verbal numbers and that of Arabic it is assumed that any normal person should be able to tell
numbers (Caramazza and McCloskey, 1987). An example one-digit multiplication tables, to use the four basic arith-
of a lexical mathematics error would be to represent two metical operations, to solve simple arithmetical problems,
hundred twenty-one as 215, whereas the representation to memorize seven digits after a single presentation, and
of five thousand six hundred as 50006 would be an error to use diverse numerical information in the everyday life.
of mathematics syntax. Nonetheless, normative studies are scarce.
Moreover, Ferro and Botelho (1980) and Caramazza
and McCloskey (1987) have observed patients with a Normative Studies
defect exclusively in the analysis of mathematics signs.
Warrington (1982) describes dissociations between the Deloche et al. (1994) developed a standardized test-
ability to develop simple arithmetical operations and arith- ing battery for the evaluation of brain-damaged adults in
the area of calculation and number processing. With the Table 8. Performance of 300 Normal University Students in Some
purpose of obtaining some norms, the battery was admin- Calculation Tests (Adapted from Ardila et al., 1998a)
istered to 180 participants stratified by education (up to Test M SD Range
9 years of formal education, 10 or 11 years, and more
than 11 years), age (2039 years, 4059 years, and 6069 WAIS: Arithmetic 11.8 7.7 38
years) and gender. This battery, named as EC301, includes (scaled score)
Digits (scaled score) 11.6 2.2 517
three notational systems for numbers: Arabic digits, writ- Mental arithmetics 5.3 1.9 08
ten verbal forms, and spoken verbal forms. (maximum score 8)
Analysis of error rates indicated the effect of some Arithmetical problems 9.5 3.2 116
demographic factors, principally, education (in counting, (maximum score 16)
transcoding, written verbal numbers, magnitude compar-
isons, and arithmetical operations subtests); incidentally,
gender (in digit numbers, and total mental calculations chological and neuropsychological tests directed to assess
scores). No age effect is mentioned in the age range in- not only calculation abilities, but also language, memory,
cluded in this study (2069 years). perceptual abilities, concept formation, and praxis abili-
In the normative study, 88% of the participants pre- ties. Two arithmetical tests were used: (1) mental arith-
sented at least one error in the EC301 test battery. The metical operations (two additions, two subtractions, two
easiest subtest turned out to be reading and writing arith- multiplications, and two divisions). Maximum possible
metical signs; this subtest was failed by only 1% of the score was 8 points (1 point for each correct answer). And
total participant sample. The hardest subtest was writ- (2) arithmetical problems. Sixteen arithmetical problems
ten multiplication; in this subtest 36% of the partci- were orally presented. The participants were allowed to
pants presented at least one error. So, errors were com- use pencil and paper if so wanted. Maximum possible
mon in normal participants, but level of difficulty was score was 16 points (1 point for each correct answer). In
variable. addition, two WAIS scores were analyzed: Arithmetic and
The educational effect deserves some comments. The Digits subtests. Thus, in total four calculation ability tests
lowest educational group was equal or below 9 years, were considered. Means, standard deviations, and ranges
but observing the mean educational level and the stan- are presented in Table 8. Noteworthy, a very significant
dard deviation, seemingly, all the participants in this group dispersion was observed in the scores. Some university
had 89 years of education (the mean education for this students were unable to solve mentally even a single ad-
groups was 8.40, and the standard deviation was 0.50). dition. Other participants had a virtually perfect perfor-
So, as a matter of fact, the whole sample had at least 8 mance.
years of education. Nonetheless, a significant educational Ostrosky et al. (1997) selected 800 normal popula-
effect was observed in several tasks. It can be conjectured tion participants in five different states of the Mexican
that if they had included participants with an even lower Republic. The obtained sample included 665 participants
educational level, the education effect would be stronger (83.12%) from urban areas, and 135 participants (16.88%)
and may have appeared in a larger amount of calculation from rural areas. Ages ranged from 16 to 85 years (mean
subtests. age = 47.77; SD = 20.14). Education ranged from 0 to
The gender effect was demonstrated only in Task 6 24 years (mean education = 6.8; SD = 6.1). Fifty-two
(Mental calculation) in the Arabic digit condition and in percent of the sample was women. Ninety-five percent
the total score. Males performed better than females. Sur- of the sample was right-handed. Four age groups were
prisingly, no age effect was found in any of the subtests. formed: (1) 1630 years, (2) 3150 years, (3) 5165 years,
It can be assumed that calculation abilities, at least for and (4) 6685 years. In addition, each age group was di-
simple tasks, remain relatively stable up to the 1960s. A vided into four different educational levels: (1) illiterates
decline in calculation abilities would be evident only after (0 years of education), (2) 14 years of education; (3)
the age of 70 years. 59 years of education, and (4) 1024 years of formal
Ardila et al. (1998) analyzed the calculation abili- education. The NEUROPSI neuropsychological test bat-
ties in a normal population sample composed exclusively tery (Ostrosky et al., 1997) was individually administered.
of young people with a high level of education. A com- It includes three items related with calculation abilities:
prehensive neuropsychological test battery was assembled (1) Digits backward, up to six digits (maximum score =
and individually administered to a 300-participant sample, 6 points), (2) serial 3 subtraction from 20 to 5 (maxi-
aged 1725 years. All of them were right-handed male mum score = 5), and (3) Calculation Abilities subtest. In
university students. The battery included some basic psy- this subtest, three very simple arithmetical problems to
Table 9. Means and Standard Deviations Found in the Different NEUROPSI Neuropsychological Tests (n 800)
(Adapted from Ostrosky et al., 1997)
Test 1630 years 3150 years 5165 years 6685 years Maximum score
Illiterates
Digits backward 2.2 (1.1) 2.8 (1.1) 2.9 (1.0) 2.7 (0.9) 6
20 minus 3 2.2 (1.6) 3.8 (1.5) 3.1 (1.8) 2.9 (1.8) 5
Calculation abilities 1.0 (1.1) 1.4 (1.1) 1.6 (1.1) 0.9 (1.1) 3
One to four years of education
Digits backward 2.6 (1.0) 2.7 (0.7) 3.0 (1.0) 2.8 (0.8) 6
20 minus 3 3.5 (1.6) 3.6 (1.4) 4.3 (1.3) 4.4 (0.9) 5
Five to nine years of education
Digits backward 3.4 (0.7) 3.4 (1.2) 3.6 (0.8) 3.4 (0.8) 6
20 minus 3 4.3 (1.3) 4.6 (0.6) 4.4 (0.9) 4.6 (0.2) 5
Ten to 24 years of education
Digits backward 4.3 (0.9) 4.4 (0.9) 4.0 (0.9) 3.9 (1.0) 6
20 minus 3 4.7 (0.8) 4.7 (0.7) 4.9 (0.4) 4.8 (0.6) 5
be mentally solved are presented (How much is 13 + cluded in this study, the age effect in this subtest cannot
15; John had 12 pesos, received 9 and spent 4. How be interpreted as a score decrease associated with age. In
much does he have; and How many oranges are there in all the educational groups, performance in the oldest par-
two and half dozens) (maximum score = 3). Normative ticipants was higher than that in the youngest participants.
results are presented in Table 9. It is observed that in gen- The age effect simply means that the performance in this
eral scores increase with educational level and decrease subtest was associated with the participants age. Scores
with age. It is interesting to note that the highest scores, tended to increase up to the 50s, and further remained
particularly in the illiterate group, are obtained not in the stable or presented a very mild decrease.
youngest group (1630 years) but in the second age group In conclusion, (1) calculation abilities present a very
(3150 years). Despite representing a very easy calcula- significant dispersion in the general normal population.
tion test, even some people with relatively high education (2) Even very simple arithmetical tasks are failed by a per-
failed some points. This observation emphasizes the sig- centage of the normal population, including people with
nificant dispersion in calculation abilities found in normal a high educational level. (3) Educational effect represents
populations. a robust effect in calculation tests. Lowest performance
Educational effect presented a very robust effect is observed in illiterate people. Interestingly, difficulties
(Table 10). In the highest educational group, scores in the are observed not only in school-trained arithmetical abili-
three subtests are about the double those in the illiterate ties (e.g., arithmetical operations), but also in nondirectly
group. Age effect, however, was notoriously weaker and school-trained numerical abilities, such as repeating digits
was observed only in the second subtest (20 minus 3). Most backward. (4) Age effect is notoriously weaker than edu-
important, even though people up to 85-year-old were in- cational effect. In the range 2069 no age effect is readily
demonstrated (Deloche et al., 1994). Including people up
to 85 and illiterate people, the age effect was disclosed
Table 10. F Values for Education and Age Variables, and Interactions only in some numerical tests, such as mentally subtract-
Between Education and Agea ing 3 from 20. But the age effect cannot be interpreted
Test Education (E) Age (A) EA as a score decrease associated with age, but rather as a
tendency to score increase up to the 50s. Further, scores
Attention: digits backward 108.00*** 1.85 2.09 may remain stable or slightly decrease. And (5) there is
20 minus 3 63.46*** 6.27*** 3.37*** a gender effect in calculation abilities demonstrated in
Calculation abilities 95.57*** 3.21 2.60
tests such as mental calculation and solving simple nu-
a Levelsof significance are pointed out (adapted from Ostrosky et al.,
merical problems. Curiously, this gender effect is stronger
1997). in people with high educational level and weaker in
***p < .0001. illiterates.
Cultural and Educational Variables In other words, for the illiterate person it is notoriously
in Calculation Abilities easier to solve the operation If you go to the market and
initially buy 12 tomatoes and place them in a bag. Later
Some studies have approached the analysis of calcu- on you decide to buy 15 additional tomatoes. How many
lation abilities in different cultural contexts and in people tomatoes will you have in the bag? than the operation:
with different educational backgrounds (e.g., Cauty, 1984; How much is 12 + plus 15?
Grafman and Boller, 1987; Levy-Bruhl, 1947). Rosin Grafman and Boller (1987) proposed that some arith-
(1973) analyzed the way in which illiterates perform arithmetical skills appear genetically linked (e.g., equivalence
metical tasks. It was observed that calculation was labo- or certain counting skills), and some are educational linked
rious and strongly relied in memorizing each step. For (e.g., arithmetical calculation and the tool used to cal-
counting, fingers were used with large numbers requir- culate: fingers, abacus, calculator, computer, or the brain).
ing representing the hands. Often, doubling and halving It is reasonable to expect that some basic numerical strate-
the figures were used for arithmetical operations (as ob- gies will be found in different cultural groups. The best
served in the initial Egyptian division and multiplication example is the use of fingers in counting.
systems). For actual trading and marketing, the operations
could be initially performed visually using physical enti-
ties, and the results retained in memory. Gender Differences in Calculation Abilities
Posner (1982) analyzed the development of mathe-
matical concepts in West African children aged 510. A Gender differences in calculation abilities have been
mild effect of experiential factors on the ability to judge the recognized since long time ago (see Halperin, 1992). It is
magnitude of numerical quantities was observed; counting usually accepted that men outperform females not only in
was noted in all children, usually relying on size cues for calculation tasks, but also in those tests requiring spatial
small qualities. To perform even simple arithmetical op- manipulation. On the quantitative portion of the Scholas-
erations and to solve numerical problems was particularly tic Aptitude Test (SAT-M) there is a difference of about
difficult. 50 points between males and females (National Educa-
Casual observation of illiterates discloses that they tion Association, 1989). SAT is a highly standardized test,
can use simple numerical concepts and they easily han- which is administered nationally to college-bound high
dle money in daily activities (at least in a country such school seniors in the United States.
as Colombia, where bills of different value have different To account for these gender differences, however, has
colors, albeit, not different sizes). Illiterates readily rec- been quite polemic (McGlone, 1980). As mentioned pre-
ognize the bigger and smaller bill, and can perform viously, gender differences may be evident even in simple
simple computations (e.g., a 5000-peso bill is equivalent arithmetical operations and in solving arithmetical prob-
to two 2000-peso bills plus one 1000-peso bill). However, lems. Gender differences, nonetheless, are not found in all
to perform subtractions is particularly painstaking, and il- mathematical tests. Stones et al. (1982) analyzed gender
literates easily get confused (e.g., when shopping). They differences at 10 different colleges. Ten different math-
usually cannot multiply or divide, excepting by 10 (e.g., ematical ability tests were administered. Gender differ-
3, 30, 300, etc.), and two, doubling and halving figures ences were found in some individual tests. Females scored
(e.g., 200, 100, 50, etc.). This frequent ability to multiply significantly higher on tests of mathematical sentences
and divide by 10 and 2 is used to perform simple arith- and mathematical reasoning, perhaps reflecting the use of
metical operations. Illiterates also use a very important verbal strategies in solving these problems. Males scored
amount of everyday numerical facts: dates (e.g., today is significantly higher than females in geometry, measure-
April 5, 1999), time (e.g., I work eight hours a day: from ment, probability, and statistics, perhaps reflecting the use
8 AM to 4 PM; I am 52 years old), weight (e.g., the cow of visualspatial strategies in these areas. Thus, the gender
weights 350 kg), distance measures (e.g., from my house effect is not a homogenous effect, but varies according to
to the park there are five blocks), and so on. Illiterates the specific calculation tasks. Sometimes, the inverse pat-
can also use simple fractions (e.g., half, quarter, tenth). tern (females outperforming males) can be observed.
In brief, illiterates can develop some calculation abilities Gender differences are observed not only in nor-
(i.e., counting, magnitude estimation, simple adding and mal but also in special populations. DD may be
subtracting). More complex arithmetical skills evidently more frequent in boys than in girls. Gender differ-
benefit and depend on schooling. Noteworthy for illiterate ences are also found in mathematically gifted chil-
people it is notoriously easier to perform concrete math- dren (Bensbow, 1988). There is a significantly higher
ematical operations than abstract arithmetical operations. percentage of males than females in mathematically gifted
Table 11. F Values and Level of Significance for the Gender Differences in the Four Educational Groups
Found in the Three Numerical Items of the NEUROPSI Neuropsychological Test Battery (n = 800) (Adapted
from Ostrosky et al., 1997)
Education
0 year 14 years 59 years 1024 years
F p F p F p F p
Digits backward 2.11 .14 0.98 .32 3.98 .05 6.23 .01
20 minus 3 5.89 .02 0.07 .91 2.41 .12 8.15 .01
Calculation abilities 2.76 .09 0.27 .60 2.22 .01 7.14 .01
children. Furthermore, gender differences in mathemati- have very high spatial abilities (Halperin, 1992). Thus,
cal abilities are progressively higher when moving to more there is ground to suppose that numerical abilities and
extreme scores: gender differences are minimal in those spatial abilities are sharing some common factor. Nonethe-
children one standard deviation above the mean, but higher less, numerical abilities required more than spatial skills.
in those children two standard deviations above the mean Correlations between spatial and numerical abilities, even
and even higher three or four standard deviations above though highly significant, are usually in a moderate
the mean scores. This observation has been confirmed range.
in different countries and remained stable over 15 years Ostrosky et al. (1997) analyzed in Mexico the gender
(Bensbow, 1988). We can assume it represents a quite ro- differences found in the NEUROPSI neuropsychological
bust observation. test battery. Only few differences were statistically sig-
Noteworthy, gender differences increase with age. nificant. No Gender Age interaction effect was found.
No significant differences are observed in elementary Gender partially interacted with education. In the three
school and middle school children. A moderate male su- NEUROPSI tests that include numerical information (Dig-
periority is found in high school, and a large and very its backward, 20 minus 3, and Calculation abilities) statis-
significant advantage is observed in college male students tically significant gender differences were observed. Per-
(Hyde et al., 1990). It means, when numerical knowledge formance was higher in men than in women. Differences
becomes more complex, gender differences become more were particularly evident in the Calculation Ability subtest
significant. (to solve three simple arithmetical problems). Noteworthy,
It has been proposed that gender differences in nu- gender differences in calculation abilities were robust in
merical abilities are a consequence of differences in spatial participants with a high level of education, and minimal
abilities (e.g., Anderson, 1990). Indeed, a strong corre- in illiterates or people with a limited education (Table 11).
lation has been demonstrated between a persons math- Summing up, gender differences in numerical abil-
ematical talent and his or her scores on spatial percep- ity represents a solid observation, confirmed in different
tion tests, almost as if they were one and the same ability studies across different countries. The hypothesis that dif-
(Dehaene, 1997). Fennema and Sherman (1977) report a ferences in calculation abilities are due to differences in
correlation of .50 between scores on a spatial relations spatial abilities has been usually supported. Nonetheless,
tests and achievement in mathematics. Hills (1957) found correlations between both numerical and spatial abilities,
that score on spatial visualization and spatial orientation even though highly significant, only account for a moder-
were correlated with performance in college mathematics ate percentage of the variance. It has to be assumed that
courses in about .23. Ardila et al. (1998) found a cor- not only spatial, but also other types of abilities, are also
relation of about .25 between different arithmetical abil- involved in numerical skills.
ity tests (Mental Arithmetic, Arithmetic Problems, WAIS
Arithmetic subtest) and several spatial tests (Rarcliffs TYPES OF ACALCULIA
test, Perceptual speed, Block design, WAIS Picture com-
pletion). These three calculation ability tests correlated Several classifications have been proposed for acal-
.35, .40, and .28, respectively, with a visuoperceptual fac- culias (e.g., Ardila and Rosselli, 1990; Grafman, 1988;
tor. It means a visuoperceptual factor can account for a Grafman et al., 1982; Hecaen et al., 1961; Lindquist,
significant percentage of the variance in numerical ability 1936; Luria, 1976). The most traditional classification
tests. Interestingly, mathematically gifted children tend to distinguishes between a primary acalculia and a secondary
acalculia (Berger, 1926). This distinction became broadly Table 12. Classification of Acalculias
accepted, and it is usually assumed that acalculia can re- Primary acalculia: Anarithmetia
sult from either a primary defect in computational abilities Secondary acalculia: Aphasic acalculia
(primary acalculia) or a diversity of cognitive defects (lan- In Brocas aphasia
guage, memory, etc.) impairing normal performance in In Wernickes aphasia
In conduction aphasia
calculation tests. Generally, it is considered that acalculia Alexic acalculia
can be correlated with executive function defects (defects In central alexia
in planning and controlling the calculation sequence, im- In pure alexia
pairments in understanding and solving arithmetical prob- Agraphic acalculia
lems, etc.), and visuoperceptual recognition of numerical Frontal (executive dysfunction) acalculia
Spatial acalculia
written information (defects in reading numbers, errors
in reading arithmetical signs, etc.). In other words, acal-
culia can be observed in cases of anterior and posterior
brain damage. Luria (1976) established a distinction be- It is usually assumed that calculation ability repre-
tween optic (visuoperceptual) acalculia, frontal acalculia, sents a rather complex type of cognition requiring the
and primary acalculia, emphasizing that calculation dis- participation of different cognitive abilities. Brain dam-
turbances can result from quite diverse brain pathology. age, nonetheless, may result in relatively restricted dis-
The most influential classification of acalculias was orders in performing arithmetical operations. Benson and
proposed by Hecaen et al. (1961). On the basis of the Denckla (1969) observed that verbal paraphasias may rep-
performance in different calculation tasks of 183 patients resent a source of calculation disturbances in aphasic pa-
with retrorolandic lesions, they distinguished three ma- tients. Ferro and Botelho (1980) found a case of limited
jor types of calculation disorders: (1) alexia and agraphia alexia for arithmetical signs. Warrington (1982) reported
for numbers, (2) spatial acalculia (or acalculia of a spa- a dissociation between arithmetical processing and the re-
tial type), and (3) anarithmetia (primary acalculia). Alexia trieval of computation facts. Benson and Weir (1972) de-
and agraphia for numbers would obviously induce calcu- scribed a patient who, following a left parietal lesion, was
lation disturbances. It may or may not be associated with able to read and write numbers and arithmetical signs and
alexia and agraphia for words. Spatial acalculia represents maintained his rote arithmetical knowledge (e.g., multipli-
a disorder of spatial organization where the rules for set- cation tables), but was unable to carry over when per-
ting written digits in their proper order and position are forming arithmetical operations. Hittmair-Delazer et al.
disrupted; spatial neglect and number inversions are fre- (1994) described a patient affected by an inability to re-
quently found in this disorder. Anarithmetia (or anarith- call and use arithmetical facts of one-digit multiplication
metria or anarithmia) corresponds to primary acalculia. It and division. This impairment contrasted with the preser-
implies a basic defect in computational ability. Anarith- vation of a wide range of complex notions (cardinality
metia does not suppose an isolated defect in numerical judgments, recognition of arithmetical signs, written cal-
concepts and arithmetical operations, but excludes alexia culations, solving arithmetical problems, additions, and
and agraphia for numbers and spatial acalculia. Interest- subtractions). Cippotti et al. (1995) found a patient with
ingly, in their analysis of acalculia Hecaen and colleagues an arteriovenous malformation in the left parietal region,
only included patients with retrorolandic lesions; frontal- who was able to read letters, words, and written number
type acalculia was not considered. names, but was unable to read aloud single Arabic numer-
Ardila and Rosselli (1990) proposed a new classifica- als. His ability to produce the next number in a sequence
tion of acalculias. A basic distinction between anarithme- and answers to simple additions and subtractions was rel-
tia (primary acalculia) and acalculia resulting from other atively spared when the stimuli were presented as number
cognitive defects (secondary acalculias) was included. names but impaired when the stimuli were presented as
Secondary acalculias can result from linguistic defects Arabic numerals. Semenza (1988) reported a patient with
(oral or written), spatial deficits, and executive function a specific deficit for arithmetical procedures due to the
(frontal) disturbances, such as attention impairments, per- systematic application of disturbed algorithms. This pa-
severation, and disturbances in handling complex mathe- tients difficulty stemmed from an inability to monitor the
matical concepts (Table 12). There is, however, a certain sequence of operations that calculation procedures spec-
degree of overlap among these acalculia subtypes. Thus, ify. Cipolotti et al. (1991) observed a patient with the clas-
aphasic, alexic, and agraphic acalculias significantly over- sical signs of Gerstmann syndrome. A significant impair-
lap. Primary acalculia is frequently associated with apha- ment in number processing and number knowledge was
sia, alexia, and agraphia. demonstrated. Nonetheless, the patient showed a largely
preserved ability to deal with numbers below 4 in all tasks a overlap between anarithmetia and alexia and agraphia
and all modalities, whereas she was totally unable to deal for numbers. In a sample of 73 patients with anarithme-
with numbers above 4. Lampl et al. (1994) described a tia, they found that 62% had aphasia, 61% constructional
patient with a selective acalculia for addition, multipli- errors, 54% visual field defects, 50% general cognitive
cation, and division but with intact ability to subtract and deficits, 39% verbal alexia, 37% somatosensory defects,
distinguish mathematical signs. Patients with frontal dam- 37% rightleft confusion, and 33% ocumolomotor de-
age may present selective impairments in using numerical fects. Their sample, however, was too heterogeneous, and
information applied to temporal facts (e.g., How many acalculia could easily be correlated with other neurologi-
years ago did WWII end?) while normally performing cal and neuropsychological defects.
arithmetical operations. Dehaene and Cohen (1997) de- Noteworthy, half of the Hecaen et al.s acalculic pa-
scribed two patients with pure anarithmetia, one with a left tients also presented a general cognitive deterioration. A
subcortical lesion and the other with a right inferior pari- significant correlation between arithmetical abilities and
etal lesion and Gerstmann syndrome. The subcortical case general cognitive performance has been proposed (Ardila
suffered from a selective deficit of rote verbal knowledge et al., 1998). Furthermore, arithmetical ability impair-
(e.g., arithmetical tables), whereas the semantic knowl- ments have been postulated to represent an early sign of
edge of numerical quantities was intact. The inferior pari- dementia (Deloche et al., 1995).
etal case suffered from a category-specific impairment of A memory defect (amnesia for quantities) has been
quantitative numerical knowledge, with preserved knowl- conjectured in rendering acalculic patients unable to carry,
edge of rote arithmetical facts. The potential dissociation borrow, and retrieve arithmetical facts (Cohen, 1971;
of different calculation elements supports the assumption Grewel, 1952). Patients with primary acalculia as a matter
that numerical ability represents a multifactor skill, requir- of fact present a decreased digit span. Their performance
ing the participation of different abilities and quite diverse in the WAIS Digits subtest is usually abnormally low. They
brain areas. frequently state that they get confused with numbers, and
quantities are difficult to understand. Blatant difficulties in
manipulating and memorizing quantities in consequence
Anarithmetia can at least partially account for their computational de-
fects. It is not easy to find cases of primary acalculia with-
Anarithmetia corresponds to primary acalculia. It out additional aphasic, alexic, and agraphic defects. As a
represents a basic defect in the computational ability. The matter of fact, few cases of pure anarithmetia have been
patients with anarithmetia present with a loss of numerical described to date. Some authors have even challenged the
concepts, inability to understand quantities, defects in us- existence of a primary acalculia not associated with other
ing syntactic rules in calculation (e.g., to borrow), and cognitive deficits (e.g., Collington et al., 1977; Goldstein,
deficits in understanding numerical signs. However, they 1948).
may be able to count aloud and to perform some other Anarithmetia is observed in cases of left angular
rote numerical learning (e.g., the multiplication tables). gyrus damage. This localization for primary acalculia has
They may conserve some numerical knowledge but fail in been widely accepted since Henschen (Gerstmann, 1940;
comparing numbers (magnitude estimation) and perform- Grafman, 1988; Henschen, 1925; Levin et al., 1993; Luria,
ing arithmetical operations (Rosselli and Ardila, 1995). In 1973). Rosselli and Ardila (1989) analyzed the errors
primary acalculia the calculation defect must be found in made by a sample of patients with left parietal injuries.
both oral and written operations. That is, anarithmetia is They found that these patients exhibited defects in oral and
a fundamental calculation defect, and is not restricted to written calculations, most of the patients confused arith-
a specific type of output (oral or written). Anarithmetia metical symbols, and all presented errors in transcoding
could be interpreted as an acquired defect in understand- tasks, in successive operations, and in solving mathemati-
ing how the numerical system works. cal problems. It could be proposed that in case of primary
The patient with anarithmetia usually presents errors acalculia, global quantification ability (discriminating be-
in the management of mathematical concepts and incor- tween collections containing different number of objects)
rectly uses arithmetical symbols. The patient also fails in and probably correspondence construction (comparing el-
solving arithmetical problems. Although it is uncommon ements in different collections) are preserved. However,
to find cases of pure anarithmetia caused by focal lesions in the basic principles used in counting elements [viz., (1)
the brain, it is routine to find some anarithmetia in cases one-one principle; (2) stable order principle; and (3) car-
of dementia (Ardila and Rosselli, 1986; Grafman et al., dinal principle] may be impaired. Counting aloud as a
1989; Parlatto et al., 1992). Hecaen et al. (1961) found numerical rote learning nonetheless can be preserved in
Table 13. Calculation Ability Test Performance in a Patient with Pri- numbers and using numerical concepts. He was aware and
mary Acalculia Associated with a Left Angular Gyrus Infarct (Adapted critical of his deficits. Brain MRI showed a small ischemic
from Ardila et al., 2000)
lesion involving the left angular gyrus.
Score When formally tested, a primary acalculia was ob-
served. Digit span forward and backward score was 3
Counting: forward 10/10 Normal for each (first percentile). Significant difficulties were ob-
Backward 9/10 Abnormal
Reading number up to 3 digits 5/5 Normal
served in the WAIS Arithmetic subtest. Forward count-
More than 3 digits 3/5 Abnormal ing was flawless, but he made one omission (1/10) when
Writing numbers 10/10 Normal counting backward. Reading numbers with three or fewer
Greater and smaller relations 5/5 Normal digits was normal. However he demonstrated inversions
Transcoding: verbal to numerical 5/5 Normal (4908 4098) and omissions (10003 1003) reading
Numerical to verbal 3/5 Abnormal
WAIS-R Digits Scaled score = 4 Abnormal
numbers with more than three digits. Writing numbers un-
Arithmetic Scaled score = 5 Abnormal der dictation was normal. Transcoding from verbal to a nu-
Mental arithmetical operations merical code and vice versa, was nearly normal. Only few
Adding small quantities 5/5 Normal literal paragraphias (homophone-orthographic and non-
Subtracting small quantities 3/3 Normal homophonic errors) and some decomposition errors were
Adding and subtracting larger 0/4 Abnormal
quantities
observed in writing (10003 un mil cero tres; one thou-
Multiplications (2 digits) 0/3 Abnormal sand zero three). He understood greater and smaller
Written arithmetical operations relations when comparing two quantities. Oral arithmeti-
Adding and subtracting (3 digits) 4/4 Normal cal operations were correct in adding or subtracting small
Multiplying and dividing (3 digits) 0/4 Abnormal quantities, but he confused adding and subtracting signs.
Arithmetical signs: Reading 4/4 Normal
Interpreting 3/4 Abnormal
Mental multiplications with two digit figures were impos-
Successive operations: adding 10/10 Normal sible. Reading arithmetical signs was correct except for
(1, 4, 7 . . .) adding instead of subtracting when performing written op-
Subtracting (100, 87, 74 . . .) 4/5 Abnormal erations. Addition and subtraction of three-digit quantities
Aligning numbers in columns 10/10 Normal was normal, but multiplication and division were abnor-
Solving arithmetical problems 0/5 Abnormal
mal. Successive operations were correct except a corrected
error when subtracting 13 from 100. Aligning of numbers
in columns during the mathematical operations was nor-
primary acalculia. In aphasic patients, however, count-
mal. He failed in solving arithmetical problems, because
ing elements can be preserved (Benson and Ardila, 1996;
he got confused. Interestingly, this patient strongly com-
Seron et al., 1992). In any event, computational strategies
plained of difficulties understanding antonymous words
required in arithmetical operations (adding, subtracting,
and calculating.
multiplying, and dividing) and mathematical problem-
solving ability are severely disrupted in primary acalculia.
Aphasic Acalculia
Illustration
Calculation difficulties are generally found in apha-
Table 13 presents the performance in a Calculation sic patients, correlated with their linguistic defects. As a
Ability Test in a patient with anarithmetia. MRI images result, patients with Wernickes aphasia exhibit their ver-
are presented in Fig. 2. This patient was a 58-year-old, bal memory defects in the performance of numerical cal-
right-handed male, with high-school-level education. Un- culations. Patients with Brocas aphasia have difficulties
til his cerebrovascular accident, he worked as a success- handling the syntax when applied to calculations. In con-
ful businessman and prestigious politician. Twenty-eight duction aphasia, repetition defects may affect successive
months before the current evaluation he suddenly lost operations and counting backward, which, like repetition,
language production and understanding. Speech therapy require subvocal rehearsal. This means that, ultimately,
was initiated and his language has generally improved, al- the calculation defects could very well have originated
though he remained with significant word-finding disrup- and been correlated with general linguistic difficulties in
tions. Substantial difficulties were present in discriminat- aphasic patients (Grafman et al., 1982). Numerical defects
ing antonyms; such as rightleft, updown, openclose, are simply a result of the linguistic deficits in aphasic pa-
to go in to go out, beforeafter, and overbelow. In addi- tients. Aphasic patients present in consequence an acalcu-
tion, he reported important impairments in understanding lia resulting from the language defect (aphasic acalculia).
Fig. 2. Brain MRI in a patient with primary acalculia. A left angular gyrus infarct is observed.
The overall error rate in various calculation tasks is clearly verbal sequencing activity similar to say backward the
correlated with the severity of the language deficit, global days of the week, and would thus mirror some of calcula-
aphasics being the most impaired patients. The qualita- tion difficulties of the conduction aphasics.
tive error pattern, however, may vary between different Use of morphology and syntax represents one of
types of language disorders, and some numerical aspects, the central impairments in patients with Brocas aphasia
for example, multiplication facts, may be preferentially (Delazer et al., 1999). This is clearly observed in transcod-
mediated by verbal processing (Delazer et al., 1999). ing tasks from a verbal (e.g., three hundred and forty-two,
etc.) to a numerical code (342, etc.) and also from numeri-
cal to verbal. The patient has defects in interpreting gram-
Acalculia in Brocas Aphasia
matical elements pointing to the position of the number
within the class (e.g., when reading three hundred thou-
Dahmen et al. (1982) studied calculation deficits in
sand two hundred is difficult to understand that hundred
patients with Brocas and Wernickes type of aphasia. Us-
in the first and in the second case do not mean the same;
ing a factor analysis, they were able to identify two differ-
hierarchy errors are evident). Brocas aphasia at least par-
ent factors: (1) numericsymbolic and (2) visualspatial.
tially be interpreted as a disorder in language sequencing
The milder calculation defects found in patients with
and consequently, in calculation tasks, numerical sequenc-
Brocas aphasia are derived from the linguistic alterations,
ing will also be altered. Reading not only Arabic numbers
while with Wernickes aphasia, defects in visualspatial
but also number words, and writing number words is ab-
processing significantly contribute to calculation difficul-
normal. Mental and written calculations are significantly
ties. The syntax of calculation is impaired in Brocas apha-
defective (Delazer et al., 1999).
sics. These patients present stack errors (e.g., 14 is read
as 4) that could be interpreted as an agrammatism in the nu-
merical system. They also have difficulties counting back- Acalculia in Wernickes Aphasia
ward and in successive operations (e.g., 1, 4, 7, or 20, 17,
14). Counting forward represents an automatic rote learn- Lexical and semantic errors are most significant
ing, whereas counting backward represents a controlled in Wernickes aphasia. Patients with Wernickes aphasia
present semantic and lexical errors in saying, reading, and (when counting backward from 80 to 70, the patients per-
writing numbers. Luria (1973) suggests that calculation formed 80, 77, 78, 76, 75, 70, 80 . . .). Errors in reading
errors in patients with acousticamnesic aphasia (one sub- (e.g., 49 29) and writing numbers to dictation (e.g.,
type of Wernickes aphasia according to Luria) depend on 3041 3091) were also observed. In transcoding from
their defects in verbal memory. This is particularly notice- numerical to verbal code 3/8 errors (order errors and letter
able in the solution of numerical problems, when the pa- omissions) were noted. In transcoding from numerical to
tient is required to remember certain conditions of a prob- verbal code 3/8 errors were also found (order errors and
lem. The verbal memory span is limited, and the patient hierarchy errors). Aligning numbers in columns was cor-
loses the thread and mixes the conditions of the problem. rect. Simple mental arithmetical operations were errorless.
Lexical errors also play a significant role in the acal- Written operations were difficult and the patient failed in
culia found in Wernickes aphasia. When the patient is 50% of the cases. When reading arithmetical signs, he con-
asked about numerical facts (i.e., How many days are fused plus (+) and multiplication () signs, and stated that
there in a year?), paraphasic errors become evident. he does not know what the minus () sign means. Solv-
The meanings of all words (including number words) are ing arithmetical problems was impossible because of the
weakened. Lexical errors are abundant in different types significant language understanding defect.
of tasks. Benson and Denckla (1969) stressed the presence
of verbal paraphasias as an important source of calculation Acalculia in Conduction Aphasia
errors in these patients.
Wernickes aphasics present semantic errors in the Patients with conduction aphasia (afferent motor
reading and writing of numbers (Delazer et al., 1999; aphasia) frequently present significant calculation errors.
Deloche and Seron, 1982). When writing numbers by dic- They may fail in performing both mental and written oper-
tation, patients with Wernickes aphasia may write com- ations. They have serious flaws in performing successive
pletely irrelevant numbers (e.g., the patients are required operations and in problem solving. In reading numbers,
to write the number 257; they repeat 820, and finally, write errors of decomposition, order, and hierarchy can appear.
193), exhibiting a loss of the sense of the language (nu- They usually fail in carrying over, in the general use of
merical paragraphias). Lexicalization (e.g., 634 is written calculation syntax, and even in reading arithmetical signs
600304) is frequently observed. In reading, they show nu- (Rosselli and Ardila, 1989). Taken together all these poten-
merical paralexias (e.g., 37 is read as 27). Decomposition tial errors, the calculation defect associated with conduc-
errors are frequent (e.g., 1527 is read 15-27) in reading. tion aphasia could be interpreted as anarithmetia. How-
Mental operations, successive operations, and the so- ever, it should be addressed that the topography of the
lution of numerical problems appear equally difficult for damage in conduction aphasia can be close to the topogra-
these patients as a result of their verbal memory, lexical, phy of the damage in anarithmetia. Conduction aphasia, as
and semantic difficulties (Rosselli and Ardila, 1989). Ver- well as anarithmetia, has been correlated with left parietal
bal memory defects are evident in mathematical problem brain injury. The association between conduction aphasia
solving when the patient has to retain different elements and some degree of anarithmetia is not coincidental.
of the problem.
Acalculia in Other Types of Aphasia
Illustration Calculation disturbances are also observed in other

types of aphasia (Benson and Ardila, 1996). In extrasyl-
A 46-year-old right-handed man, professional lawyer vian (transcortical) motor aphasia the patient may have dif-
sustained a vascular accident involving the temporal ficulties initiating and maintaining numerical sequences.
branches of the left middle cerebral artery. In the Boston Problem-solving ability may be significantly impaired,
Diagnostic Aphasia Examination the profile of a typi- and the patient may even fail in understanding what the
cal Wernickes aphasia was found: severe auditory com- problem is about. In extrasylvian (transcortical) sensory
prehension disturbances, naming difficulties, paraphasias, aphasia, significant calculation defects are usually found
and language repetition errors. Language was fluent, abun- associated with the language-understanding difficulties
dant, prosodic, and without articulatory errors. No gram- and echolalia. Temporalparietal damage results in a va-
matical omissions were found, but a significant empty riety of language disturbances and significant calculation
speech was evident. If testing for calculation ability, it was defects. Mental and written calculation can be difficult,
observed that the patient could count forward, but when and errors are observed in writing number words (Delazer
counting backward difficulties and errors were recorded et al., 1999).
Alexic Acalculia even impossible. As a result of the visual exploration de-

fects, aligning numbers in columns and carrying over
Impairments in calculation may also be correlated are tasks the patient usually fails. It is important to stress
with general difficulties in reading. This represents an that reading is performed from left to right (in Western lan-
alexic acalculia or alexia for numbers, and has been recog- guages, at least), but the performance of arithmetical op-
nized since Henschen (1925). Four basic types of alexia erations goes from right to left. This disparity may create
have been described: central alexia, pure alexia, frontal problems in those patients with visual attention problems.
alexia, and spatial alexia. Calculation errors observed in
frontal alexia were analyzed when describing acalculia in
Illustration
the Brocas aphasia, and errors in calculation in spatial
alexia will be analyzed when describing spatial acalculia.
A 45-year-old right-handed man with 8 years of edu-
cation presented with severe chest pains followed by loss
Acalculia in Central Alexia of consciousness. He was hospitalized with transient car-
diac arrest. During the following days, right homonymous
Central (parietaltemporal alexia, or alexia with hemianopsia, right hemi-body extinction, difficulties in vi-
agraphia) alexia includes an inability to read written num- sual exploration, optic ataxia, verbal alexia, and inability
bers and numerical signs. Usually, the ability to perform to recognize objects and colors were observed. CT scans
mental calculation may be considerably better. Quite of- showed a small hypodensity lesion in the left occipital
ten, central alexia is associated with anarithmetia. Reading lobe.
and writing difficulties plus computational disturbances When testing for reading, it was observed that the
may result in severe acalculia. Frequently, in these pa- patient could read letters and syllables. However, he was
tients, reading digits and numbers may be superior to unable to read any word composed of more than three
reading letters and words. Occasionally, the patient may letters. A letter-by-letter or syllable-by-syllable reading
be unable to decide if a symbol corresponds to a letter or was observed. A very significant amount of morphologic
a number. Written mathematical operations are seriously verbal paralexias was observed (e.g., casa cascara).
impaired, and mental execution is superior. No significant word-finding difficulties or language-
Although the distinction between alexia with understanding impairments were observed in conversa-
agraphia for numbers and anarithmetia is conceptually tional language. He could write by dictation, even though
valid, in reality it may be difficult to establish. The pos- he partially overlaps words when writing. A few letter
terior brain topography of the two syndromes is similar omissions were noted when writing. He could read num-
as pointed out by Hecaen et al. (1961). Usually alexia for bers (Arabic and Roman) up to three digits. With longer
numbers and arithmetical signs is associated with alexia numbers, he only read the initial part (e.g., 7528 was read
for letters, some agraphia, and some aphasic disorders. as 75, 2). When reading arithmetical signs, he confused
plus (+) and multiplication (). Transcoding from numer-
ical to verbal code was normal for numbers up to three
Acalculia in Pure Alexia
digits. Transcoding from a verbal to a numerical code was
extremely difficult, as a result of his inability to read long
Pure alexia (alexia without agraphia, or occipital
words. Performance of simple mental arithmetical opera-
alexia) is mainly a verbal alexia in which letter reading is
tion was normal. Performance of written operations was
significantly superior to word reading. As expected, these
slow and abnormal, because of the significant visual scan-
patients present greater difficulties reading numbers com-
ning difficulties. Counting forward and backward was nor-
posed of several digits (compound numbers) than read-
mal. Solving simple arithmetical problems was errorless.
ing single digits. When reading compound numbers, the
Aligning numbers in columns was painstaking, slow, and
patient exhibits decomposition (27 becomes 2,7) (digit-
poorly performed because of the visual scanning defects.
by-digit reading) and hierarchy errors (50 becomes 5) as a
result of the omission for the right-side information. When
reading words, letters placed on the left are generally un- Agraphic Acalculia
derstood better than letters placed on the right. Likewise,
in reading numbers, only the first or the first twothree Calculation errors may appear as a result of an in-
digits are read correctly, and a certain degree of right hemi- ability to write quantities. The calculation deficit may be
neglect is observed (5637 becomes 563). Because of the a function of the type of agraphia. In the agraphia asso-
alexia, performing written operations is painstaking and ciated with Brocas aphasia, the writing of numbers will
Fig. 3. Apractic agraphia illustration.
be nonfluent, with some perseveration and omissions. In write letters. Writing numbers becomes slow and difficult,
transcoding tasks, from the numerical code to the verbal and permanent self-corrections appear.
code, grammatical and letter omissions appear. The patient In cases of motor agraphia, the difficulties observed
presents difficulties in the production of written numerical in writing letters and words will be also found when
sequences (e.g., 1, 2, 3), particularly backward (e.g., 10, writing numbers. In paretic agraphia, numbers are large
9, 8) (Ardila and Rosselli, 1990). and clumsily formed. In hypokinetic agraphia, difficul-
In Wernickes aphasia, there is a fluent agraphia ties in starting the motor activity are evident, as microg-
for numbers. Because of verbal comprehension defects, raphy and progressive narrowing of numbers appear. In
the patient produces errors in writing numbers to dic- hyperkinetic agraphia, numbers are usually large, hard
tation and even writes totally irrelevant numbers (428 to read, and distorted; frequently the patient is unable to
becomes 2530). Lexical errors (numerical verbal para- write.
graphias) and fragmentation (25 becomes 20 . . . 5) are ob-
served. Language-understanding defects impair the ability Illustration
to write quantities to dictation. Solving arithmetical prob-
lems is impossible because of the language-understanding A 33-year-old right-handed man with a high school
defects and verbal memory defects. level of education suddenly during the morning developed
In conduction aphasia, there may be a very signif- right hemiparesis and impossibility to speak. CT scans
icant agraphic defect in the writing of numbers by dic- demonstrated a parietalinsular infarct. Aphasia and hemi-
tation. The patients may be unable to convert the num- paresis rapidly improved during the following days. After
ber that they have heard and even repeated to themselves a formal language evaluation, a diagnosis of a conduction
in graphic form. Order, hierarchy, and inversion errors aphasia associated apractic agraphia was proposed. Count-
are observed (Rosselli and Ardila, 1989). Some degree of ing forward and backward was intact. Reading numbers
apractic agraphia is frequently found in conduction apha- and magnitude estimation (what number is larger) was also
sia, and some degree of apractic speech is frequently found correct. Transcoding from verbal to numerical code was
in certain conduction aphasics. correct even though the patient had significant difficul-
Writing number defects are observed not only in ties writing some numbers. Transcoding from numerical
aphasic but also in nonaphasic forms of agraphia. Aprac- to verbal was notoriously difficult because of the writ-
tic agraphia becomes evident in writing not only word but ing impairments (Fig. 3). Mental successive adding and
also quantities. Self-corrections and approximations are subtracting was errorless. Finding the number or arith-
found, and frequently the patients fail to convert the num- metical sign lacking in an arithmetical operation (e.g., 12
bers they hear in a correct graphic form. Apractic agraphia + . . . = 19; 35 . . . 12 = 23) was appropriate. Solving sim-
impairs the performance of motor sequences required to ple arithmetical problems was normal. Written operations
and aligning numbers in columns were difficult because occurred 1 or 10 years ago. Of course, this deficit is re-
of the defects in writing the numbers. lated to the severe defects in temporal memory and in time
concepts observed in this group of patients (Fuster, 1993).
Frontal (Executive Dysfunction) Acalculia

Illustration
Patients with prefrontal injuries frequently develop
calculation difficulties that are not easily detected. Patients After a seizure, an anterior left hemisphere tumor
with damage in the prefrontal areas of the brain may dis- was disclosed in a 52-year-old right-handed woman with
play serious difficulties in mental operations, successive a college-level education. No language abnormalities were
operations (particularly backward operations; e.g., 100 found in the Boston Diagnostic Aphasia Examination. In
7), and solving multistep numerical problems. Written testing for reading, she could read letters, words, sen-
arithmetical operations are notoriously easier than men- tences, and texts. Understanding of written language was
tal operations. Difficulties in calculation tasks in these normal. However, she read 4/10 pseudowords as real
patients correspond to different types: (1) attention diffi- words, and 4/4 times she could not decide which of two
culties, (2) perseveration, and (3) impairment of complex words was correctly written. Spontaneous writing and
mathematical concepts. Attention deficits are reflected in writing by dictation was normal. Writing numbers with
the patients difficulty in maintaining concentration on the one or several digits was normal. Only one error was
problem. Attention difficulties result in defects in main- recorded when reading numbers (10003 was read One
taining the conditions of the tasks and impulsiveness in million and three). Transcoding (verbal to numerical; and
answers. Perseveration is observed in the tendency to con- numerical to verbal) was also normal, except for one er-
tinue presenting the very same response to different con- ror (twelve thousand three hundred sixty-nine was written
ditions. Perseveration also appears in writing and reading 2369). Magnitude estimation (to decide which number
numbers. Perseveration results in incorrect answer (e.g., is bigger, e.g., 189 and 201), reading of arithmetical signs,
when subtracting 7 from 100: 93, 83, 73, etc.). Impair- counting forward and backward, and simple arithmetical
ment in the use of complex mathematical concepts re- operations were performed without difficulty. Successive
sults in inability to analyze the conditions of numerical additions (1, 4, 7, . . . etc.) were errorless but successive
problems and developing an algorithm for its solution. subtractions (100, 87, 74 . . . etc.) were impossible. Writ-
When trying to solve mathematical problems, the patient ten arithmetical operations with three digits (adding, sub-
may have difficulties simultaneously handling diverse in- tracting, multiplying, and dividing) were normal. The pa-
formation from the same problem and may even be un- tient could successfully find 5/10 times what was lacking
able to understand the nature of the problem. Instead of in some arithmetical operations written on a card (e.g.,
solving the mathematical problem, the patient with frontal 12 + . . . = 15; 93 . . . 13 = 80), and in 5/10 times she
acalculia may simply repeat it. The above defects are re- simply answered, I cannot figure it out, or For me, it
flected in the abnormal handling of complex mathematical looks OK. She failed 3/5 very simple arithmetical prob-
concepts. lems: How many centimeters are there in two and half
The most profound defects are found in solving nu- meters?; Paul receives 8 pesos per working hour. How
merical problems, whereas elementary arithmetic is usu- much will he get working four hours?; Mary and John
ally much better preserved. Mental arithmetic is sig- get 150 in a day. Mary receives twice as much as John.
nificantly more abnormal than written operations, as in How much does each one receive? Aligning numbers in
general, mental tasks are harder than tasks using external columns was normal.
support, and using pen and pencil helps keep track of the
material in operating memory.
Interestingly, patients with frontal lobe pathology Spatial Acalculia
may present notorious disturbances in the use of tempo-
ral measures. Time is measured using quantities (2 hr, Spatial acalculia is observed in patients with right
34 years, etc.). This specific type of numerical knowledge hemispheric damage, particularly parietal lobe pathology.
is significantly disturbed in this group of patients. They It frequently coexists with hemi-spatial neglect, spatial
may be unsure if the discovery of America was carried alexia and agraphia, constructional difficulties, and other
out about 50 years ago, or about 100 years ago, or about spatial disorders (Ardila and Rosselli, 1994; Hecaen et al.,
500 years ago. Patients may not know if their accident 1961).
In cases of spatial acalculia, mental calculation is Spatial acalculia is most frequently observed in right
superior to written calculation. No difficulties in count- hemisphere pathology (Ardila and Rosselli, 1994). Hemi-
ing or in performing successive operations are observed. neglect, topographic agnosia, constructional apraxia, and
A certain degree of fragmentation appears in the reading general spatial defects are usually correlated with spatial
of numbers (523 becomes 23) resulting from left hemi- acalculia. Patients with spatial acalculia perform much
spatial neglect. In large quantities, the patient reads the last better in orally presented arithmetical tasks than in written
or two last digits, with a notorious tendency to omit left- ones.
sided information. Reading complex numbers, in which
the spatial position is critical, is affected, particularly when
the number includes several digits that are repeated (e.g., Illustration
1003 becomes 103). Inversions can be noted (32 becomes
23, or 734 becomes 43) (Ardila and Rosselli, 1990, 1994). A 68-year-old right-handed woman with high school
The difficulties observed in writing numbers are com- education was hospitalized because of a sudden loss
mon across all written tasks. Such difficulties include ex- of sensitivity in her left hemi-body. At the incoming
clusive use or simply overuse of the right half of the page; neurological exam, left hypoesthesia in face and arm,
digit iterations (227 becomes 22277) and feature itera- left homonymous hemianopia, disorientation in time and
tions (particularly when writing the number 3 extra loops place, and left hemi-neglect were found. CT scans revealed
are written); inability to maintain the horizontal direction an ischemic lesion involving the temporal and parietal
in writing; spatial disorganization; and writing over seg- branches of the right middle cerebral artery.
ments of the page already used. When performing written Neuropsychological testing indicated a significant
arithmetical operations, the patient understands how much left spatial hemi-neglect. Neglect was observed in draw-
should be carried over (or borrowed) but cannot find ing, reading, writing, and performing spontaneous activ-
where to place the carried-over quantity. Also, the inabil- ities. Severe spatial and constructional defects were also
ity to align numbers in columns prevents such patients found. A severe spatial acalculia was noted. The patient
from performing written arithmetical operations. When could not align numbers in columns (Fig. 4). Adding and
performing multiplication, the difficulty in remembering multiplying were impossible because the patient could not
multiplication tables becomes obvious, a defect correlated find where to place the carried or borrowed quantities.
with the general difficulty in making use of automatic lev- She confused plus (+) and multiplication () signs
els of language. These patients frequently mix procedures and could not read numbers on the left (e.g., 9,231 was
up (e.g., when they should subtract, they add). This is re- read as 31). Because of these defects, the patient failed
lated to another frequently found defect: they do not seem in performing all the written numerical operations that
surprised by impossible results (reasoning errors). For in- were presented. Nonetheless, her ability to perform arith-
stance, the result of subtraction is larger than the original metical operations orally and to solve numerical problems
number being subtracted. This type of error in arithmetical was nearly normal. A diagnosis of spatial acalculia asso-
reasoning has also been noted in children with DD. ciated with spatial alexia, spatial agraphia, hemi-spatial
Ardila and Rosselli (1994) studied calculation errors neglect, and general spatial defects was presented. Dur-
in a sample of 21 patients. Spatial defects that interfered ing the following weeks neglect mildly improved. Still,
with the reading and writing of numbers and with the loss she continues unable to perform any written arithmeti-
of arithmetical automatisms (e.g., multiplication tables) cal operation. Defects in spatial orientation continued to
were found. The processing system seems abnormal in be significant. One year later she began to attend a re-
these patients while the numerical calculation system is habilitation program that lasted for about 1 year. Im-
partially preserved. Difficulties in calculation procedures provement was significant. At the end of this program,
and problems in the recall of arithmetical principles were the patient could perform simple written arithmetical
observed; however, arithmetical rules were intact. The operations.
authors concluded that the numerical changes observed In summary, it is possible to find very different types
in patients with right hemisphere injury are due to (1) of acquired disorders in calculation skills. Some of them
visualspatial defects that interfere with the spatial orga- represent disability derived from defects (oral and writ-
nization of numbers and mechanical aspects of the math- ten, production and comprehension) in language. Others
ematical operations, (2) inability to evoke mathematical are closely correlated with either spatial defects (spatial
facts and remember their appropriate uses, and (3) inabil- acalculia), executive function deficits (frontal acalculia),
ity to normally conceptualize quantities and to process or primary defects in the performance of arithmetical tasks
numbers. (anarithmetia).
Fig. 4. Spatial agraphia illustration.
Commentary parietal lobe is involved in linguistic processing. Spatial

mechanisms are represented in both the right and the left
Calculation ability might be interpreted as a type parietal lobes in nonhuman primates. In humans, how-
of cognition involving in their origins at least, some ever, language is represented in a region (posterior pari-
type of body knowledge (autotopognosis) (Gerstmann, etal lobe) of the left hemisphere that, in the right hemi-
1940), spatial concepts, language, and executive func- sphere, is involved in spatial functions, and was involved
tion abilities. Finger agnosia (as a restricted form of auto- in spatial functions in both hemispheres in human an-
topagnosia) supports the association between calculation cestors (Lynch, 1980). In consequence, the evolution of
abilities and body knowledge. The association between language involved adaptations in the neural substrate of
spatial knowledge mediated through language and calcu- spatial behavior (LeDoux, 1984). It is understandable that
lation abilities have been emphasized by different authors the left parietal lobe can play a significant role in un-
(e.g., Ardila et al., 1989a,b; Hartje, 1987; Luria, 1966, derstanding spatial concepts mediated through language.
1976). Luria emphasizes that an inability to use verbally Boles (1991), presenting different tasks (recognition of
mediated spatial concepts (semantic aphasia) is always words, products, locations, dichotic digits, etc.) and using
associated with acalculia. a factor analysis, was able to identify different lateralized
The role of parietal lobe in body knowledge and parietal functions: lexical functions (e.g., word numbers)
the disorders of the body scheme in cases of parietal were associated with left hemisphere, whereas primary
pathology have often been emphasized in the literature spatial functions (e.g., locations of dots) were correlated
(e.g., Botez, 1985; Critchley, 1953). Parietal damages with right hemisphere activity.
have been associated with asomatognosia in general, Luria (1966) emphasized that defects in spatial con-
and hemiasomatognosia, alloesthesia, finger agnosia, au- ceptualization underlying the acalculia observed in left-
totopagnosia, asymbolia for pain, apraxia, and the so- parietal-damaged patients. He proposed that left parietal
called VergerDejerine syndrome (Hecaen and Albert, temporaloccipital damage could produce components of
1978). spatial apraxia, agnosia, semantic aphasia, and acalculia.
Asymmetry in cerebral organization of cognition rep- Luria considered that the same cognitive defects were
resents the most outstanding characteristic of the human present in semantic aphasia and acalculia. In both syn-
brain. LeDoux (1982, 1984) proposed that the primary dromes, defects in understanding verbally mediated spa-
functional distinction between human hemispheres in- tial concepts are evident. Therefore, acalculia is always
volves the differential representation of linguistic and spa- associated with semantic aphasia. Spatial knowledge is
tial mechanisms. The right posterior parietal lobe is in- crucial in understanding the numerical system. Cardinality
volved in spatial processing, whereas the left posterior requires the understanding of after, larger, and other
spatially tinged relationships. Quantities and arithmetical can be often correlated with an extrasylvian (transcortical
operations are intrinsically spatial concepts. The numeri- or dynamic) motor aphasia.
cal system, furthermore, assumes an spatial organization However, the association between language distur-
(units, tens, hundreds, etc.) bances and acalculia is not simple. Numbers are coded
In brief, normal performance of calculation tasks re- verbally (e.g., three, two hundred twelve) and in such re-
quires different types of cognition. Language, spatial abil- gard, calculation requires language. But quantities are also
ities, body knowledge, and executive (frontal lobe) func- coded numerically as Arabic digits (e.g., 3, 212). Further,
tion are necessarily included. Calculation abilities can be language can be coded in two modalities (oral and writ-
impaired as a result of language (oral and written), spatial, ten; i.e., phonologic and orthographic). Numbers in con-
and executive function disturbances. General consensus sequence can be coded in three different symbolic sys-
in this regard can be found. The crucial unsettled point is tems (the two language symbolic systems plus the Arabic
whether acalculia can be found in isolation as a pure pri- numerical system). For small quantities, a second numer-
mary disorder with no other cognitive deficit. Or, is acalcu- ical system is sometimes used (Roman numerals). Perfor-
lia always associated with body knowledge disturbances mance of arithmetical operations using Roman numerals,
(rightleft disorientation, finger agnosia, and in general however, is not evident.
autotopagnosia), and disturbances in the linguistic use of Grewel (1952, 1969) considers that the three semi-
spatial concepts (semantic aphasia)? The historical analy- otic systems used for representing quantities (verbal
sis of the evolution of numerical concepts and the clinical oral, verbalwritten, and Arabic digits) are independent
observation seems to provide some support to the second and possess their own distinctive features. Each numer-
point of view. ical system includes several components: (1) the set of
symbols used (lexicon), (2) the semantic representation
of the numbers (semantics), and (3) the syntactic rules
ASSOCIATED DISORDERS used for combining and manipulating numbers (syntax).
Grewel proposes that the three semiotic systems may be
Calculation ability can be regarded as a multifacto- independently affected in cases of brain pathology be-
rial skill. An important association between calculation cause they represent different symbolic systems with a
abilities and performance in different cognitive areas, in- different syntax. Clinical observation demonstrates that
cluding language, memory, constructional abilities, spa- sometimes acalculic patients may fail in transcoding tasks
tial skills, attention, and executive functions, has been (transcoding from the numerical to the verbal system
found. Acalculia is quite often correlated with diverse dis- and vice versa) (Deloche and Seron, 1982, 1984, 1987).
turbances in different cognitive domains. At least the verbal and numerical system may become
dissociated. Further, it seems to exist a general relation-
ship between lexical/syntactical language processing pre-
Acalculia and Aphasia served/impaired abilities in Brocas or Wernickes apha-
sics, and their types of transcoding impairment (Deloche,
Acalculia quite often overlaps with aphasia. Some au- 1993). In Brocas aphasia, lexical processing is better
thors have even stated that primary acalculia is always as- preserved, whereas syntactical processing is impaired.
sociated with a particular type of aphasia known as seman- In Wernickes aphasia the opposite situation is usually
tic aphasia (e.g., Luria, 1976). Frequently, acalculia is also found.
associated with diverse types of language disturbances Deficits in very specific linguistic elements have
very specially, with conduction and Wernickes aphasia been often demonstrated in acalculia. Benson and Denckla
(Rosselli and Ardila, 1989). According to Hecaen et al. (1969) observed that verbal paraphasias may represent a
(1961), 62% of the patients with anarithmetia (primary source of calculation disturbances in aphasic patients. A
acalculia) also presented aphasia. Aphasia was demon- selective disturbance in the oralverbal system was re-
strated in 84% of the patients with alexic acalculia. The sponsible for the calculation defect in this case. As men-
associated aphasia in Hecaen et al.s study was a fluent tioned previously, it has been also observed that the ver-
type of aphasia. bal processing system of number can become dissociated
Anomia, word-finding difficulties, language- from the numerical calculation system (e.g., Dagenbach
understanding defects, and paraphasias are quite and McCloskey, 1992; Pesenti et al., 1994). Dehaene and
frequently correlated with primary acalculia. Aphasia is Cohen (1997) described a patient with a selective deficit of
obviously not expected in spatial acalculia observed in rote verbal knowledge (e.g., arithmetical tables), whereas
cases of right hemisphere pathology. Frontal acalculia the semantic knowledge of numerical quantities was
intact. They also described a second patient with preserved after. The specific semantic of the cardinal depends
knowledge of verbal rote arithmetical facts and distur- upon the particular object it is applied to.
bances at the semantic level of quantities. No question, The syntactic rules used for combining and manipu-
the semantic knowledge of numerical quantities depends lating numbers suppose the understanding about the orga-
on different brain areas than the knowledge of verbal rote nization of the numerical system (Grewel, 1952, 1969).
arithmetical facts (multiplication tables, simple additions The syntactic rules are required for the mathematical
and subtractions, etc.). thought and the comprehension of the underlying oper-
Rossor et al. (1995) described a patient with severe ations (Boller and Grafman, 1983). In primary acalculia,
language problems and good calculation abilities. The disturbances in using syntactic rules and impairments in
patient answered correctly simple subtraction, additions, understanding how the numerical system is organized are
multiplications, and multidigit operation without signif- assumed.
icant difficulty. This patient seemingly did not rely on Delazer et al. (1999) analyzed the pattern of errors
verbal abilities in any of the operations, or compensated in a large sample of aphasic patients. They found as ex-
impaired verbal abilities with nonverbal skills. This report pected that most severe calculation impairments are ob-
supports the assumption that, even though verbal abilities served in global aphasics. Brocas and Wernickes apha-
are involved in numerical tasks, at least in some numerical sics scored similarly at the quantitative level, and amnesic
tasks, there is a functional independence between numer- aphasics showed only mild calculation difficulties (unfor-
ical and verbal abilities. The opposite pattern has been tunately, conduction aphasia patients were not included
also observed: intact language function and impaired cal- in this study). Calculation procedures were mainly im-
culation ability (e.g., Warrington, 1982). All these dis- paired in Wernickes aphasia. Syntactic errors were more
sociations support the assumption that different elements frequent in Brocas aphasia whereas lexical errors were
participate in normal calculation ability. mainly observed in Wernickes aphasia. The authors pro-
At the semantic level, it may be argued that the dif- pose that the retrieval of multiplication facts is preferen-
ferent number representation systems (oral, orthographic, tially mediated by verbal processing. In cases of aphasia,
numerical) share a single semantic representation. It may it will be most impaired.
be conjectured that indeed all the three representation Because of its left parietal topography, primary acal-
systems have in common a single semantics. Quantity culia may be associated not only with aphasia but also with
conceptualization may be proposed to be the semantic alexia and agraphia. The association with parietotemporal
core of the different numerical system representations. alexia (central alexia or alexia with agraphia) is evident.
Deloche (1993), however, has argued that the assumption The association with apractic agraphia is also evident (see
of a unique semantic representation for each number the next section Acalculia and Apraxia). However, unusual
seems unrealistic, considering the constellations of us- cases of acalculia with unexpected localizations have been
ages and meaning of numbers (e.g., items of cardinality, sometimes reported. For example, Tohgi et al. (1995) de-
prices, private bank account number, room numbers scribed a patient with agraphia and acalculia associated
in a hotel, bus line numbers in a city, nursery rhymes, with a left frontal (F1, F2) infarction. Although the patient
idiomatic expressions, etc.). Accordingly, there is not could add and subtract numbers, he could neither multiply
a single semantics for numbers, but semantics depends nor divide because of a difficulty in retrieving the multipli-
upon the specific usage of the numbers. Alternatively, it cation tables and calculation procedures. Lucchelim and
might be conjectured that numbers contain two different De Renzi (1993) report the case of a 22-year-old man who
semantic aspects: (1) the cardinality of the number and had an infarct in the left anterior cerebral artery that de-
(2) the bigger and smaller relationships (magnitude stroyed the medial cortex of the frontal lobe. The patient
comparison). Cardinality is invariant, but bigger and manifested primary acalculia. Accounting for these un-
smaller relations are relative and depend upon the usual acalculia localizations is not easy. Disturbances in
objects that are counted, or they are related with. Two some brain circuitry required for normal numerical oper-
pages is smaller than one book. Five dollars is big ations may be conjectured.
money for a candy, but it is small money for a shirt. Bus
line numbers has not any bigsmall semantic for the
passenger, but may have some technical (bigsmall Acalculia and Apraxia
or beforeafter or whatever) semantic for the city
traffic organization. We are proposing that cardinality The association between acalculia and apraxia is
may represent the invariant semantics of all notational not frequently mentioned excepting in the dementia
numerical systems. Cardinality simply means before syndrome (e.g., Cummings and Benson, 1992) and in
traumatic brain injury cases. In those instances, apraxia Interestingly, mental calculation impairments represent an
and acalculia appear together as a result of an extended important factor in predicting cognitive deficits in these
brain involvement and wide disturbances in cognition. dementia patients (Roudier et al., 1991).
This association is not particularly informative in un- Nonetheless, acalculia in dementia has not been a sig-
derstanding brain organization of calculation abilities, nificant research topic, and usually acalculia is not men-
because acalculia and apraxia appear amid a whole array tioned as a diagnostic criterion of dementia. The most
of diverse cognitive disturbances. frequently used clinical criteria in the diagnosis of AD
However, there is a twofold association between acal- (e.g., NINCDS; American Psychiatric Association, 1994;
culia and apraxia: (1) primary acalculia quite often is as- McKhann et al., 1984) do not list the numerical ability
sociated with ideomotor apraxia, and (2) spatial acalculia defects as a clinical symptom of dementia. Clinical crite-
significantly correlates with constructional apraxia. The ria of dementia usually emphasize memory disturbances,
frequent association between ideomotor apraxia and acal- language deficits, attention defects, and visuoperceptual
culia can be easily overlooked. As a matter of fact, many impairments. As a matter of fact, very little research con-
patients with primary acalculia also present ideomotor cerning the functions of the mathematical abilities in AD
apraxia (Rosselli and Ardila, 1989). This is not a coin- is found. Considering, however, the significant frequency
cidental association. Ideomotor and ideational apraxia are of calculation disturbances in case of brain pathology, it
observed in cases of left parietal damage, and primary may be assumed a high frequency of acalculia in cases of
acalculia too. Nonetheless, it is difficult to know how fre- dementia.
quent ideomotor apraxia is associated with acalculia be- Recently, a few papers have been published point-
cause of the lack of large samples of acalculic patients. ing out that calculation abilities represent indeed a very
Hecaen et al. (1961), however, reported that ideational important defect in AD. Marterer et al. (1996) found a sig-
or ideomotor apraxia was observed in 36.5% of the acal- nificant correlation between arithmetical impairment and
culic patients. Interesting to note, agraphia observed in the degree of dementia. Deloche et al. (1995) reported that
the Gerstmann syndrome is an apractic agraphia apraxia, calculation and number processing scores significantly
not an aphasic agraphia (Benson and Cummings, 1985; correlate with the MMSE (Folstein et al., 1975) scores and
Strub and Geschwind, 1983), and hence, it represents a language performance tests. However, their case analyses
segmentary ideomotor apraxia. In consequence, ideomo- indicated heterogeneous patterns of preserved/impaired
tor apraxia should be regarded as a frequently associated abilities with regard to other cognitive areas. Heteroge-
impairment in cases of acalculia. Unfortunately, ideomo- neous patterns of calculation deficits in brain-damaged pa-
tor and ideational apraxia can be easily overlooked in a tients have been also reported by different authors (Ardila
routine neurological or neuropsychological exam. and Rosselli, 1990; Boller and Grafman, 1985; Delazer
The association between spatial acalculia and con- et al., 1999; Grafman et al., 1982; Rosselli and Ardila,
structional apraxia is quite evident. Patients with right 1989).
hemisphere pathology, particularly right parietal damage, Rosselli et al. (1990) analyzed the calculation abil-
present general spatial and visuoconstructive defects. Spa- ities in AD. Twenty right-handed patients meeting the
tial acalculia is just a single manifestation of the general DSM-IV (American Psychiatric Association, 1994) cri-
spatial defects observed in these patients. Hecaen et al. teria for AD were studied. Age ranged from 64 to
(1961) observed visuoconstructive impairments in 95% 88 years. A neuropsychological test battery including lan-
of the patients presenting spatial acalculia. Spatial agnosia guage, memory, constructional abilities, attention, mathe-
and general visuospatial defects were recorded in 62.5% matics, and abstraction tests was administered. In addition,
of their patients. Ardila and Rosselli (1994) found a signif- the MMSE was also used. Mathematical subtests corre-
icant correlation between spatial acalculia, spatial alexia, lated higher than the MMSE with the scores in the different
spatial agraphia, and constructional apraxia. neuropsychological tests (Table 14). Highest correlations
of the mathematical subtests were observed with language
Acalculia and Dementia repetition, nonverbal memory, and attention tasks. It was
proposed that mathematical ability tests represent an ex-
There is a significant correlation between calcula- cellent predictor of general intellectual performance in
tion abilities and general intellectual performance (Ardila AD. It was further proposed that disturbances in arith-
et al., 1998). Consequently, it is understandable that cal- metical ability should be included as a diagnostic crite-
culation deficits have been reported as an early sign of rion for AD. These results support the hypothesis that a
Alzheimers Disease (AD) (Grafman et al., 1989; Deloche significant association exists between arithmetical impair-
et al., 1995; Mantovan et al., 1999; Parlatto et al., 1992). ments and the severity of dementia (Deloche et al., 1995;
Table 14. Correlations Among the Different Neuropsychological Tests, duction, block design, and line orientation tests. It is note-
the MMSE, and Two Arithmetic Tests in the Dementia Group (Adapted worthy that repetition is one of the language skills well
from Rosselli et al., 1998)
preserved in patients with AD, and in consequence, repe-
Test MMSE Arithmetic Additions tition impairments are considered significantly abnormal
(Benson and Ardila, 1996; Cummings and Benson, 1992).
General cognitive functioning Repetition of sentences involves preserved verbal mem-
MMSE .715 .751
Mathematics
ory span. An important relationship between mathemat-
WAIS-R: Arithmetic .715 .520 ical scores and short-term memory such as visual repro-
Mental Additions .751 .520 duction and immediate memory measured with the digit
Executive functioning span subtest was also observed. Short-term visual mem-
Similarities .167 .197 .307 ory (Visual Reproduction subtest from the Wechsler Mem-
Trail Form B .492 .315 .148
Phonological fluency .287 .113 .451
ory ScaleRevised) had a stronger correlation than did
Attention short-term memory (Logical Memory from the Wechsler
Trail Form A .805 .566 to.557 Memory ScaleRevised) with arithmetical ability tests
Digits .548 .580 .585 and MMSE.
Language Mantovan et al. (1999) observed that patients with
Boston Naming Test .361 .413 .150
Reading Comprehension .273 .009 .368
AD present significant difficulties in complex written cal-
Sentence Repetition .768 .600 .731 culation, but not so severe in the retrieval of arithmeti-
Semantic Fluency .280 .342 .462 cal facts. Low consistency and high variability in the er-
Visuoconstructive ror types suggest that difficulties of patients with AD in
ReyOsterrieth Figure: Copy .354 .279 .307 complex calculation arise from a monitoring deficit and
WAIS-R: Block Design .623 .431 .405
MEMORY
not from incomplete or distorted calculation algorithms.
WAIS-R: Information .351 .486 .395 Deficits in monitoring procedures may be an early and
WMS: Logical Memory .423 .393 .149 common symptom of AD.
WMS: Visual Reproduction .751 .727 .546 In summary, arithmetical ability tests seem to be ex-
cellent predictors of general cognitive performance. It may
be even proposed that arithmetical ability disturbances
should be included as an additional criterion of AD.
Marterer et al., 1996). Calculation abilities in Rosselli
et al.s research were assessed using a problem-solving
test (Arithmetic subtest of the WAIS-R) and an arithmeti- Gerstmann Syndrome
cal mental task (Consecutive Additions by 3s). Both tests
presented strong correlations (.715 and .751, respectively; In 1940 Gerstmann described a syndrome associated
p < .001) with the MMSE scores. These very robust cor- to lesions in the left angular gyrus that characteristically
relations with the MMSE may suggest that arithmetical included deficits of finger agnosia, rightleft disorien-
abilities represent very good predictors of general cog- tation, agraphia, and acalculia (Gerstmann, 1940). Ever
nitive performance in AD. Besides, correlations between since, the existence of a Gerstmann syndrome has not been
arithmetical abilities and neuropsychological test perfor- free of debate and questioning in the literature (Benton,
mance were in general higher than correlation between the 1977; Botez, 1985; Poeck and Orgass, 1966; Strub and
MMSE and neuropsychological test performance. Arith- Geschwind, 1983). In part this debate emerges because
metical ability tests may in consequence be considered to this syndrome usually unfolds either as an incomplete
be even better predictors of general cognitive performance tetrad or in association to other cognitive deficits, particu-
than the MMSE. larly, aphasia, alexia, and perceptual disorders (Frederiks,
Deloche et al. (1995) found in patients with mild AD 1985). The presence of Gerstmann syndrome, complete
a correlation of .74 between total calculation and num- or incomplete, suggests a left posterior parietal damage
ber processing scores, and the MMSE score. They ob- and more specifically, a damage to the left angular gyrus.
served that patients with calculation and number process- Even earlier the name angular syndrome was proposed
ing deficits also showed impaired language performance. in lieu of the more widely recognized Gerstmann syn-
By the same token, in the Rosselli et al. study, AD pa- drome (Strub and Geschwind, 1983). The appearance of
tients who performed significantly lower on arithmetical a Gerstmann syndrome with electrical stimulation of the
tests also did poorly on attention tests, sentence repeti- cerebral cortex in the posterior parietal area supports its
tion, semantic verbal fluency, information, visual repro- angular localization (Morris et al., 1984). Mazzoni et al.
(1990) described a case of a pure Gerstmann syndrome body parts using verbal labels right and left was not
associated with an angular gyrus traumatic damage. defective, but whose performance was always poor when
According to Strub and Geschwind (1983), the angu- mental rotation to a command was required. The authors
lar localization in Gerstmann syndrome lesion would not suggested that a defect in horizontal translation, that is,
have an extension toward the occipital lobe, as Gerstmann mental rotation, accounted for the rightleft disorienta-
proposed, but rather toward the supramarginal gyrus and tion in their patient. Furthermore, that acalculia and other
the inferior parietal gyrus. In this case the agraphia would signs associated with Gerstmann syndrome could also
correspond, then, to an apractic agraphia and not to an evolve from a deficit in the performance of these mental
aphasic one, and consequently the agraphia is not neces- rotations. Similarly, this deficit in mental rotations could
sarily be associated with alexia (Benson and Cummings, potentially be reflected in the impaired understanding of
1985). In fact, in cases of incomplete Gerstmann syn- comparisons, for example, time and place adverbs, found
drome agraphia is usually the missing element, a possible in semantic aphasia. One could infer that a single under-
reflection of the fact that apractic agraphia is not exactly lying deficit, defective mental rotations, may account for
angular, but inferior parietal instead. Furthermore, some- rightleft disorientation, finger agnosia, acalculia, and se-
times agraphia without alexia is observed, again as a result mantic aphasia. Their simultaneous appearance in a single
of an inferior parietal lesion. clinical syndrome is not coincidental. Notwithstanding,
agraphia would still remain unexplained by this unifying
underlying mechanism. It is of interest and perhaps not
RightLeft Orientation surprising that it is precisely agraphia the missing sign in
the Gerstmann syndrome.
All known languages distinguish different words to
refer to right and left; all of them also include some
other spatial relationship words, such as up and down Finger Agnosia
(Greenberg, 1978; Hagege, 1982). Defects in understand-
ing and using these words with a spatial meaning are ob- Finger agnosia, as initially described by Gerstmann
served in cases of left angular gyrus damage. Usually, in 1924 (Gerstmann, 1940), includes the inability to dis-
rightleft disorientation is mentioned in cases of Gerst- tinguish, name, or recognize the fingers not only in the
mann syndrome. Nonetheless, these patients do not only own hands but also in examiners hand or in a drawing
present difficulties in recognizing rightleft, but also other of a hand. The patient presents difficulties to selectively
spatial relationships (e.g., updown, overbelow). Right move the fingers, both by verbal command and by imita-
left disorientation is usually more evident, because it rep- tion. Most evident errors are observed in the index, middle,
resents a relatively hard distinction. Even normal people and ring fingers. Usually the patient has difficulties rec-
sometimes confuse right and left. Rightleft discrimina- ognizing his/her errors and consequently does not try to
tion is also relatively late acquired during child language correct them. Later on, Gerstmann (1940) included finger
development. As a matter of fact, the difficulty observed agnosia, plus rightleft disorientation, agraphia, and acal-
in children during language development, as well as in culia into a single syndrome. Interestingly, finger agnosia
Gerstmann syndrome, is not limited to discriminating right is associated with toe agnosia (Tucha et al., 1997).
and left, but also other spatial relationships. Spatial con- Some authors have proposed that finger agnosia rep-
cepts are strongly reflected in some language elements resents a mild form of autotopagnosia (e.g., Hecaen and
(e.g., place adverbs, prepositions). In contemporary lan- Albert, 1978). However, it has been reported that auto-
guages the underlying spatial content of prepositions is topagnosia and finger agnosia can appear dissociated, and
evident (e.g., to, from, for). These spatial concepts medi- consequently would represent different defects (De Renzi
ated through language may be disrupted in cases of brain and Scotti, 1970). Finger agnosia is a relatively frequent
damage. defect, whereas autotopagnosia represents a quite unusual
Pathogenesis of rightleft disorientation is not com- syndrome. It has been proposed that finger agnosia might
pletely understood. Patients with left posterior damage be a polymorphic phenomenon that includes apraxic, ag-
present more evident difficulties than do right posterior nostic, and aphasic aspects. In consequence, different sub-
damaged patients (Ratcliff, 1979). Rightleft disorienta- types of finger agnosia can be distinguished: visual finger
tion implies difficulties in the application of spatial con- agnosia, finger constructional apraxia, apractic defects in
cepts in the bodys lateral orientation. finger selection, and finger aphasia (anomia) (Schilder and
Gold et al. (1995) observed a patient with Gerstmann Stengel, 1931). The role of parietal lobe in body knowl-
syndrome whose ability to name or point to lateralized edge and the disorders of the body scheme in cases of
parietal pathology have been particularly emphasized in and (7) constructions with attributive relations. He also
the literature (e.g., Botez, 1985; Critchley, 1953). Pari- stated that these spatial disorders not only incidentally
etal damage has been associated with asomatognosia in accompany semantic aphasia, but that semantic aphasia
general, hemiasomatognosia, alloesthesia, finger agnosia, itself was a defect in the perception of simultaneous struc-
autotopagnosia, asymbolia for pain, apraxia, and the so- tures transferred to a higher symbolic level (Luria, 1976).
called VergerDejerine syndrome (Hecaen and Albert, In other words, patients with semantic aphasia have dif-
1978). ficulty understanding the meaning of words tinged with
Historically, calculation abilities seem to develop spatial or quasi-spatial meaning.
from counting, and in child development this begins with There seems to exist a rationale for finding a com-
the sequencing of fingers (correspondence construction) mon brain area and activity for acalculia, finger agnosia,
(Hitch et al., 1987). Finger nomination is usually se- rightleft disorientation, and semantic aphasia. Ardila
quenced in a particular order and this represents a basic et al. (1989a, 2000) proposed to replace agraphia for se-
procedure found in different cultures worldwide, both an- mantic aphasia as a part of the angular gyrus syndrome;
cient and contemporary (Ardila, 1993; Cauty, 1984; Levy- or simply to consider semantic aphasia as a fifth sign of
Bruhl, 1947). In fact, in many contemporary languages a the Gerstmann syndrome. Thus, Gerstmann (or angular)
10- or 20-based system is evident. From the Latin digi- syndrome would include acalculia, finger agnosia (or a
tus, digit can mean both number and finger. Accordingly, more extended autotopagnosia), rightleft disorientation
a strong relationship between numerical knowledge and (and also difficulties with other spatial words), and se-
finger gnosis begins to become evident and some com- mantic aphasia. Sometimes, agraphia without alexia will
monality in brain activity or anatomy can be expected. be observed, but agraphia would result from an inferior
Rightleft discrimination and finger gnosis are strongly parietal lesion, not exactly from an angular pathology.
interdependent and can even be interpreted as components
of the autotopagnosia syndrome.
TESTING FOR ACALCULIA
Semantic Aphasia Testing for calculation abilities in brain-damaged pa-

tients has four main purposes:
Commonly, investigators have reported the presence
1. To find out abnormal difficulties in calculation
of Gerstmann syndrome without aphasia as one of its
tasks. If difficulties are severe enough, the diag-
components (Roeltgen et al., 1983; Strub and Geschwind,
nosis of acalculia may be proposed.
1983; Varney, 1984). However, the existence of a possible
2. To distinguish the specific pattern of difficulties
semantic aphasia has not been specifically explored nor
the patient presents, and the particular type of
ruled out (Ardila et al., 1989a). In 1926, Head described
acalculia it corresponds to. Disturbances in dif-
a language alteration whereby he defined an inability to
ferent elements of the calculation system can be
recognize simultaneously the elements within a sentence,
proposed.
and he called it semantic aphasia. In the following years
3. To find out associated deficits. Acalculia is usually
only a small group of researchers referred to this type
associated with diverse cognitive disturbances.
of aphasia (e.g., Conrad, 1932; Goldstein, 1948; Zucker,
4. To describe the types of errors observed in the pa-
1934). Nearly 40 years later, Luria (1966, 1973, 1976) re-
tient. This information will be particularly useful
took this concept and extensively analyzed it. Since then,
in developing rehabilitation procedures.
however, only a few more authors have shown special in-
terest in studying semantic aphasia, and just a handful of Despite the general agreement that a comprehensive
references have appeared in the literature over the last two cognitive evaluation should include testing for calcula-
decades (Ardila et al., 1989a; Benson and Ardila, 1996; tion abilities, there is a significant limitation in the avail-
Brown, 1972; Hier et al., 1980; Kertesz, 1979). able testing instruments. Lezak (1995) overtly states that
Luria (1973, 1976) considered that language defi- An assessment of cognitive functions that does not in-
ciencies observed in semantic aphasia included the fol- clude an examination of calculation skills is incomplete
lowing: (1) sentences with a complex system of successive (p. 647). The WAIS-III Arithmetic subtest (Wechsler,
subordinate clauses; (2) reversible constructions, particu- 1997) is probably the most widely used instrument when
larly of the temporal and spatial type; (3) constructions testing for calculation abilities in neuropsychology. This
with double negative; (4) comparative sentences; (5) pas- subtest, however, has two significant limitations: (1) it is
sive constructions; (6) constructions with transitive verbs; assessing just a single aspect in numerical processing (to
mentally solve arithmetical problems) and (2) it is very dif- EC301

ficult to administer to patients with language and memory
defects. Deloche et al. (1989, 1994) attempted to develop
A few models of testing for calculation abilities a calculation ability testing instruments, with norms by
have been developed (Ardila and Rosselli, 1994; Deloche age, educational level, and gender. This instrument, the
et al., 1994; Grafman et al., 1982; Harvey et al., 1993; EC301 calculation ability test battery, is composed of
Luria, 1966; Rosselli and Ardila, 1989; Warrington, 1982). 31 subtests that cover 8 compound arithmetical functions
However, norms regarding the effects of age, gender, and and 5 single functions. Up-to-dately, the EC301 can be
very special education are not easily available. Even the considered as the best instrument in testing for calcula-
WAIS-III Arithmetic subtest does not control educational tion abilities. The following sections are included in the
level. EC301:
There is, however, a kind of general agreement about
how to test for calculation abilities. Testing for acalcu- 1. Counting (three subtests).
lia is rather similar everywhere. There are certain tasks 1.1. Spoken verbal counting (four items): by
that are supposed to be most important for targeting cal- ones from 1 to 31, by tens from 10 to 90, by
culation abilities: reading and writing numbers, perform- threes from 3 to 33, and counting backward
ing arithmetical operations, aligning numbers in columns, by ones from 22 to 1.
solving arithmetical problems, and so on. That is the 1.2. Digit counting requires counting by ones
way calculation abilities are tested in neuropsychology from one to 31 (1,2,3, . . . 31).
worldwide. 1.3. Written verbal counting has two items:
Lezak (1995) considers the following calculation counting by ones from 1 to 16 (one,
ability tests as most frequently used in assessing numerical two, three . . . sixteen), and counting by tens
skills: from 10 to 90 (ten, twenty . . . ninety).
2. Dot enumeration (five subtests). In all subtests,
1. WAIS Arithmetic subtest.
responses have to be produced in Arabic digit
2. Arithmetical problems. For example, those prob-
form. In Subtests 1 and 2, items contain some
lems used by Luria (1976) in assessing frontal
dots and are either canonical (dominoes) or non-
lobe pathology. A typical example is as follows:
canonical patterns, organized in subgroups with
There are 18 books in two shelves. In one shelf
no more than five dots. In Subtests 3, 4, and 5,
there is the double of books than in the other. How
participants have to point to the dots while count-
many books are there in each one? There are
ing. The arrangement of the dos varies according
different arithmetical problems included in psy-
to spatial organization.
chological test batteries, such as the Stanford-
3. Transcoding (seven tasks). Considering that
Binet Intelligence Scale (Terman and Merril,
numbers can be represented in three different
1973).
codes (spoken verbal, written verbal, and Ara-
3. Wide Range Achievement Test (WRAT) Arith-
bic) the six possible transcoding plus repetition
metic subtest (Jastak and Wilkinson, 1984).
are included. Each subtest contains six items, and
4. WoodcockJohnson Psycho-Educational Bat-
the structure of the items is similar (thousands,
teryRevised (WJ-R) (Woodcock and Johnson,
hundreds, tens, units).
1989).
4. Arithmetical signs (2 subtests).
The best-designed and almost only acalculia stan- 4.1. Reading aloud arithmetical signs.
dardized neuropsychological test battery was developed 4.2. Writing from dictation.
by a group of European neuropsychologists headed by 5. Magnitude comparison (two subtests). Eight
Deloche (Deloche et al., 1989). Preliminary normative re- pairs of numbers are included in each subtest
sults in 180 participants stratified by education, age, and asking the participant to indicate which number
gender were further available (Deloche et al., 1994). This is larger.
battery was named as EC301. 5.1. Arabic digit form.
In this chapter, a general description of the EC301 5.2. Written verbal form.
test battery will presented. Later, a model for testing cal- 6. Mental calculation (two subtests). The four ba-
culation abilities in clinical neuropsychology will be pro- sic operations are presented with two items of
posed. In the final part of this paper, an analysis of the medium difficulty.
errors found in calculation abilities disturbances will be 6.1. Spoken verbal.
introduced. 6.2. Arabic digit condition.
7. Calculation approximations (1 subtest). The par- a one-to-one correspondence. As a matter of fact, it is a

ticipant is required to estimate the result of dif- purely visuoperceptual process, usually not included in
ferent arithmetical operations. The response has testing for calculation abilities. Enumeration represents
to be selected from a multiple choice of four- the most elementary type of truly numerical knowledge.
number array. Counting is a sophisticated form of enumeration: a unique
8. Placing numbers on an analogue line (two sub- number name is paired with each object in a collection, and
tests). Participants have to point to the place of the final number name that is used stands for the cardinal
a number among four alternatives (correct and value of that collection. Testing for calculation abilities
three distractors) indicated by ticks on a verti- can begin at the counting level.
cal scale with 0100 as marked end points. Five
items are included in each subtest. 1. Counting.
8.1. Arabic digit form. 1.1. Counting real objects. A card with 10 ran-
8.2. Spoken verbal numbers. domly distributed dots can be used. The par-
9. Writing down an operation (one subtest). The ticipant is required to count aloud the dots.
ability to organize two numbers (two-digit and 1.2. Counting forward. To count from 11 to
three-digit) spatially is tested. The participant 20 can be used because this is usually the
is required to organize on a sheet of paper the most irregular segment in different numer-
numbers to perform the four basic arithmetical ical systems.
operations, but without the operation is actually 1.3. Counting backward. Even though count-
done. ing forward is clearly verbal rote learning,
10. Written calculation (three subtests). There are counting backward does not represent a rote
two items in each subtest. The three subtests in- learning. Attention plays a significant role
clude addition, subtraction, and multiplication. to count backward from 30 to 20.
11. Perceptual quantity estimation (one subtest). 2. Cardinality estimation corresponds to the task
Participants are shown six photographs of real named as Placing numbers on an analogue line
objects or sets of elements. They have to estimate by Deloche et al. (1994). The participant has to
either weights, lengths, or number of elements. point to the place of a number among four alter-
12. Contextual magnitude judgment (one subtest). natives (correct and three distractors) indicated
The task contains five items. Participants have by ticks on a vertical scale with 0100 as marked
to indicate of a number of objects or per- end points. Some five items can be included in
sons represent a small, medium, or large quan- each subtest.
tity, according to some context (e.g., 35 per- 2.1. Arabic digit form (10, 75, 40, 9, 87).
sons in a bus is a medium number of persons, 2.2. Spoken verbal numbers (eleven, ninety,
whereas 20 pages in a letter has to be rated as twenty-five, sixty, thirty-five).
a large letter, despite the fact that 35 is larger 3. Reading numbers.
than 20). 3.1. Arabic digits. To read numbers with differ-
13. Numerical knowledge (one subtest). There are ent levels of complexity: 3, 27, 298, 7,327,
six questions for each of which there is only one 10003.
correct answer (such as the numbers of days of 3.2. Roman numerals. Usually general popula-
the week). tion people can read Roman numeral up to
12, because they are frequently included in
The EC301 is currently in normalization process in clocks. Roman numerals heavily rely on the
different countries. spatial position III, IV, VI, XI, IX.
4. Writing numbers. Different levels of complexity
can be used (7, 31, 106, 1,639, 50,002,). Quanti-
A Proposed Model for Testing Calculation Abilities ties are dictated to the patient: one-digit numbers,
two-digit numbers, numbers requiring the use of
We are proposing a model for testing calculation abil- a positional value, and so on.
ities, that is just and extension of the EC301 (Deloche 5. Transcoding.
et al., 1989, 1994). Global quantification (or numerosity 5.1. From a numerical to a verbal code. Num-
perception) represents the most elementary quantification bers using a numerical code are written
process. However, global quantification does not represent (e.g., 7, 23, 109, 9,231, 1,000,027), and
yet a truly numerical process, because it does not suppose the patient is asked to write them with
letters (e.g., seven, twenty-three, one hun- difficulties, but also because of attention defects
dred nine, etc.). and perseveration. These successive arithmetical
5.2. From a verbal to a numerical code. Num- operations should be performed aloud.
bers using a verbal code are written (e.g., 10.1. Adding (e.g., 1, 4, 7 . . . 37).
three, forty-nine, three hundred seventy- 10.2. Subtracting (100, 93, 86 . . . 16).
six, nine thousand two hundred seventeen, 11. Mental calculation. Adding, subtracting, multi-
seven millions seven hundred nine), and the plying, and dividing two-digit quantities.
patient is asked to write these quantities 11.1. Adding: 12 + 25, 31 + 84, 27 + 16, 76 +
with numbers (e.g., 3, 49, 376, etc.). 40, 89 + 36.
6. Reading and writing arithmetical signs. 11.2. Subtracting: 35 14, 74 65, 44 29,
6.1. Reading arithmetical signs written on a card 91 59, 70 39.
(+, , : , , =). If the patient answers 11.3. Multiplying: 14 3, 17 8, 29 3, 14
correctly, the question And what does it 13, 27 12.
mean? can be presented. 11.4. Dividing: 15/3, 75/5, 60/15, 89/16.
6.2. Writing arithmetical signs by dictation 12. Written calculation. Adding, subtracting, multi-
(plus, minus, divided, by, equal). plying, and dividing two- and three-digit quanti-
7. Numerical rote learning. Usually it is accepted ties.
that multiplication tables represent a numeri- 12.1. Adding: 398 + 724.
cal rote learning, required for multiplications. 12.2. Subtracting: 721 536.
Nonetheless, numerical rote learning includes 12.3. Multiplying: 127 89.
as well adding and subtracting one-digit quanti- 12.4. Dividing: 465/27.
ties, required in adding and subtracting. Numer- 13. Aligning numbers in columns. A series of num-
ical rote learning can be preserved in cases of bers (e.g., 27, 2, 2,407, 12,057, 1,421,967) are
parietal acalculia, but may be impaired in cases dictated to the patient. The patient is required to
of subcortical damage and right hemisphere place them in a column, as for adding.
pathology. 14. Arithmetical operations using a different numer-
7.1. Multiplication tables (3, 7, and 9). ical base. Arithmetical operations generally use a
7.2. Adding one-digit quantities (3 + 2, 5 + 3, decimal base. There is, however, a type of every-
7 + 6, 9 + 4, 8 + 6). day numerical knowledge that does not use the
7.3. Subtracting one-digit quantities (7 5, 5 decimal system, namely, time measures. Hours
2, 9 6, 8 5, 9 5). use a 12-base and minutes and seconds a 60-base.
8. To complete an arithmetical operation. This is Potentially, this is an excellent calculation ability
a very traditional test in evaluating calculation test. Sometimes, however, this task may appear
abilities in patients with brain pathology. Differ- hard for normal people (How long time is there
ent arithmetical operations are written on a card, between 8:30 and 11:15? How long time is
but one number or arithmetical sign is missing. there between 10:45 AM and 2:10 PM? How
The patient must say what number or arithmeti- long time is there between 3:55 and 7:10?).
cal sign is missing. 15. Fractions. The use of fractions represents a rela-
8.1. Numbers (15 + . . . = 37; 15 . . . = 45; tively difficult numerical ability. The numerical
12 . . . = 4 . . . /5 = 3). value of fraction is inverse to the absolute number
8.2. Arithmetical signs (12 . . . 18 = 30; 17 . . . 6 value (i.e., one sixth is larger than one seventh).
= 11; 60 . . . 4 = 15; 9 . . . 6 = 54). Adding and subtracting fractions may be difficult
9. Magnitude comparisons. The patient is asked even for normal people.
to judge which number is larger and which is 15.1. To compare fractions. (What is larger be-
smaller. An error in this task does not mean that tween one-half and one-third?; one-third
the patient does not understand what number is and one-fourth?; one-seventh and one-
larger. Numbers due to flaws in differentiating eighth?).
between tens, hundreds, and so on. Errors may 15.2. Adding and subtracting (How much is
also be due to the difficulties understanding the one-half plus one-half?; How much
words larger and smaller (e.g., 24 42, 29 is one-half plus one-third?; How much
61, 103 97, 1,003 795, 428 294). is one-fifth plus one-third?).
10. Successive arithmetical operations. Successive 16. Digit span. It has been demonstrated that pa-
operations can be failed as a result of calculation tients with acalculia have significant difficulties
memorizing digits. In cases of brain pathology, lasts?; How long time does it take to walk one
some specific types of amnesia are observed: block (100 m) at a regular walking speed?].
patients with aphasia have difficulties memoriz- 21. Magnitude estimation. The relative value of
ing words, patients with apraxia have difficulties numbers depends upon the context. Ask the pa-
memorizing sequences of movements, and patient to state if an object or a person represents a
tients with acalculia have difficulties memorizing too much or too little quantity according to
numbers. For the acalculic patient, numbers are a specific semantic context. (e.g., A book with
confusing and the meaning of digits is weak. In 20 pages, is it too much or too little?; The
consequence, digit repetition represents a good weight of an adult person is 70 pounds, is it too
task to test for numerical representation. much or too little?; One hundred people in a
16.1. Digits forward. (Series of digits are pre- bus, is it too much or too little?; To spend two
sented at 1-s interval. The initial series has hours brushing the teeth, is it too much or too
three digits 4, 2, 7. If the patient fails, a little?).
different 3-digit series is presented 9, 3, 22. Numerical problems. Problems requiring the use
6. If the patient repeats it errorless, a 4- of one or several numerical operations are pre-
digit series is presented. When the patient sented to the patient (e.g., How many oranges
fails two consecutive times in repeating a are there in two and half dozens?; John had
list with certain number of digits, the test 12 dollars, received 9 and spent 14. How much
is stopped.) money does he have now?; Mary and John get
16.2. Digits backward. (Beginning with a 2-digit 150 in a day. Mary receives the double than John.
series, the patient is required to repeat it How much each one receives?; The sum of the
backward. Two opportunities are allowed ages of a father and a son is 48. If the father has
for each series. When the patient fails two the triple of age of the son, how old are each one
consecutive times a list with a certain num- of them?; There are 18 books in two shelves;
ber of digits, the test is stopped.) in one of them there is double number than in the
17. General numerical knowledge. (e.g., How many other. How many books are there in each one?).
days are there in one week?; How many 23. Using money. Performing arithmetical opera-
weeks are there in one year?; At what tem- tions with money represents one of the most im-
perature the water boils?; How many conti- portant everyday uses of arithmetical operations.
nents are there?; What is the population in this Low-educated people and elderly people with
country?) mild dementia can do better performing arith-
18. Personal numerical knowledge. Everybody uses metical operations with money than in abstract.
a diversity of personal numerical knowledge that Hence, the use of money may represent a poten-
can include the own phone number, address, per- tially useful calculation test. Different bills and
sonal identification, car tag, social security num- coins can be provided to the participant (e.g., five
ber, and so on. Ask about three items (e.g., phone $1 bills, two $5 bills, one $10 bill, and one $20
number, address, social security number). bill; five pennies, five nickels, three dimes, and
19. Quantity estimation. To estimate the weights, three quarters). The patient can be required to se-
lengths, or number of elements. Interestingly, lect certain quantity (Give me 17 dollars and 77
this is a task that presents a significant disper- cents, give me 23 dollars and 18 cents, give
sion in scores when administered to normal pop- me 39 dollars and 94 cents).
ulation people. (How much an egg weights?;
Table 15 summarizes the different tests and subtests
How long an average car is?; How many peo-
that can be used when testing for calculation abilities.
ple can normally travel in a bus?).
20. Time estimation. Time estimation is an abil-
ity not usually including in testing for acal- Analysis of Errors
culia. Some brain-damaged patients, particu-
larly patients with frontal lobe pathology, may A diversity of errors potentially can be observed in
present very significant difficulties estimating performing calculation tasks (Table 16). Nonetheless, six
time. [How long time ago did the WWII end?; major errors groups can be separated: (1) errors in digits,
How long time ago was the discovery of Amer- (2) errors in carrying over, (3) borrowing errors, (4)
ica?; How long time ago the current presi- errors in basic principles, (5) errors in algorithms, and (6)
dent was elected?; How long a regular movie errors in arithmetical symbols.
Table 15. Summary of the Areas that Potentially Can Be Included in a ciding the larger of two numbers (magnitude comparison).
Test for Calculation Abilities This type of error may result in impairments in appreci-
1. Counting ating the tens, hundreds, and so on, or difficulties under-
1.1. Counting real objects standing and confusions between larger and smaller.
1.2. Counting forward This defect in magnitude comparison is observed in cases
1.3. Counting backward
2. Cardinality estimation
of primary acalculia.
2.1. Arabic digit form Patients with Wernickes aphasia frequently present
2.2. Spoken verbal numbers lexicalization errors: Two hundred fifty is written 20050.
3. Reading numbers This type of errors is observed in writing. In reading, a
3.1. Arabic digits frequent type of errors observed in these patients, but also
3.2. Roman numerals
4. Writing numbers
in patients with pure (occipital) alexia is decomposition.
5. Transcoding The number 15 is read as one five and the number 537
5.1. From a numerical to a verbal code is read fifty-three seven. In pure alexia this situation can
5.2. From a verbal to a numerical code result in a digit-by-digit reading. Inversion errors (12 is
6. Reading and writing arithmetical signs read or written as 21) is observed in spatial acalculias, but
6.1. Reading arithmetical signs
6.2. Writing arithmetical signs
also can be found in fluent aphasia.
7. Numerical rote learning Substitution errors in oral language, writing, or read-
7.1. Multiplication tables ing simply mean that a number is substituted by a different
7.2. Adding one-digit quantities one. The errors refer to the number hierarchy (e.g., 30 be-
7.3. Subtracting one-digit quantities comes 3), order (5 becomes 6), or the so-called stack
8. To complete an arithmetical operation
8.1. Numbers
errors (e.g., 15 becomes 50).
8.2. Arithmetical signs Omissions and additions are not unusual types of
9. Magnitude comparisons errors in cases of brain pathology. Omissions can be ob-
10. Successive arithmetical operations served right-sided (most frequently in pure alexia) or at
10.1. Adding the left (most frequently in left hemi-spatial neglect). Ad-
10.2. Subtracting
11. Mental calculation
ditions are found in digits (23 becomes 233) but also in
11.1. Adding strokes when writing a digit, most typically, the number 3
11.2. Subtracting is written with extra loops.
11.3. Multiplying In transcoding tasks (from numerical to verbal or
11.4. Dividing from verbal to numerical) different types of errors may
12. Written calculation
12.1. Adding
be found, including decomposition, additions, omissions,
12.2. Subtracting and substitutions. Transcoding is a relatively difficult cal-
12.3. Multiplying culation task.
12.4. Dividing
13. Aligning numbers in columns
14. Arithmetical operations using a different numerical base Errors in Carrying Over
15. Fractions
15.1. To compare fractions
Primary difficulties in carrying over, that is, inabil-
15.2. Adding and subtracting
16. Digit span ity to understand the calculation principle for carrying
16.1. Digits forward over is observed in anarithmetia. Nonetheless, most pa-
16.2. Digits backward tients with calculation defects may fail in carrying because
17. General numerical knowledge of language difficulties, spatial difficulties, and so on. In
18. Personal numerical knowledge
spatial acalculia, difficulties in correctly placing the car-
19. Quantity estimation
20. Time estimation ried over quantity represent the most important difficulty.
21. Magnitude estimation The patient knows what should be carried over but does
22. Numerical problems not know where to place it.
23. Using money
Borrowing Errors
Errors in Digits
Correctly borrowing supposes an understanding of
Errors in digits represent a fundamental type of error the arithmetical rules of permutability. Evidently, in
in cases of acalculia. The patient may have difficulties de- anarithmetia, not only carrying over, but also borrowing
Table 16. Classification of the Types of Errors in Calculation Tests (Adapted from Ardila and Rosselli, 1994; Levin et al., 1993; Rosselli and Ardila,
1989, 1997 Spiers, 1987)
Type of error Description Type of acalculia
Errors in digits
Number value Inability to distinguish the larger of two numbers because of Anarithmetia
flaws in differentiating between tens, hundreds, etc. or defects in
understanding larger, smaller
Lexicalization Numbers are written as they sound without integrating tens, hundreds, etc. Aphasic
(Wernickes)
Decomposition Numbers are read without considering the number as a whole Anarithmetia; aphasic (Wernickes);
alexic (pure)
Inversions Numbers are copied, written, and repeated in the reverse order. This Spatial posterior aphasias
rotation can be partial
Substitutions A number is substituted by another due to paralexia, paraphasia, or
paragraphia, affecting the result of an operation
1. Hierarchy errors: The number is substituted by another from a different Posterior aphasics
position of series (e.g., 5 50)
2. Order errors: The substitution is by another from the same series Aphasia (posterior and anterior);
(e.g., 5 6) anarithmetia
3. Stack errors (e.g., 14 40) Aphasic; anarithmetia
Omission One or several digits are omitted (can be on the left or right side) Spatial; pure alexia
Additions A digit is inappropriately repeated upon writing it. Addition of traits Spatial
to a digit (usually 3)
Addition of numbers previously presented can be a perseveration Frontal
Errors in Transcoding When numbers are passed from one code to another code (numerical Aphasic; anarithmetia
to verbal or vice versa), decomposition, order, omission, or addition
errors are noted
Errors in carrying
over
Omission The patient does not carry over Anarithmetia
Incorrect carrying Any carrying-over error with acalculia All acalculias
over
Incorrect placement Carries over correctly but adds in the wrong column Spatial
Borrowing errors
Borrowing zero Difficulties or confusion if the arithmetical operation has a zero Anarithmetia
Borrowing The last digit on the left is not reduced despite verbalizing the loan Frontal acalculia
Defective borrowing Adding the borrowed quantity incorrectly; borrowing unnecessarily Anarithmetia
Errors in basic
principles
Multiplication tables Incorrect recall; usually the patient tries to correct it with additions Spatial
in series
Zero Fundamental errors are observed when a zero is present Anarithmetia
Errors in Algorithms
Incomplete Initiates operation correctly but is not capable of finishing Frontal
Spatial 1. Inappropriate use of space on the paper limiting the correct response Spatial
2. Inappropriate use of columns in arithmetical operations Spatial
Incorrect sequence Initiates the operation from left to right Anarithmetia
Inappropriate Numbers are organized spatially on the page for a different operation than Anarithmetia; Spatial
algorithms the desired one (e.g., in multiplication, the numbers are placed
to be divided)
Mixed procedures Different operations are used in the same problem (e.g., Adds in one column Spatial
not multiplies in another)
Reasoning errors The subject does not realize that the result is impossible (e.g., The result is Spatial
larger than what was subtracted)
Errors in symbols
Forgetting The patient cannot remember nor write the four arithmetical signs Anarithmetia
Substitution The sign is substituted by another that is different from what was asked Spatial Acalculia
is defective. It is not infrequent that patients with frontal This is simply impossible and absurd, and this type of er-
lobe pathology omit to borrow. To borrow and to carry ror is known as reasoning error. The reason for this type
over, suppose a normal working memory, an internal proof error is unclear.
cessing of the numerical information, and an appropriate
level of attention. Errors in Arithmetical Signs
Errors in Basic Principles Errors in reading and using arithmetical signs are as-
sociated with anarithmetia. For even some types it may be
There are certain basic principles that have to be used the major manifestation of primary acalculia. Patients with
all the time to successfully perform an arithmetical task. a spatial type of acalculia may confuse the directionality
Patients with right-hemisphere damage associated with of the plus (+) and times () arithmetical signs. This
spatial acalculia quite frequently have defects in recalling confusion will obviously result in a failure in performing
the multiplication tables. This defect is also observed in the correct arithmetical operation.
some cases of subcortical pathology. Using the zero repre-
sents a relatively complex numerical concept. Abnormali-
REHABILITATION OF CALCULATION
ties in using the zero are found in primary acalculia. Some
DISORDERS
other patients can also fail when the zero is present. For
example, in spatial acalculia the patient can get confused
The majority of brain-injured patients (especially in
in that there are several zeros (e.g., 100003), but they also
vascular and traumatic injury cases) present some sponta-
have difficulties when any digit is repeated several times
neous cognitive recovery during the first months after the
(e.g., 122225).
injury. Afterwards, the spontaneous recovery curve be-
comes slower and requires the implementation of rehabil-
Errors in Algorithms itation programs to achieve some additional improvement
(Lomas and Kertesz, 1978).
Patients with frontal damage typically have difficul- Two strategies have been proposed to explain the re-
ties finishing an arithmetical operation or solving a prob- habilitation of cognitive difficulties: the reactivation of
lem, even though they can begin to perform the arithmeti- the lost cognitive function and the development of an al-
cal operation or solving the arithmetical problem. This ternative strategy that achieves the same result through
difficulty in ending a task, as a matter of fact, is observed an alternative cognitive procedure. The majority of the
in different areas. For instance, when writing, the patient rehabilitation models developed for aphasias, alexias, and
can begin to write, but does not end the writing. agraphias have emphasized the second strategy, which im-
In cases of spatial defects, an inappropriate use of plies a cognitive reorganization (Seron et al., 1992) or a
the space on the paper limits the possibility of obtaining a functional system reorganization (Luria, 1973). A solid
correct response in solving an arithmetical task. Patients model applicable to calculation rehabilitation still does
with right-hemisphere pathology and spatial acalculia can not exist. The rehabilitation of calculation abilities is fre-
present as well difficulties in spatially organizing the num- quently neglected in the neuropsychology literature and
bers for performing a particular operation. The distribution ignored in most rehabilitation programs. Acalculia is usu-
of the numbers is different, for example, when multiply- ally evaluated and rehabilitated as a language-dependent
ing and dividing. An abnormal spatial distribution of the function. There are few investigative efforts directed at
quantities impedes a successful performance. studying the rehabilitation of calculation deficits.
Arithmetical operations follow a sequence from right The remediation of arithmetical facts has been re-
to left, that is the opposite sequence to that used in writ- ported using a case study design (McCloskey et al.,
ing and reading. This is a fundamental principle when 1991b; Girelli et al., 1996). The patients in both stud-
performing an arithmetical operation. Errors in the di- ies presented a selective multiplication deficit. McCloskey
rectionality are associated with misunderstanding of the et al. (1991b) reported a selective training effect for re-
positional value of units, tens, hundreds, and inability to peated multiplication problems that extended to the prob-
understand the whole numerical system. This type of di- lems complements but did not improve performance in
rectionality error is found in primary acalculia. general (e.g., the training of 3 6 would improve 6 3
There is a very interesting type of error sometimes but not 3 8). Girelli et al. (1996) described two acalculic
found in cases of right hemisphere pathology. When sub- patients who totally lost multiplication facts. After the re-
tracting, for example, the result is larger than the minuend. mediation procedure both patients presented a decrease
in the error rate of trained and untrained multiplication formance of arithmetical procedural sequences and often
problems. The performance in the trained set, however, make mistakes in recognizing arithmetical symbols.
improved more than the performance in the untrained one. Patients who have primary acalculia combined with
In the course of the training the error types changed dras- semantic aphasia (and, according to some authors, this
tically from unsystematic errors to errors clearly related combination is constant) (Ardila, 1993; Ardila et al.,
to the correct answer. 1989a; Luria, 1973), the comprehension defects extend
The cognitive model of number processing has also themselves to logicgrammar relationships in language.
been used in the development of a training program in a In cases of semantic aphasia, numbers lose their relation-
13-year-old with a severe difficulty in number-transcoding ship with the conceptual system and are perceived in a
tasks (Sullivan, 1996). The transcoding of numbers from concrete and isolated manner (Tsvetkova, 1996). These
written verbal to Arabic and from spoken verbal to Ara- patients present numerous errors in the larger than and
bic was impaired whereas numeral comprehension was smaller than tests, perceiving the number 86 as larger
preserved. After the training program the participant dis- than 112, because they consider the independent value of
played significant gains in the Arabic transcoding tasks. each number. Although the concrete denomination of dig-
The lack of consistency in the type of syntactic errors its (reading digits) is preserved, it is impossible for them
made by the patient and the control participants made the to use abstract numerical concepts. They are unable to
authors unable to interpret the generalization of their re- recognize the number of tens and hundreds included in
sults in terms of the cognitive model used. the number (e.g., in 800) or of understanding the content
A good neuropsychological assessment is the first of relationships such as 30 = 10 + 10 + 10. Accord-
step in the rehabilitation of acalculia. A test battery devel- ingly, Tsvetkova proposes a structured rehabilitation plan
oped specifically to evaluate acalculia should be used to for anarithmetia aimed at recovering the understanding
analyze the disturbances of calculation in a patient with of the composition of numbers and their positional value.
brain injury. Variables such as the patients educational Initially, the patient relearns the concept of numbers by
level and occupational activity should be carefully con- performing tasks consisting of putting together real ob-
sidered. Once the presence of acalculia is determined, a jects (tokens or sticks) and illustrations that contain the
quantitative and qualitative analysis of the patients er- corresponding number. The tasks consist of dividing the
rors in different sections of the test battery should be per- objects into groups (initially, alike, and afterwards, differ-
formed. The justification of the rehabilitation plan should ent), counting the number of objects in each group, finding
be based on the limitation that acalculia has in the patients the illustration that represents the corresponding number,
occupational and social life. The patient should preserve placing it in each group, deciding how many of these num-
sufficient cognitive capability that will allow the develop- bers are found in the given amount, and finally, writing the
ment of a new compensatory behavior or an alternative number on a sheet of paper.
strategy (De Partz, 1986; Seron, 1984). Once the patient has reacquired the concept of digits
In the following sections, techniques that have been and tens, he/she moves on to developing the concept of nu-
developed to rehabilitate patients with primary and sec- merical composition, interrelationships between numbers,
ondary acalculias are presented. The majority of the meth- and the possibility of operating with them. Then, numer-
ods described have been implemented in individual cases. ical denomination exercises (beginning with the second
Until now, no study exists that evaluates the effectiveness tenth) and comprehension exercises between the name of
of these techniques in large samples of patients. the number and its position are initiated. The patient be-
gins to understand that the name of the number indicates
Primary Acalculia Rehabilitation its positional value and the left-to-right reading indicates
to him/her a decreasing positional value (e.g., 154: one
Primary acalculia or anarithmetia is associated with hundred and fifty-four). During this time, the positional
parietal or parietaloccipital injuries (Ardila and Rosselli, composition of the numbers and their quantitative signif-
1992). Tsvetkova (1996) considers that underlying pri- icance depending on their place in the series is worked
mary acalculia is an alteration in the spatial perception on. Tsvetkova emphasizes the importance of using con-
and representation of numbers along with defects in verbal crete mediators like tokens. When the patient also presents
organization of spatial perception. The alteration in spa- anomia for numbers, one should work on the reestablish-
tial coordination systems constitutes a central underlying ment of the naming of numbers.
defect in this type of acalculia (Luria, 1973; Tsvetkova, The understanding of numbers constitutes a recur-
1996). These patients present defects in numerical con- ring learning process to relearn the arithmetical opera-
cepts, in understanding number positions, and in the per- tions. This relearning should always be initiated in the
most explicit and concrete way possible, using external and inaccuracy in visual motor coordination (optic ataxia).
aides such as outlines, drawings, and so on. Verbalizing Treatment should then include exercises that permit spa-
aloud the steps that should be followed is generally useful. tial analysis and visual motor ability training. Rehabil-
As patients improve, they go from speaking aloud to mur- itation tasks are implemented following a program that
muring, then to speaking to themselves. Training for progressively increases difficulty, beginning with simple
particular problems (i.e., 9 0) can lead to the recupera- movements designed for reaching for or indicating objects
tion of arithmetical rules (n 0 = 0) (McCloskey et al., followed by copying figures in two dimensions, and con-
1991a). The training for particular problems or operations cluding with the construction of three-dimensional figures
within the rehabilitation sessions leads then to the com- (Sohlberg and Mateer, 1989). The training in the repro-
prehension of arithmetical principles and rules (Girelli et duction of designs of different forms, colors, and sizes
al., 1996). can be initiated with aides. For instance, the patient is
Understanding calculation direction should be asked to finish a design already started until he/she can
worked on simultaneously with the relearning of arith- finally perform the task completely and independently
metical operations. If the patient with anarithmetia also (Ben-Yishay, 1983). Sohlberg and Mateer (1989) pro-
presents aphasic problems, language should be rehabili- pose, as a procedure to evaluate the generalization of
tated first. Calculation rehabilitation can be implemented the task, obtaining a base line over the performance of
once an appropriate level of linguistic comprehension and 10 designs, noting the accuracy, time of execution, and
production is achieved (Tsvetkova, 1996). the number of aides required. The therapist can choose
In conclusion, the patient with anarithmetia should 5 out of 10 designs for training. When the execution
relearn the basic concepts that underlie the numerical sys- desired from these five designs is achieved, the perfor-
tem. These basic concepts range from knowing numbers mance in the five unused designs during training is eval-
to handling them within the system of operations. uated with the goal of observing the effects of training.
This generalization should be looked for in untrained vi-
Rehabilitation of Secondary Acalculias sual motor tasks that require the same underlying skill
(Gouvier and Warner, 1987). When a visual search defect
In patients with difficulties in recognizing numbers (ocular apraxia) exists, visual pursuit tasks may help to
as a result of a perceptual deficit, the rehabilitation compensate.
process is directed at the recovery of steadfastness and Rosselli and Ardila (1996) describe the writing and
the generalization of visual perception. These patients reading rehabilitation of a patient with Balints syndrome,
frequently present low scores on reading numbers tests with severe ocular apraxia. They used visual movement
and on transcoding numbers from one code to another, exercises such as (1) demonstrating the visual pursuit
with numerous rotation errors. (Rosselli and Ardila, of objects; (2) placing the index fingers at a distance of
1989). The visualperceptual difficulties affect the 15 cm from the sides of the face and requiring the patient
execution of written numbers tasks, in contrast to an to look toward the left and right index fingers 10 times con-
adequate performance of mental arithmetical operations. secutively, and practicing convergence exercises; and (3)
When writing is preserved in these patients (as is the case from a central point at a distance of 30 cm, the patient must
of alexia without agraphia), writing numbers in the air bring the right or left index finger toward his/her nose, per-
helps their recognition. manently maintaining visual contact. In addition, visual
Tsvetkova (1996) proposes using the number recon- kinesthetic exercises were included in the rehabilitation
struction technique with these patients. The technique plan; the patient was shown letters he had to reproduce in
includes number reconstruction by starting from certain the air, and, later, he had to say the name of the letters.
visual elements (e.g., completing eight, starting from the Likewise, when following words, the patient should simul-
number 3), looking for certain elements within a number taneously perform the movements of writing these words.
(e.g., looking for the number 1 in the number 4), and fi- In place of letters, numbers may be used. Within the vi-
nally, performing a verbal analysis of the similarities and sual searching exercises described by Rosselli and Ardila
differences that can be observed between numbers. At the (1996), looking for words and letters in letter groups that
same time that the number reconstruction technique is progressively become more complex was included. Time
used, spatial orientation exercises, comprehension of the and precision were recorded.
right-to-left relationship, and visual analysis of geometri- Patients with aphasic acalculia that receive therapy
cal objects and forms should be developed. for their oral disorder usually improve significantly and
Patients with alexia without agraphia generally in parallel fashion with the improvement of the calcula-
present spatial integration difficulties (simultanagnosia) tion disturbance (Basso, 1987). Acalculia rehabilitation
in these patients parallels language rehabilitation using culties in adequately exploring their environment, several
denomination techniques, auditory verbal memory tech- rehabilitation programs have been directed to develop this
niques, and semantic conceptual classification techniques. ability (Weinberg et al., 1977). Within the rehabilitation
When acalculia is fundamentally derived from defects in techniques for hemi-inattention during reading, the fol-
phonological discrimination, prominent errors in oral nu- lowing is discussed: (1) placing a vertical line on the left
merical tasks are found. Therefore, within the rehabil- margin of the paragraph to be read, and (2) numbering
itation program, visual stimuli should be used initially the beginning and end of each line. As the treatment ad-
(Tsvetkova, 1996). vances, the clues are eliminated until the patient is finally
Patients with frontal lesions generally present perse- capable of reading without help. Upon diminishing the
verations and attention difficulties that prevent an adequate hemi-spatial neglect in general, spatial defects in reading
performance on calculation tests. These patients usually diminish simultaneously (Ardila and Rosselli, 1992). In
do not present errors in naming or recognizing numbers. the recovery of spatial agraphia, it has been suggested us-
Tsvetkova (1996) proposes the idea of providing control ing sheets of lined paper, which limit the writing space. It
strategies to patients that will allow them to direct their at- is also suggested to draw vertical lines that mark spaces
tention and reduce perseveration. These control strategies between letters and words.
refer to descriptions of the steps that the patient should fol- Rosselli and Ardila (1996) describe the rehabilitation
low to satisfactorily complete the task. When faced with of a 58-year-old woman with spatial alexia, agraphia, and
the problem of forming the number 12, by starting from acalculia associated to a vascular injury in the right hemi-
other numbers, the following steps can be described to the sphere. The rehabilitation process was based in the reha-
patient: (1) forming the number 12 by starting from other bilitation of unilateral spatial neglect and associated spa-
numbers with the help of addition, the patient is asked to tial difficulties. The patient could adequately perform oral
use the maximum number of combinations; (2) achieving calculations but was completely incapable of performing
the same number by starting from other numbers with the written arithmetical operations with numbers composed of
help of subtraction, the patient is asked to use the maxi- two or more digits. In a special test of written arithmetical
mum number of combinations; and (3) achieving the same operations (additions, subtractions, multiplications, and
number starting from other numbers with the help of mul- divisions), an initial score of 0/20 was obtained. Left hemi-
tiplication. The maximum number of combinations should inattention, a mixing up of procedures and the impossi-
be used. The patient is trained to verbalize and follow the bility of adequately orienting the columns were observed.
necessary steps. Because these patients do not generally The rehabilitation techniques implemented included the
present defects in mathematical procedures, it is not nec- following:
essary to provide them with special instructions for the
execution of each operation. 1. Using short paragraphs with a red vertical line
The use of permanent verbalization is a useful tech- placed on the left margin and with the lines num-
nique with patients with visualperceptual difficulties. bered on the left and right sides, the patient, using
Spatial acalculia is associated with hemi-inattention (uni- her index finger, had to look for the numbers cor-
lateral spatial neglect), which can be observed in right as responding to each line. The clues (vertical line
well as left injuries (Rosselli et al., 1986). Although it and numbers) were progressively eliminated.
is notoriously more frequent and severe in cases of right 2. In a text with no more than 12 lines, the patient
brain injury, unilateral spatial neglect or hemi-inattention had to complete the missing letters (i.e., perform
refers to the inability to respond (attend to stimuli) pre- sequential and ordered spatial exploration).
sented in the contralateral visual field to the brain injury. 3. Letter cancellation exercises were repeated con-
These patients tend to present number omissions on the stantly, and clues to facilitate their execution were
opposite side of the brain lesion. The hemi-spatial neglect included. Time and precision were recorded.
constitutes one of the factors that interferes most with an 4. In spontaneous writing exercises using lined pa-
adequate cognitive recovery. Although hemi-spatial ne- per with a thick colored line in the left margin,
glect is frequently associated with hemianopsia (visual the patient had to look for the vertical line when
loss in the contralateral field to the lesion), it should be finishing each line. Later, the line was eliminated,
evaluated independently. Frequently, cancellation tasks, but the patient had to verbalize (initially aloud
copies of drawings, visual search tasks, bisection of a line, and later to herself) and explore to the extreme
and a drawing of clock are used, as well as tasks that help left before beginning to read the next line.
to overcome the neglect. On the basis of the hypothesis 5. To facilitate the relearning of numbers through
that patients with unilateral spatial neglect present diffi- dictation, squares were used to place the
numbers in space, and the concepts of hierar- very important that the instructor record all errors com-
chy were practiced permanently (units, tens, hun- mitted by the child, with the purpose of analyzing the
dreds, etc.). cognitive processing steps that have problems (Rosselli,
6. To provide training in arithmetical operations, she 1992).
was given in writing additions, subtractions, mul- Counting by tens starting at a number other than 10
tiplications, and divisions with digits separated in is the first step in developing a technique that utilizes the
columns by thick colored lines and the tops of basic structure of the decimal system in teaching addition
the columns were numbered (from right to left). (Neibart, 1985). In children with dyscalculia, the train-
The patient had to verbalize the arithmetical pro- ing to count by tens should start at 10 (10, 20, 30 . . . )
cedures and, with her right index finger, look for and after several repetitions move to another number (3,
the left margin before she could pass to the next 13, 23). According to Neibart, the use of block is im-
column. Later, the patient herself would write the portant for the child to discover the concept that is
operations she was dictated. then internalized. Once the student has mastered count-
ing by 10 at any number, the addition of 10 (23 + 10) is
The techniques described previously were proven
developed.
useful 8 months after the treatment was started. The pa-
Sullivan (1996) developed a training program for a
tient presented significant improvement but in no way a
child with transcoding number difficulties. He was unable
complete recovery.
to transcribe a number given in written verbal or in spo-
ken verbal (six hundred forty thousand sixty-four) into an
Rehabilitation of Developmental Dyscalculia Arabic code (640, 064). The training program consisted of
introducing a syntactic frame (H = hundreds; T = tens, O
Strang and Rourke (1985) recommend that the reme- = ones) for Arabic numeral production. It was shown how
dial programs for children with dyscalculia include, when to use the syntactic frame. For example, for the numeral six
possible, systematic and concrete verbalizations of the op- hundred eighty-seven, he was told that because the num-
erations and arithmetical procedures. The operations that ber contains a hundred quantity the frame must have three
involve mechanical arithmetic should be converted into slot. He was then shown how to fill the frame with the Ara-
verbal tasks that permit the child to take apart the oper- bic number (6/H, 8/T, 7/0). After he had understood how
ations, and, in this way, facilitate his/her learning. The to fill a syntactic frame successfully, he was able to gener-
teaching method should be clear, concrete, precise, and ate the syntactic frames on his own. After the training he
systematic. showed a significant improvement in transcoding number
Once the child has developed an adequate recognition tasks.
of the numbers, one should begin to work with calculation The childs parents should be taught about the learn-
difficulties. Initially, one should choose an arithmetical op- ing strategy that is being used so that in family activities
eration that presents a problem, and describe it verbally (e.g., shopping), they practice the same arithmetical activi-
in such a way that the child can repeat the description ties, and the generalization of the remedial program can be
independently of whether or not he/she understands the promoted. Because these children with DD present atten-
underlying mathematical concept. Later, the child should tion and visualperceptual discrimination difficulties, it is
verbalize the steps that should be followed to perform convenient to involve the child in tasks that require a de-
the operation in question (e.g., Step 1, name the mathe- tailed description of visual stimuli. In addition, one should
matical sign; Step 2, direct eyes toward the right, etc.). work on the parts in an organized manner. The difficulties
Once the different steps have been verbalized, the child in the interpretation of social situations can be improved
should write them and repeat them orally as many times by creating artificial situations (e.g., movies, photographs,
as needed. Then, the instructor should use concrete aides etc.) so that the child can interpret the images and the con-
(table, equipment, and places) to explain the mathematical text. The instructor can give clues to perceive the fictional
concept. circumstances and the meanings of the gestures (Ozols
Squared sheets of paper should always be used. and Rourke, 1985).
At times, the use of colors helps the discrimination of The prognosis of DD depends on variables such
the rightleft. Each time the child is presented with an as the severity of the disorder, the degree of the
arithmetical problem, it should be read out to mini- childs deficiency in the execution of neuropsycholog-
mize the possibility of his/her forgetting visual details. ical tests, the promptness of the initiation treatment,
It is useful to have a pocked calculator at hand so that and the collaboration of the parents in the remedial
the child can revise the results of the operations. It is program.
CONCLUSION Bensbow, C. P. (1988). Sex differences in mathematical reasoning ability

in intellectually talented preadolescents: Their nature, effects, and
possible causes. Behav. Brain Sci. 11: 169232.
Calculation ability represents a very complex type of Benson, D. F. (1979). Aphasia, alexia and agraphia. New York: Churchill
cognition, including linguistic (oral and written), spatial, Livingstone.
memory, body knowledge, and executive function abili- Benson, D. F., and Ardila, A. (1996). Aphasia: A Clinical Perspective.
New York: Oxford University Press.
ties. Considering its complexity, it is not surprising to find Benson, D. F., and Cummings, J. L. (1985) Agraphia. In: Frederiks, J.
how frequent it is impaired in cases of focal brain pathol- A. M. (ed.), Handbook of Clinical Neurology: Clinical Neuropsy-
ogy and dementia. Neuropsychology has usually recog- chology (Vol. 45), Elsevier, Amsterdam, pp. 457473.
Benson, D. F., and Denckla, M. B. (1969). Verbal paraphasias as a source
nized the importance of developmental and acquired cal- of calculations disturbances. Arch. Neurol. 21: 96102.
culation disturbances. However, there is some paucity in Benson, D. F., and Geschwind, N. (1970). Developmental Gerstmann
the research devoted to the neuropsychological analysis syndrome. Neurology 20: 293298.
Benson, D. F., and Weir, W. F. (1972). Acalculia: Acquired anarithmetria.
of calculation disturbances. The difficulty in finding cases Cortex 8: 465472.
of isolated acalculia may have played some role in the Benton, A. (1977). Reflection on the Gerstmann syndrome. Brain Lang.
relative paucity of research in this area. 4: 4562.
Benton, A. (1992). Gerstmanns syndrome. Archives of Neurology 49:
445449.
Ben-Yishay, Y. (1983). Working approaches to the remediation of cog-
REFERENCES nitive deficits in brain damaged persons. Rehabilitation Monograph
No. 6, NYU Medical Center Institute of Rehabilitation Medicine,
New York.
American Psychiatric Association (1987). Diagnostic and Statistical Berger, H. (1926). Uber Rechenstorunger bei Herderkraunkunger des
Manual of Mental Disorders, DSM-III-R, American Psychiatric As- Grosshirns. Arch. Psychiatr. Nervenkr. 78: 236263.
sociation, Washington, DC. Boles, D. B. (1991). Factor analysis and the cerebral hemispheres: Pilot
American Psychiatric Association (1994). Diagnostic and Statistical study and parietal functions. Neuropsychologia 29: 5991.
Manual of Mental Disorders, DSM-IV, American Psychiatric As- Boller, F., and Grafman, J. (1983). Acalculia: Historical development
sociation, Washington, DC. and current significance. Brain Cogn. 2: 205223.
Anderson, J. R. (1990). Cognitive Psychology and Its Implications (2nd Boller, F., and Grafman, J. (1985). Acalculia. In: Fredericks, F. A. S.
Ed.), Freeman, New York, pp. 177193. (ed.), Handbook of Clinical Neurology: Clincial Neuropsychology
Antell, S. E., and Keating, D. (1983). Perception of numerical invariance (Vol. 45), Elsevier, Amsterdam, pp. 472482.
by neonates. Child Dev. 54: 695701. Botez, M. (1985). Parietal lobe syndromes. In: Frederiks, J. A. M. (ed.),
Ardila, A. (1993). On the origins of calculation abilities. Behav. Neurol. Handbook of Clinical Neurology: Clinical Neuropsychology (Vol.
6: 8998. 45), Elsevier, Amsterdam, pp. 6386.
Ardila, A., Lopez, M. V., and Solano, E. (1989a). Semantic aphasia Brown, J. W. (1972). Aphasia, agnosia and apraxia. Springfield:
reconsidered. In: Ardila, A., and Ostrosky-Solis, F. (eds.), Brain Thomas.
Organization of Language and Cognitive Processes, Plenum Press, Burbaud, P., Degreze P., Lafon, P., Franconi, J. M., Bouligand, B.,
New York, pp. 177193. Bioulac, B., et al. (1995). Lateralization of prefrontal activation
Ardila, A., Rosselli, M., and Rosas, P. (1989b). Neuropsychological as- during internal mental calculation: A functional magnetic resonance
sessment in illiterates: Visuospatial and memory abilities. Brain imaging study. J. Neurophysiol. 74: 21942200.
Cogn. 11: 147166. Calpadi, E. J., and Miller, D. J. (1988). Counting in rats: Its functional
Ardila, A., Galeano, L. M., and Rosselli, M. (1998). Toward a model of significance and the independent cognitive processes that constitute
neuropsychological activity. Neuropsychol. Rev. 8: 177189. it. J. Exp. Psychol. Anim. Behav. Process. 14: 317.
Ardila, A., Concha, M., and Rosselli, M. (2000). Angular gyrus Caramazza, A., and McCloskey, M. (1987). Dissociation in calculation
(Gerstmanns) syndrome revisited: Acalculia, finger agnosia, right processes. In: Deloche, G., and Seron, X. (eds.), Mathematic Dis-
left disorientation, and semantic aphasia. Aphasiology 14: 743 abilities: A Cognitive Neuropsychological Perspective, Erlbaum,
754. Hillsdale, NJ, pp. 221234.
Ardila, A., and Ostrosky, F. (1984). Some final remarks. In: Ardila, Cauty, A. (1984). Taxonomie, syntaxe et economie des numerations
A., and Ostrosky-Solis, F. (eds.), The Right Hemisphere: Neurol- parlees. Amerindia 9: 111146.
ogy and Neuropsychology, Gordon and Breach Science, London, Childe, V. G. (1936). Man Makes Himself, Pitman, London.
pp. 265273. Cipolotti, L., Butterworth, B., and Denes, G. (1991). A specific deficit
Ardila, A., and Rosselli, M. (1986). La vejez: Neuropsicologia del for numbers in a case of dense acalculia. Brain 114: 2619
fenomeno del envejecimiento [Neuropsychology of aging], Prensa 2637.
Creativa, Medellin (Colombia). Cipotti, E. K., Warrington, W. K., and Butterworth, B. (1995). Selective
Ardila, A., and Rosselli, M. (1990). Acalculias. Behav. Neurol. 3: 39 impairment in manipulating Arabic numerals. Cortex 31: 7387.
48. Clark, J. M., and Campbell, I. D. (1991). Integrated versus modular
Ardila, A., and Rosselli, M. (1992). Neuropsicologia Clinica [Clinical theories of number skills and acalculia. Brain Cogn. 17: 204239.
neuropsychology], Prensa Creativa, Medellin (Colombia). Cohen, R. (1971). Dyscalculia. Arch. Neurol. 4: 301307.
Ardila, A., and Rosselli, M. (1994). Spatial acalculia. Int. J. Neurosci. Collington, R., LeClerq, C., and Mahy, J. (1977). Etude de la semologie
78: 177184. des troubles du calcul observes au cours des lesions corticales. Acta
Badian, N. A. (1983). Dyscalculia and nonverbal disorders of learn- Neurol. Belg. 77: 257275.
ing. In: Miklebust, H. R. (ed.), Progress in Learning Disabilities Conrad, K. (1932). Versuch einer psychologischen Analyse des Parietal
(Vol. 5), Grune and Stratton, New York, pp. 129146. Syndromes. Montschrift fur Psychiatrie und Neurologie, 34: 11
Basso, A. (1987). Approaches to neuropsychological rehabilitation: Lan- 20.
guage disorders. In: Meier, M., Benton A., and Diller, L. (eds.), Critchley, M. (1953). The Parietal Lobe, Edward Arnold, London.
Neuropsychological Rehabilitation, Guilford Press, New York, Croxen, M. E., and Lytton, H. (1971). Reading disability and difficulties
pp. 167182. in finger A sequential processing deficit? Ann. Dyslexia 42: 6989.
Cummings, J. L., and Benson, D. F. (1992). Dementia: A clinical Ap- Fuster, J. M. (1993). The prefrontal cortex. New York: Raven Press,
proach (2nd Ed.), Butterworth-Heinemann, Boston. 1989, 2nd edition.
Dagenbach, D., and McCloskey, M. (1992). The organization of arith- Geary, D. C. (1993). Mathematical disabilities: Cognitive, neuropsycho-
metic facts in memory: Evidence from a brain damage patient. Brain logical and genetic components. Psychol. Bull. 114: 345362.
Cogn. 20: 345366. Gelman, R., and Meck, E. (1983). Preschoolers counting: Principles
Dahmen, W., Hartje, W., Bussing, A., and Sturm, W. (1982). Disorders before skill. Cognition 13: 343359.
in calculation in aphasic patientsSpatial and verbal components. Gerstmann, J. (1940). The syndrome of finger agnosia, disorientation for
Neuropsychologia 20: 145153. right and left, agraphia and acalculia. Archives of Neurology and
Davis, H., Albert, M., and Barron, R. W. (1985). Detection of number Psychiatry 44: 398408.
or numerousness by human infants. Science 228: 1222. Girelli, L., Delazer, M., Semenza, C., and Denes, G. (1996). The re-
Davis, H., and Perusse, R. (1988). Numerical competence in animals: mediation of arithmetical facts: Evidence from two rehabilitation
Definitional issues, current evidence and a new research agenda. studies. Cortex 32: 4966.
Behav. Brain Sci. 11: 561615. Gold, M., Adair, J. C., Jacobs, D. H., and Heilman, K. M. (1995). Right-
Davis, H., Bryson, S. and Hoy, C. (1992). Case study of language and left confusion in Gerstmanns syndrome: A model of body centered
numerical disability: A sequential processing deficit? Annals of spatial orientation. Cortex, 61: 267284.
Dyslexia 42: 6989. Goldstein, K. (1948). Language and Language Disturbances, Grune and
Dehaene, S. (1997). The Number Sense. How the Mind Creates Mathe- Stratton, New York.
matics, Oxford University Press, New York. Gouvier, W., and Warner, M. (1987). Treatment of visual imperception
Dehaene, S., and Cohen, L. (1997). Cerebral pathways for calculation: and related disorders. In: Williams, J. M., and Long, Ch. (eds.), The
Double dissociation between rote verbal and quantitative knowl- Rehabilitation of Cognitive Disabilities, Plenum Press, New York.
edge of arithmetic. Cortex 33: 219250. Grafman, J. (1988). Acalculia. In: Boller, F., Grafman, J., Rizzolatti, G.,
Delazer, M., Girelli, L., Semenza, C., and Denes, G. (1999). Numerical and Goodglass, H. (eds.), Handbook of Neuropsychology (Vol. 1),
skills and aphasia. J. Int. Neuropsychol. Soc. 5: 213221. Elsevier, Amsterdam, pp. 121136.
Deloche, G. (1993). Aphasia and acalculia. In: Blanken, G., Dittmann, J., Grafman, J., and Boller, F. (1987). Cross-cultural approaches to the study
Grimm, H., Marshall, J. C., and Wallesch, C. W. (eds.), Linguistic of calculation processes. In: Deloche, G., and Seron, X. (eds.),
Disorders and Pathologies, Walter de Gruyter, Berlin, pp. 325332. Mathematical Disabilities: A Cognitive Neuropsychological Per-
Deloche, G., Hannequin, D., Carlomagno, S., Agniel, A., Dordain, M., spective, Erlbaum, Hillsdale, NJ, pp. 257272.
Pasquier, F., et al. (1995). Calculation and number processing in Grafman, J., Passafiume, D., Faglioni, P., and Boller, F. (1982). Calcula-
mild Alzheimers disease. J. Clin. Exp. Neuropsychol. 17: 634639. tion disturbances in adults with focal hemisphere damage. Cortex
Deloche, G., and Seron, X. (1982). From three to 3: A differential analysis 18: 3750.
of skills in transcoding quantities between patients with Brocas Grafman, J., Kampen, D., Rosenberg, J., Salazar, A. M., and Boller,
aphasia and Wernickes aphasia. Brain 105: 719733. F. (1989). The progressive breakdown of number processing and
Deloche, G., and Seron, X. (1984). Some linguistic components of acal- calculation ability: A case study. Cortex 25: 121133.
culia. In: Rose, F. C. (ed.), Advances in Neurology, Vol. 42: Progress Graham, D. J. (1987). An associative retrieval model of arithmetic mem-
in Aphasiology, Raven Press, New York, pp. 215222. ory: How children learn to multiply. In: Sloboda, J. A., and Rogers,
Deloche, G., and Seron, X. (1987). Numerical transcoding: A general D. (eds.), Cognitive Processes in Mathematics, Clarendon Press,
production model. In: Deloche, G., and Seron, X. (eds.), Mathe- Oxford, pp. 123141.
matical Disabilities: A Cognitive Neuropsychological Perspective, Greenberg, J. H. (1978). Universals of Human Language, Stanford Uni-
Erlbaum, Hillsdale, NJ, pp. 137170. versity Press, Stanford, CT.
Deloche, G., Seron, X., and Bergego, C. (1989). Traitement des nombres Grewel, F. (1952). Acalculia. Brain 75: 397407.
et calcul: Donnes theoriques et perpectives therapeutiques. Ann. Grewel, F. (1969). The acalculias. In: Vinken, P. J., and Bruyn, G. (eds.),
Readapt. Med. Phys. 32: 627637. Handbook of Clinical Neurology (Vol. 3), North Holland, Amster-
Deloche, G., Seron, X., Larroque, C., Magnien, C., Metz-Lutz, M. N., dam, pp. 181196.
Riva, I., et al. (1994). Calculation and number. Processing: Assess- Grigsby, J. P., Kemper, M. B., and Hagerman, R. J. (1987). Develop-
ment Battery: Role of Demographic factors. J. Clin. Exp. Neuropsy- mental Gerstmann syndrome without aphasia among fragile X syn-
chol. 16: 195208. drome. Neuropsychologia 25: 881891.
DePartz, M. P. (1986). Re-education of a deep dyslexic patient: Rationale Gross-Tsur, V., Shalev, R. S., Manor, O., and Amir, N. (1995). De-
of the method and results. Cogn. Neuropsychol. 3: 149177. velopmental right-hemisphere syndrome: Clinical spectrum of
De Renzi, E., and Scotti, G. (1970). Autotopagnosia: Fiction or reality? the non-verbal learning disability. J. Learn. Disabil. 28: 80
Arch. Neurol. 23: 221227. 85.
Fennema, E., and Sherman, J. (1977). Sex-related differences in mathe- Gross-Tsur, V., Manor, O., and Shalev, R. S. (1996). Developmental
matics achievement, spatial visualization, and sociocultural factors. dyscalculia: Prevalence and demographic features. Dev. Med. Clin.
J. Educ. Res. 14: 5171. Neurol. 38: 2533.
Fergusson, D. M., Horwood, L. J., and Lawton, J. M. (1990). Vulnerabil- Hagege, C. (1982). La Structure des Langues, Presses Universitaires de
ity to childhood problems and family social background. J. Child France, Paris.
Psychol. Psychiatry 31: 11451160. Halperin, D. F. (1992). Sex Differences in Cognitive Abilities (2nd Ed.),
Ferro, J. M., and Botelho, M. H. (1980). Alexia for arithmetical signs: Erlbaum, Hillsdale, NJ.
A cause of disturbed calculation. Cortex 16: 175180. Hartje, W. (1987). The effects of spatial disorders on arithmetical skills.
Fisher, N. J., Rourke, B. P., Bieliauskas, L., Giordani, B., Berent, S., and In: Deloche, G., and Seron, X. (eds.), Mathematical disabilities,
Foster, N. L. (1996). Neuropsychological subgroups of patients with Erlbaum, Hillsdale, NJ, pp. 121125.
Alzheimers disease. J. Clin. Exp. Neuropsychol. 18: 349370. Harvey, S. L., Goldstein, F. C., and Spiers, P. A. (1993). Acalculia.
Folstein, M. F., Folstein, S. E., and McHugh, P. R. (1975). Mini-mental In: Heilman, K. M., and Valenstein, E. (eds.), Clinical Neuropsy-
state: A practical method for grading the cognitive state of patients chology (3rd Ed.), Oxford University Press, New York, pp. 91
for the clinician. J. Psychiatr. Res. 12: 189198. 122.
Frederiks, J. A. M. (1985). Disorders of the body schema. In: Frederiks, Hecaen, H., and Albert, M. L. (1978). Human Neuropsychology, Wiley,
J. A. M. (ed.), Handbook of Clinical Neurology, Vol. 45: Clinical New York.
Neuropsychology, Elsevier, Amsterdam, pp. 373394. Hecaen, H., Angelerges, T., and Houllier, S. (1961). Les varietes clin-
Fuson, K. C. (1988). Childrens Counting and the Concepts of Number, iques des acalculies au cours des lesions retrorolandiques. Rev. Neu-
Springer-Verlag, New York. rol. (Paris) 105: 85103.
Henschen, S. E. (1925). Clinical and anatomical contributions on brain Lewis, C., Hitch, G. J., and Walker, P. (1994). The prevalence of specific
pathology. Arch. Neurol. Psychiatry 13: 226249. arithmetic difficulties and specific reading difficulties in 9 to 10 year
Hernadek, M. C. S., and Rourke, B. P. (1994). Principal identifying fea- old boys and girls. J. Child Psychol. Psychiatry 35: 283292.
tures of the syndrome of nonverbal learning disabilities in children. Lezak, M. D. (1995). Neuropsychological Assessment (3rd Ed.), Oxford
J. Learn. Disabil. 27: 144148. University Press, New York.
Hier, D. B., Mogil, S. I., Rubin, N. P., and Komros, G. R. (1980). Semantic Lindquist, T. (1936). De Lacalculie. Acta Med. Scand. 38: 217277.
aphasia: A neglected entity. Brain Lang. 10: 120131. Lomas, J., and Kertesz, A. (1978). Patterns of spontaneous recovery in
Hills, J. R. (1957). Facor analyzed abilities and success on college math- aphasic groups: A study of adult stroke patients. Brain Lang. 5:
ematics. Educ. Psychol. Meas. 17: 615622. 388401.
Hitch, G., Cundick, J., Haughey, M., Pugh, R., and Wright, H. (1987) Loring, D. W. (ed.) (1999). INS Dictionary of Neuropsychology, Oxford
Aspects of counting in childrens arithmetics. In: Sloboda, J. A., and University Press, New York.
Rogers, D. (eds.), Cognitive Processes in Mathematics, Clarendon Loveland, K. A., Fletcher, J. M., and Bailey, B. (1990). Verbal and non-
Press, Oxford, pp. 2641. verbal communications of events in learning disabilities subtypes.
Hittmair-Delazer, M., Semenza, C., and Denes, G. (1994). Concepts and J. Clin. Exp. Neuropsychol. 12: 433447.
facts in calculation. Brain 117: 715728. Lucchelim, F., and De Renzi, E. (1993). Primary dyscalculia after medial
Hittmair-Delazer, M., Sailer, U., and Benke, Th. (1995). Impaired arith- frontal lesion of the left hemisphere. J. Neurol. Neurosurg. Psychi-
metic facts but intact conceptual knowledgeA single case study atry 56: 304307.
of dyscalculia. Cortex 31: 139147. Luria, A. R. (1966). Higher Cortical Functions in Man, Basic Books,
Holender, D., and Peereman, R. (1987). Differential processing of phono- New York.
graphic and logographic single digit numbers by the two hemi- Luria, A. R. (1973). The Working Brain, Basic Books, New York.
spheres. In: Deloche, G., and Seron, X. (eds.), Mathematic Dis- Luria, A. R. (1976). Basic Problems in Neurolinguistics, Mouton, New
abilities: A Cognitive Neuropsychological Perspective, Erlbaum, York.
Hillsdale, NJ, pp. 4386. Lynch, J. C. (1980). The functional organization of posterior parietal
Hours, F. (1982). Les Civilisations du Paleolithique, Presses Universi- association cortex. Behav. Brain Sci. 3: 485534.
taires de la France, Paris. Mandell, A. M., Knoefel, J. E., and Albert, M. L. (1994). Mental Status
Hurford, R. (1987). Language and Number, Basil Blackwell, Oxford. Examination in the elderly. In: Albert, A. L., and. Knoefel, J. E.
Hyde, J. S., Fennema, E., and Lamon, S. J. (1990). Gender differences (eds.), Clinical Neurology of Aging, Oxford University Press, New
in mathematical performance: A meta-analysis. Psychol. Bull. 107: York, pp. 277313.
139155. Mantovan, M. C., Delazer, M., Ermani, M., and Denes, G. (1999). The
Jastak, S., and Wilkinson, G. S. (1984). Wide Range Achievement Test breakdown of calculation procedures in Alzheimers disease. Cor-
Revised, Jaskat Assessment Systems, Wilmington, DE. tex 35: 2138.
Kinsbourne, M., and Warrington, E. K. (1963). The developmental Ger- Marterer, A., Danielczyk, W., Simanyi, M., and Fisher, P. (1996). Calcu-
stmann syndrome. Arch. Neurol. 8: 490501. lation abilities in dementia of the Alzheimers type and in vascular
Kertesz, A. (1979). Aphasia and associated disorders. New York: Grune dementia. Arch. Gerontol. Geriatr. 23: 189197.
and Stratton. Martinet, A. (1975). Evolution de Langues et Reconstruction, Presses
Klein, A., and Starkey, P. S. (1987) The origins and development of Universitaires de France, Paris.
numerical cognition: A comparative analysis. In: Sloboda, J. A., and Mazzoni, M., Pardossi, L., Cantini, R., Giornetti, V., and Arena, R.
Rogers, D. (eds.), Cognitive Processes in Mathematics, Clarendon (1990). Gerstmann syndrome: A case report. Cortex 25: 459468.
Press, Oxford, pp. 125. McCloskey, M., and Caramazza, A. (1987). Cognitive mechanisms in
Koehler, O. (1951). The ability of birds to count. Bull. Anim. Behav. 9: normal and impaired number processing. In: Deloche, G., and
4145. Seron, X. (eds.), Mathematical Disabilities: A Cognitive Neuropsy-
Kosc, L. (1970). Psychology and psychopathology of mathematical abil- chological Perspective, Erlbaum, Hillsdale, NJ, pp. 201220.
ities. Stud. Psychol. 12: 159162. McCloskey, M., Caramazza, A., and Basili, A. (1985). Cognitive mecha-
Lampl, Y., Eshel, Y., Gilal, R., and Savora-Pihnas, I. (1994). Selec- nisms in number processing and calculation: Evidence from dyscal-
tive acalculia with sparing of the subtraction process in a pa- culia. Brain Cogn. 4: 171196.
tient with left parietotemporal hemorrhage. Neurology 44: 1759 McCloskey, M., Sokol, S. M., and Goodman, R. A. (1986). Cognitive
1761. processes in verbal number processing: Inference from the per-
Langer, J. (1986). The Origins of Logic: One to Two Years, Academic formance of brain-damaged subjects. J. Exp. Psychol. Gen. 115:
Press, New York. 313330.
LeDoux, J. E. (1982). Neuroevolutionary mechanisms of cere- McCloskey, M., Aliminosa, D., and Macaruso, P. (1991a). Theory based
bral asymmetry in man. Brain Behav. Evol. 20: 197 assessment of acquired dyscalculia. Brain Cogn. 17: 285308.
213. McCloskey, M., Aliminosa, D., and Sokol, S. M. (1991b). Facts, rules,
LeDoux, J. E. (1984). Cognitive evolution: Clues from brain asymmetry. and procedures in normal calculation: Evidence from multiple
In: Ardila, A., and Ostrosky, F. (eds.), The Right Hemisphere: Neu- single-patient studies of impaired arithmetic fact retrieval. Brain
rology and Neuropsychology, Gordon and Breach Science, London, Cogn. 17: 154203.
pp. 5160. McGlone, J. (1980). Sex differences in human brain asymmetry: A crit-
Leon-Portilla, M. (1986). Tiempo y realidad en el pensamiento Maya ical survey. Behav. Brain Sci. 5: 215264.
[Time and reality in Maya thinking], Universidad Nacional Au- McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D. L., and
tonoma de Mexico, Mexico. Stadlan, E. M. (1984). Clinical diagnosis of Alzheimers disease: A
Levin, H., Goldstein, F. C., and Spiers, P. A. (1993). Acalculia. In: report of the NINCDS-ADRDA work group. Department of Health
Heilman, K. M., and Valenstein, E. (eds.), Clinical Neuropsychol- and Human Services Task Force on Alzheimers Disease. Neurology
ogy (3rd Ed.), Oxford University Press, New York. 34: 939944.
Levy-Bruhl, L. (1947). Las funciones mentales en las sociedades inferi- McNeil, J. E., and Warrington, E. (1994). A dissociation between ad-
ores [Mental functions in lower societies], Lautaro, Buenos Aires, dition and subtraction with written calculation. Neuropsychologia
Buenos Aires. Originally published in 1910. 32: 717728.
Lewandowsky, M., and Stadelmann, E. (1908). Ueber einen be- Mechner, F. (1958). Probability relations within response sequences un-
merkenswerten Fall von Himblutung und uber Rechenstorungen der ratio reinforcement. J. Exp. Anal. Behav. 1: 109121.
bei Herderkrankurg des Gehirns. J. Psychol. Neurol. 11: 249 Morris, H. H., Luders, H., Lesser, R. P., Dinner, D. S., and Hahn,
265. J. (1984). Transient neuropsychological abnormalities (including
Gerstmanns syndrome) during cortical stimulation. Neurology 34: Rosselli, M. (1992). Discalculia. In: Rosselli, M., and Ardila, A. (eds.),
877883. Neuropsicologia Infantii, Prensa Creativa, Medellin (Colombia).
National Education Association (1989). The NEA 1989 Almanac of Rosselli, M., and Ardila, A. (1989). Calculation deficits in patients
Higher Education, National Education Association, Washington, with right and left hemisphere damage. Neuropsychologia 27: 607
DC. 618.
Neibart, M. (1985). Ten is the key. Acad. Ther. 20: 593598. Rosselli, M., and Ardila, A. (1995). Rehabilitacion de la alexia y la
OBoyle, M. W., Benbow, C. P., and Alexander, J. E. (1995). Sex dif- agrafia. In: Ostrosky, F., Ardila, A., and Dochy, R. (eds.), La Reha-
ferences, hemispheric laterality, and associated brain activity in the bilitacion en Neuropsicologia, Trillas, Mexico.
intellectual gifted. Dev. Neuropsychol. 11: 415443. Rosselli, M., and Ardila, A. (1996). Rehabilitation of calculation disor-
Osterrieth, P. A. (1944). Le test de copie dune figure complexe. Arch. ders. In: Leon-Carrion J. (ed.), Neuropsychological Rehabilitation:
Psychol. 30: 206256. Fundamentals, Directions and Innovations, St. Lucie Press, Delray
Ostrosky, F., Ardila, A., and Rosselli, M. (1997). Neuropsi: Evaluacion Beach, FL, pp. 353370.
Neuropsicologica Breve en Espanol, Bayer, Mexico, DF. Rosselli, M., and Ardila, A. (1998). Acalculia. In: Neurobase, Arbor
Ostrosky, F., Ardila, A. and Rosselli, M. (1999). Neuropsi: A brief Publishing Corporation, San Diego, CA.
neuropsychological test battery in Spanish with norms by age and Rosselli, M., Ardila, A., and Rosas, P. (1990). Neuropsychological as-
educational level. Jorunal of the International Neuropsychological sessment in illiterates. II: Language and praxic abilities. Brain
Society, 5: 413433. Cogn. 12: 281296.
Ozols, E., and Rourke, B. P. (1985). Dimensions of social sensitivity in Rosselli, M., Rosselli, A.Vergara, I. and Ardila, A. (1986). Topography
two types of learning disabled children. In: Rourke, B. P. (ed.), of the hemi-inattention syndrome. International Journal of Neuro-
Neuropsychology of Learning Disabilities, Guilford Press, New science, 27: 165172.
York, pp. 135145. Rossor, M. N., Warrington, E. K., and Cipolotti, I. (1995). The isolation
Parker, S. T., and Gibson, K. H. (1979). A developmental model for the of calculation skills. J. Neurol. 242: 7881.
evolution of language and intelligence in early hominids. Behav. Roudier, M., Marcie, P., Podrabinek, N., Lamour, Y., Payan, C., Fer-
Brain Sci. 2: 367408. manian, J., et al. (1991). Cognitive functions in Alzheimers dis-
Parlatto, V., Lopez, O. L., Panisset, M., Iavarone, A., Grafman, J., and ease: Interaction of cognitive domains. Dev. Neuropsychol. 7: 231
Boller, F. (1992). Mental calculations in mild Alzheimer disease: 241.
A pilot study. Int. J. Geriatr. Psychiatry 7: 599602. Rourke, B. P. (1987). Syndrome of non verbal learning disabilities: The
PeBenito, R. (1987). Developmental Gerstmann syndrome: Case report final common path of white matter disease dysfunction. Clin. Neu-
and review of literature. J. Dev. Behav. Pediatr. 8: 229232. ropsychol. 1: 209234.
PeBenito, R., Fisch, B. C., and Fisch, M. L. (1988). Developmental Rourke, B. P. (1988). The syndrome of non verbal learning disabilities:
Gerstmann syndrome. Arch. Neurol. 45: 977982. Developmental manifestation in neurological diseases, disorders,
Pesenti, M., Seron, X., and Van Der Lider, M. (1994). Selective impair- and dysfunctions. Clin. Neuropsychol. 2: 293330.
ments as evidence for mental organization of arithmetical facts: BB, Rourke, B. P. (1989). Nonverbal Learning Disabilities: The Syndrome
a case of preserved subtractions. Cortex 25: 661671. and the Model, Guilford Press, New York.
Pirozzolo, F. J., and Payner, K. (1978). Disorders of oculomotor scanning Rourke B. P. (1993). Arithmetic disabilities, specific and otherwise: A
and graphic orientation in developmental Gerstmann syndrome. neuropsychological perspective. Journal of Learning Disabilities,
Brain Lang. 5: 119126. 26: 214226.
Poeck, K., and Orgass, B. (1966). Gerstmanns syndrome and aphasia. Sakurai, Y., Momose, T., Iwata, M., Sasaki, Y., and Kanazawa, I. (1996).
Cortex 2: 421437. Activation of prefrontal and posterior superior temporal areas in
Poncet, M., Pelissier, J. F., Sebahoun, M., and Nasser, C. J. (1971). A visual calculation. J. Neurol. Sci. 39: 8994.
propos dun case dautotopoagnosie secondaire a un lesion parieto- Schilder, P., and Stengel, E. (1931). Asymbolia for pain. Arch. Neurol.
occipitale de lhemisphere majeur. Encephale 60: 114. Psychiatry 25: 598600.
Pontius, A. A. (1985). Finger misrepresentation and dyscalculia in an Semenza, C. (1988). Impairment in localization of body parts following
ecological context: Toward an ecological (cultural) evolutionary brain damage. Cortex 24: 443449.
neuro-psychiatry. Percept. Mot. Skills 57: 11911208. Seron, X. (1984). Re-education strategies in neuropsychology: Cognitive
Pontius, A. A. (1989). Color and spatial error in block design in stone- and pragmatic approaches of the disorders. In: Rose, F. C. (ed.),
age Auca Indians: Ecological underuse of occipital-parietal sys- Advances in Neurology, Vol. 42: Progress in Aphasiology, Raven
tem in men and of frontal lobes in women. Brain Cogn. 10: 54 Press, New York, pp. 223241.
75. Seron, X., Deloche, G., Ferrand, I., Cornet, J. A., Frederix, M., and
Posner, J. K. (1982). The development of mathematical knowledge in Hirsbrunner, T. (1992). Dot counting by brain-damaged subjects.
two West African societies. Child Dev. 53: 200208. Brain Cogn. 17: 116137.
Premack, D. (1976). Intelligence in Ape and Man, Erlbaum, Hillsdale, Shalev, R., Weirtman, R., and Amir, N. (1988). Developmental dyscal-
NJ. culia. Cortex 24: 555561.
Queixalos, F. (1985). Lorientation spatiale dans la grammaire Sikuani. Shalev, R. S., Auerbach, J., and Gross-Tsur, V. (1995a). Developmental
J. Societe Americanistes 71: 115128. dyscalculia behavioral and attentional aspects: A research note. J.
Queixalos, F. (1989). Numeracion tradicional Sikuani [Traditional Child Psychol. Psychiatry 36: 12611268.
Sikuani numbering]. Glotta 3: 2831. Shalev, R. S., Manor, O., Amir, N., Wertman, E., and Gross-Tsur, V.
Rasanen, P., and Ahonen, T. (1995). Arithmetic disabilities with and (1995b). Developmental dyscalculia and brain laterality. Cortex 31:
without reading difficulties: A comparison of arithmetic errors. Dev. 357365.
Neuropsychol. 11: 275298. Seidenberg, M., Beck, N., Geisser, M., Giodani, B., Sackellares, J. C.,
Ratcliff, G. (1979). Spatial thought, mental rotation and the right hemi- Berent, S., et al. (1986). Academic achievement of children with
sphere. Neuropsychologia 13: 191206. epilepsy. Epilepsia 27: 753759.
Roeltgen, D. P., Sevush, S., and Heilman, K. M. (1983). Pure Ger- Siegler, R. S. (1987). Strategy choices in subtraction. In: Sloboda, J.
stmanns syndrome from a focal lesion. Arch. Neurol. 40: 46 A., and Rogers, D. (eds.), Cognitive Processes in Mathematics,
47. Clarendon Press, Oxford.
Rosenberger, P. B. (1989). Perceptual-motor and attentional correlates Sohlberg, M. M., and Mateer, C. A. (1989). Introduction to Cognitive
of developmental dyscalculia. Ann. Neurol. 26: 216220. Rehabilitation: Theory and Practice, Guilford Press, New York.
Rosin, R. T. (1973). Gold medallions: The arithmetic calculations of an Sokol, S. M., McCloskey, M., Cohen, N. J., and Aliminosa, D. (1991).
illiterate. Counc. Anthropol. Educ. Newsl. 4: 19. Cognitive representations and processes in arithmetic: Inferences
from the performance of brain-damaged subjects. J Exp Psychol Tohgi, H., Saitoh, K., Takahashi, H., Utsugisawa, H., Hatano, K., and
Learn Mem Cogn., 17: 355376. Sasaki, T. (1995). Agraphia and acalculia after left prefrontal
Sokol, S. M., Macaruso, P., and Gollan, T. H. (1994). Developmental (F1,F2) infraction. J. Neurol., Neurosurg. Psychiatry 58: 629632.
dyscalculia and cognitive neuropsychology. Dev. Neuropsychol. 10: Tucha, O., Steup, A., Christian, S., and Lange, K. (1997). Toe agnosia
413442. in Gerstmann syndrome. Journal of Neurology, Neurosurgery and
Spiers, P. A. (1987): Acalculia revisited: Current issues. In: Deloche, G., Psychiatry, 63: 399403.
and Seron, X. (eds.), Mathematic Disabilities: A Cognitive Neu- Turner, R. (1984). The Great Cultural Tradition. The Foundation of Civ-
ropsychological Perspective, Erlbaum, Hillsdale, NJ, pp. 126. ilization, McGraw-Hill, New York.
Spreen, O., and Strauss, E. (1991). A Compendium of Neuropsycholog- Tsvetkova, L. S. (1996). Acalculia: Aproximacion neuropsicologica al
ical Tests, Oxford University Press, New York. analisis de la alteracion y la rehabilitacion del calculo. In: Ostrosky,
Starkey, P. (1992). The early development of numerical reasoning. Cog- F., Ardila, A., and Dochy, R. (eds.), Rehabilitacion neuropsicolog-
nition 43: 93126. ica, Planeta, Mexico, pp. 114131.
Stones, I., Beckmann, M., and Stephens, L. (1982). Sex-related differ- Varney, N. R. (1984). Gerstmann syndrome without aphasia: A longitu-
ences in mathematical competencies of pre-calculus college stu- dinal study. Cortex 10: 378387.
dents. Sch. Sci. Math. 82: 295299. Warrington, E. K. (1982). The fractionation of arithmetical skills: A
Strang, J. D., and Rourke, B. P. (1985). Arithmetic disabilities sub- single case study. Q. J. Exp. Psychol. 34A: 3151.
types: The neuropsychological significance of specific arithmetic Wechsler, D. (1997). WAIS-III: Administration and Scoring Manual, The
impairment in childhood. In: Rourke, B. P. (ed.), Neuropsychol- Psychological Corporation, San Antonio, TX.
ogy of Learning Disabilities, Guilford Press, New York, pp. 87 Weinberg, J., Diller, L., Gordon, W. A., Gerstmann, L., Lieberman, A.,
101. Lakin, P., et al. (1977). Visual scanning training effect on read-
Strub, R., and Geschwind, N. (1983) Localization in Gerstmann syn- ing related tasks in acquired right brain damage. Arch. Phys. Med.
drome. In: Kertesz, A. (ed.), Localization in Neuropsychology, Aca- Rehabil. 58: 479483.
demic Press, New York, pp. 295322. Weintraub, S., and Mesulam, M. M. (1983). Developmental learning
Sullivan, K. S. (1996). Remediation of Arabic numeral processing in a disabilities of the right hemisphere. Arch. Neurol. 40: 463468.
case of developmental dyscalculia. Neuropsychol. Rehabil. 6: 27 Woodcock, R. W., and Johnson, M. B. (1989). WoodcockJohnson
54. Psycho-Educational BatteryRevised, DLM Teaching Resources,
Swadesh, M. (1967). El lenguage y la vida humana [Language and hu- Allen, TX.
man life], Fondo de Cultura Economica, Mexico. Woodruff, G., and Premack, D. (1981). Primitive mathematical concepts
Temple, C. (1989). Digit dyslexia: A category-specific disorder in de- in the chimpanzee: Proportionality and numerosity. Nature 293:
velopmental dyscalculia. Cogn. Neuropsychol. 6: 93116. 568570.
Temple, C. (1991). Procedural dyscalculia and number fact dyscalculia: Wynn, K. (1990). Childrens understanding of counting. Cognition 36:
Double dissociation in developmental dyscalculia. Cogn. Neuropsy- 155193.
chol. 8: 155176. Wynn, K. (1992). Childrens acquisition of the number words and count-
Terman, L.M. and Merrill, M.A. (1973). Stanford-Binet Intelligence ing system. Cognit. Psychol. 24: 220251.
Scale. Manual for the Third Revision, Form L-M. Boston: Zucker, K. (1934). An analysis of disturbed function in aphasisa. Brain,
Houghton-Mifflin Co. 57: 6072.
Archives of Clinical Neuropsychology 22 (2007) 297307
Toward the development of a cross-linguistic naming test

Alfredo Ardila
Downloaded from http://acn.oxfordjournals.org at Florida International University on March 22, 2010

Department of Communication Sciences and Disorders, Florida International University,
HLS139 Miami, FL 33199, USA
Abstract
Developing a cross-linguistic naming test has represented a challenge in language evaluation. In this paper, it is proposed that
a cross-linguistic naming test should fulfill at least the following three criteria: (1) include only universal words found across
different languages. The basic cross-linguistic core vocabulary is usually referred as the Swadesh word list; (2) include different
semantic categories (e.g., living and nonliving elements); and (3) avoid the confounding of perceptual difficulties. Departing from
the Swadesh word list, a cross-linguistic naming test was developed, including six different semantic categories: (a) body-parts (10
words), (b) natural phenomena (non-touchable) (5 words), (c) external objects (potentially known through the sight and the touch)
(5 words), (d) animals (5 words), (e) colors (5 words), and (f) actions (10 words). A total of 40 color pictures were selected to
represent these basic words. It is emphasized that this test has two major advantages: on one hand, it is readily available in hundreds
of different languages; and, on the other hand, it is not a fixed test, but it includes photographs that can be replaced. Theoretically,
norms are not required, and it represents a low-ceiling test. Word frequency can be used as a criterion of the level of difficulty.
The next step will be to find the performance profile in different language pathologies, as well as the decline pattern in cases
of dementia.
2007 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved.
Keywords: Anomia; Naming; Swadesh word list; Language testing; Aphasia; Cross-cultural neuropsychology
1. Introduction
Naming is a basic language ability. Word-finding difficulty (anomia) represents the most frequent aphasia sign
(Benson & Ardila, 1996; Goodglass, 1993; Luria, 1976). Different tests have been developed to assess naming ability;
some of them, however, have become more extended than others (e.g., Boston Naming Test [BNT], Kaplan, Goodglass,
& Weinstraub, 1983; Object Naming Test, Newcombe, Oldfield, Ratcliff, & Winfield, 1971; Peabody Picture Vocabulary
Test, Dunn & Dunn, 1981).
It has long been recognized that naming body-parts, external objects, and colors depend on the activity of different
brain areas (e.g., Moore & Price, 1999; Spitzer et al., 1998), and that these naming functions can be differentially
affected by specific focal lesions (Hecaen & Albert, 1978; Luria, 1966, 1976). Action naming is occasionally included
in some aphasia test batteries (e.g., Boston Diagnostic Aphasia Examination, Goodglass & Kaplan, 1983). A spe-
cific Action Naming Test was developed by Ardila and Rosselli (1994) with the purpose of testing a patient
presenting a restricted ability to name actions. In addition to these four distinctions in naming (naming body-parts,
external objects, colors, and actions), it has been further found that much finer distinctions can be established with
regard to the naming defects observed in cases of brain pathology, which can be limited to a rather specific seman-
E-mail address: ardilaa@fiu.edu.
0887-6177/$ see front matter 2007 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.acn.2007.01.016
298 A. Ardila / Archives of Clinical Neuropsychology 22 (2007) 297307
tic categories (e.g., peoples names, living things, geographical names, etc.) (e.g., Harris & Kay, 1995; Goodglass,
Wingfield, Hyde, & Theurkauf, 1986; Lyons et al., 2002; Warrington & Shallice, 1984) and even as specific as
medical terms (Crosson, Moberg, Boone, Rothi, & Raymer, 1997). It is reasonable to expect that a naming test
includes naming of different semantic categories (e.g., living and nonliving things) becauseas mentioned above,
the naming of these categories depend on different brain areas and can be differentially affected in cases of brain
pathology.
The major naming test used in aphasia assessment is the BNT (Kaplan, Goodglass, & Weinstraub, 1983). The BNT
has been adapted to several languages, including Spanish (Garcia-Albea, Sanchez-Bernardos, & del Viso-Pabon, 1986),
German (Merten, 2004), Dutch (Marien, Mampaey, Vervaet, Saerens, & De Deyn, 1998), etc., has had widespread
clinical and research applications. Nonetheless, the BNT has at least the following limitations: (a) pictures are frequently

difficult to recognize, particularly in people with posterior brain pathology, elders, and illiterates; (b) it was developed in
a particular cultural context, and it is culturally and linguistically biased. Some figures correspond to typical American
elements (e.g., pretzel), and/or are well known only for people living in some world areas (e.g., beaver); (c) it does not
distinguish semantic categories, and as a matter of fact, the semantic categories are rather randomly used (e.g., animals,
musical instruments, external objects, etc.). In consequence, it is not appropriate to distinguish category-dependent
naming defects; (d) it is based in the assumption that the pictures are presented in an increasing level of difficulty.
This assumption is not clearly substantiated. Clinical observations do not always support it. Furthermore, the order
of difficulty also varies from language to language. For instance, the level of difficulty for the different pictures is
relatively different in Spanish than in English (personal observation).
In cases of aphasia, naming difficulties are manifested by (1) slowness or impossibility to find the correct word,
(2) circumlocutions (including descriptions, e.g., it is very big and strong, and the use of superordinate words, e.g.,
it is an animal), and (3) paraphasias; different types of paraphasias are usually distinguished, though the two major
types correspond to semantic paraphasias (e.g., bench table) and phonological paraphasias (e.g., pencil percil)
(Ardila & Rosselli, 1993; Benson & Ardila, 1996; Luria, 1976). Phonetic deviations can be also recorded, but phonetic
deviations are regarded as speech, not language defects.
Theoretically, a naming test should allow scoring for (a) naming speed; frequently, subtle naming defects are observed
not as an overt failure to find the correct word, but as slowness in naming; (b) paraphasias (at least, phonological and
semantic); (c) circumlocutions (i.e., superordinate terms and descriptions; e.g., well, this is a big animal that may be
found in Africa); (d) word-retrieval ability when using phonological cueing (e.g., it is a/pe/. . .); and (e) failure in
naming.
Two major factors can affect the naming ability. Naming ability is significantly correlated with the individ-
uals educational level and age (Lezak, 2004; Spreen & Strauss, 1998). Lexical knowledge significantly correlates
with the individuals educational level (e.g., Lecours et al., 1987, 1988; Manly et al., 1999; Rosselli, Ardila, &
Rosas, 1990); vocabulary tends to decrease particularly after the seventh decade of the life (e.g., Mackay, Connor,
Albert, & Obler, 2002; Ramsay, Nicholas, Au, Obler, & Albert, 1999; Tsang & Lee, 2003). Illiterate elders
present in consequence, the lowest ability to find names. Some times a gender effect has been reported in nam-
ing, but this gender effect is inconsistent (e.g., Grabowski, Damasio, Eichhorn, & Tranel, 2003; Pineda et al.,
2000).
In addition to the impairments in naming associated with abnormal brain conditions, difficulties in finding names
may be due to a diversity of factors. The three major confoundings in testing naming ability are as follows: (1)
Perceptual difficulties: The item is not recognized because of visualperceptual defects. To overcome this difficulty,
the BNT allows semantic cuing (i.e., to tell the semantic category). (b) Lack of familiarity with the item: Many of the
items included in current naming tests have different levels of familiarity for people living in different countries and
cultural contexts. For instance, the pretzel is a typical American snack, virtually unknown in most countries. (c) Word
frequency: A frequent and easy word in a particular culture can be more unusual and difficult in another culture. A
dart is easier to name for English speakers, and harder for Spanish speakers, whereas a domino is easier to name for
Spanish speakers than for English speakers (personal observation).
Developing a cross-linguistic naming test requires using words that are found across different languages (basic
universal or core vocabulary). Just few words are recognized to be universal. They refer to those elements that every
personregardless of time, place, and living conditionshas been exposed to (www.ethnoculture.com). This basic
core vocabulary was proposed by Swadesh (1952, 1967), and it is usually known in linguistics as the Swadesh word
list.
A. Ardila / Archives of Clinical Neuropsychology 22 (2007) 297307 299
Given that naming represents a basic language ability, to develop a cross-linguistic naming test is important because
of different reasons: (1) it would allow to use similar testing instruments across different languages; (2) it would allow
to compare language defects in speakers of different languages; (3) the problems of translation and test equivalence
would be partially overcome; and finally (4) developing a new naming test represents an opportunity to fix some of the
problems existing in currently available naming tests mentioned above.
2. The Swadesh word list (Basic Core Vocabulary)
There are two major versions of this basic universal vocabulary. The original 100-word lexicostatistical list was
published by Swadesh in 1952 and supposedly represents the Basic Core Vocabulary existing in any language. The

200-word list was used for Austronesian and Indo-European, with observed replacements rates, and was published by
Kruskal, Dyen, and Black in (1973). The Swadesh word list is meant to be a list of 100 (200) key concepts that all
languages, irrespective of cultural differences, are most likely to have words for, and are least likely to have borrowed
from other languages.
Using this list of stable words, glottochronologists believed they could calculate the approximate amount of time
that had passed between the split-up of two related languages (Swadesh, 1967). Unfortunately, language does not seem
to change at a constant rate. The word list and the formulas to use it will, for instance, accurately show that the Romance
languages started to split up about 2,000 years ago.
In consequence, there are different versions of the Swadesh word list; some are shorter (100 words) and some are
longer (200 words, and even sometimes 207 and 250 words). The shorter Swadesh word list includes the following:
1. Grammatical words: I/me, you, we, this, that, who, what, not.
2. Quantifiers: All, many, one, two.
3. Adjectives: Big, long, small, hot, cold, full, new, good, round, dry.
4. Human distinctions: Woman, men, person, name.
5. Animals: Fish, bird, dog, louse.
6. Highly frequent elements: Tree, seed, leaf, root, bark, skin, flesh, blood, bone, grease/oil, egg, horn, tail, feather.
7. Body-parts: Hair, head, ear, eye, nose, mouth, tongue, claw, foot, knee, hand, belly, neck, breast, heart, liver.
8. Actions: Drink, eat, bite, see, hear, know, sleep, die, kill, swim, fly, walk, come, lie (down), sit, stand, say.
9. Natural phenomena: Sun, moon, star, water, rain, stone, sand, earth, cloud, smoke, fire, ash, burn, path, mountain,
night.
10. Colors: Red, green, yellow, black, white.
The Swadesh word list is currently available in hundreds of languages. In consequence, a naming test using the
Swadesh word list would be automatically available in hundreds of different languages (see, www.rosettaproject.org/
live/search/contribute/swadesh/view/index html).
3. A cross-linguistic naming test
A naming test should clearly distinguish different semantic categories, at least external elements, body-parts, colors,
and actions. External elements may be further subdivided (e.g., natural phenomena and external objects potentially
known through sight and touch). These five semantic categories (body-parts, natural phenomena (non-touchable),
external objects, animals, colors, and actions) were selected because of two major reasons: (1) There are category-
specific naming defects, and consequently a naming test should include different semantic categories; (2) Only these
five semantic categories can be clearly separated if departing from the Swadesh word list. That means, these are
universal semantic categories.
Furthermore, a naming test is expected to avoid at best the confounding of perceptual defects. Two strategies
can potentially be used to minimize the effects of visuoperceptual impairments: (1) to select realistic elements,
easy to be visually recognized. Color photographs should be preferred to black and white ink drawings. (2) To
provide the semantic category; for example, instead of asking about the name, to ask about the name of this
animal.
With the above considerations in mind, 40 words were selected from the Swadesh word list. All of them, except
two, were taken from the short 100-word list. The two words taken from the longer Swadesh word list were snake
and worm. This was done to have five animal names. One of the animal names in the short list was not used (louse)
because of the difficulty in representing it.
In developing the test, the following strategies were used:
Table 1
Cross-linguistic naming test

English Frequency Order of difficulty Spanish Frequency Order of difficulty
Body-parts (10)
Ear 305 3 Oreja 2,742 33
Eye 240 2 Ojo 147 3
Nose 1,940 27 Nariz 2,367 31
Mouth 1,014 15 Boca 559 9
Tooth 1,756 25 Diente 2,121 28
Tongue 2,878 32 Lengua 901 14
Knee 1,966 29 Rodilla 3,684 37
Belly 5,676 40 Estomago 3,003 35
Neck 1,598 24 Cuello 1,714 24
Foot 484 8 Pie 329 4
Natural phenomena (5)
Sun 1,058 17 Sol 362 5
Moon 3,020 34 Luna 2,254 30
Cloud 2,456 31 Nube 1,307 20
Fire 719 12 Fuego 667 11
Mountain 1,590 23 Montana 883 13
External objects (5)
Tree 695 11 Arbol 582 10
Leaf 1,852 26 Hoja 881 12
Bone 1,955 28 Hueso 1,574 23
Egg 1,544 22 Huevo 4,142 38
Feather 5,302 37 Pluma 1,035 16
Animals (5)
Dog 823 14 Perro 1,121 17
Bird 1,115 18 Ave 2,802 34
Snake 5,513 39 Serpiente 3,472 36
Fish 1,017 16 Pez 2,600 32
Worm 5,336 38 Gusano 4,970 40
Colors (5)
Red 791 13 Rojo 1,494 22
Green 1,116 19 Verde 1,998 26
Yellow 2,162 30 Amarillo 2,004 27
White 3,436 35 Blanco 1,286 19
Black 502 9 Negro 2,161 29
Actions (10)
To drink 1,129 20 Beber 1,002 15
To eat 662 10 Comer 445 8
To hear 258 4 Oir 130 2
To sleep 1,381 21 Dormir 387 6
To swim 2,952 33 Nadar 4,796 39
To fly 3,892 36 Volar 1,321 21
To walk 463 7 Caminar 1,559 25
To lie down 447 6 Acostarse 1,129 18
To sit 317 5 Sentarse 444 7
To say (talk) 34 1 Decir (hablar) 42 1
Note. Word frequency in English (Kilgarriff, 1997; http://www.itri.brighton.ac.uk/Adam.Kilgarriff/bnc-readme.html#lemmatised) and Spanish

(Juilland & Chang-Rodriguez, 1964). The frequency order of each word is presented (1 corresponds to the most frequent word in the language).
Order of difficulty refers to the relative difficulty in this naming test.
Fig. 1. Cross-linguistic naming test. Original in color. Pictures corresponding to colors have been omitted.
1. Forty words corresponding to five different categories were selected: (a) body-parts (10 words), (b) natural phe-
nomena (non-touchable) (5 words), (c) external objects (potentially known through sight and touch) (5 words), (d)
animals (5 words), (e) colors (5 words), and (f) actions (10 words). A total of 40 color photographs were selected
to represent these basic words (see Table 1 and Fig. 1). Only those words that could be visually represented in a
clear way were included (grammatical words, quantifiers, and adjectives were excluded). For some categories (e.g.,
colors), all the available Swadesh list words were included. For other categories (e.g., actions), only words that
were easily represented were selected.
2. The items corresponding to each category (e.g., body-parts) were presented in sequence.
3. For each category, there were different instructions: (a: body-parts) What body-part is this? (b & c: natural phenomena
and external objects) What is this? (d: animals) What is the name of this animal? (e: colors) What color is this? (f:
actions) What is this person doing? (except for flying: What is this bird doing?). The examiner may point to the
element to be named. The semantic category is provided in order to minimize the effect of potential visuoperceptual
difficulties.
4. To avoid the confounding of perceptual errors, 10 cm 15 cm color photographs taken from Internet sites were

included in the test.
Table 2
Scoring sheet
Item Answer <5 s; 520 s Phonological paraphasia Semantic paraphasia Superordinate Description Phonological cueing Failure
Ear
Eye
Nose
Mouth
Tooth
Tongue
Knee
Belly
Neck
Foot
Sun
Moon
Cloud
Fire
Mountain
Tree
Leaf
Bone
Egg
Feather
Dog
Bird
Snake
Fish
Worm
Red
Green
Yellow
White
Black
Drink
Eat
Hear
Sleep
Swim
Fly
Walk
Lie down
Sit
Say (talk)
Total
Table 3
Summary of scores
Body-parts /10= %
Natural phenomena /5= %
External objects /5= %
Animals /5= %
Colors /5= %
Actions /10= %
Total correct /40= %
Number of phonological paraphasias =. Number of semantic paraphasias =. Superordinate/descriptions =.

Some general guidelines for the administration of this naming test were used:
1. There is no a discontinue rule. All 40 pictures are presented to the participant.

2. The pictures are presented one by one, but all the items corresponding to a specific category are presented in
sequence. In order to minimize the perceptual confounding, the instructions are different for each category (see
above). No semantic cueing is provided.
3. Time is taken. Immediate responses (within 5 s) are separated from slow responses (up to 20 s). Time in naming
is used for qualitative analysis, but not considered in the correct score. After 20 s, the initial syllable is presented
(phonological cueing). Noteworthy, phonological cueing can be problematic when comparing different languages,
because of the differences in the phonological length of the words. However, words retrieved using phonological
cueing are not included in the correct score. An answer sheet is used (Table 2). A dichotomous variable for timing
(<5 and 620 s) is used simply for simplicity in recording, but timing could also be used as a continuous variable.
Results can be summarized in a simple table (Table 3).
4. Three types of naming errors are distinguished: phonological paraphasias, semantic paraphasias, and circumlocu-
tions (i.e., using superordinate words and/or descriptions). Failure is marked if, after the phonological cueing, the
participant fails to find the correct name.
5. Both semantic and phonological paraphasias may be scored in two different ways: (a) Total number of paraphasias,
refers to the total number of paraphasic errors presented by the subject (e.g., if, when naming the dog, the participant
answers both cat and horse, then two semantic paraphasias are counted). (b) Paraphasic naming refers to the
number of pictures in which paraphasias are observed (e.g., when naming the dog the participant answers cat
and horse there is only one picture that was being named in a paraphasic way). Both scoring strategies have
clinical value. In this test, the second procedure is used, because it more accurately reflects the failure in finding
specific words.
4. The issue of norms
In cognitive testing, it is usually assumed that norms are always required. Otherwise, no comparison is reliable.
This idea, however, is more a desideratum than a reality. Furthermore, it does not seem to be a completely realistic
idea. As a matter of fact, in the future, the search for norms may be coordinated with the search for understanding the
sources of variation.
To obtain norms in English or Spanish (each one with about 400 million speakers) seems realistic. But English and
Spanish are just two out of the three largest existing languages accounting together for no more than 15% of the worlds
population. Worldwide, there are about 6,800 different languages (http://www.ethnologue.com/), most of them with a
limited number of speakers. To obtain norms for all these 6,800 different languages is simply unrealistic. Furthermore,
most of the world languages are small languages, and obtaining a reliable database would mean testing a high percentage
of the speakers. If we assume that the average language has 1 million speakers (the real number is lower), and we want
to normalize the neuropsychological instruments using just 200 stratified subjects in each language, it would mean that
about 1.5 million participants would be required. This is a nonrealistic endeavor for contemporary neuropsychology.
It seems more realistic to determine the linguistic factors potentially affecting cognitive test performance.
Theoretically, no norms are required for the cross-linguistic naming test. It is assumed that these 40 words correspond
to a basic vocabulary that any person living anywhere in any time has used. This assumption is taken from the historical
linguistics (see above). The question becomes, at what age is this basic vocabulary expected to be acquired? All the
words included in naming tests are root words (e.g., dog), not inflected (e.g., dogs) or derived words (e.g., doggy). The
vocabulary size for the total number of words is expected to be about 10,000 for first graders. Roots words grow from
about 2,000 in first graders to about 5,500 in third graders (Anglin, 1993; Hoff, 2001). As a consequence, the proposed
naming test would be anticipated to be useful for adults. In children, the specific pattern of development of this basic
core vocabulary would have to be established.
Despite that norms are not required, and word frequency may be used as a criterion of the participants vocabulary
size, piloting the pictures and the potential errors is obviously needed. The set of pictures presented in this paper was
piloted in a small sample (n = 10) of college students. Different pictures were initially used, and those presenting some
confusion in the answers were discharged. Finally, only those pictures with 100% correct answers were retained.

Some potential problems can be encountered: (a) a particular element may have different names in a language.
For instance, belly in English may be called stomach, tummy, gut, abdomen, or paunch, and in Spanish
may be named as estomago, barriga, panza, or vientre. Of course, all these answers should be regarded as
correct. In other words, alternative (but correct) naming is acceptable (e.g., saying may be telling or speaking).
(b) The patient may answer to something different from what is intended. For instance, when the picture of the belly is
presented, the patient may answer navel. The examiner may show what is supposed to be named (and how do you
name all this area of the body?). (3) The picture is perceptually confusing. Obviously, the solution is to get a better
quality photograph. The point is not the specific photograph, but to have an unambiguous picture easily recognizable
by any normal person. Photographs can be changed if required, to adapt the test to the specific conditions (e.g., the
ethnicity of the people in the pictures representing actions and body-parts).
It is worth noting that, even though the test uses a basic vocabulary, the dispersion in word frequency is significant,
extending from some extremely frequent words, up to relatively infrequent words, corresponding to about the 5,000th
word in frequency. Table 1 presents the word frequency for the English and Spanish versions of the test. The most
frequent word in both English and Spanish is say (decir). The order of frequency, as a matter of fact, is quite alike
in both English and Spanish. Some discrepancies in frequency between English and Spanish are easily understandable;
for instance, in English the word feather only refers to a birds feathers; whereas in Spanish the word pluma not
only means a birds feathers but also a pen. In any case, it may be conjectured (subject to empirical verification) that
any person having about a 5,0006,000-word vocabulary should obtain a perfect score in this naming test.
Knowing the frequency of the words has an additional advantage in testing the language: it allows conjecturing about
the size of the participants vocabulary. In cases of aphasia, it allows speculation about the severity of the vocabulary
loss.
5. Cross-linguistic use
This test may be available right away to be used in hundreds of different languages (www.rosettaproject.org/
live/search/contribute/swadesh/view/index html). Table 4 presents the test in eight different languages. Furthermore,
it has to be emphasized that any of the pictures can be replaced by another picture representing the same word. The
point is not the exact picture, but the quality of the picture, as realistic as possible, to avoid perceptual confoundings.
The set of pictures that we have been using can be replaced by a totally new set of pictures with the same names.
Nevertheless, some cautions are in order. A clinician working with an interpreter for an unfamiliar language will
need to be careful to ascertain that synonym responses are truly synonyms and not superordinates, circumlocutions,
paraphasias, or other errors. Those working with illiterates and with test-nave participants will need to be particularly
careful to be certain that a testing set has been established and understood. Although the intention of the test design is
to have a test that will produce a perfect score (for naming success) in normal adults, this remains to be empirically
verified for each language and population of use. Furthermore, more sophisticated scoring, such as response time and
frequency of error type, will require norms for clinical interpretation. Although the test is likely to be specific for
anomia, it is not likely to be highly sensitive. The low ceiling of the test can be expected to produce many false negative
results, that is, individuals with significant loss of naming abilities may nevertheless obtain normal scores on this test.
This test has been developed primarily for the purpose of detecting naming difficulties in adults. Other applications
may require further development and empirical support. For example, use of this test to track vocabulary development
in children cannot be assumed to be comparable across languages.
Table 4
Cross-linguistic naming test in eight different languages
English Spanish Esperanto Latin Turkish Basque Russian Sikuania
Ear Oreja Orelo Auris Kulak Belarri Uxa Muxu

Eye Ojo Okulo Oculus Goz Begi Glaz Itaxu
Nose Nariz Nazo Nasus Burun Sudor Nos Pumu
Mouth Boca Buso Os Agz Aho Rot Koibo
Tooth Diente Dento Dens Dis Hortz Zub Wono
Tongue Lengua Lango Lengua Dil Mihia Jazyk Ebanu
Knee Rodilla Genuo Genu Diz Belaun Koliena Matabaka

Belly Estomago Ventro Venter Karn Sabel Zhivot Koto
Neck Cuello Nuko Collum Ense Lepo Ceja Wosi
Foot Pie Piedo Pes Ayak Oin Noga Taxu
Sun Sol Suno Sol Gune Eguzki Sontse Huamekotia
Moon Luna Luno Luna Ay Hilargia Luna Juameto
Cloud Nube Nubo Nimbus Bulut Hodei Oblako Pahubo
Fire Fuego Fajro Ignis Ates Su Ogon Iso
Mountain Montana Monto Mons Da Mendi Gora Tsu
Tree Arbol Arbo Arbor Agac Arbol Diereva Nae
Leaf Hoja Folio Folium Yaprak Orri List Boxu
Bone Hueso Osto Os Kemik Hezur Kost Si
Egg Huevo Ovo Ovum Yumurta Arrautza Jaitso Tobo
Feather Pluma Plumo Penna Tuy Luma Piero Korope
Dog Perro Hundo Canis Kopek Txakur Sobaka Awiri
Bird Pajaro Birdo Avis Kus Txori Ptitsa Baratsui
Snake Serpiente Serpento Serpens Yylan Suge Zmeja Homo
Fish Pez Fiso Piscis Balk Arrain Ryba Duhai
Worm Gusano Vermo Vermis Kurt Har Gusenitsa Oro
Red Rojo Ruga Ruber Kyrmyzy Gorri Krasno Tsobia
Green Verde Verda Viridis Yei Berde Ziliony Rauna
Yellow Amarillo Flava Flavus Sary Ori Zhiolty Wajana
White Blanco Blanka Albus Beyaz Zuri Bliedno Niopo
Black Negro Nigra Niger Siyah Beltz Tchiorne Tsaebia
Drink Beber Trinki Bibere Icmek Edan Pit Apa
Eat Comer Mangi Edere Yemek Jan Iedit Xane
Hear Oir Audi Audire Duymak Entzun Slushat Humet
Sleep Dormir Dormi Dormire Uyumak Lo Spat Mahita
Swim Nadir Nagi Natare Yuzmek Igeri Plavat Hua
Fly Volar Flugi Natare Ucmak Hegan Lietat Puna
Walk Caminar Piediri Gradi Yurumek Ibili Guliat Pona
Lie down Acostarse Kusi Iacere Yatmak Gezur Lgat Boka
Sit Sentarse Sidi Sedee Oturmak Eseri Sadit Ekataba
Say Decir Diri Dicere Demek Esan Skazat Hai
a Sikuani (or Tucano) is an Ameridian language from the Amazonian jungle.
6. Conclusion
To develop a cross-linguistic naming test has been a major challenge in the language disorders area. The obvious
solution seems to be taking the most basic universal vocabulary found across different world languages. This solution
apparently may have an obvious problem: the test ceiling will be very low. Many language testse.g., the Token
Testhowever, have a low ceiling. The low ceiling as matter of fact is both, a disadvantage, but also an advantage in
language testing. It is an advantage because any error can be considered as significant. Nonetheless, word frequency
in the Swadesh word list presents a significant dispersion, implying that there is some heterogeneity in the level of
difficulty (see Table 1).
The following evident step in a cross-linguistic test is to obtain the specific profile of performance in different
language pathologies, and furthermore, to find out how this basic universal vocabulary develops in children and
disintegrates in case of dementia. This is an endeavor in language testing for the future.
Understanding the variables that can affect cognitive test performance seems to be as important as obtaining a large
number of norms in different linguistic and cultural groups.
Acknowledgments
My most sincere gratitude to Tedd Judd, Alexandra Reiss, and Ruben Echemendia, for their most valuable sug-

gestions and comments, and to Nuria Pardo for her help in selecting the pictures and developing the initial pilot
study.
References
Anglin, J. M. (1993). Vocabulary development: A morphological analysis. Monographs of the Society for Research in Child Development, 58(10)
Ardila, A., & Rosselli, M. (1993). Language deviations in aphasia: A frequency analysis. Brain and Language, 44, 165180.
Ardila, A., & Rosselli, M. (1994). Averbia as a selective naming disorder: A single case report. Journal of Psycholinguistic Research, 23, 139148.
Benson, D. F., & Ardila, A. (1996). Aphasia: A clinical perspective. Oxford University Press.
Crosson, B., Moberg, P. J., Boone, J. R., Rothi, L. J., & Raymer, A. (1997). Category-specific naming deficit for medical terms after dominant
thalamic/capsular hemorrhage. Brain and Language, 60, 407442.
Dunn, L. M., & Dunn, L. M. (1981). Peabody Picture Vocabulary TestRevised. Circle, MN: American Guidance Service.
Garcia-Albea, J. E., Sanchez-Bernardos, M. L., & del Viso-Pabon, S. (1986). Test de Boston para el Diagnostico de la Afasia: Adaptacion Espanola.
In H. Goodglass & E. Kaplan (Eds.), La Evaluacion de la Afasia y de Transtornos Relacionados. Madrid: Editorial Medica Panamericana.
Goodglass, H. (1993). Understanding aphasia. San Diego, CA: Academic Press.
Goodglass, H., & Kaplan, E. (1983). The assessment of aphasia and related disorders. Philadelphia: Lea & Febiger.
Goodglass, H., Wingfield, A., Hyde, M. R., & Theurkauf, J. C. (1986). Category specific dissociations in naming and recognition by aphasic patients.
Cortex, 22, 87102.
Grabowski, T. J., Damasio, H., Eichhorn, G. R., & Tranel, D. (2003). Effects of gender on blood flow correlates of naming concrete entities.
Neuroimage, 20, 940954.
Harris, D. M., & Kay, J. (1995). Selective impairment of the retrieval of peoples names: A case of category specificity. Cortex, 31, 575582.
Hoff, E. (2001). Language development (2nd ed.). Belmont, CA: Wadsworth/Thompson Learning.
Juilland, A., & Chang-Rodriguez, E. (1964). Frequency dictionary of Spanish words. The Hague: Mounton & Co.
Kaplan, E. F., Goodglass, H., & Weintraub, S. (1983). The Boston Naming Test. Philadelphia: Lea & Febiger.
Kilgarriff, A. (1997). Frequencies in the dictionary. International Journal of Lexicography, 10, 135155.
Kruskal, J. B., Dyen, I., & Black, P. (1973). Some results from the vocabulary method of reconstructing language trees. In I. Dyen (Ed.), Lexicostatistics
in genetic linguistics (2nd ed., pp. 3055). The Hague: Mouton & Co.
Lecours, A. R., Mehler, J., Parente, M. A., Beltrami, M. C., Canossa de Tolipan, L., Castro, M. J., et al. (1988). Illiteracy and brain damage 3: A
contribution to the study of speech and language disorders in illiterates with unilateral brain damage (initial testing). Neuropsychologia, 26,
575589.
Lecours, R. L., Mehler, J., Parente, M. A., Caldeira, A., Cary, L., Castro, M. J., et al. (1987). Illiteracy and brain damage. I: Aphasia testing in
culturally contrasted populations (control subjects). Neuropsychologia, 25, 231245.
Lezak, M. D. (2004). Neuropsychological assessment (4th ed.). New York: Oxford University press.
Luria, A. R. (1966). Higher cortical functions in man. New York: Basic.
Luria, A. R. (1976). Basic problems of neurolinguistics. The Hague: Mouton & Co.
Lyons, F., Hanley, J. R., & Kay, J. (2002). Anomia for common names and geographical names with preserved retrieval of names of people: A
semantic memory disorder. Cortex, 38, 2335.
Mackay, A. I., Connor, L. T., Albert, M. L., & Obler, L. K. (2002). Noun and verb retrieval in healthy aging. Journal of the International
Neuropsychological Society, 23, 764770.
Manly, J. J., Jacobs, D. M., Sano, M., Bell, K., Merchant, C. A., Small, S. A., et al. (1999). Effect of literacy on neuropsychological test performance
in nondemented, education-matched elders. Journal of the International Neuropsychological Society, 5, 191202.
Marien, P., Mampaey, E., Vervaet, A., Saerens, J., & De Deyn, P. P. (1998). Normative data for the Boston naming test in native Dutch-speaking
Belgian elderly. Brain and Language, 65, 447467.
Merten, T. H. (2004). Development of a German Short Version of the Boston Naming Test. Neurological Rehabilitation, 10, 305311.
Moore, C. J., & Price, C. J. (1999). A functional neuroimaging study of the variables that generate category-specific object processing differences.
Brain, 122, 943962.
Newcombe, F., Oldfield, R. C., Ratcliff, G. G., & Winfield, A. (1971). Recognition and naming of object-drawings by men with focal brain wounds.
Journal of Neurosurgery and Psychiatry, 34, 329340.
Pineda, D. A., Rosselli, M., Ardila, A., Mejia, S. E., Romero, M. G., & Perez, C. (2000). The Boston Diagnostic Aphasia ExaminationSpanish
version: The influence of demographic variables. Journal of the International Neuropsychological Society, 6, 802814.
Ramsay, C. B., Nicholas, M., Au, R., Obler, L. K., & Albert, M. L. (1999). Verb naming in normal aging. Applied Neuropsychology, 6, 5767.
Rosselli, M., Ardila, A., & Rosas, P. (1990). Neuropsychological assessment in illiterates. II: Language and praxic abilities. Brain and Cognition,
12, 281296.
Spitzer, M., Kischka, U., Guckel, F., Bellemann, M. E., Kammer, T., Seyyedi, S., et al. (1998). Functional magnetic resonance imaging of category-
specific cortical activation: evidence for semantic maps. Brain Research Cognitive Brain Research, 6, 309319.
Spreen, O., & Strauss, E. (1998). A compendium of neuropsychological tests (2nd ed.). New York: Oxford University press.
Swadesh, M. (1952). Lexicostatistic dating of prehistoric ethnic contacts. Proceedings of the American Philosophical Society, 96, 152163.
Swadesh, M. (1967). El Lenguaje y la Vida Humana [Language and human life]. Mexico: Fondo de Cultural Economica.
Tsang, H. L., & Lee, T. M. (2003). The effect of ageing on confrontational naming ability. Archives of Clinical Neuropsychology, 18, 8189.

Warrington, E. K., & Shallice, T. (1984). Category specific semantic impairments. Brain, 107, 829854. www.itri.brighton.ac.uk/Adam.
Kilgarriff/bnc-readme.html#lemmatised www.rosettaproject.org/live/search/contribute/swadesh/view/index html www.ethnoculture.com
www.ethnologue.com
Normal aging increases cognitive heterogeneity: Analysis of

dispersion in WAIS-III scores across age
Alfredo Ardila
Department of Communication Sciences and Disorders, Florida International University, HLS139, Miami, FL 33199, USA
Accepted 12 August 2007
Abstract
Individual differences in cognitive decline during normal aging need further delineation. The purpose of this study was to find the
score dispersions in the WAIS-III subtests at different ages. Norms presented in the Administration and Scoring Manual [Wechsler,
D. (1997). WAIS-III: Administration and scoring manual. San Antonio: The Psychological Corporation] were used. The WAIS-III
was standardized and normalized using 2450 American adults divided into 13 age ranges and 4 education groups. Means and standard
deviations for the different WAIS-III subtests were deduced and the ratio Percentage of the mean = (standard deviation/mean) 100
was calculated. It was hypothesized that during normal aging, whereas mean scores decrease, score dispersions increase, pointing
to an increased heterogeneity in intellectual abilities in older individuals. In all subtests, except Digit Span, it was found that score
dispersions indeed increased during aging. However, in some subtests, increase in dispersion was less than 20% (Block Design,
Object Assembly, and Information), whereas in others, increase in dispersion was over 200% (Matrix Reasoning, L-N Sequencing,
Digit-Symbol, Picture Completion, and Picture Arrangement). It was proposed that cognitive heterogeneity during normal aging
is related to those abilities measured with these latter subtests, basically, executive functions, attention, and selected non-verbal
abilities. In other abilities (e.g., visuoconstructive abilities and fund of general information), normal aging is associated with a more
homogenous pattern of decline.
2007 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved.
Keywords: Aging; WAIS-III; Cognitive changes
Diverse social, physical, psychological, and cognitive changes develop with aging. Cognitive changes include
attention, memory, language, visuospatial and visuomotor abilities, perception, and executive functions (Albert &
Knoefel, 1994; Ardila, 2003; Cummings & Benson, 1992; Mendez & Cummings, 2003; Rosselli & Ardila, 2001).
Usually, memory is considered to be the cognitive domain most sensitive to the aging effect (Albert, 1997, 2002). Age-
related cognitive changes, however, are not homogenous; variations in the magnitude and rate depend on the specific
cognitive function (Albert, 1994). Neuropsychological test performance tends to present a negative correlation with
age, but some subjects can maintain appropriate cognitive functioning even during the eighth decade of life (Cerella,
1990; von Faber et al., 2001).
Consequently, age is a significant variable influencing intelligence measures (Albert & Heaton, 1988; Wechsler,
1997). Some of the intelligence subtest scores (e.g., WAIS subtests), however, are influenced by age, while others
Tel.: +1 305 348 2750.

E-mail address: ardilaa@fiu.edu.
0887-6177/$ see front matter 2007 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.acn.2007.08.004
remain relatively independent from this variable. The classical age-intelligence pattern describes non-verbal IQ as
strongly influenced by age as compared to verbal IQ. Verbal subtests have been considered to measure crystallized
intelligence (Cattell, 1963; Ryan, Sattler, & Lopez, 2000), a type of intelligence that is maintained with little variance
over a lifetime (Mackay, Connor, Albert, & Obler, 2002). Through this type of intelligence the subject expresses
previous verbal learning and general fund of knowledge (Horn, 1985; Horn & Cattell, 1966).
Non-verbal (or performance) subtests are part of fluid intelligence that allows the use of current information in
the solution of new problems (Cattell, 1963). This type of intelligence requires visuomotor and/or spatial skills and
is very sensitive to aging: younger subjects outperform older ones in those tasks or problems that require the use of
fluid intelligence (Ardila & Rosselli, 1989; Ryan et al., 2000). Older subjects perform well on tasks that use previous
knowledge in the solution of the problem, but may have serious difficulties in tasks that require unfamiliar solutions.
A fundamental factor impacting scores on non-verbal tests is speed. When time limits are used, older people are
at disadvantage (Albert, 1994; La Rue, 1992). Age is usually associated with motor slowness, decreased speed in
perceptual information processing, and an increase in reaction times. Other factors such as flexibility of thinking and
meaningfulness of the test may influence test performance as well (Albert, Wolfe, & Lafleche, 1990). It is assumed
that flexibility of thinking decreases only after the seventies (Willis, Geo, Thomas, & Garry, 1988). Wecker, Kramer,
Hallam, & Delis (2005) administered to 719 individuals, 2089-year old, three different tasks that required verbal and
non-verbal cognitive switching. It was found that mental flexibility is affected by age independently from age-related
changes in component skills.
The most common test of intelligence in adults is the Wechsler Adult Intelligence Scale (WAIS) (Wechsler, 1944,
1977). Two additional versions have been developed: WAIS-R (Wechsler, 1981) and WAIS III (Wechsler, 1997).
WAIS scales are sensitive to the aging effect. For instance, to obtain an IQ of 100 (average) at age 74 requires
answering correctly half of the questions needed to obtain the same IQ at age 20. Due to this age effect, the WAIS
scales adjust scores according to the subjects age. The latest WAIS version (WAIS-III) includes some new subtests
directed to overcome some of the limitations found in the previous Wechsler intelligence scale versions: Instead
of 11 subtests, there are 14 (seven verbal subtests and seven performance subtests) subtests. In the Verbal Scale,
the Letter-Number Sequencing subtest is added. In the Performance Scale, Matrix Reasoning and Symbol Search
areas are added. Besides, the Digit-Symbol subtest is used under different conditions: Digit-SymbolCoding (as
previously used), Digit-SymbolIncidental Learning (paringto recall the symbols matched with the numbers; and
free recallto recall the symbols used in the coding section), and Digit-SymbolCopy (to copy the symbols that were
used in the Digit-SymbolCoding).
Regardless, there are a significant amount of publications analyzing different aspects of the WAIS-III (currently there
are over 100 published papers); seemingly only one paper has approached the question of WAIS-III scores and aging.
Ryan et al. (2000) investigated the effect of age on the Wechsler Adult Intelligence Scale-III using the normative scores.
They found that there were few differences in the verbal abilities of the younger and older age groups. The Information
subtest showed the most stability across the age ranges. Performance subtests presented significant changes across age
ranges. They concluded that, with advancing age, performance on subtests measuring speed of information processing
and perceptual organization change substantially, whereas verbal subtests show minimal or no change. They interpreted
these results as supporting the view that measures of fluid intelligence show more of a decline with advancing age than
do measures of crystallized intelligence. No analysis of score dispersions across age ranges, however, is presented.
In summary, it is well known that cognitive decline occurs with age. However, individual differences in cognitive
decline are not so clear. An obvious way to find individual differences during normal aging is to analyze the dispersion
in test scores occurring at different ages.
The purpose of this study was to find the score dispersions in different WAIS-III subtests at different ages.
It was hypothesized that, whereas mean scores decrease, dispersions increase, pointing to an increased hetero-
geneity intellectual ability during normal aging. Dispersion was understood not simply as the standard deviations
of the test scores, but as the relationship between the standard deviations and the means of the different test
scores.
1. Procedure
Norms presented in the Administration and Scoring Manual (Wechsler, 1997) were used. The WAIS-III standardiza-
tion data were not available, and only the information contained in the WAIS-III Administration and Scoring Manual
was used in the analyses presented in this paper. Means and standard deviations for the different WAIS-III subtests
were deduced using the following procedure:
(1) In each subtest, the raw score corresponding to the scaled score 10 was considered as the mean. When the scaled
score 10 corresponded to a score range (e.g., 3438) the mean for the range was used (e.g., 36).
(2) For calculating the standard deviations, scaled scores equal to 7 and 13 were selected (i.e., minus one and plus one
standard deviation). The mean raw score minus the raw score corresponding to a scaled score 7; and the raw score
corresponding to a scaled score 13 minus the mean raw score were added and divided by two; the resulting value
was considered as the standard deviation for that particular subtest. When the scaled scores 7 and 13 corresponded
to a range (e.g., 2225) the mean value was considered as the scaled score (e.g., 23.5).
(3) Finally, the ratio Percentage of the mean = (standard deviation/mean) 100 (that is, the percentage of the mean
represented by the standard deviation) was calculated. This percentage of the mean ratio informs about the
heterogeneity (dispersion) of the scores. As a matter of fact, the standard deviation is not informative by itself; it
only has meaning with regard to the mean. A standard deviation of 3 can be very low if the mean is 50, but very
high if the mean is 5. At the best of my knowledge, the ratio Percentage of the mean is not frequently used.
The WAIS-III was standardized and normalized using 2450 American adults divided into 13 age ranges (1617,
1819, 2024, 2529, 3034, 3544, 4554, 5564, 6569, 7074, 7579, 8084, and 8589), and taken from four
different geographical areas (Northeast, North Central, South and West). About 79% (1925 participants) of the sample
is referred to as Whites, about 11.4% (279 participants) is referred to as African-Americans, about 7.4% (181
participants) is referred to as Hispanics, and about 2.6% (65 participants) is referred to as Others. Four education
groups are distinguished (<8, 911, 12, 1315 and >16). About 11.5% of the sample had an education less or equal to
8 years, and about two thirds of the subjects had 12 or more years of education. Education at the different age ranges
is not mentioned (for a detailed description of the sample and sample selection, see Wechsler, 1997).
2. General results
Table 1 presents the general results of the verbal subtests. Because the WAIS-III standardization data were not
available only descriptive statistics (mean scores, standard deviations, and the ratio Percentage of the mean) were used.
2.1. Vocabulary
Middle age ranges show higher dispersion. Higher scores (about 30% higher) are observed during the 4554 years
range. Scores in the first and last age ranges are similar. Dispersions are variable across-ages.
2.2. Similarities
Similarities subtest scores tend initially to mildly increase. Highest scores are found in the 3054 range. Later, a
slow decrease in scores is observed. Lowest mean score in the age range of 8589 is about 30% lower than at the 3054
range. The ratio Percentage of the mean increases about 50%. This means, some increased heterogeneity in scores is
evident.
2.3. Arithmetic
Scores increase up to the 4554 years. Further slow decrease is observed. Standard deviations remain relatively
unchanged.
2.4. Digits
Raw scores present a slow but continued decrease observed beginning in the early 30s. Dispersions remain quite
stable.
Table 1
Calculated raw scores in the different verbal WAIS-III subtests
1617 1819 2024 2529 3034 3544 4554 5564 6569 7074 7579 8084 8589
Vocabulary
Mean 33.5 36.0 36.0 40.0 42.5 44.4 46.5 41.0 41.0 41.0 40.5 35.5 35.0
S.D. 11.7 12.7 12.7 12.2 12.5 13.0 13.0 13.8 13.8 13.5 12.5 13.5 13.0
% of the mean 34.9 35.2 35.2 30.5 29.4 29.3 27.9 33.6 33.6 32.9 30.9 38.0 37.1
Similarities
Mean 22.0 22.5 22.5 22.5 23.5 23.5 23.5 21.0 20.0 20.0 19.5 17.5 16.5
S.D. 5.5 5.5 5.5 5.7 5.7 5.7 6.5 6.7 6.5 6.0 6.0 6.2 6.0
% of the mean 25.0 24.4 24.4 25.3 24.2 24.2 27.6 31.9 32.5 30.0 30.7 35.4 36.3
Arithmetic
Mean 12.0 12.0 12.5 12.5 13.0 13.5 14.5 13.5 12.5 12.5 12.0 11.0 10.0
S.D. 3.5 3.5 4.0 4.5 4.0 4.0 4.0 4.0 4.0 4.0 3.7 3.0 3.0
% of the mean 29.2 29.2 32.0 36.0 30.7 33.7 27.5 33.7 31.2 31.2 32.4 27.2 33.3
Digits
Mean 17.5 17.5 17.5 17.5 17.0 17.0 17.0 16.0 16.0 15.0 15.0 14.5 14.0
S.D. 4.2 4.2 4.5 4.5 4.2 4.2 4.2 4.5 4.2 3.2 4.0 3.7 3.2
% of the mean 24.0 24.0 25.7 25.7 24.7 24.7 24.7 28.1 26.2 21.3 26.6 25.5 22.8
Information
Mean 14.5 14.5 14.5 14.5 15.5 16.5 17.5 16.5 16.5 16.5 15.5 13.5 13.5
S.D. 5.5 5.7 5.7 5.7 6.0 5.7 5.7 6.0 6.0 5.7 6.0 5.5 5.5
% of the mean 37.9 29.3 29.3 29.3 38.7 34.5 32.5 36.3 36.3 34.5 38.7 40.7 40.7
Comprehension
Mean 17.5 18.5 18.5 19.5 20.5 21.5 22.5 20.5 20.5 20.5 19.5 17.5 16.5
S.D. 5.7 5.7 6.0 5.7 6.0 5.5 5.7 6.0 6.0 5.7 5.7 6.0 5.7
% of the mean 32.5 30.8 32.4 29.3 29.2 25.8 25.3 29.2 29.2 27.8 29.2 34.2 34.5
L-N Sequencing
Mean 11.0 11.0 11.0 11.0 11.0 11.0 10.0 9.0 9.0 8.0 8.0 7.0 6.0
S.D. 2.5 3.0 3.0 2.7 2.5 2.5 3.0 2.5 2.7 2.5 2.5 3.0 3.0
% of the mean 22.7 27.2 27.2 24.5 22.7 22.7 30.0 27.7 30.0 31.2 31.2 42.8 50.0
Means and standard deviations (in parenthesis) as well as the percentage of the mean ratios are presented.
2.5. Information
As in the Vocabulary subtest, an inverted U-shaped curve between the 1617 and 8589 years is shown for the
Information subtest. Higher scores (about 20% higher) are observed during the 4554 years range. Scores in the first
and last age ranges are similar. Standard deviations are relatively high.
2.6. Comprehension
Comprehension is another subtest presenting an inverted U-shaped curve between the 1617 and 8589 years.
Higher scores (about 20% higher) are observed during the 4554 years range. Scores in the first and last age ranges
are similar. Dispersions are stable.
2.7. L-N Sequencing
Mean scores remained stable up to the 3544 age range. Further, continuous decline is observed. Scores at the oldest
age range are about 40% of the scores at 3344 range. Dispersions steadily increase from the 5564 up to the oldest
range.
Table 2 presents the general results of the performance subtests.
Table 2
Calculated raw scores in the different performance WAIS-III subtests
1617 1819 2024 2529 3034 3544 4554 5564 6569 7074 7579 8084 8589
Picture Completion
Mean 20.5 20.5 20.5 20.5 20.5 20.5 20.0 19.5 17.5 16.5 16.5 15.5 14.5
S.D. 3.0 3.0 3.0 3.0 2.7 2.7 3.0 4.2 5.2 4.7 4.2 4.7 4.7
% of the mean 14.6 14.6 14.6 14.6 13.1 13.1 15.0 21.5 29.7 28.4 25.4 30.0 32.4
Digit-Symbol
Mean 77.0 81.0 80.0 78.0 77.0 75.0 70.0 61.0 54.0 51.0 47.0 42.0 35.0
S.D. 15.7 16.0 16.2 15.5 16.0 16.5 15.2 15.0 15.0 14.7 14.5 15.0 14.2
% of the mean 20.3 19.7 20.2 19.8 20.7 22.0 21.7 26.0 27.7 34.6 30.8 35.7 40.5
Block Design
Mean 41.5 41.5 41.5 41.5 41.0 40.5 36.5 33.5 29.5 28.5 26.5 25.0 24.0
S.D. 13.5 13.5 13.5 13.5 13.2 13.0 12.0 12.0 11.7 10.0 9.0 8.0 8.0
% of the mean 32.5 32.5 32.5 32.5 32.1 32.1 33.8 35.8 39.6 35.0 33.9 32.0 33.3
Matrix Reasoning
Mean 17.5 16.5 16.5 16.5 16.5 15.5 13.5 12.5 10.5 9.0 9.0 7.5 7.0
S.D. 4.0 4.7 4.7 4.7 5.0 5.2 5.5 6.0 4.5 4.5 4.0 3.7 3.2
% of the mean 22.8 28.4 28.4 28.4 30.3 33.5 40.7 48.5 42.8 50.0 44.4 49.3 45.7
Picture Arrangement
Mean 15.5 15.5 15.5 15.5 15.5 14.5 13.5 12.5 10.5 8.5 8.0 6.5 4.5
S.D. 4.2 4.2 4.2 4.5 4.2 4.7 5.0 4.7 5.7 5.0 4.2 4.5 3.7
% of the mean 27.0 27.0 27.0 29.0 27.0 32.4 37.0 37.6 54.2 58.8 52.5 69.2 82.2
Symbol Search
Mean 34.0 35.5 34.5 34.5 33.0 32.5 29.0 26.0 22.5 22.0 19.5 16.0 15.0
S.D. 9.7 9.5 9.2 9.2 9.0 8.5 7.5 8.0 8.7 7.5 6.0 7.0 7.0
% of the mean 28.5 26.7 26.6 26.6 27.2 26.1 25.8 30.7 38.6 34.1 30.7 43.7 46.6
Object Assembly
Mean 34.5 35.0 35.0 35.0 35.0 33.5 29.0 27.5 23.5 22.5 21.5 18.5 18.5
S.D. 10.0 10.2 10.2 10.2 10.2 11.5 7.5 10.0 9.5 8.0 7.7 6.5 6.2
% of the mean 28.9 29.1 29.1 29.1 29.1 34.3 25.8 36.3 28.3 35.5 35.8 35.1 33.5
Means and standard deviations (in parenthesis) as well as the percentage of the mean ratios are presented.
2.8. Picture Completion
Highest scores are found in the 1634 age range. At 8589 age range, scores are about 70% of these highest scores.
Dispersions virtually duplicates between the youngest and oldest groups (14.6 and 32.4).
2.9. Digit-Symbol
Scores at the oldest range are less than half of the scores found in younger. Dispersion duplicates.
2.10. Block Design
Scores at the oldest age range are about 60% of the scores at the youngest age range. Dispersion in scores, however,
remains virtually identical.
2.11. Matrix Reasoning
Scores at the oldest age range are about 40% of the scores at the youngest age range. Dispersion is about twice in
the oldest than in the youngest participants.
2.12. Picture Arrangement
Scores at the oldest age range are about 29% of the scores at the youngest age range. Dispersion is about three times
in the oldest than in the youngest participants.
2.13. Symbol-Search
Scores at the highest age range are about 45% of the scores at the lowest age range. Dispersion is about one and
half times higher in the oldest than in the youngest participants.
2.14. Object Assembly
Scores at the oldest age range are about 50% of the scores at the youngest age range. Dispersion presents just a mild
increase when becoming older.
3. Age-dependent changes in cognition
Two groups were selected to analyze the age-dependent changes in WAIS-III subtest scores: (1) the age group in
which the highest scores were found; and (2) the age group where the lowest scores were found. Lowest scores for
all the subtests were found in the oldest group (Table 3). Highest scores, however, were observed at different ages.
In general, for verbal abilities highest scores were observed during the 40s and 50s, whereas for non-verbal abilities
highest scores were found in the 1634 years range. Dispersion increased in most subtests, but important differences in
the magnitude of this increase were noted. The sequence of decline and the sequence of heterogeneity were analyzed.
3.1. Sequence of decline
In some subtests, scores were apparently associated with age. In others, the association with age was mild. Con-
sequently, a hierarchy of age-related cognitive decline can be proposed (Table 4). The scores in Digits, Information,
Vocabulary, Picture Completion, and Similarities subtests decreased less than 30% during aging (highest and lowest
scores), whereas in Symbol Search, Digit-Symbol, Matrix Reasoning, and Picture Arrangement, decreases were over
Table 3
Variations in the means and dispersions (percentage of the mean)
Subtest First Group (G1) at the Second Group (G2) at Comparing G1 and G2
age of the highest score 8589 years
Age Mean % of the Mean % of the Variation in Variation in the
mean mean the mean % of the mean
Vocabulary 4554 46.5 27.9 35.0 37.1 75% 133%

Similarities 3054 23.5 24.227.6 16.5 36.3 70% 140%
Arithmetic 4554 14.5 27.5 10.0 33.3 69% 121%
Digits 1629 17.5 24.025.7 14.0 22.8 80% 92%
Information 4554 17.5 32.5 13.5 40.7 77% 116%
Comprehension 4554 22.5 25.3 16.5 34.5 73% 136%
L-N Sequencing 1644 11.0 22.727.4 6.0 50.0 54% 200%
Picture Completion 1644 20.5 13.114.6 14.5 32.4 71% 234%
Digit-Symbol 1819 81.0 19.7 35.0 40.5 43% 206%
Block Design 1629 41.5 32.5 24.0 33.3 58% 102%
Matrix Reasoning 1617 17.5 22.8 7.0 45.7 40% 200%
Picture Arrangement 1634 15.5 27.029.0 4.5 82.2 29% 293%
Symbol Search 1819 35.5 26.7 15.5 46.6 44% 174%
Object Assembly 1834 35.0 29.1 18.5 33.5 53% 115%
Note: (G1) age group in which the highest scores were found; (G2) age group where the lowest scores were found (8589 years). Variations (in
percentages) for the means and standard deviations are presented.
Table 4
Hierarchy of age-related cognitive decline
Subtest Percentage of score difference
between the highest and lowest score
Digits 80% (Mild decline with aging)

Information 77%
Vocabulary 75%
Comprehension 73%
Picture Completion 71%
Similarities 70%
Arithmetic 69%
Block Design 58%
L-N Sequencing 54%
Object Assembly 53%
Symbol Search 44%
Digit-Symbol 43%
Matrix Reasoning 40%
Picture Arrangement 29% (Strong decline with aging)
Table 5
Changes in dispersion in the different WAIS-III subtests
Percentage of dispersion at the oldest range
with regard to the highest score range
Digits 92% (No heterogeneity with aging)

Block Design 102%
Object Assembly 115%
Information 116%
Arithmetic 121%
Vocabulary 133%
Comprehension 136%
Similarities 140%
Symbol Search 174%
Matrix Reasoning 200%
L-N Sequencing 200%
Digit-Symbol 206%
Picture Completion 234%
Picture Arrangement 293% (Strong heterogeneity with aging)
50% (that is, the lowest score was less than half of the highest score). This sequence of decline can be useful in
understanding the effects of aging on cognition.
3.2. Sequence of heterogeneity
It was found that score dispersions increased in all subtestsexcepting Digits; however, in some subtests increase
in dispersion was less than 20% (Block Design, Object Assembly, and Information), whereas in others, increase
in dispersion was over 200% (Matrix Reasoning, L-N Sequencing, Digit-Symbol, Picture Completion, and Picture
Arrangement) (Table 5). Consequently, there is a test-dependent heterogeneity associated with aging.
4. Conclusions
Interpreting cross-sectional studies when analyzing age-related changes in cognition can be problematic, because
there is a cohort effect potentially obscuring the results. People at different ages may significantly differ in education,
life experiences, environmental conditions and even nutrition. Obviously, longitudinal studies may provide the most
accurate way to look for age trends, but from the practical point of view they may be extremely difficult to carry out.
Different limitations to this study should be pointed out: it represents a secondary analysis of some available
data; statistical analyses are simply descriptive; many details regarding the sample selection are not available; means
and standard deviations are deduced from the information presented in the WAIS-III Manual; etc. Regardless these
limitation, however, some tentative conclusions can be drawn.
Highest scores in different subtests were observed at different ages. In general, highest scores were found at an
earlier age in those tests tapping non-verbal fluid abilities. However, decline in scores occurred earlier in these non-
verbal than in verbal subtests. Frequently, but not necessarily, these subtests were timed tests. Highest scores in the
non-verbal subtests were found below the mid 30s; later, score decline was found. In one subtest (Matrix Reasoning)
score decline was apparent as early as the 1819 years range. For verbal tests, highest scores were most frequently
found around the late 40s and early 50s.
Taking the age when the highest score is found, the two subtests with the highest dispersions were Information
and Block Design, whereas the subtest with the lowest dispersions was Picture Completion. Dispersion in the last one
was less than half of the dispersion found in Information and Block Design. For the rest of the tests, dispersions were
similar at the age of the highest scores.
Some subtests were extremely sensitive to the aging effect, whereas in others, aging had just a mild effect on scores.
Decline in scores (difference between the highest and lowest) ranged between 20% (Digits) and over 70% (Picture
Arrangement).
In all the subtestsexcept Digits, it was found that score dispersions increased during aging. However, in some
subtests increase in dispersion was less than 20% (Block Design, Object Assembly, and Information), whereas in
others, increase in dispersion was over 200% (Matrix Reasoning, L-N Sequencing, Digit-Symbol, Picture Completion,
and Picture Arrangement). It was proposed that cognitive heterogeneity during normal aging is related to those abilities
measured with these latter subtests, basically, executive functions, attention, and some non-verbal abilities. In other
abilities (e.g., working memory, visuoconstructive abilities, and fund of general information) normal aging is associated
with a more homogenous pattern of decline.
Some association between age-related dispersion and age-related decline can be conjectured. For example, the
greatest dispersion appears to be seen on the same measures that show the greatest decline (e.g., Picture Arrangement),
whereas the lowest dispersion was generally observed in those measures with the lowest decline (e.g., Digits).
A note of caution should be introduced. Comparing scores at different age ranges has a potentially major confounding
effect: educational level at each age-rage is not reported in the WAIS-III Manual (Wechsler, 1997). Quite likely, however,
younger people had a higher education than older participants. Consequently, differences observed across age-ranges
may be due to an educational rather than age effect. Observed age differences probably are significantly inflated due
to this uncontrolled educational artifact. Nonetheless, the focus of the previous analysis was the score dispersions,
not the mean scores. Education is clearly correlated with increased scores in cognitive test performance (e.g., Ardila,
Ostrosky-Solis, Rosselli, & Gomez, 2000), but seemingly, education has not ever been reported to be correlated with
decreased score dispersion. Consequently, the educational differences do not seem to be affecting the age-related
changes in score dispersions.
In conclusion, current results support the assumption that aging is not only associated with a general decline
in cognitive test scores, but also with increase in test scores heterogeneity. With aging, individuals become more
intellectually heterogeneous. This observation can be especially useful not only from a clinical but also theoretical
point of view for accurately understanding the effects of aging on cognition. Analyzing the scores dispersion in different
cognitive tests may significantly further our understanding of age-related changes in cognition.
References
Albert, M. S. (1994). Age-related changes in cognitive function. In M. L. Albert & J. E. Knoefel (Eds.), Clinical neurology of aging. New York:
Oxford University Press.
Albert, M. S. (1997). The ageing brain: Normal and abnormal memory. Philosophical Transactions of the Royal Society of London. B Biological
Sciences, 352(1362), 17031709.
Albert, M. S. (2002). Memory decline: The boundary between aging and age-related disease. Annals of Neurology, 51, 282284.
Albert, M. S., & Heaton, R. K. (1988). Intelligence testing. In M. S. Albert & M. B. Moss (Eds.), Geriatric neuropsychology (pp. 1032). New
York: The Guilford Press.
Albert, M. L., & Knoefel, J. E. (Eds.). (1994). Clinical neurology of aging. New York: Oxford.
Albert, M. S., Wolfe, J., & Lafleche, G. (1990). Differences in abstraction ability with age. Psychology and Aging, 5, 94100.
Ardila, A. (2003). El proceso de envejecimiento normal [Nomal aging process]. In J. C. Arango, S. Fernandez Guinea, & A. Ardila (Eds.), Las
demencias: aspectos clnicos, neuropsicologicos y tratamiento (2nd ed., pp. 313). Mexico: Editorial Manual Moderno.
Ardila, A., Ostrosky-Solis, F., Rosselli, M., & Gomez, C. (2000). Age related cognitive decline during normal aging: The complex effect of education.
Archives of Clinical Neuropsychology, 15, 495514.
Ardila, A., & Rosselli, M. (1989). Neuropsychological characteristics of normal aging. Developmental Neuropsychology, 5, 307320.
Cattell, R. B. (1963). Theory of fluid and crystallized intelligence: A critical experiment. Journal of Educational Psychology, 54, 122.
Cerella, J. (1990). Aging and information processing rate. In J. E. Birren & K. W. Schaie (Eds.), Handbook of the psychology of aging (3rd ed., pp.
7092). San Diego: Academic Press.
Cummings, J. L., & Benson, D. F. (1992). Dementia: A clinical approach (2nd ed.). London: Butterworths.
Horn, J. L. (1985). remodeling old models of intelligence. In B. B. Wolman (Ed.), Handbook of intelligence: Theories, measurements and applications.
New York: Wiley.
Horn, J. L., & Cattell, R. B. (1966). Refinement and test of the theory of fluid and crystallized intelligence. Journal of Educational Psychology, 57,
253270.
La Rue, A. (1992). Adult development. In A. E. Puente & R. J. McCaffrey (Eds.), Handbook of neuropsychological assessment: A biopsychosocial
perspective. New York: Plenum.
Mackay, A. I., Connor, L. T., Albert, M. L., & Obler, L. K. (2002). Noun and verb retrieval in healthy aging. Journal of the International
Mendez, M., & Cummings, J. L. (2003). Dementia: A clinical approach. New York: Oxford.
Rosselli, M., & Ardila, A. (2001). Normal and abnormal aging. In M. Ponton & J. Leon-Carrion (Eds.), Neuropsychology and the hispanic patient
(pp. 341360). Mahwah, NJ: Lawrence Erlbaum Associates.
Ryan, J. J., Sattler, J. M., & Lopez, S. J. (2000). Age effects on Wechsler Adult Intelligence Scale-III subtests. Archives of Clinical Neuropsychology,
15, 311317.
von Faber, M., Bootsma, van der Wiel, A., van Exel, E., Gussekloo, J., Lagaay, A. M., et al. (2001). Successful aging in the oldest old: Who can be
characterized as successfully aged? Archives of Internal Medicine, 161, 26942700.
Wechsler, D. (1944). The measurement of adult intelligence. Baltimore: The Williams and Wilkins Company.
Wechsler, D. (1977). Manual: Wechsler Adult Intelligence Scale. Orlando, FL: Psychological Corporation.
Wechsler, D. (1981). Wechsler Adult Intelligence Scale-Revised. New York: Psychological Corporation.
Wechsler, D. (1997). WAIS-III: Administration and scoring manual. San Antonio: The Psychological Corporation.
Wecker, N. S., Kramer, J. H., Hallam, B. J., & Delis, D. C. (2005). Mental flexibility: Age effects on switching. Neuropsychology, 19, 345352.
Willis, L., Geo, R. A., Thomas, P., & Garry, P. J. (1988). Differential decline in cognitive function: The possible role of health status. Developmental
Neuropsychology, 4, 2328.
DEVELOPMENTAL NEUROPSYCHOLOGY, 28(1), 539560
Copyright 2005, Lawrence Erlbaum Associates, Inc.
The Influence of the Parents

Educational Level on the Development
of Executive Functions
Alfredo Ardila
Department of Communication Sciences and Disorders
Mnica Rosselli
Department of Psychology
Florida Atlantic University
Esmeralda Matute and Soledad Guajardo

Institute of Neuroscience
University of Guadalajara, Mexico
Information about the influence of educational variables on the development of ex-

ecutive functions is limited. The aim of this study was to analyze the relation of the
parents educational level and the type of school the child attended (private or pub-
lic school) to childrens executive functioning test performance. Six hundred
twenty-two participants, ages 5 to 14 years (276 boys, 346 girls) were selected
from Colombia and Mexico and grouped according to three variables: age (56,
78, 910, 1112, and 1314 years), gender (boys and girls), and school type (pri-
vate and public). Eight executive functioning tests taken from the Evaluacin
Neuropsicolgica Infantil; Matute, Rosselli, Ardila, & Ostrosky, in press) were in-
dividually administered: Semantic Verbal Fluency, Phonemic Verbal Fluency, Se-
mantic Graphic Fluency, Nonsemantic Graphic Fluency, Matrices, Similarities,
Card Sorting, and the Mexican Pyramid. There was a significant effect of age on
all the test scores and a significant effect of type of school attended on all but Se-
mantic Verbal Fluency and Nonsemantic Graphic Fluency tests. Most childrens
test scores, particularly verbal test scores, significantly correlated with parents ed-
Requests for reprints should be sent to Alfredo Ardila, Department of Communication Sciences and
Disorders, Florida International University, 12230 NW 8th Street, Miami, FL 33182. E-mail:
alfredoardila@cs.com
540 ARDILA, ROSSELLI, MATUTE, GUAJARDO
ucational level. Our results suggest that the differences in test scores between the
public and private school children depended on some conditions existing outside
the school, such as the parents level of education. Implications of these findings
for the understanding of the influence of environmental factors on the development
of executive functions are presented.
The term executive function is a relatively new term in the neurosciences. The ob-
servation that the frontal lobes were involved in behaviors such as problem solv-
ing, planning, inhibiting responses, strategy development with implementation,
and working memory resulted in the comprehensive term executive function
(Lezak, 1983). These skills improve as the brain matures during childhood and ad-
olescence (Anderson, Anderson, Northam, Jacobs, & Catroppa, 2001). The in-
complete development of the frontal lobes in childhood implies that a limited abil-
ity to apply effective executive skills may be present (Anderson, Lajoie, & Bell,
1996). Although age (as an index of maturational status) seems to account for most
of the executive function test-performance variability in children, very little has
been studied regarding the influence of environmental factors. This study focused
on the effect of two environmental variables on executive function test perfor-
mance: the influence of the level of education of the childrens parents and the type
of school that the child attends (public or private).
Few studies have approached the development of executive functions in chil-
dren (e.g., Anderson, 2001; Espy, Kaufmann, & Glisky, 2001; Levin et al., 1991;
Soprano, 2003; Welsh & Pennington, 1988). Generally, a progressive improve-
ment in test score with maturing age has been observed. However, the increase in
scores depends on the test and is not steady across the different age ranges (Senn,
Espy, & Kaufmann, 2004). Recently, Klenberg, Korkman, and Lahti-Nuuttila
(2001) studied the development of executive functions in four hundred 3- through
12-year-old Finnish children. Data from 10 subtests measuring impulse control
and inhibition of irrelevant responses, auditory and visual attention, visual search,
planning, and verbal and visual fluency were included. The development pro-
ceeded sequentially, from motor inhibition and impulse control to functions of se-
lective and sustained attention, and finally to fluency. Significant relations be-
tween gender and development and between parent education and development
were found for several subtests. The authors concluded that development of basic
inhibitory functions precedes the development of more complex functions of se-
lective attention, and executive functions continue to develop into adolescence.
Anderson et al. (2001) used a sample of 138 children ages 11.0 to 17.11 years. So-
cioeconomic status (SES) was controlled across age groups. They found a relatively
flat development trajectory for executive functions during late childhood and early
adolescence. Progress, however, tended to be faster during early and middle child-
hood and to slow during late childhood and adolescence in this sample.
EXECUTIVE FUNCTIONS 541
Developmental studies have provided evidence that the development of execu-

tive functions is a multistage process in which different components develop at dif-
ferent times (Klenberg et al., 2001). Environmental variables have an unclear in-
fluence on this development, although performance on neuropsychological tests is
very sensitive to educational and cultural variables (Ardila, 1995; Fletcher-Janzen,
Strickland, & Reynolds, 2000; Lezak, 2004; Rosselli & Ardila, 2003). Links have
been suggested between cognitive development and SES in different countries
(Kohen, Gunn, Leventhal, & Hertzman, 2002). For example, spoken language
skills of children reared in poverty are depressed in comparison with the general
population, and in comparison with their general cognitive abilities. Hoff (2003b)
found that children whose families belong to high SES demonstrated larger pro-
ductive vocabulary than the children from mid-SES families. The influence of en-
vironmental factors has also been shown for the development of other cognitive
abilities (e.g., Ardila & Rosselli, 1994). Turkheimer, Haley, Waldron, DOnofrio,
and Gottesman (2003) suggested that the proportions of IQ variance attributable to
familial background and environment vary nonlinearly with SES.
One very important environmental influence in the childs cognitive develop-
ment is the parentslevel of education. Parents with a higher education create a more
intellectually stimulating environment for their children (Hoff, 2003a, 2003b). It has
been demonstrated that highly educated parents (professionals) have a different way
of interaction with their children particularly in respect to their language used (Hoff,
Laursen, & Tardif, 2002). College-educated mothers talk more, use a richer vocabu-
lary, and read more to their children than those mothers limited to a high school edu-
cation (Hoff-Ginsberg, 1991). Parentss educational level has also been related to
childrens school attendance and general cognitive development (e.g., Ganzach,
2000; Teachman, 1987; White, 1982). Children from parents with higher education
tend to have a larger vocabulary, more rapid language development, better perfor-
mance in cognitive tests, and higher school attendance. Portes, Cuentas, and Zady
(2000) assessed the relation of parentchild interaction to childrens intellectual
achievement. The results demonstrated that although interaction characteristics are
related to childrens intellectual achievement, that relation is moderated by context
factors that may vary in each culture. Some differences in educational systems
among different countries have to be considered. Further, the association between
parents education and private versus public schools may vary across countries.
SES is a compound variable that includes family income, parental education,
occupational status, and place of residence. A significant association between SES
and test scores has been frequently reported (e.g., Bradley & Corwyn, 2002;
Lichtenstein & Pedersen, 1997; Petrill & Deater-Deckard, 2004; Turkheimer et al.,
2003). Research shows that SES is associated with a wide array of health, cogni-
tive, and socioemotional outcomes in children, with effects beginning prior to birth
and continuing into adulthood. A variety of mechanisms linking SES to child
well-being have been proposed, with most involving differences in access to mate-
rial and social resources or reactions to stress-inducing conditions by both the chil-
dren themselves and their parents (Bradley & Corwyn, 2002). This SES effect on
test scores has been documented in diverse cultures. For example, Sigman,
Neumann, Jansen, and Bwibo (1989) observed in children growing up in rural
Kenya that cognitive scores were best predicted by a combination of factors, in-
cluding duration of schooling, food intake, physical stature, and SES. In this article
we are dealing only with parents educational level, but we are aware of the close
association between parents education and SES. As a matter of fact, parent educa-
tion is a major component of the SES variable.
The purpose of this research was to analyze the relation of the parents educa-
tional level and the type of school (private or public school) on children s execu-
tive functioning test performance. We predict a significant correlation between the
parents education and childrens performance on executive functioning measures
Likewise, it was anticipated that children from private schools would perform
better than children from public schools. Attendance at private or public school
may be linked with differences in the level of education of the parents, their inter-
ests, and their income. Most likely, the level of education and the income of chil-
drens parents who attended private school are higher that those children whose
parents attended public school. We anticipated that the effect of type of education
would be independent of the age of the child. In other words, children from private
schools would have higher test scores than children from public school across age
groups. It was also expected based on previous research (e.g., Anderson et al.,
2001; Klenberg et al., 2001; Rosselli, Ardila, Bateman, & Guzman, 2001) that test
scores would improve with age.
METHOD
Participants
Six hundred twenty-two 5- to 14-year-old children (276 boys, 346 girls) were se-
lected from private and public schools in Colombia (city of Manizales, population
about 500,000 inhabitants) and Mexico (city of Guadalajara, population about
3,300,000 inhabitants; and city of Tijuana, population about 1,100,000 inhabit-
ants). The mean educational levels of the public school childrens fathers and
mothers were 11.26 (SD = 4.23) and 10.82 years (SD = 3.99), respectively; the
mean educational levels of the private school childrens fathers and mothers were
15.65 (SD = 3.10) and 15.55 (SD = 2.72) years, respectively. The sample included
579 right-handers and 43 left-handers. Table 1 presents the gender, age, and type of
school distribution of the sample.
All participants were screened to identify any history of neurological or psy-
chiatric problems, mental retardation, and learning disabilities, using a struc-
TABLE 1
General Characteristics of the Sample
Boys Girls
Age Public Private Public Private

Group School School Total School School Total Total
56 23 25 48 29 30 59 107
78 26 25 51 30 36 66 117
910 38 23 61 42 37 79 140
1112 40 23 73 49 29 78 141
1314 29 24 53 37 27 64 117
Total 156 120 276 187 159 346 622
tured interview for parents included in the Evaluacin Neuropsicolgica Infantil

(ENI; Matute et al., in press). Although no formal testing was done to rule out
intellectual or learning disabilities, we screened for grade retention, ascertained
that no chronological agegrade level disparity was present, and that the chil-
drens reading and math performance agreed with their chronological grade lev-
els according to the school records. Handedness was assessed using the handed-
ness test included in the ENI.
Instruments
The scores of eight executive function subtests from the ENI (Matute et al., in
press) were analyzed. The ENI is a new neuropsychological test battery developed
in the Spanish language, targeting different cognitive domains (Matute, Montiel,
Rosselli, & Ardila, 2003; Matute, Rosselli, Ardila, & Morales, 2004; Rosselli,
Matute, Ardila, & Montiel, 2003).
The following are the executive functioning tests that were used.
1. Semantic Verbal Fluency: Children were instructed to name all animals they
could in 1 min. The score was the total number of animals named correctly.
2. Phonemic Verbal Fluency: Children were instructed to say as many words
starting with M as they could in 1 min, omitting all proper nouns (names of people,
places, etc.) as well as morphological variations of the same word. The score was
the total number of correct words.
These two verbal tests assess concept formation as well as an ability to simulta-
neously remember to think in an abstract manner and to shift responses when re-
quired (Anderson et al., 2001; Matute et al., 2004).
3. Semantic Graphic Fluency (meaningful figures): Children were instructed
to draw as fast as possible (within 3 min) all the different meaningful figures they
could on a page featuring 35 contiguous 1-in. squares in a 5 7 array (Matute et al.,

2004). The examiner instructed the child to make each drawing as simple as possi-
ble. One point was given for drawing of a shape that represented something defi-
nite. The points were added to obtain the total score.
4. Nonsemantic Graphic Fluency (geometric designs): Children were in-
structed to draw, as fast as possible, linear geometric figures connecting five points
with four lines within a square presented on a sheet of paper containing a 7 5 ar-
ray of 35 dot matrices (adapted from Regard, Strauss, & Knapp, 1982), but only 3
min were allowed as in Lee, Loring, Newell, and McCloskey (1994) and Spreen
and Strauss (1998). All 35 dot matrices were identical and contained five symmet-
rically arranged dots; four of them were black, and each one of them was located in
a different corner. The fifth point was white and was located at the center of the
square. All lines must connect dots and at least one has to touch the white dot.
Those figures that were not formed by four lines or those in which the white dot
was not connected were scored as intrusions, whereas those that the child had
made previously were scored as perseverations. One point was given to each cor-
rect drawing.
These two graphic fluency tests were designed to measure the individuals abil-
ity to generate a series of novel designs (Regard et al., 1982) and have been used as
a measure of executive functions in children (Matute et al., 2004).
5. Similarities: The child was requested to find commonalities between pairs of
words. This type of task assesses verbal concept formation (Lezak, 2004), and has
been associated with left temporal and frontal functioning. Eight pairs of words
were presented, one at a time. The word pairs ranged in level of difficulty from the
simplest (dog and cat), to the most difficult (liberty and justice). Abstract general-
izations or categorizations (i.e., A dog and a cat are animals) were scored with 2
points; specific and descriptive answers (i.e., A dog and the cat have four legs)
were scored with 1 point, and incorrect answers received a zero. Maximum score
was 16.
6. Matrices: This multiple-choice subtest consisted of a series of visual pat-
tern-matching and analogy problems pictured in nonrepresentational designs. It
requires the child to conceptualize spatial and design relations such as in Ravens
Progressive Matrices (Raven, Court, & Raven, 1976). This test consists of eight
items ranging from simple and concrete to complex and abstract. One point is
given to each option correctly selected by the child. The maximum score is 8.
7. Card Sorting: This is a sorting subtest that requires the child to shift concepts
of color, shape, and number. The format of this subtest is similar to the Wisconsin
Card Sorting Test (WCST; Heaton, 1981) and taps attention, concept formation,
cognitive flexibility, and working memory. The child is given 54 cards on which
are printed one to three symbolsa circle, square, or rhombus, in pink, yellow, and
blue. The childs task is to place the cards one by one under three stimulus
cardsone pink square, two yellow rhombuses, and three blue circlesaccording
to a principle (color, shape, or number) that the child must deduce from the pattern
of the examiner responses (correct or incorrect) to the childs placement of the
cards. The test begins with color as the basis for sorting, followed by form, and
ending with number. The test continues until the child has made three runs of 10
correct placements, or if the child has placed all 54 cards. The score corresponds to
the number of successful runs (categories). Maximum score is 3.
8. Pyramid of Mexico: This test requires identification and organization of
steps to achieve a goal. The goal is to arrange, with the fewest moves, three blocks
with different colors (green, white, and red) and sizes (large, medium, and small),
according to a desired construction model. Eight different desired constructions
are presented one by one on color cards. The child has to construct the design with
the minimum number of moves. The score corresponds to the number of items
constructed with the minimum number of moves. The maximum score of suc-
cesses is 11. This test is similar in principle to the Tower of London test (Shallice,
1982), and its purpose is to measure problem-solving ability.
For validity purposes, the Spanish Escala de Inteligencia Wechsler Para Nios
Revisada (WISCR; Wechsler, 1974) was also administered to 30 of the partici-
pants to calculate the validity of the ENI subtests, Matrices and Similarities. Corre-
lations were obtained between these tests and two subtests (Block Design and Sim-
ilarities) from the WISCR. The Pearson correlation (two-tailed) of ENI
Similarities with WISCR Similarities subtest was 0.67 (p = .0001). The scores of
the ENI Matrices were correlated, using Pearson correlation, to the WISCR
Block Design subtest (r = .52; p = .001). The Block Design was used because it is
one of the best predictors of abstract spatial abilities.
Procedure
The purpose of this research was initially explained to the school principals. After
accepting to participate, the teachers from the different grades were approached.
Teachers were informed in detail about the different points of the testing, such as
purpose, confidentiality, parents consent, time to complete the testing, and so
forth. Class lists were reviewed with the teachers. Children were randomly se-
lected from several classrooms, using these lists provided by the schoolteachers.
Their parents were contacted and interviewed. Children with no history of school
failure and no history of neurological or psychiatric disorders were chosen and in-
dividually tested at their schools or homes in two sessions of about 90 min each.
The whole battery ENI was administered to each child, but only the executive
function tests were analyzed in this article. Each child received a small gift (e.g., a
box of crayons, candy) for participation in the study.
Statistical Analyses
A multivariate analysis of variance (MANOVA) was performed using age (56-,
78-, 910-, 1112-, and 1314-year-old groups), sex (boys, girls), and school
type (public, private) as independent variables and executive function test scores as
dependent variables. The parents educational level was used as a covariate. Main
effects and interactions were analyzed. The overall MANOVA (Hotelling T2) was
calculated and univariate analyses were obtained for each of the eight dependent
measures. Partial eta squares (2) were used to estimate the effect sizes measure for
the multivariate and univariate Fs. Post hoc analysis was done using Tukey hon-
estly significant difference test to compare mean differences within groups.
Pearson correlation coefficients were obtained to determine the associations be-
tween the parents level of education and the different executive test scores. In ad-
dition, to further evaluate the influence of environmental factors on those tests that
significantly correlated with the parental education, a hierarchical regression anal-
ysis was used, entering school and parental education as predictors for each test
and controlling for age.
RESULTS
MANOVAs were initially used to analyze the main effects and interactions of the
three independent variables over all dependent measures (test scores) using parent
level of education as a covariate. The main effects of age, sex, and type of school
were statistically significant, F(4, 601) = 22.25, p < .0001, 2 = 0.44; F(1, 601) =
3.31, p < .001, 2 = 0.10, and F(1, 601) = 3.08, p < .01, 2 = 0.10, respectively.
These main effects were found when all the dependent variables were analyzed to-
gether in the multivariate model. Also, the parents level of education was a signifi-
cant covariate, F(1, 601) = 4.31, p < .0001, 2 = 0.13. A significant interaction was
found between age group and type of school, F(4, 601) = 2.17, p = .001, 2 = 0.07.
The interactions between sex and age, F(4, 601) = 0.68, p = .926, 2 = 0.02, and sex
and type of school, F(1, 601) = 1.26, p = .262, 2 = 0.04 were not significant. For
all the subtests, performance increased with age and was higher in private than in
public school children. The most important variable was age, but type of school
was also a significant variable for all the tests. No differences were observed be-
tween boys and girls. That is, type of school remained a significant factor, inde-
pendent of age or gender of the child, and after controlling for parent education.
Univariate ANOVAs were used to analyze the age, gender, and type of school
effects on each of the dependent measures. Table 2 presents the between-subject
effects for the different executive function tests. Age had a significant main effect
on all the dependent variables, and type of school had a significant main effect on
all measures except Card Sorting. Significant interactions between age and type of
TABLE 2
Analysis of Variance for Each Executive Function Test
Age Gender School Age Gender Gender School Age School
Test F 2 F 2 F 2 F 2 F 2 F 2
Verbal fluency
Semantic 50.4** .42 4.4* .01 44.3** .14 .43 .00 1.13 0.0 2.8* .04
Phonologic 168.0** .53 1.7 .00 28.7** .05 .97 .00 1.0 .00 1.34 .00
Graphic fluency
Semantic 101.7** .40 30.3** .05 55.6** .09 .91 .00 3.8* .01 .83 .00
Nonsemantic 91.5** .39 .3 .00 20.0** .03 1.1 .00 1.13 .00 2.6* .01
Similarities 118.9** .44 .1 .00 29.5** .05 .6 .00 .14 .00 1.7 .01
Matrices 64.3** .34 .2 .00 6.87** .00 .5 .00 .15 .00 1.7 .01
Card Sorting 27.1** .15 1.5 .00 0.09 .00 1.2 .00 .76 .00 3.2* .02
Mexican Pyramid 17.63** .11 2.6 .00 26.96** .04 1.3 .01 .17 .00 3.5* .03
*p < .05. **p < .01.
547
school were found for Semantic Verbal Fluency, Nonsemantic Graphic Fluency,
Card Sorting, and Mexican Pyramid tests. Differences between private and public
school were highest in the youngest group for Semantic Verbal Fluency and Mexi-
can Pyramid (see Figures 14). The scores in the Semantic Verbal Fluency test of
the youngest group of children from private schools was about 50% higher than the
scores for the youngest public school children F(1, 106) = 11.96, p = .001. In the
oldest group, this difference was only about 10%, F(1, 116) = 0.40; p = .52. The di-
rection of the interaction was similar for the Mexican Pyramid. In this test, the
youngest group from private school scored significantly higher than the same age
group from the public school, F(1, 106) = 10.46, p = .002, whereas in the oldest
group this difference disappeared, F(1, 116) = 0.07, p = .78. The direction of the in-
teraction for the Nonsemantic Graphic Fluency test scores was in the opposite di-
rection; the difference in this test between the youngest public and private school
children is small, F(1, 106) = 0.82, p = .36, but it is large between the public and
private school groups in the oldest sample, F(1, 116) = 7.31, p = .008.
The main effect of gender was observed in Semantic Verbal Fluency and Se-
mantic Graphic Fluency tests. In the former variable, boys outperformed girls; the
opposite pattern was found in the latter variable (see Table 2). A significant inter-
action between gender and school type for Semantic Graphic Fluency was found.
Girls superiority was observed mainly in the private school group. Interactions be-
tween age, gender, and type of school were not significant for any of the dependent
measures.
Considering that type of school (public or private) is significantly associated
with the parents educational level, we decided to look at the correlations between
the childrens tests scores and their parents years of education. We calculated the
correlations between parents educational level and childrens test scores. Only the
youngest and oldest children were select in this analysis to minimize and maxi-
mize the potential school effect (zero years of school vs. 7 years of school; Table
3). Six significant correlations were found between the parents education and test
scores. The highest correlations in the younger group were found with Verbal Flu-
ency. The highest correlations in the oldest group were found with Matrices and
Similarities.
The hierarchical regression analysis showed in Table 4 corroborates the predic-
tive value of parental education and the type of school on the fluency tests and sim-
ilarities. After 38% of the variance attributable to the childs age was removed, the
parents education and type of school accounted for an additional 12% of the Se-
mantic Verbal Fluency score variance. The variance associated to the parents edu-
cation after controlling for age on Phonemic Verbal Fluency, Semantic Graphic
Fluency, and Nonsemantic Graphic Fluency was 3%, 7%, 3%, and 5 %, respec-
tively. The parents education better predicts Semantic Verbal and Graphic Flu-
ency, Nonsemantic Graphic Fluency, and Similarities test scores than the type of
school.
FIGURE 1 Test scores of children from public and private schools in different age groups.
Note. 1 = 56 years; 2 = 78 years; 3 = 910 years; 4 = 1112 years; 5 = 1314 years.
549
550
551
552
TABLE 3
Correlations Between Parents Years of Education and Scores in the
Different Executive Functioning Tests in the 5- to 6-Year-Old Children
Group and in the 13- to 14-Year-Old Group
5- to 6-Year-Olds 13- to 14-Year-Olds

Test Parents Parents
Semantic Verbal Fluency 0.46* 0.30*

Phonemic Verbal Fluency 0.33* 0.18*
Semantic Graphic Fluency 0.30* 0.30*
Nonsemantic Graphic Fluency 0.31* 0.30*
Similarities 0.27* 0.34*
Matrices 0.11 0.30*
Card Sorting 0.20* 0.03
Mexican Pyramid 0.13 0.16
*p < .05 (2-tailed).
DISCUSSION
In accordance with our predictions based on previous research there was a signifi-
cant association between age and performance on executive function tests for both
boys and girls. The most important finding in this study, however, was the robust
difference observed between children in private and public schools on executive
function test scores. This effect showed a significant interaction with age for most
measures. Differences between public and private school children were robust for
younger children but decreased with age and schooling for some tests (e.g., Verbal
Fluency tests). Hence, the score difference between the public and the private
school in these measures represented indeed not a school, but a previous-to-school
effect. For the Nonsemantic Graphic Fluency, the differences between children
from public and private school increased with age. This finding suggests a more
school-dependent effect for graphic nonverbal measures than for verbal fluency
measures. The differences between the public and private school children in Pho-
nemic Verbal Fluency, Semantic Graphic Fluency, and Matrices tests scores ob-
served in the younger group remained similar across age groups.
This study found a significant correlation of the parents education level with
most of the measures of childrens functioning. Moreover, the school-type variable
covaried with the parents education, and in many cases parental education better
predicts the executive function test performance than the type of school. These re-
sults suggest that the differences in test scores between the public and private
school children depended on some conditions existing outside the school, such the
parents level of education. Researchers (see Hoff et al., 2002; Hoff, 2003a for a re-
view) have shown a strong association between the parents SES, types of lan-
guage used, and ways of interactions, such as more reading behavior. In addition, a
TABLE 4
Hierarchical Regression Analysis Predicting the Variance in a Childs
Executive Test Scores Attributable to Parental Education and Type of
School, Removing the Effects of Age
Test B SE B p R2 R2 Change
Semantic Verbal Fluency
Step 1 .38
Age 2.39 .18 .62 .001
Step 2 .50 .12
Age 2.46 .16 .64 .001
Parents education .33 .07 .24 .001
Type of school 1.56 .59 .13 .01
Phonemic Verbal Fluency
Step 1 .50
Age 2.34 .09 .70 .001
Step 2 .53 .03
Age 2.38 .09 .72 .001
Type of school 1.03 .32 .11 .01
Semantic Graphic Fluency
Step 1 .36
Age 3.02 .17 .59 .001
Step 2 .42 .07
Age 3.12 .16 .61 .001
Type of school 2.62 .55 .19 .001
Nonsemantic Graphic Fluency
Step 1 .36
Age 2.74 .15 .60 .001
Step 2 .39 .03
Age 2.82 .15 .62 .001
Type of school 0.81 .51 .06 ns
Similarities
Step 1 .43
Age 1.68 .08 .66 .001
Step 2 .49 .06
Age 1.72 .07 .68 .001
Type of school 1.46 .26 .06 ns
whole array of complex and uncontrolled environmental conditions is associated

with attending a private or public school (Aldana, Amzquita, & Becerra, 1983;
Reimers, 2002). Attending a private or public school may have different associa-
tions in Mexico, in the United States, or in any other country. Parents may select a
private school because they suppose it is either academically stronger, there are so-
cial class considerations, or simply because of family traditions or distance from

home. Children in private school do not receive any scholarships.
The parents educational level is not an isolated variable. It reflects many asso-
ciated living conditions (Baydar, Brooks-Gunn, & Furstenberg, 1993). In this
study there was a significant association between the parent education and all flu-
ency tests. The fact that this effect was found, suggests that the influence of the par-
ents educational level is a strong factor contributing to the differences observed
between the private and public school in childrens test performance. The influ-
ence of the mothers educational level over the childs literacy at an early age was
described by Baydar et al. (1993).
Parents education was significantly different between the two school groups.
The mean educational level of the public school childrens parents was about 11.04
(SD = 4.11), roughly corresponding to a high school level (in Mexico and Colom-
bia the equivalent to the U.S. high school is completed after 11 years of education).
In the private school group the childrens parents educational level was 15.60 (SD
= 2.91), roughly corresponding to a college level of education.1 Dispersions within
groups were not particularly high (standard deviations were about 3 years), and
hence both groups were relatively homogenous. However, a few parents from pub-
lic schools had a college level of education.
We cannot assume that the lower education (public school) group was a so-
cially deprived group, or that it was a group presenting nutritional or serious
health problems. The public school parents were not living in significant poverty
or social marginality, but were lower middle-class individuals. The average educa-
tion of our childrens parents in the public school sample was above the average
education for the Mexican adult population, M = 7.6 years of schooling (Instituto
Nacional de Estadstica Geograa e Informticia, 2000). We also cannot conclude
that the quality of education offered by private or public schools accounted for the
differences.
A crucial question is obvious: Why did an additional five or more years of ed-
ucation in the childrens parents make a significant contribution to the childrens
executive function test performance? Noteworthy, this educational difference
corresponds to whether or not university was attended. The decision of continu-
ing university studies depends on a whole array of variables, including financial
resources, system of values (what is and what is not important in life), personal
interests, previous academic performance, family tradition, intellectual level, and
ability to delay the reward (i.e., those children beginning to work immediately
1In the Mexican educational system, as well as in other Latin American systems derived from the
Spanish educational system, there is no difference between college and university. After high school,
students go to university. Usually, after some five university years they obtain a professional degree
(e.g., psychologist).
after high school will receive money sooner than those attending college; Ayala
Rubio, 2001).
It may be conjectured that, once finishing high school, students divide into two
major groups: those more academically oriented (pursuing a higher educational
level) and those less academically oriented (not pursuing a higher educational
level). These two groups, of course, are the extremes of a continuum. Those indi-
viduals pursuing college education likely have different values and attitudes (i.e.,
different cosmovision) than those who are excluded from university education.
This difference may be reflected in the environment parents provide to their chil-
dren, the types of activities they perform with the child, and the values they trans-
mit to their offspring. Furthermore, the parents of public and private school chil-
dren, who either attended college or not, may be different in their value systems,
their personal interests, and in their intellectual ability.
Based on previous research we could expect that those children attending private
schools (generally, offspring of parents with a college education) had a more
school-oriented home environment and more intellectual environmental conditions
long before entering school. Teachman (1987) found that parents use resources to
create a home environment conductive to higher attainment in education.
In this study, differences in vocabulary, measured with the Semantic Verbal Flu-
ency test, were evident when beginning school; children entering private schools
scored about 50% higher than those children entering public schools. This ex-
tended vocabulary most likely came mainly, but not only, from the parents.
Children entering private schools also had a richer awareness of verbal issues. For
instance, scores in the Phonemic Verbal Fluency testa test requiring phonologi-
cal awarenesswas about three times higher in children when starting in a private
than in a public school. Very likely, children entering a private school already had
some prereading training, and some of them had early reading ability. The parents
and their previous living conditions likely provided this prereading ability.
We were surprised to find how alike the educational level of the mothers and fa-
thers was for both groups. The mean difference in the educational level between fa-
thers and mothers was less than 1 year of education. We may, in consequence assume
a home effect, not a father or mother effect in the differences that were found.
Another point deserves consideration. The parents educational level was asso-
ciated especially with the scores in verbal tests. For some nonverbal tests (e.g., Ma-
trices and Card Sorting) no significant association was found. Interestingly, Ma-
trices has been regarded as a nonverbal abstracting ability test (e.g., Anastasi,
1988). The Card Sorting represents a simpler version of the WCST. The WCST is a
widely used test in neuropsychology to assess abstract behavior and executive dis-
orders (Lezak, 2004). This study suggests a stronger parents educational effect for
verbal executive function tests than for nonverbal executive function tests.
In summary, this study revealed differences in executive function test perfor-
mance between private and public school children. The results suggest that these
differences are not due to the school characteristics, per se, but to some conditions
potentially associated with attending a private or public school, such as the parents
educational level. The parents educational level may be associated with home en-
vironmental circumstances that establish early skills when solving problems or
performing executive function tasks. We suggest that parents who attend college
may have a different value system (cosmovision) than those who never received a
college education. This different value system may be correlated with a more intel-
lectually stimulating environment for their children that ultimately will result in
their higher performance in some executive functioning tests. It is likely that the in-
fluence of the parents level of education is also reflected in other neuropsycho-
logical tests, particularly verbal tasks. More research is needed to analyze the in-
fluence of home environmental conditions on different cognitive tests.
The value of this study may be twofold. First, there is a very little research on
the influence of environmental variables on the development of executive func-
tions. This study may help to understand variables that contribute to the variability
found in executive functioning test scores among children of different ages. Sec-
ond, most of the studies on the development of executive functions have focused
on children younger than 6 years of age (e.g., Anderson et al., 2001). Current study
analyzed the development of executive functions from ages 5 to 14. Results may
provide important developmental information within a broader age range.
Some limitations to this study should be noted. First, only three different sam-
ples were taken (two from Mexico and one from Colombia), limiting general-
izability of current results. We did not deal with SES, but simply educational
level. We assume, however, that educational level may be as informative about in-
tellectual functioning as the SES. Furthermore, it may be easier to operationalize.
SES, in addition to school attendance, also includes other elements, such as in-
come, that are not directly related to intellectual functioning. The family income
does not necessarily correlate with level of education.
Other limitations to the generalization of the results of this study are related to
the validity of the executive function tests. Some of these tests (e.g., Mexican Pyra-
mid) are unpublished and have not been validated. In this study, however, we ad-
dressed this limitation by establishing validity coefficients with well-known mea-
sures for three of the executive function tests. Furthermore, the question, Do
current results apply to executive functions in particular, or to neuropsychological
functioning in general? remains unanswered. We used just a few tests supposedly
measuring executive functioning. But, of course, they are not simply evaluating ex-
ecutive functioning, but are also evaluating other domains, such as vocabulary,
visuoconstructive abilities, and so forth. Further research in obviously required.
Finally, it could be argued that not all the administered tests are completely
evaluating executive functions. As a matter of fact, it is difficult to separate intel-
lectual functions in general from executive functions. Executive functions include
self-regulating, stimulus control, and purposeful rational thought, thus affecting
test performance. Many of the executive tasks attempt to tap into the individuals
ability to formulate strategies, to separate essential from nonessential, and adapt to
conditions that become novel. It has been emphasized that there is no one single
measure of executive functioning, but rather different tests represent different ex-
ecutive components, and as such, more than one measure should be used in a com-
prehensive assessment (Bamdad, Ryan, & Warden, 2003).
ACKNOWLEDGMENT
Our sincere gratitude to Dr Erika Hoff for her most significant suggestions and in-
valuable help when writing this article.
REFERENCES
Aldana, E., Amzquita, E., & Becerra, C. (1983). Trabajo de los nios matriculados en las escuelas
publicas de Bogot [Public school children work in Bogota]. Bogot, Colombia: Instituto SER de
investigacin.
Anastasi, A. (1988). Psychological testing (6th ed.). New York: Macmillan.
Anderson, V. (2001). Assessing executive functions in children: Biological, psychological, and devel-
opmental considerations. Pediatric Rehabilitation, 4, 119136.
Anderson, V., Lajoie, G., & Bell, R. (1996). Neuropsychological assessment of the school-age child.
Melbourne, Australia: University of Melbourne, Department of Psychology.
Anderson, V. A., Anderson, P., Northam, E., Jacobs, R., & Catroppa, C. (2001). Development of execu-
tive functions through late childhood and adolescence in an Australian sample. Developmental
Ardila, A. (1995). Directions of research in cross-cultural neuropsychology. Journal of Clinical and
Experimental Neuropsychology, 17, 143150.
Ardila, A., & Rosselli, M. (1994). Development of language, memory and visuospatial abilities in 5- to
12-year-old children using a neuropsychological battery. Developmental Neuropsychology, 10,
97120.
Ayala Rubio, S. (2001). La desercin escolar: un callejn sin salida? [School desertion: A dead end].
In E. Matute & M. Romo-Beltran (Eds.), Los retos en la educacion del siglo XXI [The challenges of
the education of the twenty first century]. Guadalajara, Mxico: University of Guadalajara Editors.
Bamdad, M. J., Ryan, L. M., & Warden, D. L. (2003). Functional assessment of executive abilities fol-
lowing traumatic brain injury. Brain Injury, 17, 10111020.
Baydar, N., Brooks-Gunn, J., & Furstrnberg, F. F. (1993). Early warning signs of functional illiteracy:
Predictors in childhood and adolescence. Child Development, 64, 815829.
Bradley, R. H., & Corwyn, R. F. (2002). Socioeconomic status and child development. Annual Review
of Psychology, 53, 371399.
Espy, K. A., Kaufmann, P. M., & Glisky, M. L. (2001). New procedures to assess executive functions in
preschool children. Clinical Neuropsychology, 15, 4658.
Fletcher-Janzen, E., Strickland, T. L., & Reynolds, C. R. (Eds.). (2000). The handbook of cross-cultural
neuropsychology. New York: Plenum.
Ganzach, Y. (2000). Parents education, cognitive ability, educational expectations and educational at-
tainment: Interactive effects. British Journal of Educational Psychology, 70, 419441.
Heaton, R. (1981). Wisconsin Card Sorting Test: Manual. Lutz, FL: Psychological Assessment Re-
sources.
Hoff, E. (2003a). Causes and consequences of SES-related differences in parent-to-child speech. In M.
H. Bornstein & R. H. Bradley (Eds.), Socioecononomic status, parenting, and child development (pp.
147160). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
Hoff, E. (2003b). The specificity of environmental influence: Socioeconomic status affects early devel-
opment via maternal speech. Child Development, 74, 13681378.
Hoff, E., Laursen, B., & Tardif, T. (2002). Socioeconomic status and parenting. In M. H. Bornstein
(Ed.), Handbook of parenting (2nd ed., pp. 231252). Mahwah, NJ: Lawrence Erlbaum Associates,
Inc.
Hoff-Ginsberg, E. (1991). Mother-child conversations in different social classes and communicative
settings. Child Development, 62, 782796.
Instituto Nacional de Estadstica Geograa e Informticia. (2000). XII Censo general de poblacin y
vivienda [XII General census of population and housing]. Mexico D.F., Mexico: Author.
Klenberg, L., Korkman, M., & Lahti-Nuuttila, P. (2001). Differential development of attention and ex-
ecutive functions in 3- to 12-year-old Finnish children. Developmental Neuropsychology, 20,
407428.
Kohen, D. E., Gunn, J. B., Leventhal, T., & Hertzman, C. (2002). Neighborhood income and physical
and social disorder in Canada: Associations with young childrens competencies. Child Develop-
ment, 73, 18441860.
Lee, G. P., Loring, D. W., Newell, J., & McCloskey, L. (1994). Figural fluency on the Five Point Test:
Preliminary normative and validity data. International Neuropsychological Society Program and
Abstract, 1, 51.
Levin, H. S., Culhane, K., Hartmann, J., Evankovich, K., Mattson, A. J., Harward, H., et al. (1991). De-
velopmental changes in performance on test of purported frontal lobe functioning. Developmental
Lezak, M. D. (1983). Neuropsychological assessment. New York: Oxford University Press.
Lezak, M. D. (2004). Neuropsychological assessment (4th ed.). New York: Oxford University Press.
Lichtenstein, P., & Pedersen, N. L. (1997). Does genetic variance for cognitive abilities account for ge-
netic variance in educational achievement and occupational status? A study of twins reared apart and
twins reared together. Sociological Biology, 44, 7790.
Matute, E., Montiel, T., Rosselli, M., & Ardila, A. (2003). Writing, spelling and phonemic skills across
ages in Spanish speaking children. Journal of the International Neuropsychological Society, 29,
220221.
Matute, E., Rosselli, M., Ardila, A., & Morales, G. (2004). Verbal and nonverbal fluency in Spanish
speaking children. Developmental Neuropsychology, 26, 647660.
Matute, E., Rosselli, M., Ardila, A., & Ostrosky, F. (in press). ENI: Evaluacin neuropsicolgica
infantil [ENI: Child neuropsychological assessment]. Mexico: Universidad de Guadalajara.
Petrill, S. A., & Deater-Deckard, K. (2004). Task orientation, parental warmth and SES account for a
significant proportion of the shared environmental variance in general cognitive ability in early
childhood: Evidence from a twin study. Developmental Sciences, 7, 2532.
Portes P. R., Cuentas, T. E., & Zady, M. (2000). Cognitive socialization across ethnocultural contexts:
Literacy and cultural differences in intellectual performance and parent-child interaction. Journal of
Genetic Psychology, 161, 7998.
Raven, J. C., Court, J. H., & Raven, J. (1976). Manual for Ravens progressive matrices. London: Lewis.
Regard, M., Strauss, E., & Knapp, P. (1982). Childrens production on verbal and non-verbal fluency
tasks. Perceptual and Motor Skills, 55, 839844.
Reimers, F. (2002). Distintas escuelas, diferentes oportunidades: Los retos para la igualdad de
oportunidades en Latinoamrica [Different schools, different opportunities: The challenge for equal
opportunities in Latinoamrica]. Madrid, Espaa: Muralla.
Rosselli, M., & Ardila, A. (2003). The impact of culture and education on nonverbal neuropsycho-
logical measurements: A critical review. Brain and Cognition, 52, 326333.
Rosselli, M., Ardila, A., Bateman, R., & Guzman, M. (2001). Neuropsychological test scores, aca-
demic performance, and developmental disorders in Spanish-speaking children. Developmental
Rosselli, M, Matute, E., Ardila, A., & Montiel, T. (2003). The effect of age over childrens perceptual
skills. Journal of the International Neuropsychological Society, 29, 295296.
Senn, T. E., Espy, K. A., & Kaufmann, P. M. (2004). Using path analysis to understand executive func-
tion organization in preschool children. Developmental Neuropsycholology, 26, 445464.
Shallice, T. (1982). Specific impairments of planning. Philosophical Transactions of the Royal Society
of London, 298, 199209.
Sigman, M., Neumann C., Jansen, A. A., & Bwibo, N. (1989). Cognitive abilities of Kenyan children in
relation to nutrition, family characteristics, and education. Child Development, 60, 14631474.
Soprano A. M. (2003). Evaluation of executive functions in children. Review of Neurology, 37, 4450.
Spreen, O., & Strauss E. (1998). A compendium of neuropsychological tests (2nd ed.). New York: Ox-
ford University Press.
Teachman, J. D. (1987). Family background, educational resources and educational attainment. Ameri-
can Sociological Review, 52, 548557.
Turkheimer, E., Haley, A., Waldron, M., DOnofrio, B., & Gottesman, I. I. (2003). Socioeconomic sta-
tus modifies heritability of IQ in young children. Psychological Science, 14, 623628.
Wechsler, D. (1974). Escala de Inteligencia de Wechsler para Nios Revisada. Madrid, Spain: TEA
Editores.
Welsh, M. C., & Pennington, B. F. (1988). Assessing frontal lobe functioning in children: Views from
developmental psychology. Developmental Neuropsychology, 4, 199236.
White, K. R. (1982). The relation between socioeconomic status and academic achievement. Psycho-
logical Bulletin, 91, 461481.
Intent. J. Nrwosuence. Vol 89. pp. 189-205 D 1997 OPA (Overseas Publichers Association)
Reprints avilable directly from the publisher Amsterdam B. V. Published in The Netherlands under license by
Photocopying permitted by license only Gordon and Breach Science Publishers
Printed in Malaysia
Int J Neurosci Downloaded from informahealthcare.com by Florida International University, Medical Library on 03/14/11
SPECIFIC LEARNING DISABILITIES:

A NEUROPSYCHOLOGICAL PERSPECTIVE
ALFRED0 ARDILA
Instituto Colombiano de Neuropsicologia. Bogota, Colombia
(Received 16 September 1996)
A dispersion in cognitive abilities is expected in normal populations. Specific learning disabilities

would represent an extreme polarity in a continuum of normal cognitive dispersion. Three proposi-
For personal use only.
tions relative to learning disabilities are advanced in the present paper. First, specific learning disabil-
ities are expected to be found for diverse cognitive functions, even though some of these specific
learning disorders have yet to be described in scientific literature. Second, it is noted that specific fac-
tors can be affected in cases of learning disabilities. Lastly, a parallel between focal neuropsycholog-
ical syndromes and specific learning disabilities is proposed. Developmental learning difficulties
would represent dysfunctional or dysmaturational defects; whereas neuropsychological syndromes
represent a consequence of acquired structural brain pathology.
Keywords: Learning disabilities; dyslexia; individual differences; neuropsychological testing; brain

dysmaturation
Specific learning disabilities, despite extensive research, continue to be a polemic

area of study, generating controversy and confusion among the scientific commu-
nity. Disagreement among researchers is evident even in the naming of learning
disabilities; these disorders are variously referred to as specific learning disabil-
ities, developmental disabilities, learning disorders, developmental disor-
ders, or even minimal brain dysfunctions (Kinsbourne, 1985; Lubar &
Deering, 1981; Morrison & Siegel, 1991).
In order to approach these learning defects, it is important to depart from the
premise that cognitive abilities do not constitute a homogeneous phenomenon
within a particular individual. Rather, a dispersion of cognitive abilities is expected
Correspondence to Alfred0 Ardila. Tel.: (305) 593-1223. Fax (305) 592-7930.

Present address: Miami Institute of Psychology, 8180 NW 36 Street, Miami, FL 33166.
My most sincere gratitude to Dr. D.F. Benson for his most valuable suggestions while preparing this
paper, and to Dr. G . Rodriguez, for his support in editing and correcting this manuscript.
189
190 A. ARDILA
(intrasubject variability). Hence, some degree of intrasubject dispersion in cognitive

abilities is not the exception to, but rather the rule (Anastasi. 1982; Cronbach, 1990).
When clinical assessment reveals a discrepancy between normal levels of intellec-
tual functioning and significant subaverage performance in a given cognitive
domain. a specific learning disability should be assumed. Specific learning disabil-
ities, therefore, refer to selective deficits for some particular type of learning.
In neuropsychology a relatively modest interest in the analysis of individual
differences has existed (e.g.. Vernon, 1994). In a certain way, it is usually
assumed that some basic cognitive abilities should be observed in any individual
with a normal brain: Any normal person (1.e.. any individual without evident brain
damage) should recognize faces, discriminate language sounds. follow a route,
possess a basic repertoire of learned movements. be able to copy a complex fig-
ure. etc. As a matter of fact, most neuropsychological tests infer a virtually perfect
performance in normal subjects and, from a psychometric perspective, they have
a quite low ceiling.
However. in clinical application. variability in the performance of any neu-
ropsychological test is to be expected (Lezak. 1995). If, for instance, the Rey-
Osterrieth Complex Figure (Osterrieth, 1944). or the Boston Naming Test

(Goodglass. Kaplan & Weintraub. 1983) were administered to a normal popula-
tion sample. by statistical definition some 2 . 5 9 of the subjects would obtain a
score two standard deviations below the mean. Moreover, less than O S % (one out
of 300) of the subjects would score three standard deviations below the mean.
These subjects from the vantage point of neuropsychological measurement would
present an abnormal performance with a probability lr\>rlbrlou. .01. Abnormal
in neuropsychology is usually regarded as brain damage or dysfunction. These
subjects. inherently, do not present with structural brain damage. If these two tests
(the Rey-Osterrieth Complex Figure and the Boston Naming Test) were both
given. and assuming that the instruments represent noncorrelated tests. the num-
ber of abnormal subjects would increase twofold (one out of 100). If five non-
correlated tests were administered (assuming five independent factors) the
number of abnormal subjects would be spuriously increased (about one out of
40). Conversely, from a statistical perspective, (assuming a normal distribution)
an equivalent number of subjects with superior abilities two-three standard devi-
ations above the mean) would be expected to occur in these same cognitive
domains. Given. however. that neuropsychological tests have very low ceilings,
they are not appropriate instruments to detect superior levels of cognitive ability.
In brief. there are important individual differences in the performance of neu-
ropsychological tests developed to investigate brain integrity. It is expected that
some percentage of the general population will obtain scores that, from a statisti-
cal perspective, should be considered pathological (Nadler, Mittenberg,
SPECIFIC LEARNING DISABILITIES 191
DePiano & Schneider, 1994). These individual differences should be taken into
consideration in any attempt to understand brain organization of cognition in gen-
eral and learning disabilities in particular.
DYSLEXIA
Dyslexia represents the prototype of a specific learning disability. Dyslexia has

been defined by the World Federation of Neurology as a disorder manifested by
difficulty in learning to read despite conventional instruction, adequate intelli-
gence, and sociocultural opportunity. It is dependent upon fundamental cognitive
disabilities which are frequently of constitutional origin (cf. Critchley. 1985).
This is basically the same definition used for every type of specific learning dis-
ability. There are several points in the definition that merit particular considera-
tion: (1) the difficulty is specific, and general intelligence is normal; otherwise, it
would represent a nonspecific or global intellectual impairment (mental retarda-
tion); (2) it represents a cognitive defect, not a behavioral disorder; and (3) its ori-
gin can be constitutional, that is, it can appear in family groups; or simply, some
genetic factors can be involved (Lubs et al., 1994; Olson, Wise, Conners & Rack,
1990; Vogler, DeFries & Decker, 1985).
Dyslexia can be associated with some soft neurological signs, such as synki-
nesias. difficulties to recognize fingers, right-left confusions, defects in visual
scanning, etc. (Critchley, 1985; Pirozzolo, 1979). Furthermore, children with
dyslexia exhibit difficulties with sequential series learning (letters. numbers,
months, etc.). difficulties learning to read analog clocks, impairments in the use of
spatial relations (e.g., up-down), and what has been named as the developmental
Gerstmann syndrome (dysgraphia, dyscalculia, right-left recognition defects.
and finger recognition deficits) (Benson & Geschwind, 1970; Biscaldi, Fischer &
Aiple, 1994; Brachacki, Fawcett & Nicolson, 1994; Eden, Stein, Wood & Wood,
1994; Evans. Drasom & Richards, 1994; Lubar & Deering, 1981; Critchley, 1985;
Singh, Broota & Gupta, 1993; Toth & Siegel, 1994) (Table I). In brief, dyslexia is
much more than failure in learning to read.
TABLE I Some defects frequently associated with dyslexia
Synkinesias
Difficulties to recognize fingers
Right-left confusions
Disorders in eye-movements
Defective serialisation
Difficulties in learning to read the clock
Impairments in the use of spatial relations (e.g.. up-down)
Developmental Gerstmann syndrome
192 A. ARDILA
It is usually accepted that dyslexia does not represent a homogeneous disorder,

and several subtypes of dyslexia can be distinguished (Bakker, 1979; Boder,
1973; Bravo-Valdivieso, 1980; Masutto, Bravar & Fabbro. 1994; Mattis, French
& Rapin. 1975; Rourke, Bakker, Fisk & Strang, 1983; Rispens, Van der Stege &
Bode, 1994) (Table 11). Furthermore, dyslexia can depend of the idiosyncrasies of
a linguistic writing system (Bravo-Valdivieso, 1988; Wimmer. 1993). Hence,

dyslexia in Spanish is not completely equivalent to dyslexia in English or
Chinese. Cognitive abilities required to read in Spanish, English, and Chinese do
not seem to have perfect correspondence.
Despite the different nosologic classifications that have been proposed, it seems
evident that at least two distinct types of dyslexia can be identified: (1) dyslexia
resulting from difficulties in auditory processing and in matching graphic sym-
bols with language sounds; and ( 2 ) dyslexia resulting from deficits in the visuop-
erceptual processing, in the orderly ability to scan the written text, and to obtain a
simultaneous recognition of those graphemes included in a word. Additionally,
reading difficulties could also eventually be associated with some retardation in
language acquisition, and difficulties in writing may be correlated with deficits in
graphomotor abilities.
TABLE I1 Some classifications of dyslexia
1. P-type dyslexia
2. L-type dyslexia
(Bakker. 1979)
1. Dysphonetic dyslexia
2. Dyseidetic dyslexia
(Boder. 1973)
1. Developmental phonological dyslexia
2. Developmental surface dyslexia
(Ellis. 1993)
1. Audiophonic dyslexia
-7. Visuospatial dyslexia
(Johnson & Myklebust, I97 I )
1. Perceptual deficits
2. Defects in articulatory and graphomotor abilities
3. Dyslexia associated with language problerns
1. Difficulties in temporal sequencing
(Mattis et al.. 1975)
1. Auditory-linguistic subtype
2 . Visual-spatial subtype
(Pirozzolo, 1979)
I . Central auditory processing dyslexia
2. Visual-perceptual dyslexia
(Ouiros. 1964,
Consequently, it seems reasonable to conclude that: (1) there are at least two dif-
ferent cognitive defects (linguistic and visuospatial) that can eventually result in an
inability or delay in normal reading skill acquisition; and (2) dyslexia is not a spe-
cific impairment in the acquisition of reading ability. Learning-to-read difficulty is
just a particular manifestation of some, more fundamental cognitive deficits which
can eventually result in difficulties with learning to read in a normal way.

Reading represents a supramodal skill, there are cognitive abilities that serve
support functions and abilities that are fundamental in reading acquisition. It does
not seem reasonable to infer that a specific cognitive ability governs the ability to
read. Reading represents a complex functional system, requiring, and based on
different cognitive abilities (factors). Therefore, it does not seem reasonable to
attribute reading to a specific ability, just as it is unreasonable to postulate that a
specific cognitive factor governs ones ability to use computers. As a matter of
fact, different authors have supported this interpretation of dyslexia: Dyslexia is
just a single manifestation of one or several more fundamental cognitive defects
(see, Kinsbourne, 1985; Pirozzolo, 1979; Stanovich, 1994).
It is interesting that a specific intellectual profile has been reported in cases of
dyslexic children (Naidoo, 1972). This profile becomes most evident when using
standard intellectual measures such as the Wechsler Intelligence Scale. According to
this profile, known as the ACID pattern, it is expected that Arithmetic, Coding,
Information and Digit Span subtests are significantly depressed in dyslexic individu-
als, whereas all other subtest scores are normal or even superior (Rugel, 1974; Kolb
& Whishaw, 1990). However, this particular profile refers to dyslexia in general, and
does not distinguish between dyslexia subtypes. If dyslexia subtypes were separated,
there is greater likelihood that intellectual profiles would become even more specific.
NONVERBAL LEARNING DISABILITIES
A nonclassical learning disability could be taken as a second example. Johnson

and Myklebust (1971) identified a group of children who performed poorly in
arithmetic and also were impaired in interpreting social signals, pretending and
making abstractions. They labeled these disorders as nonverbal learning disabili-
ties, and emphasized the impaired social perception in these children. Later stud-
ies supported this proposal (e.g., Brumbach, Staton & Wilson, 1983; Denckla,
1983; Loveland, Fletcher & Baile, 1990; Rourke, 1989; Rourke & Finlayson,
1978; Voeller, 1986, 1991; Tranel et a]., 1987). Weintraub and Mesulam (1983)
analyzed this nonverbal learning syndrome that begins early in life in a 14-patient
sample and is characterized by emotional and interpersonal difficulties, shyness,
visuospatial disturbances, and inadequate paralinguistic communicative abilities
(e.g., inappropriate prosody).
194 A. ARDILA
Examination revealed neurologic and neuropsychological signs consistent with

right-hemisphere dysfunction. In many of these individuals, a family history of
similar symptoms was disclosed. The authors proposed that, just as developmen-
tal involvement of the left hemisphere may produce dyslexia, dysfunction of the
right hemisphere may lead to chronic emotional difficulties, disturbances in inter-
personal skills, and poor visuospatial abilities. Tranel, Hall, Olson, and Tranel
(1987) pointed to a similar spectrum of behaviors in these individuals (chronic
social isolation, shyness, withdrawal, visual spatial deficits, poor eye contact,
impaired emotional prosody and gestures); these authors also attributed the syn-
drome to right hemisphere dysfunction.
This proposal is particularly important for any theoretical interpretation of learning
disabilities: learning disabilities represent a diverse collection of syndromes and can
go far beyond the classic disturbances conventionally taught in academic institutions.
DSM-IV APPROACH TO LEARNING DISABILITIES

It is important to take into account how learning disabilities are considered in the
Diagnostic and Statisticul Maizual of Mental Disorders, fourth edition (DSM-IV)
(Table 111). DSM-IV (American Psychiatric Association, ( 1994) includes a section
designated as Learning Disorders (formerly named Disorders in Academic
TABLE 111 Classification of developmental disorders

(Disorders Usually First Diagnosed in Infancy. Childhood,
or Adolescence; American Psychiatric Association. 1994).
Only Learning Disorders, Motor Skill Disorders. and
Communication Disorders, could be considered as specific
learning disabilities.
Mental retardation
Learning Disorders
Reading disorder
Mathematic disorder
Disorder of written expression
Motor skills disorders
Developmental coordination disorder
Communication disorders
Expressive language disorder
Mixed receptive-expressive language disorder
Phonological disorder
Stuttering
Pervasive developmental disorder
Attention deficit and disruptive disorder
Feeding and earing disorders of infancy or early childhood
Tics disorders
Elimination disorders
Other disorders of infancy. childhood, or adolescence
SPECIFIC LEARNING DlSABILITIES 195
Abilities) in the chapter devoted to the Disorders Usually First Diagnosed in

Infancy, Childhood, or Adolescence. DSM-IV presents that a diagnosis of a
learning disorder should be made, when the individuals achievement on indi-
vidually administered, standardized tests in reading, mathematics, or written
expression is substantially below the expected for age, schooling, and level of
intelligence (p. 46). Substantially below is further defined as, a discrepancy of

more than two standard deviations between achievement and IQ. Three subtypes
of learning disorders are distinguished in the DSM-IV: Reading Disorder,
Mathematics Disorder, and Disorder of Written Expression. The chapter of
Disorders Usually First Diagnosed in Infancy, Childhood, or Adolescence
additionally recognizes Motor Skills Disorder (Developmental Coordination
Disorder), and a group of four Communication Disorders (Expressive Language
Disorder, Mixed Receptive-Expressive Language Disorder. Phonological
Disorder, and Stuttering).
Several observations could be mentioned:
(1) The criterion used in the DSM-IV for classifying learning disabilities rep-
resents a purely school-performance based criterion. Thus, if reading difficulties
exist, the disability as a consequence a Reading Disorder (dyslexia) is diag-

nosed; if difficulties are observed in writing, then a Disorder of Written
Expression exists; and finally, if defects are observed in calculation abilities, the
diagnosis of Mathematics Disorder is made.
(2) Because of a reliance on purely academic criteria, this classification by neces-
sity is contingent upon which learning modalities are considered as most functional in
our current educational system. For instance, constructional difficulties, musical dis-
abilities, and defects in social skill acquisition are not considered learning disorders.
(3) Finally, it is evident that academic abilities are not a good referent point for
neuropsychology. To read or to perform arithmetical operations represents com-
plex functional systems. In neuropsychology it is well known that different types
of reading and calculation disorders can be observed in case of brain pathology;
that is, there exist different subtypes of alexias and acalculias (e.g., Ardila &
Rosselli, 1990; Benson & Ardila, 1996; Ellis, 1993). The human brain is special-
ized for a particular type and level of information processing, potentially useful
for reading or performing calculations (Benson, 1994). But the brain is not per se
specialized for reading or for performing calculations (Ardila, 1993).
MORE LEARNING DISABILITIES
During recent years the idea that, not only most classical learning disabilities
(i.e.. dyslexia, dysphasia, and dyscalculia) can be distinguished, but also many
other types of learning difficulties, has tended to prevail (Rourke, 1988). The
I96 A. ARDILA
existence of a variety of specific learning difficulties have been proposed in per-

ceptual recognition (e.g., prosopodysgnosia. or simply, developmental prosopag-
nosia; McConachie, 1976). in memory abilities (specific developmental memory
difficulties or developmental dysmnesias; De Renzi & Lucchelli, 1990), etc.
Moreover. if learning disabilities were interpreted as a result of certain brain dys-
maturation or dysfunction. it could be assumed that, theoretically, as many devel-

opmental learning disabilities as basic cognitive abilities must exist. For instance,
although it has yet to be described in the scientific literature, a dysmusia or devel-
opmental amusia must theoretically exist. Furthermore, it should be possible to
find learning difficulties equivalent to every known neuropsychological syn-
drome. Or, more exactly, a parallel between learning disabilities and neuropsy-
chological syndromes should exist. Thus, developmental learning difficulties
represent dysfunctional or dysmaturation defects; whereas neuropsychological
syndromes represent a consequence of a structural brain pathology (Table IV).
TOWARD AN EXPLANATION OF LEARNING DISABILITIES

Obtaining an acceptable explanation of learning disabilities has not been an easy

task. Efforts were initially focused on the analysis of dyslexia (Fletcher et al.,
1994). Abnormalities in language lateralization and defects in handedness were
TABLE 1V Some examples of relations between focal neuropsychological syndromes and specific
learning disabilities. In both cases the same factor has been assumed to be impaired. It is pointed the
author who described the syndrome as a learning disability.
Syidronir Lecrniiny disability Foctor Author

Anarithmetia Discalculia Numerical Several
Prosopagnosia Developmental Visuoperceptua McConachie.
prosopagnosia I976
Spatial Spatial-type Visuospatial Johnson &
agnosia learning disabilities Myklebust. 1971
Amusia Dysmusia Musical Not described yet
Alexia with Linguistic Visual-
agraphia dyslexia phonological Several
integration
Alexia wlo Visuospatial Visual perception Several
agraphia dyslexia
Apraxia Developmental Praxic Ajuriaguerra &
dysprdxia Stambak. 1969
Left Not described Verbal fluency Not described
convexital Divergent Yet
prefrontal thinking
Gerstniann Dev. Gerstmann Spatial knowledge Benson &
\yndrome syndrome through language Geschwind. 1970
Amnesia Dev dismnesia Memory De Renzi &
Lucchelli, 1990
mentioned (Orton, 1937). Nonetheless, it is a self-evident assumption that a learn-

ing-disabled child must present some central nervous system dysfunction, specif-
ically in those brain areas normally involved in the dysfunctional ability
(Witelson, 1977). This assumption has led to the search of brain abnormalities
responsible for the specific difficulty in learning. Galaburda and Kemper (1979)
reported evidence of neural developmental abnormalities (neural migration

defects) in the left hemisphere language association areas, of an individual suffer-
ing dyslexia. This observation has been further corroborated (Galaburda et al.,
1985; Humphreys, Kaufmann & Galaburda, 1990).
Other brain abnormalities have also been demonstrated in dyslexic individuals
(Duane, 1992; Galaburda, 1993, 1994; Hier, Lemay, Rosenberger & Perlo, 1978;
Hynd et al., 1990; Jernigan, Hesselink, Sowell & Tallal, 1991; Leonard et al.,
1993; Schultz et al., 1994). Using magnetic resonance imaging, Leonard et al.
(1993) compared three subject groups: dyslexics, unaffected relatives, and con-
trols. They reported that all groups had left-sided asymmetry for the temporal
bank and right-sided asymmetry for the parietal bank. The group with dyslexia
had exaggerated asymmetries, owing to a significant shift of right planar tissue
from the temporal to parietal bank. They also had a higher incidence of cerebral
anomalies bilaterally. Activation levels in the thalamus, particularly in the medial
geniculate body, have been proposed to be different between normal readers and
dyslexic children (Roush, 1995).
Employing brain imaging techniques, it has also been possible to demonstrate
in dyslexic subjects some anomalous patterns of brain activation during verbal
tasks (Rumsey, Zametkin, Andreason & Hanahan, 1994). Thus, using positron
emission tomography, Rumsey et al. (1992) observed that normal readers acti-
vated left temporoparietal cortex during a phonological task, but not during a non-
phonologic attention task. Dyslexic subjects failed to activate those
temporoparietal regions activated in normals during phonological detection, but
did not differ from controls in these regions during rest or attentional tests.
Flowers, Wood, and Naylor (1991) using measures of regional cerebral blood
flow, observed that dyslexics fail to present a normal cortical activation in the left
hemisphere during verbal tasks (verbal memory, auditory perception and spelling
analysis). Hagman et al. (1992) using positron emission tomography, also demon-
strated in dyslexic adults a failure to present normal temporal activation during
verbal tasks.
Taken together, all these research studies demonstrate the existence of subtle
abnormalities (structural and functional) in dyslexic subjects brain. Theoretically,
similar abnormalities (albeit, in other brain sites), should also be found for other
learning disabilities (e.g., dysphasia, dyscalculia, visuospatial disabilities, etc.).
These abnormalities would be corroborating that, individual differences in
198 A. ARDILA
cognition are correlated with individual differences in subtle brain morphology

and activity patterns. As a consequence. a parallel between specific learning dis-
abilities and neuropsychological syndromes is expected (see Table IV).
Geschwind and Behan ( 1982). and later Geschwind and Galaburda (1987) pro-
posed an integrative theory about learning disabilities. They attempted to relate
dyslexia, left-handedness. and immune disorders. They hypothesized that an ele-

vated level of prenatal testosterone. acting independently on the embryonic thy-
mus and the embryonic brain. is the casual link underlying the proposed
association among the three conditions. The action on the thymus is responsible
for an increased risk of immune disorders. whereas the effects on the brain lead to
an increase probability of anomalous lateralization. Left-handedness and dyslexia
are thought to result from an altered left-hemisphere development. This hypothe-
sis attracted a great deal of attention and significantly stimulated research in the
area (e.g.. Bishop, 1990; Hugdahl. Synnevaag & Satz. 1987; Tonnessen, Lokken,
Hoien & Lundberg. 1993). The Geschwind-Behar-Galaburda hypothesis. how-
ever. should be regarded as a research proposal requiring further analysis and
investigation.
WHAT CAN BE LOCALIZED IN THE BRAIN?
Contemporary neuroimaging techniques have provided valuable information

about brain activity during the perfotmance of different cognitive tasks. Based on
measures of focal brain metabolism, positron emission tomography (PET) allows
one to visualize levels of brain activity and focal involvement during different
conditions. Some examples presented as illustrations.
It has been reported that during the performance of a relatively simple task,
(such as recognition of syllables) an increase in activation is observed in a very
restricted area of the left temporal lobe corresponding to the primary auditory cor-
tex (Black & Behrmann. 1994). During the passive viewing of a black and white
checkerboard, the active regions occur along the boundary of the calcarine fissure
(primary visual area) (Hirsch. 1994). In contrast. in performing a facematching
task a broader regional activation in both occipital lobes is observed, particularly
in the right occipital lobe (Chertkow & Bub. 1994). Viewing and hearing nouns,
as well as generating verbs activates the left occipital. temporal and frontal areas
respectively (Raichle, 1994). In all cases. changes in the level of activation are
limited to some rather specific brain areas. It essence, these findings advance the
hypothesis that specific brain areas are involved in these relatively simple tasks.
When a "complex" task is utilized. (e.g., reading aloud) a more complex matrix
of activated cortical areas is recorded. Different brain areas provide specific
SPECIFIC LEARNING DISABILITIES I99
contributions to performance during a reading task: occipital (visual perception),

temporal (language decoding), and Brocas frontal area (language control and
production) (Pettersen. Fox, Snyder & Raichle, 1989)). For each function, a lim-
ited cortical region is specifically activated, whereas other brain areas are only
partially active (Price et al., 1994).While speaking, activation not only occurs in
the mouth area along the primary motor cortex in the left hemisphere, but also in
the superior temporal lobe, and in the supplementary motor area.
Briefly, therefore, it could be stated that: (1) during relatively simple tasks
(e.g., listening syllables, watching a checkerboard), brain activity changes are
restricted to rather specific brain areas; and (2) during complex cognitive tasks
(e.g., reading), a participation of multiple brain areas is observed. Each of these
areas makes its particular contribution to the whole system.
Clinical observation supports the assumption that apparently different tasks
(e.g., to solve numerical problems and to recognize fingers) can be impaired as a
consequence of brain pathology originating from a common neural substrate.
Children with difficulties in learning to read, may also present other rather differ-
ent difficulties (see above). Conversely, acquired brain pathologies of different
topographic localization can impair performance in exactly the same task (e.g.,
the ability to solve numerical problems can be disrupted in cases of quite diverse
brain damage locations; Ardila & Rosselli, 1990).A certain common factor evi-
dently should be contained in those apparently different, but simultaneously
altered, tasks in the case of a single focal pathology (e.g.. to solve numerical prob-
lems and to recognize the fingers). Brocas aphasia can serve as an illustrative
example.
It is usually recognized that Broca aphasia has two different distinguishing
clinical features: (a) a motor component (lack of fluency, disintegration of the
speech kinetic melodies, verbal-articulatory defects, etc.); and (b) Agramatism
(e.g., Benson & Ardila, 1996; Luria, 1976; Goodglass, 1993; Kertesz, 1985). If
both defects are simultaneously observed (i.e., they are very highly correlated), it
simply means that both deficits are just two different manifestations of a single
underlying defect. It is not easy to understand which central defect could be the
single factor responsible for these two clinical manifestations (e.g., inability to
sequence expressive elements). However, a single common factor underlying
both defects should be assumed. Most likely, Brocas area is not specialized in
producing language per se, but rather a certain activity in the neural system that is
fundamental to the performance of skilled movements required for language pro-
duction. and also, language syntax. It is interesting to note that deaf-mute subjects
(who, as a consequence of never having produced highly skilled verbal articula-
tory movements) present with a virtually total impossibility to learn, understand,
and use language grammar (Poizner, Klima & Bellugi, 1987). Perhaps a lack of
200 A . ARDILA
normal verbal articulatory development is inherently associated with a lack of

normal grammatical development.
It has been assumed that the cerebral cortex is organized in modules, which
probably range in size from a hundred thousand to a few million neurons (Carlson,
1994). Each module receives information from other modules, perform certain
processing and then pass the information to other(s) modules (De Valois & De
Valois, 1988: Livingstone & Hubel. 1988). It has been proposed that in the pri-
mary visual cortex there are some 2.500 modules, each measuring approximately
0.5 x 0.7 mm. The neurons in each module are devoted to the analysis of various
features in one specific portion of the visual field. However, it is reasonable to
expect that several modules can simultaneously participate in each specific type of
information processing, i.e., that modules could be grouped into supramodules. or
simply, higher order modules. The activity or information processing performed
by a high order module would roughly correspond to a factor.
Higher order modules may potentially be useful for apparently different types
of abilities (e.g.. skilled movements required for language production and lan-
guage syntax). Potentially, they might be also useful for other nondeveloped or
not analyzed yet, activities. The same information processing level can be used in
different functions.
Reading can illustrate the complexity of brain organization in any psychologi-
cal process (functional system). Reading is based in certain fundamental abilities
or factors (Luria, 1966, 1973, 1976) (e.g., complex shape perception, cross-modal
learnings, etc.) already existing 5,000 years ago, and by nature, existing in illiter-
ate individuals. What might be localizable in the brain is not reading per se, but
certain basic abilities or factors (information processing levels depending on
some specific assemblies or high order modules) required to read. albeit not only
to read. Learning to read supposes the existence of the neural activation and long-
term potentiation (Levy & Steward, 1983; Perkins & Teyler, 1988) of several
higher order modules, functioning as supramodule information processing sys-
tems (visual information related with the perception of letter shapes, cross-modal
association between letters and language sounds, etc.).
Identical information processing levels, i.e., the same operation, might be use-
ful for apparently different types of cognition. As an example, it has been reported
that painting. playing chess. mechanics, and music abilities may be impaired in
cases of right hemisphere damage. Damage in those same areas in an Eskimo or
Amazonian Indian would imply an impossibility to move around the snow or the
jungle (Ardila. 1993). Brain assemblies (basic circuitry) or modules able to per-
form a specific information processing level (factor) can be potentially useful
for many different types of cognition. However, any type of complex cognition
will require the orchestration of multiple factors (multiple brain areas; that is, a
SPECIFIC LEARNING DISABILITIES 20 1
particular brain system supporting the corresponding functional system)

(Luria, 1966,1973,1976).
Focal brain pathology, or brain dysfunction in cases of specific learning dis-
abilities, would imply the abnormal activity of a specific brain area, associated
with a particular information processing level. Pathology affecting a given corti-
cal area can cause the entire functional system (e.g., reading) to fail. Hence, in
complex psychological processes the whole system can fail as a result of pathol-
ogy or dysfunction at different levels. It is not surprising then, to find that there
are different types of alexia and different types of developmental dyslexia.
SOME TENTATIVE CONCLUSIONS
It seems evident that when attempting to explain specific learning disabilities

some assumptions must be taken into consideration:
1. From a psychometric perspective, there is a dispersion in cognitive abilities,
among different individuals (intersubject variability), and also within each indi-
vidual (intrasubject variability). A normal IQ does not imply that homogeneity in

intellectual abilities exists. Learning disabilities would represent an extreme polar-
ity on a continuum of normal dispersion. Therefore, from a psychometric perspec-
tive, specific learning disabilities are not only comprehensible but also evident.
2. In our current educational system, there are certain highly valuable types of
learning. This is particularly true for language, reading, and mathematics.
Conversely this is generally not true for spatial or musical learning.
Developmental dysphasia, dyslexia and dyscalculia represent important obstacles
for an appropriate academic performance in our contemporary educational sys-
tem. Spatial and musical abilities most often are regarded as less important, or
secondary abilities. In the academic environment, it is not likely that a child will
have to leave school or be referred for professional treatment because of limited
spatial or musical abilities.
3. In dyslexia it has been demonstrated that subtle abnormalities exist in the
brain of these learning disabled individuals. It seems reasonable to assume that
functionally similar abnormalities may be found in other types of specific learn-
ing disabilities.
4. A parallel between specific learning disabilities and neuropsychological syn-
dromes may be proposed. Neuropsychological syndromes represent acquired cog-
nitive defects associated with structural brain pathology; specific learning
disabilities represent developmental dysfunctions of specific brain areas.
5. Simple psychological activities are correlated with the increased activation
of relatively limited brain areas. Complex psychological processes (functional
202 A. ARDILA
systems") require the participation of multiple brain areas ("brain systems").

Moreover, focal brain pathology or dysfunction may result in the failure of the
whole system.
References
Ajuriaguerra. J . & Stambark. M. ( 1969). Developmental dyspraxia and psychomotor disorders. In P.
J. Vinken & G. W. Bruyn (Eds.), Haridbook qf cVinical neurology, r,ol 4: Disorders of speech.
percepriori arid Tyrnholir behmYors North-Holland Amsterdam (pp. 443-464).
American Psychiatric Association. ( 1994).Dingriosfic and Storisticol Moriirnl sf Merirnl Disorders. 4a
edition. American Psychiatric Association Washington DC.
Anastasi. A. ( 1982). P.Tyrhologicril restirig. Macmillan Publishing Co. New York, 5a edition.
Ardila, A. ( 1993). Historical evolution of spatial abilities. Behmiorril Nerirolog.v. 6, 83-88.
Ardila, A. & Rosselli. M. ( 1990).Acalculias. Behrrrioral Neiiruloyy, 3, 39-48.
Ardila. A . & Rosselli. M. ( 1992). Neirropsiroloyia Cliriica [Clinical neuropsychology]. Prensa
Creativa Medellin (Colombia).
Ardila. A. & Rosselli. M. ( 19941.Development of language. memory and visuospatial abilities in five -
t o I?-year-old children using a neuropsychological battery test. De\,elopnerrtal Neirropsycholoyy.
10.07-1 20.
Ardila. A., Rosselli. M . & Bateman. J. R. 11993). Structure of cognitive activity using a neuropsycho-
IoFical test battery. Behtrrioirral Neurology. 7,49-58.
Ardila. A,. Rosselli, M. & Puente. A. ( 1993).Neirro~psyrho/ogic.o[Eiulitatiorr ofthe Spanish-Speakers.

Plenum New York.
Bakker. D. J. ( 1979). Hemispheric differences in reading strategies: Two dyslexias? Bulletin of the
Orton Sociery. 29. 83-1 00.
Benson, D. F. ( 1994). The Neirrology of Thinking. Oxford University Press New York.
Benson, D. F. & Ardila. A. ( 1996).Aphasici: A cliriicalpel.specti\,e. Oxford University Press New York.
Benson. D. F. & Gewhwind. N. ( 1970). Developmental Gerstmann syndrome. Neurologv, 20.
96-102.
Bihcaldi. M.. Fischer. B. & Aiple. F. ( 1994).Saccadic eye moveinenta of dyslexic and normal reading
children. Percepriori. 23. 45-63.
Bihhop. D. V. M. ( 1990).Huridedness arid de\,elopriiema/ disorderx Blackwell Oxford. England.
Black. S. E. & Behrmann. M . ( 1994). Localization in alexia. In A. Kertesz (Ed. Locrrlizorion atid lieu-
roirricrgirig iri rierrrojp.sTe/i[ilog?.(pp. 33 I-! 76). Academic Press New York.
Brachacki. G. W. Z.. Fawcett. A. J. & Nicolson, R. 1. (1994). Adults with dyslexia have a deficit in
voice recognition. Perceptirol arid Moror Skilk, 78. 305-306.
Bravo-Valdivieso. L. ( 1980). Teoritrs sobre lo Didexin y sir Eilfoqite Cieririfico [Dyslexia theories and
it5 acientific study]. Editorial Universitaria Santiago.
Bravo-Valdivieso. L. ( 1988). Las Dislexias: lnvestigacion en Latinoamerica [Dyslexia: Research in
Latin America]. In A. Ardila & F. Ostrosky (Eds.) Lerzgirnge Ornl y Esrrito. Trillas Mexico.
Brumbach. R. A,. Staton. E. D. & Wilson, H. ( 1980). Neuropsychological studies of children during
and after remission of endogenous depressive episodes. Perreptical arid Moror Skills. 50,
1163-1 167.
Carlson. N. R. ( 1993). Physiolog~qfbehaiVor. 5th edition. AIlyn and Bacon Boston.
Chertkw. H. & Bub. D. (1994). Functional activation and cognition: The 150 PET substraction
method. In A. Kertesz (Eds). Loro/i:crriori mid rieirroirrrcrgirig iri riuirrupsvchologv Academic
Press New York (pp. 151-183).
Critchley. M. ( 1985). Specific Developmental Dyslexia. En: J. A. M. Frederiks (ed) Hnndbook of
Nertrology,\to/ 46: Neirrohehrri~iortilDisorders Elsevier Amsterdam (pp. 105-1 22).
Cronbach. L. J. ( 1990). Essrriricds ofpsychologicnl restirig. Harper & Row New York. 5th edition.
Denckla. M . B. ( I983 ). The neuropsychology of social-emotional learning disabilities. Archives qf
"V~rrro/n,q~, 40. 461462.
De Renri. E. & Lucchelli. F. (1990). Developmental dysmnesia in a poor reader. Brain, 113,
1337-1 346.
De Valois, R. I. & De Valois, K. K. (1988). Spatial Vision. Oxford University Press New York.
Duane, D. D. (1991). Dyslexia: Neurobiological and behavioral correlates. Psychiatric Annals, 21,
703-708.
Eden, G. F., Stein, J. F., Wood, H. M. &Wood, F. B. (1994). Differences in eye movements and read-
ing problems in dyslexic and normal children. Vision Research, 34, 1345-1358.
Ellis, A. W. (1993). Reading, Writing and Dyslexia. Lawrence Erlbaum Associates Hillsdale.
Elwood. R. W. (1991). Factor structure of the Wechsler Memory Scale-Revised (WMS-R) in a clini-
cal sample: A methodological reappraisal. The Clinical Neuropsychologist, 5, 329-337.
Evans, B. J. W., Drasom, N. &Richards, I. L. (1994). An investigation of some sensory and refractive
visual factors in dyslexia. Vision Research, 10. 1913-1926.
Flowers, D. L., Wood, F. B. & Naylor, C. E. (1991). Regional cerebral blood flow correlates of lan-
guage processing in reading disability. Archives of Neurology, 48,637-643.
Fletcher, J. M., Stuebing. K. K., Shaywitz, B. A., Shaywitz. S. E.. Rourke, B. P. & Francis, D. J.
(1994). Validity of the concept of dyslexia: Alternative approaches to definition and classifica-
tion. In K.P. Van des Bos, L.S. Siegel, D. Bakker & D.L., Share (Eds.), Current directions in
dyslexia research Swets & Zeitlinger Lisse (Holland) (pp. 31-44).
Galaburda, A. M. (1993). Neuroanatomical basis of developmental dyslexia. Neurologic Clinics, 11.
161-173.
Galaburda, A. M. (1994). Developmental dyslexia. Revue Neuroloqique. 149, 1-3.
Galaburda, A. & Kemper, T. (1979). Cytoarchitectonic abnormalities in developmental dislexia: A
case study. Annals of Neurology. ($94-100.
Galaburda, A. M., Sherman, G . F., Rosen, G. D., Aboitiz, F. & Geschwind, N. (1985). Developmental
dyslexia: Four consecutive patients with cortical anomalies. Annals of Neurology, 18, 222-223.
Geschwind, N. & Galaburda, A. M. (1985). Cerebral lateralization: biological mechanisms, associu-
tions andpathology. The MIT Press Cambridge.

Goodglass, H. (1993). Understanding Aphasia. Academic Press New York.
Goodglass. H.. Kaplan, E. & Weintrab, D. (1983). The assessment of aphasia and related disorders.
Lea Febiger Philadelphia.
Hagman, 0. J., Wood, F., Buchsbaund, M. S., Tallal, P., Flowers, L. & Katz. W. (1992). Cerebral
brain metabolism in adult dyslexic subjects assessed with possitron emossion tomography during
performance of an auditory task. Archives of Neurology, 49,734741.
Hier, D., Lemay. M., Rosenberger, P. & Perlo, V. P. (1978). Developmental dyslexia: evidence of a
subgroup with a reversal of cerebral asymmetry. Archives of Neurology, 3590-92.
Humphreys, P., Kaufmann, W. E. & Galaburda, A. M. (1990). Developmental dyslexia in women:
Neuropathological findings in three patients. Annals of Neurology. 28, 727-738.
Hugdahl, K., Synnevaag, B. & Satz, P. (1987). Immune and autoimmune disorders in dyslexic chil-
dren. Neuropsychologia, 28,673-679.
Hynd, G. W., Semrud-Clikeman, M., Lorys, A. R., Novery, E. S. & Eliopolus, D. (1990). Brain mor-
phology in developmental dyslexia adn attention deficit disorderihyperactivity. Archives qf
Neurology, 47,919-926.
Jemigan, T. L., Hesselink, J. R., Sowell, E. & Tallal, P. A. (1991). Cerebral structure on magnetic reso-
nance iamgening in language and learning impaired children. Archives of Neurology, 48,539-545.
Johnson, D. & Myklebust, H. (1971). Learning Disabilities. Gruen & Stratton New York.
Kertesz, A. (1985). Aphasia. In J. A. M. Frederiks (Ed.) Handbook of clinical neurologv. vol 45:
Clinica Neuropsychology Elsevier Amsterdam (pp. 287-332).
Kinsbourne, M. (1985). Learning disabilities. In J. A. M. Frederiks (Ed.) Handbook of neurology. i,ol
46: Neurobehavioral disorders. Elsevier Amsterdam.
Kolb, B. & Whishaw, I. Q. (1990). Fundamentals ofHuman Neuropsychology. W.H. Freeman and Co.
New York.
Leonard, C. M.. Voeller, K. K. S., Lombardino, L. J., Moms, M.,K., Hynd, G. W., Alexander, A. W..
Andersen, H. G., Garofalakis, M. K., Honeyman, J. C., Mao, J., Agee, 0. F. & Staab. E. V.
(1993). Anomalous cerebral structure in dyslexia revealed with magnetic resonance imaging.
Archives of Neurology, 50,461-469.
Levy, W. B. & Steward, 0. (1983). Temporal contiguity requirements for long-term associative poten-
tiatioddepression in the hippocampus. Neuroscience, 8,79 1-797.
Livingstone, M. S. & Hubel, D. H. (1988). Psychophysical evidence for separate channels for the per-
ception of form. color, movement, and depth. Journal of Neuroscience, 7,3416-3468.
204 A. ARDILA
1Loveland. K. A,. Fletcher. J . M & Bailey. V. (1990).Verbal and nonverbal communication of events
in learninp-di\ahility subtypes. Joirrnol oj C/irrico/ arid E.rperinlenta/ Neirropsycho/ogy. 12,
433447.
Luhar. J. F. & Dccrinp. W. M. ( 1981 ). Rehtii~iorolA~proac/ies to Neurology. Academic Press: New York.
Luhs. H. PI., Rahin. M.. Feldman. E.. Jallad. B. J . et al (1993). Familial dyslexia: Genetic and medical
findings in ele\ en three generation families. Anrials qfDy.s/exitr, 43,44-60.
ILuria. A R. ( 19h6l. H i g h e r Cortical Firnctions i n Man. Basic Books: New York.

Luria. A. R 1 I973 1. The Working Brcriri. Basic Books: New York.
Luria. A . R. ( 1976). Rtr.\ic Prohlems of Neir~olirigriistics.Mouton: The Hague.
Masutto. C.. Bravar. L. & Fabhro. F. 1991).Neurolinguistic diferentiation of children with subtypes
of dy\leuia. J ~ J I Wqt ~LuarningI ~ Disabiliries. 27. 520-516.
hlatti\. S.. French. J. & Rapin. I. ( 1975). Dyslexia in child and young adults: Three independent neu-
rop~,cchologicals) ndroines. Der.e/opmerircil Medicine arid Child Neurology, 17. 150-163.
\L..(onachie. H. ( 1976). Developmental prosopagnosia: A single case report. Cortes. 12, 76-82.
Yacller. J . D.. Mittenherg. W.. DePiano F. A. & Schneider. B. A. (1994). Effects of patient age on neu-
I tip\> chological tebt inteipretation. Pr~?fessiorrul P.cydioloyy Research arid Practice, 25,288-295.
Yaidoo. S ( 1972 ). Spwific Dy.s/c.rio. Wiley: Nea York.
Olson. R. K.. \Vi<e. B.. Comers. F. A. & Rack, J. P. (1990). Specific deficits in component reading
;ud language skills: Genetic and environmental influences. Joirrnol of learning Disabilities. 22,
.3?9-.348.
.
Orton. S . .I. ( 1037).Retiding. Nriting ~ n Speech d Problems in Children. Norton: New York.
O\tcrricrh. P. A. ( 1 9 4 ) .Le test dr copie dune figure cornplexe. Archiiw des Psychologie. 30,206-256.
Perkin,. A T. B Tcyler. T. J . 1988).A critical period for long-term potentiation in the developing rat
\ iwal cortex. Er(rirr Kesrorclt. 439. 2 5 4 7 .
Ptrtersrn. S. E.. Fox. P. T.. Snyder. A. Z. & Raichle. M . E. ( 1989). Activation of extrastriate and
trontd cortical areas by visual words and word-like stimuli. Science, 249. 1041-1044.
Pirw.xdIo. F. J. ( I979 I. The Nrrrropsvcho/o,qy of Developmental Reading Disorders. Praeger Press:
Nen Yorli.
Poimer. 11.. Klima. E. S. & Bellugi. L!. (1987). Ulro~the Hands Re\err/ About the Brain. MIT Press:
Cambridge. MA.
Pu\iier. 14 I., Peterscn. S. E.. Fox. P. T. & Raichle. M. E. ( 1988).Localization of cognitive operations
in tlir human braiii. S ~ , i e r r (240.
~ . 1627-1631.
Prics. C J.. Wiw. R . J. S.. Watson. J . D. G.. Patterson. K., Howard. D. & Frackowiak, R. S. J. (1994).
B rdin
x, t. i \ ation during reading: The effects of exposure duration and task. Brain, 117,1255-1269.
Quiros. I . B. i IQ6-t).Dyhphshia and dyslexia in school children. Folia Phoniatrica, 16, 201-215.
Raichle. M. E. 1 19941.Visualizing the mind. Scierrtijc Atnerican, 270(4), 5 8 4 5 .
K i s p r i ~J . Van Jer Stege & Bode. H. ( 1994).The clinical relevance of dyslexia subtype research. En:
K. f. Van de\ Bo\. L. S. Siegrl. D. Bakker & D. L. Share (Eds.), Cirrrerit Directions in Dyslexia
Re,eoi-c/r Swets s( Zeitlingrr: Lisse (Holland) (pp. 3 1-44),
K i w k e . B. P. 1988J. The \yndronie of nonverbal learning disabilities: developmental manifestations
111 neurolopical disease. disorders and dysfunctions. The Clinical Neuropsyhologisr, 2.293-330.
R O L ~ XB. 1. 1 19891.Aorii~erhalLuorning Disrihilities: The Syndrome and the Model. The Guilford
Press. Neu York.
Rourke. R . P.. Bakker. D. J.. Fish. J. L. 8; Strang. J. D. (1983). Child Neuropsyc/~ology:Ai7
/nt!,(dmr i m ro Tlwo!:\. Ke.~rtrrc.h. trnd CIiriical Practice. The Guilford Press: New York.
Rourkc. R . P. B Finlaymn. M. A. J. i 1978). Neuropsychological significance of variations in patterns
ot .Jcadeniic performance: verbal and visuospatial abilities. Jorrrtial qf Abnormal Child
l s \ c ~ / l o / o <6.
~ ~171-13.:.
.
R o u \ h . b 1995).Arguing why Johnny cant read. Science. 267. 1896-1898.
Runi\eq. 1. M.. Andreason. P.. Zametkin. A. J.. Aquino, T.. King. A. C.. Hamburger. S. D., Pikus, A..
Rapaport. J. L. & Cohen. R. M. i 1992). Failure to activate the left temporoparietal cortex in
dy\lcxia. .+l.hiit.S ~ ! f l v l , r ~ d n g49.
y , 527-534.
Runw!. J. M.. Zanictkin. A. J.. Andreason. P. & Hanahan. A. P. (1994). Normal activation of fron-
toteniporal language cortex in dyslexia. as measured with oxygen 15 positron emission tomogra-
ph! . .\rc./ii~.c,.\ntiVi~rtr.r>/ogy. 51. 17-38.
Schulti. R. T.. Cho. N. K.. Staib. L. H.. Kier. L. E., et al. (1994). Brain morphology in normal and
dyslexic children: The influence of sex and age. Ai7nals of Neurology, 35. 732-742.
Singh, R.. Broota, A. & Gupta. G. C. (1993). Spatial and linguistic abilities in dyslexic children.
Journal of Personaliq and Clinical Studies, 9,55-58.
Stanovich, K. E. (1994). Are discrepancy-based definitions of dyslexia empirically defensible? In K.
P. Van des Bos, L. S. Siegel, D. Bakker & D. L. Share (Eds.), Current Directions in Dvslexia
Research. Swets & Zeitlinger: Lisse (Holland) (pp. 15-30).
Tonnessen, F. E., Lokken, A,, Hoien, T. & Lundberg, I. (1993). Dyslexia, left-handedness, and
immune disorders. Archives of Neurology, 50,411416.

Toth, G. & Siegel, L. S. (1994). A critical evaluation of the IQ-achievement discrepancy-based defin-
ition of dyslexia. In K. P. Van des Bos, L. S. Siegel, D. Bakker & D. L. Share (Eds.), Current
Directions in Dyslexia Research. Swets & Zeitlinger: Lisse (Holland) (pp. 41-70).
Tranel, D., Hall, L. E., Olson, S. & Tranel, N. N. (1987). Evidence of a right hemisphere develop-
mental learning disability. Developmental Neuropsychology, 3, 113-1 17.
Verner, P. A. (1994). Neuropsychology of Individual Differences. Academic Press: New York.
Voeller, K. K. J. (1986). Right hemisphere deficit syndrome in children. American Jourrial of
Psychiatry. 143, 1004-1009.
Voeller, K. K. J. (1991). Social-emotional learning disabilities. Psychiatric Annals, 21, 735-74 1.
Vogler, G. P., DeFries, J. C. & Decker, S. N. (1985). Family history as an indicator of risk for reading
disability. Journal of Learning Disabilities, 18,419421.
Weintraub. S. & Mesulam, M. M. (1983). Developmental learning disabilities of the right hemisphere.
Archives of Neurology, 40,463468.
Wimmer, H. (1993). Characteristics of developmental dyslexia in a regular writing system. Applied
Psycholinguistics, 14, 1-33.
Witelson, S. F. (1977). Hemispheric specialization for linguistic and nonlinguistic tactual perception
using dichotic stimulation technique. Cortex, 10, 3-17.
Clinical Linguistics & Phonetics, January 2011; 25(1): 2336
Patterns of stuttering in a Spanish/English bilingual: A case

report
Clin Linguist Phon Downloaded from informahealthcare.com by Florida International University, Medical Library on 03/10/11
ALFREDO ARDILA, ELIANE RAMOS, & ROBERT BARROCAS
Department of Communication Sciences and Disorders, Florida International University,

Miami, Florida, USA
(Received 20 May 2010; Accepted 21 July 2010)
Abstract
Stuttering patterns may differ when comparing two languages. In bilinguals, specific patterns of stuttering
in each one of the languages may potentially be found. This study reports on the case of a 27-year-old
Spanish/English simultaneous bilingual whose dominant language is English. Speech and language testing
was performed in both languages (language repetition, language understanding, vocabulary, reading,
verbal fluency, spontaneous speech and conversational speech). Some differences in the stuttering
distribution were found: stuttering in adjectives, adverbs and conjunctions occurred at least twice as
much in Spanish as in English; stuttering was also more frequent in verbs in Spanish. Some hypotheses are
presented to explain the differences in stuttering severity in both languages. It seems that when comparing
stuttering in two different languages in the same subject, it is possible to find similarities in the stuttering
pattern, suggesting general stuttering laws; but also differences, associated not only with language-specific
idiosyncrasies but also with the individuals bilingualism characteristics.
Keywords: stuttering, bilingualism, Spanish characteristics, stuttering in Spanish/English bilinguals
Introduction
About half of the world population is bilingual (Grosjean, 1982; Siguan, 2001), although the
exact percentage depends upon the definition of bilingualism that is used. In the United States,
there are over 40 million bilinguals (,15% of the population) (U.S. Census Bureau, Statistical
Abstract of the United States, 2008); 75% of these bilinguals are Spanish/English bilinguals. If
stuttering affects ,1% of the general population (American Psychiatric Association [APA],
1994; Ardila, Bateman, Nio, Pulido, Rivera, & Vanegas, 1994; Mansson, 2000) it may be
calculated that in the country there are ,3 million people presenting with stuttering. In
consequence, it can be assumed that in the United States there are ,450,000 bilingual
stutterers, and close to 350,000 of them are Spanish/English bilinguals.
Stuttering occurs in all languages and ethnic groups (Andrews, Craig, Feyer, Hoddinott,
Howie, & Neilson, 1983; Zimmermann, Liljeblad, Frank, & Cleeland, 1983), although
prevalence might differ (Perello, 1995), because linguistic variables may affect stuttering.
Correspondence: Alfredo Ardila, PhD, ABPN, Florida International University, Department of Communication Sciences and
Disorders, NHS 431B, Miami, Florida 33199, USA. Tel: (305)348-2750. Fax: (305)348-2740. E-mail: ardilaa@fiu.edu
ISSN 0269-9206 print/ISSN 1464-5076 online 2010 Informa UK Ltd.
DOI: 10.3109/02699206.2010.510918
24 A. Ardila et al.
Brown (1945) identified four basic speech/language factors that determined whether words
will be spoken dysfluently by adults who stutter. These factors are: (1) word class (content
words are more prone to stuttering than function words); (2) word length (long words are
more difficult); (3) sentence position (words that appear in early positions are more likely to
be stuttered) and (4) phone the word starts with (words starting with consonants are more
difficult than those that start with vowels). Other speech/language factors, such as utterance
rate have also been correlated with stuttering frequency (Howell, Au-Yeung, & Pilgrim,
1999). Due to linguistic differences across languages (i.e. differences in word-length, stress
patterns, phonological characteristics, etc.), specific stuttering characteristics may differ

when comparing two languages (Dworzynski, Howell, & Natke, 2003; Dworzynski &
Howell, 2004). In bilinguals, specific patterns of stuttering in each one of the languages
may potentially be found. Certainly, a bilingual stutterer may represent a potential model to
compare stuttering in two different languages.
There is a general belief among researchers in the area that stuttering is more prevalent in
bilinguals than in monolinguals (Van Borsel, Maes, & Foulon, 2001). Travis, Johnson, and
Shover (1937) suggest that a direct relationship between bilingualism and stuttering may exist.
For 26% of the bilingual stutterers in their study, the age of onset of stuttering coincided with the
introduction of a second language. Several authors seem to support that bilingualism is directly at
the origin of the fluency disorder. Pichon and Borel-Maisonny (1964) state that stuttering could
be ascribed to bilingualism in an important percentage of stutterers, and, hence, bilingualism
may be regarded as a risk factor for stuttering. Other authors (e.g. Karniol, 1992) have
presented a similar point of view. Supposedly, using two languages may represent an excessive
linguistic demand for some children. Noteworthy, certain dysfluencies can be considered
normal when speaking a second language. In a recent study, Howell, Davis, and Williams
(2009), using a clinical referral sample of 317 children aged 812 who stuttered collected in
South-East England between 19992007, found that 69 of the subjects were bilingual (21.8%);
38 children used a language other than English primarily or exclusively in the home and 36 of
these (94.7%) bilinguals who stuttered did so in both their languages. The authors concluded
that certainly there is an increased chance of stuttering onset for bilingual children. Furthermore,
bilingual children not only have an increased risk of stuttering but also a lower chance of recovery
from stuttering than monolingual speakers.
Few studies have analysed stuttering specifically in Spanish/English bilinguals. Cabrera and
Bernstein Ratner (2000) studied a 5-year-old Spanish/English bilingual boy, and reported an
association between code-switching events and the occurrence of fluency failures. Dale (1977)
studied four Cuban-American male adolescents, all of whom were born in the United States but
spoke only Spanish at home. All four subjects were quite proficient in Spanish and English, but
none of them exhibited dysfluent speech while speaking English. While conversing in Spanish,
however, varying degrees of dysfluency were observed. Bernstein Ratner and Benitez (1985)
described a 50-year-old adult male bilingual stutterer who had spoken Spanish and English
since learning to speak and used both languages almost equally. Some general observations in
this case were: (1) it was harder for this subject to initiate sentences or clauses in Spanish than in
English; (2) conjunction and clause initial dysfluencies were twice as frequent in Spanish as in
English; (3) initial noun phrases and verb phrases were associated with stuttering in English, but
not in Spanish and (4) There was a tendency for dysfluency to appear on vowel-initiated words
both in English and Spanish; nonetheless, Spanish vowels were represented almost twice as
often. The authors suggested that differences in the loci of dysfluencies between English &
Spanish were associated mainly with differences between English & Spanish sentence structure.
Bilingualism and stuttering 25
Across different studies Howell and colleagues (Au-Yeung, Gomez, & Howell, 2003;
Howell, 2004; Howell & Au-Yeung, 2007) have analysed the linguistic characteristics of
stuttering in Spanish-speakers. Au-Yeung et al. (2003) selected 46 Spanish-speakers divided
into five age groups. The authors were interested in pinpointing if the developmental change in
loci of dysfluency from mainly function words to mainly content words, observed for English
speakers who stutter, also occurs for comparable Spanish speakers who stutter. It was found
that the rate of dysfluency on function words was higher than that on content words, particularly
in the youngest speakers. Function word dysfluency rate dropped off and content word
dysfluency rate increased across age groups. Howell (2004) observed that the correlation
between word type and stress does not apply to the same extent in Spanish as in English; in
Spanish it was found that both phonetic and metrical factors are important and independent
determinants of stuttering in adults who stutter: non-stressed content words had higher
stuttering rates than non-stressed function words. In English it is difficult to dissociate the
influence of syllabic and metrical factors, given that content words tend to weigh highly on
indexes of phonetic complexity and stress is also carried almost exclusively on these word types.
Howell and Au-Yeung (2007) analysed if the phonetic complexity affected stuttering rate for
Spanish speakers. The analysis was performed using Jakielskis (1998) Index of Phonetic
Complexity (IPC) scheme in which each word is given an IPC score based on the number of
complex attributes it includes for each of eight factors. Stuttering on function words for Spanish
did not correlate with IPC score for any age group. This mirrors the finding for English that
stuttering on these words is not affected by phonetic complexity. The IPC scores of content
words correlated positively with stuttering rate for 611-year-old and adult speakers. Evidence
was obtained that the factors found to be important determinants of stuttering on content
words in English for speakers aged 12 and above also affected Spanish speakers.
It has been demonstrated that diverse phonetic/articulatory idiosyncrasies may affect stutter-
ing. Dworzynski et al. (2003) analysed the linguistic factors affecting stuttering in German
speakers. In adults, both word type (content/function) and word length increased stuttering rate
significantly. It was also found that when word difficulty (based on a combined measure of all
Browns factors) increased, stuttering rate rose. With children, only the word-length factor was
significant, and stuttering rate was not governed to the same extent by overall word difficulty.
Significant correlations between stuttering rate and phonetic complexity score were found for
content words for children over the age of 6 years and adults (Dworzynski & Howell, 2004).
Although there is not a clear explanation for this association between stuttering and bilingu-
alism, it can be conjectured that using two languages represents an additional linguistic burden
for bilingual children. Lebrun and Paradis (1984) consider that the input of linguistically mixed
utterances might trigger the development of stuttering in bilingual children with a predisposition
to stuttering, and, hence, stuttering may be more apparent in bilingual than in monolingual
children. Indeed, a diversity of factors may contribute to the apparently increased stuttering
frequency reported in bilinguals, such as the characteristics of the two languages (i.e. similarities
and differences between both languages), the type of bilingualism (simultaneous, successive,
etc.), the mastery of the two languages, etc. For instance, it is evident that language mastery plays
a significant role in bilingual stuttering: it has been observed that language ability in a second
language influences the frequency, distribution and nature of dysfluencies; and stutterers stutter
less in their more dominant language than in the language with lower proficiency (Jankelowitz &
Bortz, 1996; Lim, Lincoln, Chan, & Onslow, 2008). Van Borsel et al. (2001) have suggested that
a major investigation topic in the area is whether or not prevalence of stuttering in bilinguals is
affected by the similarities of the languages involved (i.e. the functional distance between both
languages); unfortunately, available data do not provide a clear answer to this question.
26 A. Ardila et al.
There are three possibilities for stuttering patterns in bilinguals: (1) stuttering occurs in one
language but not the other; (2) Stuttering occurs in both languages with similar speech
behaviour patterns in each one and (3) Stuttering occurs in both languages but varies from
one language to another (Nwokah, 1988). Although similar patterns and distributions of
stuttering with different degrees of stuttering severity in each language do occur in bilingual
stutterers, it is more common to find that both the severity and distribution of dysfluencies
differ from one language to another (Van Borsel et al., 2001).
The purpose of this study was to analyse the pattern of stuttering in a Spanish/English bilingual.
Although both Spanish and English are Indo-European languages with diverse similarities and
commonalities, from a phonetic/articulatory point of view several significant differences can be
found between Spanish and English. Some of them are: (1) more types of words carry stress in
Spanish than English. As a matter of fact, Spanish presents a rather clear accentual structure; every
word has an accent that is clearly salient; (2) Spanish is a syllabic language; each word is composed
of one or several syllables usually formed around one vowel (or diphthong), that are easily
distinguished; (3) the number of phonemes is lower in Spanish than in English (,23 phonemes
vs. ,34 phonemes); this difference is mainly due to the increased amount of English vowels, but
Spanish includes a significantly larger amount of diphthongs; conversely, Spanish words are on
average longer in number of phonemes and syllables than English words. As stress (Natke,
Sandriesen, van Ark, Pietrowsky, and Kalveram, 2004; Wingate, 2002), phonetic structure
(Howell, Au-Yeung, & Sackin, 1999) and word-length (Brown, 1945) are three known determi-
nants of stuttering, the participation of these factors could lead to different frequency and patterns
of stuttering in these two languages. In addition, English is a stress-timed language (syllables may
last different amounts of time, but there is a given amount of time between two consecutive
stressed syllables, and that time is roughly a constant), while Spanish is a syllable-timed language
and every syllable takes up roughly the same amount of time when pronounced (Abercrombie,
1965). This last difference clearly may also affect speech fluency. Because of these differences in
the syllabic characteristics between Spanish and English, we emphasised the analysis of the
percentage of stuttered words. Furthermore, it has been suggested that speaking rate may be faster
in Spanish than in English (Miranda & Valencia, 1997), and it is known that utterance rate
correlates with stuttering frequency (Howell et al., 1999).
This study hypothesised that similarities and differences in stuttering patterns would be
found, suggesting general stuttering laws as well as language-dependent stuttering idiosyn-
crasies. Information about the patterns of stuttering in Spanish and English is limited;
extending this information can be particularly valuable in further understanding linguistic
variables affecting stuttering.
Method
Case report
The subject is a 27-year-old, right-handed male. He is a second generation Cuban-American. His
mother and father immigrated to the United States at 8 and 10 years of age, respectively. English
was the dominant language spoken in the home. However, the subject was exposed to Spanish for
at least 8 hours per day until age 5 while under the care of his grandparents, who only spoke
Spanish. The majority of the subjects schooling was received in English, although he studied
Spanish grammar and literature for 2 years in high school. The subject does not speak any other
languages.
The subjects mother noted the onset of stuttering at ,67 years of age. The subjects father
and paternal grandfather and great-grandfather were also dysfluent. The subject received speech
therapy for ,30 minutes per week while in elementary school. He also was evaluated and attended
about three sessions at a private practice prior to entering junior high school. The subject then
received speech therapy once per week for ,4 months when he was 18 years old. In each of these
instances, therapy was provided only in English. The type of stuttering therapy received was easy
onset therapy. Additionally, the subject has engaged in several years of self-study to control the
severity of his stuttering. The result of this self-study has been the inclusion of stuttering
modification and cognitive behaviour therapy techniques to the subjects repertoire.

The subject and his family have observed a gradual decrease in the severity of his stuttering in
both English and Spanish. However, the subject is aware that he exhibits a greater number of
dysfluencies in Spanish than in English. The subject currently speaks Spanish ,10% of the time.
Testing procedure
The following tests were administered in a single session:
(1) Spontaneous speech test 1: description of a picture (Plate # 1 from the Boston
Diagnostic Aphasia Examination; Goodglass & Kaplan, 1983);
(2) Spontaneous speech test 2: reporting something (e.g. the content of two university
courses that he is currently taking);
(3) Reading: English and Spanish texts with a similar topic (literature analysis) and
similar level of difficulty, each one containing 646 words;

(4) Boston Naming Test: English and Spanish (Kaplan, Goodglass, & Weintraub, 1978,
1996);
(5) Verbal fluency in two condition: (a) Category/Semantic (to tell as many as possible words
in 1 minute; two categories were used: animals, fruits); (b) letter/phonological (to tell
as many words beginning with the letters F, A, S and M as possible in 1 minute);
(6) Language repetition: taken from the Multilingual Aphasia Examination English and
Spanish versions (Benton & Hamsher, 1976; Rey & Benton, 1991);
(7) Peabody Picture Vocabulary Test: English and Spanish versions (Dunn & Dunn,
2007; Dunn, Padilla, Lugo, & Dunn, 1986). This test was administered to have
some additional information about the subjects bilingualism level. Because of the
characteristics of the test, no oral answer is required: and
(8) Vocabulary sub-test from the WAIS-III English and Spanish (Wechsler, 1997; 1999)
version.
Order of administration was randomly selected, but Spanish and English were given in an
alternating way. The order of administration was:
(1) Boston Naming Test: English,
(2) Repetition Spanish,
(3) Spontaneous Speech 1 English,
(4) Vocabulary Spanish,
(5) Peabody Vocabulary Test English,
(6) Reading Spanish,
(7) Vocabulary English,
(8) Spontaneous Speech 1 Spanish,
(9) Repetition English,
(10) Verbal fluency Spanish,
28 A. Ardila et al.
(11) Reading English,

(12) Spontaneous speech 2 Spanish,
(13) Verbal fluency English,
(14) Boston Naming Test Spanish,
(15) Spontaneous Speech 2 English and
(16) Peabody Vocabulary Test Spanish.
Once the formal testing was completed, an informal conversation both in Spanish and in English
was conducted (Conversation condition); the topic of this informal conversation was the
participants experience with stuttering. This closing conversation was also recorded and used
as additional data for analysis. The Spontaneous Speech and Conversation conditions were
transcribed by a bilingual Spanish/English assistant. Transcriptions were double checked by the
authors. Any discrepancies found were discussed and then resolved by the first two authors in
conjunction. Transcriptions were then entered into a spreadsheet with utterances separated into
Terminable Units. (A T-Unit is equivalent to a sentence with all of its subordinate clauses.
Coordinated clauses are separated into different T-units; see Gutirrez-Clellen, Restrepo,
Bedore, Pea, & Anderson, 2000.) Each T-unit was then coded by two graduate assistants (one
for English and one for Spanish) for number of words, number of clauses, grammatical category of
each word, grammatical category of stuttered words, number of stuttering events, type of stutter-
ing event, number of fillers and whether stuttering/filler occurred in the beginning of sentences or
clauses. Counted as stuttering events were phonemic prolongations, phonemic repetitions, part-
word and whole-word repetitions. Each repetition type was counted only once regardless of the
number of repetitions. Blocks were not observed and fillers were not counted as stuttering events,
but rather analysed as an associated speech abnormality (see Table I).
Coding for Spanish transcriptions was checked by the first author and for English transcrip-
tions by the second author, and discrepancies were resolved by the two authors in conjunction.
Table I. General characteristics of the spontaneous speech and conversation conditions.
Spontaneous Spontaneous
speech 1 speech 2 Conversation All conditions
Eng Span Eng Span Eng Span Eng Span
Total number of words 91 59 289 87 872 282 1252 428

Total number of T-units 7 7 25 6 67 30 99 43
Mean length of T-units (in words) 13 8.43 11.56 14.5 13 9.72 12.6 9.95
Index of complexity (mean # clauses per unit) 1.86 1.29 2.08 1.66 2.3 1.83 2.21 1.67
Total number of fillers 11 8 50 27 70 57 131 92
Fillers per T-unit 1.86 1.14 2 3.86 1.04 1.97 1.32 2.14
Total number of stuttering moments 4 9 44 15 115 61 163 85
% of stuttered words 4 15 15 17 13 21 13 20
Stuttering moments per T-unit .57 1.28 1.76 2.5 1.72 2.03 1.64 1.97
% phonemic prolongations 11.1 0 2.27 0 1.74 6.55 2 6
% phonemic repetitions 0 0 0 6.67 .87 4.92 .6 5
% part-word repetition 25 0 11.36 20 12.2 16.4 13 15
% part-word 1 syllable 100 0 40 100 92.9 100 76 100
% whole-word repetition 75 88.89 86.36 80 85.2 72.1 65 75
% whole-word 1 syllable 100 100 100 91.67 100 79 100 83
Note: Eng English; Span Spanish.

Results
Table II presents the general results in the different formal tests that were administered. In all the
tests, scores were higher in English than in Spanish. Excepting Verballetter condition, scores
can be considered as normal. No question, English was the dominant language and Spanish the
non-dominant language. His score on the Boston Naming TestSpanish, was 34/60, usually
considered as normal in bilinguals non-dominant language (Rosselli, Ardila, Santisi, Arecco,
Salvatierra, & Conde, 2002). Dysfluencies were absent or minimal in all these tests, excepting
WAIS-III Vocabulary. Furthermore, in reading, no stuttering episode was observed either in

English or Spanish.
Table I presents the characteristics of the two spontaneous speech and conversation condi-
tions. As observed, the amount of speech produced was notoriously higher in English than in
Spanish: while the total number of words in these three conditions was 1252 in English, it was
only 428 in Spanish; that is, he used almost three-times more words in English than in Spanish;
overall speech rate was higher in English than in Spanish: , 104 words/minute in English and
42 words per minute in Spanish; this difference was related with the introduction of longer
pauses in Spanish than in English. Language complexity was higher in English than in Spanish,
according to the Mean Length of Terminable Units (T-units, Gutirrez-Clellen et al., 2000)
Table II. General results in the different tests.
Score Raw score Scaled score Dysfluencies Standard score

Boston Naming Test

Spanish 34/60 3
English 58/60 0
Peabody
Spanish 117 (125) na
English 216 (228) 114
Language repetition
Spanish 8 (percentile ,50) 2
English 14 (percentile ,95) 0
Vocabulary (WAIS-III)
Spanish 9 17
English 16 9
Verbal fluency (semantic)
Spanish
Animals 19 0
Fruits 9 0
English
Animals 29 0
Fruits 10 0
Verbal fluency (letter)
Spanish
F 5 0
A 3 0
S 5 0
M 3 0
English
F 5 0
A 2 0
S 7 0
M 4 0
30 A. Ardila et al.
and the Index of Sentence Complexity (mean # of clauses per T-unit). Stuttering was more
evident in Spanish than in English: the number of stuttering moments per T-unit was between
.571.76 in English, whereas in Spanish it was between 1.282.50. Fillers per T-unit were 1.32
in English and 2.14 in Spanish. Dysfluency types were very similar in English and Spanish, with
the greatest percentage being whole-word repetition (65% in English and 75% in Spanish)
followed by part-word repetitions (13% in English and 15% in Spanish). Phonemic prolonga-
tions and repetitions were more frequent in Spanish than in English (11 vs. 2.6%).
Figure 1 presents the percentage of words stuttered in English and Spanish, and the stuttering
moments per T-unit in English and Spanish. It is observed that, according to Percentage of
Words Stuttered, stuttering in English was observed in ,13% of the words (4%, 15% and 13%
in the three conditions); while in Spanish it was found in ,20% of the words (15%, 17% and
21%); that is, stuttering in Spanish was ,1.54-times more frequent than in English. According
to Stuttering Moments per T-Unit, stuttering in English averaged 1.64 (.57, 1.76 and 1.72);
while in Spanish they averaged 1.97 (1.28, 2.50 and 2.03); that means that Stuttering Moments
per T-Unit were 20% more frequent in Spanish than in English.
Table III shows the percentage of stuttering by grammatical category. Several general conclu-
sions can be drawn: (1) stuttering in nouns was most infrequent; it was absent in English and
(a)
25
20
15
English
10 Spanish
0
Spontaneous 1 Spontaneous 2 Conversation
(b)
3
2.5
1.5 English
Spanish
1
0.5
0
Spontaneous 1 Spontaneous 2 Conversation
Figure 1. Percentage of stuttered words (a) and stuttering moments per T-units (b) in English and Spanish.
Table III. Percentage of stuttering by grammatical category (each spontaneous condition separate and all three
conditions combined).
Spontaneous Spontaneous
speech 1 speech 2 Conversation Combined
Eng Span Eng Span Eng Span Eng Span
Nouns 0 0 0 0 0 2 0 1.15
Verbs 7.69 0 9.8 18.2 7.84 18 8.29 15.71
Auxiliaries (plus aux verbs) 14.3 0 8.33 50 28.9 25 23.94 16.7

Adjectives 0 0 5.26 33.3 5.08 11.1 4.82 17.24
Adverbs 0 0 7.14 25 5.97 35.7 6.18 34.37
Conjunctions 0 50 11.5 25 7.69 14.7 8.25 17.5
Pronouns 0 20 12.8 0 15.9 18.8 14.65 18.9
Prepositions 6.67 33.3 29.2 15.8 13.4 15.2 10.61 17.24
Determiners (articles/demonstratives) 6.67 30 34.8 22.2 8.77 33.3 14.73 29.03
Others (quantifiers, infinitival to, 0 6.67 16.7 0 28 46.7 25.86 52.63
negative)
observed in only 1.15% of Spanish nouns; (2) Stuttering in adjectives, adverbs and conjunctions
occurred at least twice as often in Spanish as in English; (3) in a less consistent way, stuttering was
observed in verbs more frequently in Spanish than in English; (4) in general, stuttering in both
languages was more frequent in function (i.e. prepositions, pronouns, auxiliary verbs, conjunc-
tions, determiners, quantifiers, particles and infinitival to) than in content (i.e. nouns, verbs,
adjectives and adverbs) words. This difference was statistically significant for English ( 2 24.11;
df 1; p .001) but not for Spanish ( 2 1.32; df 1; p ns).
All subsequent analyses were done using the conversation condition because this was the
condition where most stuttering occurred (Table IV).
To pinpoint the effect of vowel vs. consonant initiated words, word-length and phonetic
complexity, the initial segment of the conversation condition (208 words in English and 210
words in Spanish) was selected. Table V presents the association between stuttering and
Table IV. Percentage of stuttering at the beginning of a sentence or clause in both languages
(conversation conditions).
T-units Stuttered moments Stuttering beginning Percentage
English 67 (872 words) 79 21 26.6

Spanish 30 (282 words) 48 17 35.4
Table V. Percentage of stuttering in vowel and consonant initiated words in both languages
(conversation conditions).
Vowel initiated Consonant initiated
Total Stuttered Percentage Total Stuttered Percentage
English 55 6 10.91 153 13 8.50

Spanish 92 16 17.39 118 16 13.56
32 A. Ardila et al.
Table VI. Association between word-length and stuttering (conversation conditions).
English Spanish
Total Stuttered Percentage Total Stuttered Percentage
1-phoneme 16 1 6.25 14 3 21.43

2-phoneme 57 9 15.79 63 15 23.81
3-phoneme 56 7 12.5 34 4 11.76
4-phoneme 29 1 3.45 35 2 5.71
5-phoneme 15 0 0 25 5 20.00
6-phoneme 15 1 6.67 19 2 10.53
6 < phoneme 20 0 0 20 1 5.00
Total 208 19 210 32
vowel and consonant initiated words in both languages. For both languages, stuttering was
more frequent (by ,2530%) in vowel initiated words than in consonant initiated words.
However, this difference was not statistically significant for English ( 2 .27; df 1; p ns)
or Spanish ( 2 .20; df 1; p ns). Distribution was similar in English and Spanish.
The association between word-length and stuttering was further analysed (Table VI). In
English, the majority of the stuttered words (17/19 89.5%) contain one-to-three phonemes;
in Spanish one-to-three phoneme words represent 68.8% (22/32) of stuttered words.
Finally, the IPC was calculated following the procedure described by Dworzynski and
Howell (2004) and Howell and Au-Yeung (2007). Two logistic regression analyses were
performed to assess whether IPC significantly predicted the stuttering of a word in Spanish
and in English. Table VII presents the summary of this analysis both for English and for
Spanish and Figure 2 illustrates the association between phonetic complexity and stuttering
(conversation condition). Results showed that IPC was not significant for English and
marginally significant for Spanish. Nonetheless, the prediction was in a negative direction:
The higher the IPC, the lower the stuttering probability.
Discussion
Although the patient was a simultaneous Spanish/English bilingual, with English as a dominant
language, it was evident that stuttering was more frequent in Spanish than in English. Several
explanations could be proposed for this observation: (1) stuttering in general is more severe in
the less dominant language (Jankelowitz & Bortz, 1996; Lim et al., 2008); in our case, Spanish
was the native language but currently English was clearly the dominant one; (2) The subject
received therapy in English and, hence, his control of stuttering in English was more efficient
Table VII. Logistic regression analysis using IPC as the independent variable and stuttering as
the dependent variable for Spanish and English.
Variables in the equation
B SE Wald df Sign. Exp(B)
Spanish ".267 .132 4.098 1 .043 .765

English ".263 .153 2.953 1 .086 .769
2.5
1.5
IPC
English
1
Spanish
0.5
0
Average IPC Dysfluent Average IPC Fluent
Figure 2. Association between phonetic complexity and stuttering (conversation condition).
than in Spanish and (3) Because of a diversity of linguistic reasons (e.g. stress characteristics,
word length, speech rate, etc.) stuttering probability in Spanish is higher than in English.
Although there are no comparative epidemiological studies of stuttering in Spanish and
English, the prevalence of self-reported stuttering in Spanish seems too high (2% according to
Ardila et al., 1994), compared with the usually accepted prevalence of stuttering in English
(.81% according to the DSM-IV; APA, 1994). Interestingly, in the four Cuban-American
male adolescents studied by Dale (1977), all of them proficient in Spanish and English, none
exhibited dysfluent speech when speaking English, while varying degrees of dysfluency were
observed in Spanish. By the same token, in the balanced bilingual described by Bernstein Ratner
and Benitez (1985), it was found that not only was it harder for this subject to initiate sentences or
clauses in Spanish than in English, but also the conjunction and clause initial dysfluencies were
twice as frequent in Spanish as in English; and, in general, stuttering appeared more severe in
Spanish than in English. It has been suggested that stuttering prevalence varies across different
languages (e.g. Perello, 1995); for instance, it may be lower in tonal languages. It could be
suggested that, because of some linguistic reasons, Spanish represents a language with a high
prevalence of stuttering. It is not simple, however, to pinpoint which are the specific linguistic
reasons that make Spanish a language with an increased prevalence of stuttering; nonetheless, it
could be speculated that the evident syllabic structure, making Spanish an overtly syllable-timed
language (Berg, 1991), may increase the likelihood of dysfluencies. However, only comparative
studies of stuttering in different types of languages can clarify the linguistic reason accounting for
the different distribution of stuttering across languages.
These three explanations (i.e. stuttering in general is more severe in the less dominant
language; subjects control of stuttering in English was more efficient than in Spanish; stuttering
probability in Spanish is higher than in English) are not contradictory, and all of them may have
contributed to the disproportionately higher frequency of stuttering in Spanish.
Similarities and differences in the stuttering pattern in both languages were observed. For
both languages stuttering was observed specially in spontaneous speech and conversation
conditions; it was lower in the WAIS-III Vocabulary sub-test (i.e. defining words); it was
absent in English and minimal in Spanish in naming (Boston Naming Test); and it was absent
in both languages during the verbal fluency tests and reading conditions.
34 A. Ardila et al.
In general, stuttering in both languages was more frequent in function than in content
words, but more so in English; stuttering in adjectives, adverbs and conjunctions occurred at
least twice as often in Spanish than in English; these differences are congruent with the
suggestion that there are differences in the loci of dysfluencies between English and
Spanish due to their specific linguistic characteristics proposed by Bernstein Ratner and
Benitez (1985). These authors also found a different distribution of stuttering in different
grammatical categories; for instance, they reported that stuttering in conjunctions in Spanish
was twice as frequent as in English, exactly coincidental with our results.
For both languages, stuttering was more frequent in vowel-initiated words than in consonant-
initiated words, although stuttering in vowel-initiated words was slightly higher in Spanish than in
English. Also, stuttering was more likely in longer words in Spanish than in English; in English,
stuttering was observed mostly in short (one-to-three phonemes) words. Although the percentage
of stuttering in shorter words was smaller in Spanish, this is still a high percentage, especially
considering that Spanish has fewer short words than English. It is likely that this bias towards
shorter words is due to the higher occurrence of stuttering in function words in both languages.
We found that there is no significant association between IPC and stuttering probability. The
stuttered language sample that we used for this analysis was composed almost exclusively by
function words. This finding is congruent with Howell and Au-Yeungs (2007) report that
stuttering on function words for Spanish does not correlate with IPC score in children and is
also similar to the observation for English that stuttering in function words is not affected by
phonetic complexity.
Noteworthy, whether in English or Spanish, our case study does not support the four basic
speech/language factors proposed by Brown (1945) to determine whether words will be

spoken dysfluently by individuals who stutter. We found that function words were more
prone to stuttering than content words; stuttering was observed especially in short words; in
only ,30% of the cases (26.6% in English and 35.4% in Spanish) stuttering was observed at
the beginning of a sentence or clause (although our subject used a significant amount of fillers
in the beginning of sentences that potentially may have helped him avoid stuttering) and
stuttering was more frequent in vowel- than consonant-initiated words.
In conclusion, when comparing stuttering in two different languages in the very same subject, it
is possible to find similarities in the stuttering pattern, suggesting general stuttering laws; but also
differences, associated not only with the linguistic idiosyncrasies of each one of the languages, but
also with the individuals mastery of each language and bilingualism characteristics.
Acknowledgement
Our most sincere gratitude to Melissa Marsal and Tara Blickenstaff for their invaluable help
in the coding and analysis of the results of this study, and to Erika Fountain for transcription
of speech samples.
Declaration of interest: The authors report no conflicts of interest. The authors alone are
responsible for the content and writing of the paper.
References
Abercrombie, D. (1965). Studies in Phonetics and Linguistics. London: Oxford University Press.
American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (4th ed.).
Washington DC: American Psychiatric Association Press.
Andrews, G., Craig, A., Feyer, A.M., Hoddinott, S., Howie, P., & Neilson, M. (1983). Stuttering: A review of
research findings and theories circa 1982. Journal of Speech and Hearing Disorders, 48, 226246.
Ardila, A., Bateman, J.R., Nio, C.R., Pulido, E., Rivera, D.B., & Vanegas, C.J. (1994). An epidemiological study of
stuttering. Journal of Communication Disorders, 27, 3748.
Au-Yeung, J., Gomez, IV., & Howell, P. (2003). Exchange of disfluency with age from function words to content
words in Spanish speakers who stutter. Journal of Speech Language and Hearing Research, 46, 754765.
Benton, A.L., & Hamsher, K. (1976). Multilingual Aphasia Examination. Iowa City, IA: University of Iowa Press.
Berg, T. (1991). Phonological processing in a syllable-timed language with pre-final stress: Evidence from Spanish
speech. Language and Cognitive Processes, 6, 265301.
Bernstein Ratner, N., & Benitez, M. (1985). Linguistic analysis of a bilingual stutterer. Journal of Fluency Disorders,
10, 211219.
Brown, S.F. (1945). The loci of stuttering in the speech sequence. Journal of Speech Disorders, 10, 181192.
Cabrera, V., & Bernstein Ratner, N. (2000, November). Stuttering patterns in the two languages of a bilingual child.
Paper presented at the ASHA annual convention, Washington, DC.
Dale, P. (1977). Factors related to dysfluent speech in bilingual CubanAmerican adolescents. Journal of Fluency
Disorders, 2, 311314.
Dunn, L.M., & Dunn, L.M. (2007). Peabody Picture Vocabulary Test. (4th ed.). Circle Pines, MN: American
Guidance Service (AGS), Inc.
Dunn, L.M., Padilla, E.R., Lugo, D.E., & Dunn, L.M. (1986). Test de Vocabulario en Imagenes Peabody: TVIP. Circle
Pines, MN: American Guidance Service (AGS), Inc.
Dworzynski, K., & Howell, P. (2004). Predicting stuttering from phonetic complexity in German. Journal of Fluency
Disorders, 29,149173.
Dworzynski, K., Howell, P., & Natke, U. (2003). Predicting stuttering from linguistic factors for German speakers in
two age groups. Journal of Fluency Disorders, 28, 95112.
Goodglass, H., & Kaplan, E. (1983). The assessment of aphasia and related disorders (2nd ed.). Philadelphia: Lea &
Febiger.
Grosjean, F. (1982). Life with two languages. Cambridge, MA: Harvard University Press.
Gutirrez-Clellen, V., Restrepo, M.A., Bedore, L., Pea, E., & Anderson R. (2000). Language sample analysis in Spanish-
speaking children: Methodological considerations. Language, Speech, and Hearing Services in Schools, 31, 8898.
Howell, P. (2004). Comparison of two ways of defining phonological words for assessing stuttering pattern changes
with age in Spanish speakers who stutter. Journal of Multilingual Communication Disorders, 2, 161186.
Howell, P., & Au-Yeung, J. (2007). Phonetic complexity and stuttering in Spanish. Clinical Linguistic Phonetics, 21,
111127.
Howell, P., Au-Yeung, J., & Pilgrim, L. (1999). Utterance rate and linguistic properties as determinants of lexical
dysfluencies in children who stutter. Journal of the Acoustical Society of America, 1, 481490.
Howell, P., Au-Yeung, J., & Sackin, S. (1999). Exchange of stuttering from function words to content words with age.
Journal of Speech, Language and Hearing Research, 42, 345354.
Howell, P., Davis, S., & Williams, R. (2009). The effects of bilingualism on stuttering during late childhood. Archives
of Disease in Childhood, 1, 4246.
Jakielski, K. (1998). Motor organization in the acquisition of consonant clusters (Unpublished doctoral dissertation).
University of Texas, Austin.
Jankelowitz, D.L., & Bortz, M.A. (1996). The interaction of bilingualism and stuttering in an adult. Journal of
Communication Disorders, 29, 223234.
Kaplan, E., Goodglass, E., & Weintraub, S. (1996). Test de vocabulario de Boston. Madrid: Editorial Medica
Panamericana.
Kaplan, E.F., Goodglass, H., & Weintraub, S. (1978). The Boston Naming Test. Boston: Lea & Febiger.
Karniol, R. (1992). Stuttering out of bilingualism. First Language, 12, 255283.
Lebrun, Y., & Paradis, M. (1984). To be or not to be an early bilingual? In Y. Lebrun, & M. Paradis (Eds.), Early
bilingualism and child development (pp. 918). Lisse: Swets & Zeitlinger.
Lim, V.P., Lincoln, M., Chan, Y.H., & Onslow, M. (2008). Stuttering in English-Mandarin bilingual speakers: The
influence of language dominance on stuttering severity. Journal of Speech, Language, and Hearing Research,
51,15221537.
Mansson, H. (2000). Childhood stuttering: Incidence and development. Journal of Fluency Disorders, 25, 4757.
Miranda, J.O., & Valencia, R.R. (1997). English and Spanish versions of a memory test: Word-length effects versus
spoken-duration effects. Hispanic Journal of Behavioral Sciences, 19, 171182.
Natke, U., Sandrieser, P., van Ark, M., Pietrowsky, R., & Kalveram, K.T. (2004). Linguistic stress, within-word
position, and grammatical class in relation to early childhood stuttering. Journal of Fluency Disorders, 29,109122.
36 A. Ardila et al.
Nwokah, E.E. (1988). The imbalance of stuttering behavior in bilingual speakers. Journal of Fluency Disorders, 13,
357373.
Perello, J. (1995). Trastornos del habla (5th ed.). Barcelona: Masson.
Pichon, E., & Borel-Maisonny, S. (1964). Le bgaiement. Sa nature et son traitement. Paris: Masson.
Rey, G., & Benton, A. (1991). Examen de afasia multilinge. Iowa City, IA: AJA Associates Inc.
Rosselli, M., Ardila, A., Santisi, M.N., Arecco, M.R., Salvatierra, J., & Conde, A. (2002). Stroop Effect in Spanish-
English bilinguals. Journal of the International Neuropsychological Society, 8, 819827.
Siguan, M. (2001). Bilingismo y lenguas en contacto [Bilingualism and languages in contact]. Madrid: Alianza Editorial.
Travis, L.E., Johnson, W., & Shover, J. (1937). The relation of bilingualism to stuttering. Journal of Speech Disorders,
2, 185189.
Van Borsel, J., Maes, L., & Foulon, S. (2001). Stuttering and bilingualism: A review. Journal of Fluency Disorder, 26,
179205.
Wechsler, D. (1997). WAIS-III: Administration and scoring manual. San Antonio, TX: The Psychological Corporation.
Wechsler, D. (1999). WAIS-III. Escala de Inteligencia de Wechsler para Adultos III. Madrid: TEA Ediciones.
Wingate, M.E. (2002). Foundations of stuttering. San Diego, CA: Academic Press.
U. S. Census Bureau, Statistical Abstract of the United States. (2008). Retrieved from http://www.census.gov/prod/
2007pubs/08abstract/pop.pdf
Zimmermann, G., Liljeblad, S., Frank, A., & Cleeland, C. (1983). The Indians have many terms for it: Stuttering
among the BannockShoshoni. Journal of Speech and Hearing Research, 26, 315318.
Intern. J . Npuroscience, 1991. Vol. 57, pp. 13-79 C 1991 Gordon and Breach Science Publishers S.A.
Reprints available directly from the publisher Printed in the United Kingdom
Photocopying permitted by license only
NEUROPSYCHOLOGICAL DEFICITS IN CHRONIC

COCAINE ABUSERS
A L F R E D 0 ARDILA, MONICA ROSSELLI and STEVEN STRUMWASSER
Miami Institute of Psychology of the Caribbean Center for Advanced Studies
(Received Augusl 26, 19901
A basic neuropsychological assessment battery was given to thirty-seven chronic freebase cocaine ("crack")
abusers. The following tests were used: Wechsler Memory Scale (Wechsler, 1945), Rey-Osterrieth Complex
Figure (Osterrieth, 1944) (copy and immediate reproduction), Verbal Fluency (semantic and phonologic),
Boston Naming Test (Goodglass, Kaplan, & Weintrab, 1983), Wisconsin Card Sorting Test (Heaton, 198 I )
and Digit-symbol from the WISC (Wechsler, 1974). In general, performance was lower than expected
according to their age and cductational level. Subjects showed significant impairment in short-term verbal
memory and attention subtests. Neuropsychological test scores were correlated with lifetime amount of
cocaine used, suggesting a direct relationship between cocaine abuse and cognitive impairment. A pattern
of cognitive decline is proposed.
Keywords: Cocaine abuse; neuropsvchology
During the last several years, a vast importance has been given to the medical and
social analysis of drug abuse (e.g., Perez, 1987). This has been particularly true with
regard to the effects of alcohol, marijuana and LSD (Adams, Rennik, Schooff &
Keegan, 1975; Carlin & Trupin, 1977; Fletcher et al., in press; Grant, Adams, Carlin
& Rennik, 1977; Parsons & Farr, 1981; Satz, Fletcher & Sutker, 1976; Wright &
Hogan, 1972). In general, these studies have shown the existence of some cognitive
deficits with chronic drug abuse; these effects appear particularly evident with regard
to memory, attention, and abstraction abilities (Bruhn & Maage, 1975; Bruhn et al.,
1981; Carlin, 1986; Carlin, Strauss, Grant & Adams, 1978; Parsons & Farr, 1981).
This has led to the proposal of the existence of a group of toxic dementias (e.g., Ardila
& Rosselli, 1986; Marsden, 1985). Much evidence has already been established in
regard to alcoholic dementia and its manifestations of impairments in short-term
memory, attentional deficits, and spatial defects (Victor & Adams, 1985).
Research on the cognitive effects of the chronic use of alkaloids extracted from the
coca leaf (Erythroxylum coca) has been, however, particularly scarce. The acute effects
of cocaine on behavior are much better known. It seems clear that due to the stimulant
effect of cocaine on the nervous system the subject displays euphoria, restlessness, and
increased energy (Gawing & Ellinwood, 1988; Hartmann, 1988). Cocaine intoxication
can also produce some psychiatric symptoms (Gawin & Kleber, 1986; Miller, Gold
& Milman, 1989). Frequent users may develop panic attacks (Washton & Gold,
1984), paranoid ideation, depression, anxiety, and loss of motivation (Washton &
Tatarsky, 1984); violent behavior is commonly seen also among cocaine patients
(Manschreck et al., 1988). When given chronically cocaine can produce some be-
Address correspondence to Dr. Alfred0 Ardila, Instituto Colombiano de Neuropsicologia, Apartado

Aereo 17021, Bogota, Colombia, South America
73
74 A . ARDILA, M. ROSSELLI A N D S. STRUMWASSER
havior disorders similar to those seen in schizophrenic psychosis (Wyatt, Fawcett, &
Kirch, 1989).
Freebase cocaine (crack) has been associated with cerebrovascular complication
(Kaye & Fainstat, 1987; Levine et al., 1987; Golbe & Merkin, 1986). Cocaine intoxi-
cation can produce seizures, cardiac arhythmias, and respiratory arrest (Miller, Gold,
& Milman, 1989). Although the exact mechanism of cocaine on the nervous system
is not presently understood, it has been hypothesized that it produces a direct
neurotoxin effect or multiple microinfarctions due to hemodynamic changes. Drowsi-
ness, unsteady gait and seizures have been correlated with passive smoke inhalation
in children exposed to the smoke of free cocaine used by their adult caretakers
(Bateman & Heagarty, 1989). The fetal effects of crack with its use during preg-
nancy have been recently studied by Cherukuri et al. (1988).
The cognitive consequences of chronic cocaine use are less clearly defined. Adamse,
OMalley, Heaton and Gawin (unpublished) have suggested a correlation between the
amount of cocaine consumed and the severity of cognitive impairment. Washton and
Gold (1984) reported that 57% of the cocaine abusers interviewed experienced
memory problems. Press (1 983) found better (although nonsignificant) performance
of normals when compared with cocaine abusers on the Luria-Nebraska Neuropsy-
chological Battery. According to the author, verbal memory subtests were the most
significantly impaired in cocaine abusers. Low scores in the range of impairment on
Trials A and B have also been reported in freebase cocaine users (Reitan & Wolfson,
1985). Adamse et al. (unpublished) administered a neuropsychological screening
battery to 20 chronic cocaine abusers and compared their performance with normal
controls. The poorest performance was observed on the Symbol-Digit Modalities
Test, Arithmetic Test and Story Memory. The authors suggested an underlying deficit
in attention and memory.

The purpose of the present study was to analyze further the cognitive effects of
chronic cocaine abuse.
METHOD
Subjects
Thirty-seven freebase cocaine (crack) abusers (23 males and 14 females) with a
mean age of 28.86 ( S D = 5.18; range = 20-46) were studied. The subjects mean
educational level was 10.86 ( S D = 1.20). All subjects met the criteria of substance
dependence according to the DSM-111-R (1 984) but were abstinent for approximately
30 days (mean 27.35) before the evaluation. They were in an inpatient rehabilitation
program in Miami (Florida) at the moment of the neuropsychological testing. The
subjects had selected crack as their main drug for 16.64 months (on average) before
the neuropsychological evaluation. All subjects were considered heavy users (average
9.27 grams per week). All were polydrug users before using crack. However, once
they started using crack they did not use any other drug, except for marijuana
occasionally. No subject presented a major positive neurological (head trauma,
epilepsy, etc.), or psychiatric (DSM-111-R, 1984) disorder. At the time of the evalu-
ation, no subject was taking any medication that might affect the central nervous
system. Subjects were nonpaid volunteers.
Instruments
All subjects were assessed in abstraction, language, constructional and memory
CHRONIC COCAINE ABUSERS 75
abilities, using the following neuropsychological tests: (I). Wechsler Memory Scale
(WMS) (Wechsler, 1945). It includes the following subtests: Information, Orientation,
Mental Control, Digits, Logical Memory, Visual Reproduction and Associative
Learning. (2). Rey-Osterrieth Complex Figure (Osterrieth, 1944). Copy and im-
mediate reproduction. Taylors scoring system was used (Lezak, 1983). (3). Verbal
Fluency. Two different conditions were used: Semantic verbal fluency (animals), and
phonological verbal fluency (FAS). (4). Boston Naming Test (Goodglass, Kaplan, &
Weintraub, 1983).(5). Wisconsin Card Sorting Test (WCST) (Heaton, 1981). (6). Digit-
symbol from the WISC-R (Wechsler, 1974).
Procedure
Subjects were interviewed to determine eligibility. Using a structured interview,
subjects were asked about demographic data, developmental, family, and medical
history, and history of substance abuse. Subjects who met the criteria for participation
in the study were then administered a battery of neuropsychological tests. Admini-
stration time was approximately 90 minutes, divided into two sessions.
RESULTS
Results of the WMS, Rey-Osterrieth Complex Figure, and Verbal Fluency Tests, were
compared both with the norms collected by Ardila, Rosselli, and Puente (in press),
matching age and educational level, and with the norms presented by Lezak (1983).
For the rest of the tests, normative data presented in the test manuals were used. Table
I presents the results obtained on the WMS. Only for Information, Logical Memory
and Associative Learning Subtests, the drug abuser group scores are one or more
standard deviations below the normative group scores. Associative Learning subtest
was the most impaired: the mean score in Associative Learning subtest is almost two
standard deviations below the normative means used. For the cocaine-abuser group
TABLE I
Mean scores and standard deviations (in brackets) obtained in the Wechsler Memory Scale subtests
SUBTEST COCAINE ABUSERS NORMATIVE DATA

(Ardila et al., in press) (Lezak, 1983)
(Wechsler,
1945)
Information 4.73(0.83) 5.30(0.56)

Orientation 4.84(0.44) 4.80(0.50)
Mental Control 5.43(2.26) 4.95(2.52)
Logical Memory 6.59(1.96) 13.43(3.69) 9.28(3.10)
Digits 8.76(2.15) 9.05(2.22)
Visual Reproduction 6.78(2.66) 8.25(2.70) I l.OO(2.73)
Associative Learning 10.81(2.20) 15.60(3.25) 15.7q2.81)
Total 47.94 62.39

Memory Quotient 84*
79**
*According to Ardila et al.s norms

**According to Wechslers norms
76 A. ARDILA. M. ROSSELLI A N D S . STRUMWASSER
TABLE I1
Mean scores and standard deviations (in brackets) obtained in the neuropsychological tests
SUBTEST COCAINE ABUSERS NORMATIVE DATA

(Ardila et al., in press) (Lezak, 1983)
(Goodglass et al.. 1983)
Rey Figure (copy) 3 1.43(S.02) 29.33(4.06) 32.00

Rey Figure (memory) 18.36(6.55) 19.44(5.66) 22.00
Verbal Fluency (animals) 14.21(3.11) 14.03(4.72) 18.00
Verbal Fluency (FAS) 32.20(2.79) 34.26(4.30) 3 1-44
Boston Naming Test 49.94(6.03) 55.73(4.42)
the Memory Quotient calculated with regard to the normative data was one standard
deviation below the expected normative score.
Table TI presents the results for the Rey-Osterrieth Complex Figure, Verbal Flu-
ency, and Boston Naming Tests. There were no significant differences between the
experimental and normative group means in the copy of the Rey-Osterrieth Complex
Figure, and Verbal Fluency subtests. However, in the immediate reproduction of the
Rey-Osterrieth Complex Figure, and in the Boston Naming Test, the drug-abuser
group scored about one standard deviation below the expected score.
The Digit-symbol subtest mean raw scores in the cocaine group was 41.45
( S D = 8.78) that corresponds in average to a scale score of 4 (mean = 10) for
16-year-old children. On the Wisconsin Card Sorting Test the mean total number of
errors was 33.35 ( S D = 21.75); the normals mean is 21.60 errors ( S D = 16.70)
(Heaton, 1981).
The score in each of the subtests was correlated with lifetime cocaine use and with
the estimated amount of drug intake. Table I11 presents these correlations. In general,
correlations with lifetime use of crack were higher than correlations with weekly
amount of cocaine use. There was a significant correlation between lifetime drug
abuse and the scores in the Digits subtest of the WMS and Memory Quotient, and
with the Rey-Osterrieth Complex Figure immediate reproduction condition. Amount
intake correlated significantly only with Information subtest of the WMS, and with
the Boston Naming test. Other correlations were lower and nonsignificant.
TABLE I11
Correlations between cocaine use and scores in the neuropsychological subtests
Subtest lifetime (in months) P amount P

Information -0.10 0.26 - 0.29 0.04
Orientation - 0.24 0.07 0.18 0.13
Mental Control - 0.2 I 0.09 0.01 0.47
Logical Mernorq - 0.23 0.07 -0.14 0.19
Digits - 0.36 0.01 - 0.08 0 29
Visual Reproduction - 0.01 0.49 - 0.02 0.43
Associative Learning - 0.22 0.08 - 0.20 0.10
Memory Quotient - 0.34 0.01 - 0.07 0.32
Rey Figure (copy) -0.15 0.17 -0.17 0.15
Rey Figure (memory) - 0.47 0.001 -0.17 0.15
Fluency (animals) - 0.07 0.33 - 0.06 0.34
Flucncy (FAS) -0.1s 0.25 0.10 0.16
Boston Naming Test -0.14 0.19 - 0.3 1 0.03
Digit Symbol - 0.22 0.08 -0.13 0.21
Wisconsin --0.19 0.12 -0.16 0.16
DISCUSSION
Our results, although preliminary, point to a moderate but significant cognitive

impairment associated with cocaine abuse. This cognitive impairment is reflected
particularly in verbal memory (Logical Memory and Associative Learning) subtests,
and to a lesser degree, nonverbal memory (immediate reproduction of the Rey-Oster-
rieth Complex Figure and Visual Reproduction from the WMS) tests. The Memory
Quotient was significantly decreased in cocaine abusers, and negatively correlated
with lifetime cocaine use; Digit-Symbol subtest scores were also significantly de-
creased. Abstracting ability (Wisconsin Card Sorting Test) and naming (Boston
Naming Test) scores were about one standard deviation below the expected scores;
while Verbal Fluency, Rey-Osterrieth Complex figure copy, and Information scores
were virtually normal. This pattern of impairment is, in general, consistent with the
impairment usually found in stimulant-abusers (Adams et al., 1975; Carlin, 1986;
Carlin et al., 1977, 1978; Washton & Gold, 1984). The significant correlation observed
with lifetime amount of cocaine use supports the assumption that cocaine is respons-
ible for the lowered performance in the neuropsychological tests.
According to our results, freebase cocaine abuse, under the conditions of our
sample, is associated with a moderate but significant cognitive impairment. However,
this cognitive impairment follows a specific pattern: short-term verbal memory (Logi-
cal Memory, and Associative Learning substest, and Memory Quotient) and atten-
tional (Digit-Symbol and Digits subtests) abilities are most sensitive, while abstract-
ing (Wisconsin Card Sorting Test), naming (Boston Naming Test), short-term non-
verbal memory (Visual Reproduction and Rey-Osterrieth Complex Figure, immediate
reproduction) are less impaired than short-term verbal memory and attention. Verbal
fluency (semantic and phonologic), constructional abilities (Rey-Osterrieth Complex

Figure, copy), and long-term memory (Information subtest) are virtually unimpaired.
It would be expected that changing the conditions of the sample (lower or higher
amounts of cocaine intake, longer or shorter lifetime use, mixture with additional
drugs) the level of cognitive impairment would be different, although the general
pattern, similar.
Our results are consistent with previous research on neuropsychological effects of
cocaine abuse. Press (1983) underlines the verbal memory deficits in cocaine abusers,
while Adamse et al. (unpublished) suggest an underlying defect in concentration and
memory. Low scores in attentional tasks such as the Trials tests have been also
reported (Reitan & Wolfson, 1985). Consequently, it seems consistent that memory
and attention are most senstive to chronic cocaine abuse, probably followed by
abstraction and naming abilities. It is important to note that these are the general
cognitive deficits usually found in chronic drug abusers (Bruhn & Maage, 1975; Bruhn
et al., 1981; Carlin, 1986; Carlin et al., 1978; Parsons et al., 1981), eventually leading
to a toxic dementia.
REFERENCES
Adams, K.M., Rennik, P., Schooff, K., & Keegan, J. (1975). Neuropsychological measurement of drug
effects: polydrug research. Journul qf Psychedelic Drugs, 7, 15 1-1 60.
Adamse, M . , OMalley, S.. Heaton, R.K., & Gawin. F.H. (unpublished). Neuropsychological assessment
in chronic cocaine abusers.
Ardila, A., & Rosselli, M. (1986). Neuropsicologk~ de1,fendmeno del envejecimienio.Medellin (Colombia):
Prensa Creativa.
Ardila, A., Rosselli, M., & Puente A. (in press). Neurop~.rc/iolo~icul
ussesswient: A cliniculguide. New York:
Plenum.
78 A. ARDILA, M. ROSSELLI A N D S. STRUMWASSER
Bateman, D.A., & Heagarty. M.C. (1989). Passive freebase cocaine (crack) inhalation by infants and
toddlers. American Journal of Disabled Children, 143, 25-27.
Bruhn, P., Arlien-Soberg, P.. Glydensted, C., & Christensen. E.L. (1981). Prognosis in chronic toxic
encephalopathy. Acta Neurologica Scandinavica, 64, 259-272.
Bruhn, P., & Maage, N. (1975). Intellectual and neuropsychological functions in young men with heavy
and long-term patterns of drug abuse. American Journal of Psychiatry, 132, 397-401.
Carlin, A.S. (1986). Neuropsychological consequences of drug abuse. In 1. Grant & K.M. Adams (Eds.)
Neuropsychological assessment of neuropsychiatric disorders. New York: Oxford University Press,
pp. 478-4971
Carlin, A.S., Strauss, F.F., Grant, I., & Adams, K.M. (1978). Prediction of neuropsychological impairment
in polydrug abuse patients. Addictive Behavior, 5 , 229-234.
Carlin, AS.. & Trupin, E. (1977). The effects of long term chronic cannabis use on neuropsychological
functioning. International Journal of Addiction, 12:. 617-624.
Cherukuri, R., Minkoff. H., Feldman, J., Parekh, A.. & Glass, L. (1988). A cohort study of alkaloidal
cocaine (crack) in pregnancy. Obstetric-Gynecology, 72, 147-1 5 1.
Diagnostic Statistical Manual of Mental Disorders-111-R (1984). Washington: American Psychiatric
Association.
Fletcher, J.M., Page, J.B., Francis, D.J., Krauskopf, D., Trup, W.P., & Satz, P. (in press). Cognitive effects
of chronic cannabis abuse. N e n England Journal of Medicine.
Gawin. F.H., & Kleber, H. (1986). Abstinence symptomatology and psychiatric diagnosis in cocaine
abusers. Archives of General Psyrhiatrjj, 43, 107-1 13.
Gawin, F.H., & Ellinwood, E.H. (1988). Cocaine and other stimulants. Action. abuse and treatment. New
England Journal of Medicine. 318. 1173-1 182.
Golbe. L.I.. & Merkin, M.D. (1986). Cerebral infarction in a user of free-base cocaine (crack). Neurol-
ogy, 36. 1602-1604.
Goodglass. H.. Kaplan, E., & Weintraub, S . (1983). Boston naming test. Philadelphia: Lea Febiger.
Grant, I., Adams, H.K., Carlin, A.S., & Rennik, P.M. (1977). Neuropsychological deficits in polydrug
abusers. Drug and Alcohol Dependence, 2, 91-108.
Hartmann, D.E. (1988). Neuropsychological toxicologj. New York: Pergamon Press.
Heaton. R. ( I 981). Winsconsin Card Sorting Test: Manual. Odessa: Psychological Assessment Resources,
Inc.
Kaye. B.R.. & Fainstat, M. (1987). Cerebral vasculitis associated with cocaine abuse (letter). J A M A , 257.
1175.
Levine, S.R., Washington, J.M., Jefferson, M.F., Kiernan, S.W., Moen, M., Feit, H., & Welsh. K.M.
(1987). Crack cocaine-associated stroke. Neurology. 37, 1849-1853.
Lezak, M.D. (1983). Neuropsychological assessment. New York: Oxford University Press.
Manschreck. T.C., Laughery, J.A.. Weisstein, C.C.. Allen, D., Humblestone, B., Neville, M., Podlewski,
H.. & Mitra, N. (1988). Characteristics of freebase cocaine phychosis. Yale Journul of Biolugicul
Medicine. 61. 115-122.
Marsden, C.D. (1985). Assessment of dementia. In J.A.M. Frederiks (Ed.) Handbook ufclinical neurolugj~,
vol. 46: Neurohehavioral disorders. Amsterdam: Elsevier. pp. 221 -232.
Miller, N.S.. Gold. M.S., & Milman, R.L. (1989). Cocaine. American Farnil), Ph-vsician. 39. 115-120.
Osterrieth, P.A. (1944). Le test de copie dune figure complexe. Archieves de psychologie, 30, 206-356.
Parsons. E.S.. & Farr, S.D. (1981). The neuropsychology of alcohol and drug abuse. In S.B. Filskov, T.S.
Boll (Eds.), Handbook of clinical neuropsychology. vol I . New York: Wiley, pp. 320-365.
Perez. A. (1987). Cocaina: surgimiento y evolucion de un mito. Bogota: Catilogo Cientifico.
Press. R.J. ( 1983). The neurupsychologicul qfects ofcucuine und opiate use. Ann Arbor: University Micro-
films International.
Reitan, R.M.. & Wolfson, D. (1985). The Halstead-Reiran Neuropsychological Battery. Theor), and Clinical
Interpretation. Tucson, Arizona: Neuropsychology Press.
Satz. P., Fletcher, J.M.. & Sutker. L.S. (1976). Neuropsychological, intellectual. and personality correlates
of chronic marihuana use in native Costa Ricans. In R. Dornbush, A.M. Freedman, & M . Fink (Eds.)
Chronic cannabis use. Annals of the New York Academp of Sciences. 282, 266-306.
Victor. M.. 81 Adams, D.R. (1985). The alcoholic dementia. In J.A.M. Frederiks (Ed.), Handbook o/
clinical neurology, vol 46: Neurobehuvioral disorders. Amsterdam: Elsevier, pp, 335-350.
Washton. A.M., & Gold, M.S. (1984). Chronic cocaine abuse: Evidence and adverse effects on health and
functioning. Psychiatric Annals, 733-743.
Washton. A.M.. & Tatarsky, A. (1984). Adverse effects of cocaine abuse. Nationd Institute ofDrug Abuse:
Reseurch Monograph Series. 4Y. 247-254.
Wechsler. D . (1945). A standarized memory test for clinical use. The Journul of Psychology, I Y , 87-95.
Wechsler. D. (1 974). WISC-R: Manual. New York: Psychological Corporation.
Wright. M., & Hogan, T.P. (1972). Repeated LSD ingestion and performance on neuropsychological tests.
Journal of Nervous and Mental Disease, 154. 432-438.
Wyatt, R.J., Fawcett, R., & Kirch, D. (1989). A possible animal model of defect state schizophrenia. In
S.C. Schulz, & C.A. Tamminga (Eds.), Schizopkwnicc: Sciencrific progress. New York: Oxford Uni-
versity Press, pp. 184-189.
INTERNATIONAL JOURNAL OF PSYCHOLOGY, 2006, 41 (5), 324332
Cognitive testing toward the future: The example of

Semantic Verbal Fluency (ANIMALS)
Alfredo Ardila Feggy Ostrosky-Sols

Florida International University, Universidad Nacional Autonoma de Mexico,
Miami, FL, USA Lomas de Reforma, Mexico
Byron Bernal
Miami Childrens Hospital, FL, USA
W e are proposing that, in the future, tests included in psychological and neuropsychological batteries
should fulfil the following criteria. (1) Have a large enough normative database (normative criterion).
Performance of subjects of different ages and different educational levels, including illiterates, should be well
established. Normative data from different countries and cultural contexts should be available. (2) Know the
effects of brain damage on different characteristics on the test (clinical criterion). (3) Know how the brain is
activated when the test is performed (experimental criterion). (4) Know how this test correlates with other
cognitive tests (psychometric criterion). Few contemporary tests fulfil all these criteria. A notable exception is
Semantic Verbal Fluency test using the category ANIMALS. This test requires the generation of words
corresponding to a specific semantic category, such as animals, fruits, vegetables, etc. Typically, the number of
correct words produced in 1 minute is counted. Semantic verbal fluency taps lexical knowledge and semantic
memory organization. Using regional cerebral blood flow measures, it has been reported that both frontal and
temporal activation are observed while performing this test. Optimal fluency performance involves generating
words within a subcategory and, when a subcategory is exhausted, switching to a new subcategory. Although
different semantic categories have been used in this test, ANIMALS is the most frequent due to some significant
advantages: (1) it is a clear enough semantic category across languages and cultures; (2) it is a relatively easy
semantic category with only minor differences among people living in different countries, different educational
systems, or belonging to different generations; and (3) it is an easy-to-administer, short, and common test
included in different neuropsychological test batteries. It is concluded that obtaining similar information for
other cognitive tests represents a huge research endeavour for psychology and neuropsychology during the 21st
century.
N ous proposons, qua lavenir, les tests inclus dans le testing psychologique et neurologique doivent remplir
les criteres suivants: (1) avoir une base de donnees normatives assez grande (critere normatif). La
performance des sujets dage et de niveaux deducation differents, incluant les illetres, doit etre bien etablie. Les
donnees normatives provenant de pays et de contextes culturels differents doivent etre disponibles; (2) connatre
les effets du dommage cerebral sur les differentes caracteristiques sur ce test (critere clinique); (3) savoir
comment le cerveau est active quand le test est execute (critere experimental); et (4) savoir comment ce test est
correle avec dautres tests cognitifs (critere psychometrique). Peu de tests contemporains remplissent tous ces
criteres. Une exception remarquable est la Fluence Verbale Semantique utilisant la categorie ANIMAUX. Ce test
requiert la generation de mots correspondant a une categorie semantique specifique, tel que les animaux, les
fruits, les vegetaux, etc. Typiquement, la quantite de mots produits en une minute est comptee. La fluence verbale
semantique capte la connaissance lexicale et lorganisation de la memoire semantique. En utilisant des mesures
regionales de debit sanguin cerebral, il a ete rapporte quune activation frontale et temporale est observee en
executant ce test. La performance optimale au test de fluence implique la generation de mots a linterieur dune
sous-categorie et, lorsquune sous-categorie est epuisee, le passage a une nouvelle sous-categorie. Meme si
Correspondence should be addressed to Alfredo Ardila, PhD, ABPN, Florida International University, Department of
Communication Sciences and Disorders, HLS144, Miami, Florida 33199, USA (E-mail: ardilaa@fiu.edu).
# 2006 International Union of Psychological Science
http://www.psypress.com/ijp DOI: 10.1080/00207590500345542
SEMANTIC VERBAL FLUENCY 325
differentes categories semantiques ont ete utilisees dans ce test, les ANIMAUX est celle la plus utilisee a cause des
avantages significatifs suivants: (1) il sagit dune categorie semantique assez claire a travers les langues et les
cultures, (2) cest une categorie semantique relativement facile, avec seulement des differences mineures entre les
gens vivant dans des pays differents, des systemes educatifs differents ou appartenant a des generations
differentes, et (3) il sagit dun test qui est facile a administrer, bref et commun, inclus dans differentes batteries de
tests neuropsychologiques. Nous concluons que lobtention dinformations similaires pour dautres tests cognitifs
represente un defi de taille pour la recherche en psychologie et en neuropsychologie durant le 21eme siecle.
S e propone que en el futuro, las pruebas utilizadas en la evaluacion psicologica y neuropsicologica deberan
cubrir los siguientes criterios: (1) Contar con un base de datos normativa lo suficientemente grande.
Establecer claramente el desempeno de sujetos de diferentes edades y niveles educativos. Contar con datos
normativos de diferentes pases y contextos culturales (criterio normativo); (2) conocer los efectos del dano
cerebral en diferentes aspectos de la prueba (criterio clnico); (3) conocer la activacion cerebral ante la
ejecucion de la prueba (criterio experimental); y (4) conocer como se correlaciona la prueba con otras pruebas
cognitivas (criterio psicometrico). Pocas pruebas contemporaneas cubren estos criterios. Una excepcion es la
prueba de fluidez verbal semantica con la categora ANIMALES. Esta prueba requiere la generacion de palabras
correspondientes a una categora semantica como animales, frutas, vegetales, etc. Generalmente la cantidad de
palabras correctas generadas en un minuto es contabilizada. Usando medidas del flujo sanguneo cerebral se ha
reportado una activacion frontal y temporal mientras se ejecuta esta tarea. El desempeno optimo en esta prueba
implica la generacion de palabras dentro de una subcategora y cuando esta se agota, cambiar a una nueva.
Aunque diferentes categoras semanticas han sido utilizadas en esta prueba, la categora ANIMALES es la mas
frecuente debido a algunas ventajas significativas: (1) es una categora clara entre idiomas y culturas, (2) es una
categora relativamente sencilla con pequenas diferencias entre personas de diferentes pases, sistemas educativos
o entre personas pertenecientes a distintas generaciones, y (3) es una prueba sencilla de administrar, corta e
incluida en varias bateras neuropsicologicas. Se concluye que el obtener informacion semejante para otras
pruebas cognitivas representa un enorme reto para la psicologa y neuropsicologa durante el siglo XXI.
During the last decades, psychology and neuro- significant; in other tests (e.g., spelling), language
psychology have been significantly interested in may be crucial, etc.
developing reliable and valid enough cognitive 2. Clinical criterion. That is, to know reason-
testing procedures (e.g., Lezak, 2004; Mitrushina, ably well the effects of different types of brain
Boone, & DElia, 1999; Spreen & Strauss, 1998; pathology on this test: right, left, anterior, poster-
Wechsler, 1997). Important controversies, how- ior, cortical, and subcortical. In addition, informa-
ever, have been frequent (e.g., Ardila, 1999). We tion regarding test performance in major
are proposing that in the future, in addition to the pathological conditions (dementia, traumatic
fundamental psychometric test properties (e.g., brain injury, Parkinsons disease, etc.) is highly
reliability), cognitive tests should fulfil at least the desirable.
following four criteria. 3. Experimental criterion. That is, to know
reasonably well how the brain is activated in
1. Normative criterion. That is, to have a large normal individuals when this particular test is
enough normative database to understand the performed. fMRI and/or PET studies are required
variables impacting test performance. Performance to know how different brain areas participate
in subjects with different ages and educational when performing the test.
levels, including illiterates (about one fifth of the 4. Psychometric criterion. To know how this
world population; www.portal.unesco.org), should specific test correlates with other cognitive mea-
be well established. As a matter of fact, norms in sures; i.e., the communality existing with other
illiterates are crucial in order to understand assessment procedures.
cognition without education as a confounding
variable. Developmental data and performance in Few contemporary tests fulfil these optimal
elderly populations are also necessary. Normative criteria. A notorious exception, however, is
data from different countries and cultural contexts Semantic Verbal Fluency for ANIMALS. This
should also be available. This information will be paper analyses the three last criteria in the
most important to identify variables affecting the Semantic Verbal Fluency test when the category
test performance. For some tests (e.g., language ANIMALS is used. The analysis of the first
tests), educational level plays a major role; for criterion was presented elsewhere (Ardila,
others (e.g., memory tests), age may be most Ostrosky-Sols, & Bernal, 2005).
326 ARDILA, OSTROSKY-SOLIS, BERNAL
SEMANTIC VERBAL FLUENCY (ANIMALS) by bursts of words over time, which are semanti-
cally or phonemically related. Optimal fluency
Controlled oral word association or word or performance involves generating words within a
verbal fluency has become one of the most popular subcategory and, when a subcategory is exhausted,
tests in neuropsychology (Lezak, 2004; Spreen & switching to a new subcategory. Troyer,
Strauss, 1998). Two different conditions can be Moscovitch, and Winocur (1998a) devised a
used: (1) phonemic fluency (words beginning with method to quantify these processes as switches
a particular letter, usually in English F, A, and S); and average cluster size. These components are
and (2) category or semantic verbal fluency (words differentially affected by various neurological
corresponding to a specific semantic category, disorders. Clustering is related to temporal lobe
such as animals, fruits, vegetables, etc.). Typically, functioning, while switching is related to frontal
the number of correct words produced in 1 minute functions (Testa, Troster, Fields, Gleason,
is counted. A normal person can produce about 12 Salmon, & Beatty, 1998; Troyer, Moscovitch,
words beginning with a specific letter, and about Winocur, Alexander, & Stuss, 1998b).
16 words corresponding to a semantic category, Although different semantic categories have
within 1 minute (Spreen & Strauss, 1998). The been used in this test, ANIMALS is the most
level of productivity, however, depends upon the frequent. Using ANIMALS has some significant
letter or the semantic category. A very strong advantages. (1) It is a clear enough semantic
educational effect has been demonstrated in both category across languages and cultures. The
conditions, especially in the phonemic condition semantic field of other categories can vary across
(Ostrosky-Sols, Ardila, & Rosselli, 1999; different languages. For example, the semantic
Ostrosky-Sols, Ardila, Rosselli, Lopez, & field of the English word vegetables is only
Mendoza, 1998; Ostrosky-Sols, Ramrez, partially coincidental with the semantic field of the
Lozano, Picasso, & Velez, 2004; Rosselli, Ardila, Spanish word vegetales. Furthermore, in Spanish,
& Rosas, 1990). Ardila, Ostrosky-Sols, Rosselli, vegetales may be a confusing semantic category,
and Gomez (2000) found that the educational level whereas vegetables is a clear one in English. In
accounted for 38.5% of the variance in the Spanish, trees or grass can be considered as
phonemic condition and for 23.6% of the variance vegetables. Interestingly, many English-speaking
in the semantic condition. A moderate to middle people include some fruits as vegetables (e.g.,
correlation (around .30.60) between scores in tomatoes are indeed fruits, but many people
both conditions has also been reported (Ardila, consider them to be vegetables). Furniture,
Rosselli, & Bateman, 1994; Ardila, Galeano, & clothes, and other categories can be different in
Rosselli, 1998; Matute, Rosselli, & Ardila, 2004; different countries and different languages.
Ostrosky et al., 1999). ANIMALS do not present these ambiguities. (2)
Verbal fluency tests have often been interpreted ANIMALS is a relatively easy semantic category
as executive functioning tests (Bruyer & Tuyumbu, with only minor differences among people living in
1980; Parks et al., 1988; Perret, 1974; Ramier & different countries, exposed to different educa-
Hecaen, 1970; Ruff, Allen, Farrow, Niemann, & tional systems, or belonging to different genera-
Wylie, 1994). Difficulties in performing this test tions (for example, a generation effect can be
are found in cases of frontal lobe damage, usually conjectured for the category tool). The brain
left and bilateral frontal lesions. In general, it is activity, therefore, is not expected to vary as much
assumed that phonemic verbal fluency is superior as it could with a task based on phonological clues
to semantic fluency as an executive functioning when used in different languages (Chinese and
test. Semantic verbal fluency is rather tapping Spanish, for example). Nell (2000) has suggested
lexical knowledge and semantic memory organiza- that Semantic Verbal Fluency using ANIMALS
tion. Using regional cerebral blood flow measures, represents a reliable test for adults with low
it has been reported that frontal and temporal education. Lopera (n.d.), using 16 different
activation are both observed while performing this semantic categories in a 3000-participant sample,
test. PET studies have demonstrated that the found that the easiest semantic categories were
frontal lobe is activated in phonemic generation BODY-PARTS, THINGS TO EAT, and
while the temporal lobe is more active in semantic ANIMALS (Table 1). That is, there is a hierarchy
generation of words (Warburton et al., 1996). of difficulty when finding elements corresponding
Several investigators have observed that words to a particular semantic category. (3) ANIMALS
tend to be produced in semantic clusters on the Verbal Fluency is an easy to administer, short, and
semantic fluency task and in phonemic clusters on common test included in different neuropsycholo-
the phonemic fluency task. The clusters are defined gical test batteries (e.g., Boston Diagnostic
TABLE 1 Agraval, & Relkin, 2003), vascular and degenera-

Average number of words generated by Spanish speakers in tive dementias (Cooper et al., 2001; Troyer et al.,
1 minute in different semantic categories (n53000) 1998a), and amyotrophic lateral sclerosis
(Lopera, n.d.)
(Abrahams et al., 2000).
Category Words in 1 minute Two major conclusions can be drawn: (1) in
18
general, Semantic Verbal Fluency is a task that is
Body-parts
Things to eat 17 quite sensitive to diverse types of brain abnorm-
Animals 16 alities, particularly those involving the left hemi-
Kitchen elements 14 sphere; (2) in addition to quantitative scores, a
Fruits 13 diversity of pathological phenomena may be
Colours 13
observed: intrusions, perseverations, difficulties
Places or sites 12
Adjectives 12 in maintaining instructions, etc. Qualitative ana-
Parts of the house 12 lysis should complement the quantitative scoring.
Transportation means 11
School elements 11
Verbs 11 EXPERIMENTAL CRITERION
Tools 10
Personal care elements 10 Diverse studies have been published, analysing the
Furniture 10
Flowers 8
brain areas involved in semantic verbal fluency
(Abrahams et al., 2003; Fama et al., 2000; Gaillard
et al., 2003; Hugdahl et al., 1999; Lurito, Kareken,
Lowe, Chen, & Mathews, 2000; Pihlajamaki et al.,
Aphasia Examination, Goodglass & Kaplan, 1972; 2000; Schlosser et al., 1998; Tucha, Smely, &
CERAD neuropsychological test battery, Morris Lange, 1999; Vingerhoets et al., 2003; Weiss et al.,
et al., 1989; Neuropsi, Ostrosky-Sols et al., 1999). 2003). There is converging evidence that verbal
Instructions are simple, and no special instruments fluency activates the left posterior aspect of the
(except a stopwatch, a pencil, and paper) are middle temporal gyrus, and other brain areas can
required. It is also a cheap test, accessible any- additionally participate, such as the left dorsolat-
where. eral prefrontal convexity and the hippocampus.
Pihlajamaki et al. (2000) observed that the medial
CLINICAL CRITERION temporal lobe is required for the process of
retrieval by category.
A number of brain conditions can affect the Semantic fluency tasks differ somewhat in their
performance in verbal fluency tests: focal brain application from the regular manner in which they
damage, especially frontal lobe pathology are applied in the clinical neuropsychology con-
(Herrmann, Ehlis, & Fallgatter, 2003; Ravnkilde, text. The tasks are applied with some changes,
Videbech, Rosenberg, Gjedde, & Gade, 2002), which are summarized in Table 2. Vascular
Parkinsons disease (Donovan, Siegert, & responses of the brain coupled to the task have
McDowall, 1999; Troyer et al., 1998a), schizo- to be switched on and off several times to obtain
phrenia (Chen, Chen, Chan, Lam, & Lie-Mak, enough statistical power to demonstrate activa-
2000; Curtis et al., 1998), subcortical dementia tion. The duration of the free retrieval is shor-
(Testa et al., 1998; Troster et al., 1998), traumatic tened, as the generation decays considerably in the
brain injury (Bradley, Torner, Fisher, & second half of the minute. Since a second attempt
Aharon-Peretz, 2001), Huntingtons disease (Ho to generate nouns within the same category would
et al., 2002; Suhr & Jones, 1998), depression add some difficulty, as the subject may run out of
(Crowe, 1992; Okada, Okamoto, Morionobu, salient nouns, the generation of animals is usually
Yamawaki, & Yokota, 2003; Ravdin, Katzen, followed by other categories. Usually an fMRI
TABLE 2
Task differences between regular and fMR laboratories
Test type Duration Off period Several items Motion restriction Speech Controlled
Regular laboratory 1 min None Not necessarily No Overt Yes

FMR laboratory 2040 s 2040 s Yes Yes Covert No
(usually 30 s)
consists of three or four on epochs, preceded, (Brodmanns 44, 9, and SMA). However, some
ended, and separated by off epochs of the same differences appear: Verb generation also elicited
duration. Thus, generating only animals during activation in Wernickes area, left caudate nucleus,
three different periods would require division of and contralateral Brocas area. Not all subjects
this category; for example, asking for the retrieval show the same differences, but there are cases in
of wild animals in the first attempt, then insects in which the activation is almost identical.
the second, and birds for the third. No differences according to language or gender
Semantic fluency, phonological fluency, and have been reported in this task. Vingerhoets et al.
verb generation activate left lateral frontal areas. (2003) analysed activation patterns in 12 multi-
Left Brodmanns 44, 45, 46, 9, 7, and 8 are lingual right-handed men performing a word
activated most frequently, with other areas varying fluency task, a picture naming task, a comprehen-
among different subjects. These ancillary areas are sion reading task, and their respective control
the supplementary motor area (SMA), posterior tasks in three languages (Dutch, French, and
temporal areas, cingulate gyrus, contralateral English) while whole-head fMRI was applied. In
frontopolar area, and visual areas. Figures 1 and general, all language tasks revealed predominantly
2 show activation obtained with a semantic fluency overlapping regions of activation for the different
languages. Weiss et al. (2003) observed that men
task and a verb generation task in the same

subject. The similarities are limited to the activa- and women who did not differ significantly in
tion of areas directly related to fluency
Figure 2. Brain activation of verb generation. This

image was obtained with the same techniques utilized for
the Semantic Fluency Task. The skull was not removed
in the post-processing. During the on blocks, the subject
Figure 1. Brain activation of semantic fluency. These was asked to think of verbs related to a list of nouns
clusters were obtained from a paradigm consisting of 3 aurally presented at 0.2 Hz. In addition to the areas
on epochs alternating with 3 off epochs. During on the activated in the Semantic Fluency Task shown in
subject was silently generating names of animals, Figure 1, there is activation the left caudate nucleus
vegetables, and fruits (respectively). The volunteer was and stronger participation of contralateral Broca.
asked to think of a blue sky during the off periods. The Wernickes area is expected to activate because of the
largest clusters of activation are located in left Broca, stimulus delivery mode. Notice the small activation in
left mouth premotor area, and SMA. Some activation the right frontopolar area (Brodmanns 10), which,
also appears in the left interparietal sulcus, right strikingly is common to both tasks, and probably related
Brodmanns 7, and contralateral Brocas area (from to working memory or retrieving (from the Miami
the Miami Childrens Hospital, department of radiol- Childrens Hospital, department of radiology;
ogy; www.mch.com/clinical/radiology/fmri). www.mch.com/clinical/radiology/fmri).
Verbal Fluency Task performance showed a very and delayed), backwards digits, and language
similar pattern of brain activation. comprehension. Correlation with naming is lower
than correlation with language understanding.
Interestingly, phonemic verbal fluency, but not
PSYCHOMETRIC CRITERION Semantic Verbal Fluency, significantly correlates
with the Finger Tapping Test (Ardila et al., 1994,
Three studies were selected to analyse the Semantic 1998). Some correlation with arithmetic and, in
Verbal Fluency pattern of correlations with other general, with conceptual functions is also
tests (Ardila et al., 1994, 1998; Ostrosky-Sols et al., observed. These data suggest that attention, work-
1999) (Tables 3, 4, and 5). It is observed that ing memory, and immediate and delayed verbal
Semantic Verbal Fluency correlates with phono- memory are cognitive functions that contribute to
logic verbal fluency, verbal memory (immediate semantic verbal fluency.
TABLE 3
CONCLUSIONS
Statistically significant correlations over .15 (p , .01),
Semantic Verbal Fluency and other cognitive tests (Ardila
et al., 1998) Recent decades have seen an exponentially grow-

ing interest in developing valid and reliable
Test r
neuropsychological testing instruments that can
Phonologic Verbal Fluency .31 be used in different clinical settings. An enormous
WAIS: Arithmetic .27 diversity of tests has been developed (see, for
Information .22
example, Lezak, 2004). Simultaneously, the need
Digits .19
Perceptual speed: Similarities .18 to adapt these instruments to cultural and
Verbal IQ .18
Full Scale IQ .19
TABLE 5
Correlations between semantic verbal fluency and other
neuropsychological tests in 98 subjects aged 1112 years
TABLE 4 (Ardila et al., 1994)
Correlations between Semantic Verbal Fluency (ANIMALS)
and other neuropsychological tests in the NEUROPSI Subtest r
neuropsychological test battery (adapted from Ostrosky .50
Phonologic verbal fluency
et al., 1999)
Wechsler Memory Scale:
Subtest r Information .07
Orientation 2.17
Orientation: Time .29 Mental control .01
Place .16 Logical memory .25
Person .12 Digits .22
Attention: Digits backwards .40 Visual reproduction .22
Visual detection .38 Associative learning .05
20 minus 3 .37 Logical memory: delayed .19
Encoding: Verbal memory .34 Associative learning: delayed .05
Copy of a semi-complex figure .43 Visual reproduction: delayed 2.10
Language: Naming .25 Verbal memory
Repetition .14 Maximum .44
Comprehension .45 Delayed recall .46
Verbal fluency: Phonologic .61 Praxis
Reading .30 Left hand .05
Writing: Dictation .14 Right hand 2.17
Copy .18 Tapping test
Conceptual functions: Similarities .38 Right hand 2.04
Calculation abilities .33 Left hand 2.19
Sequences .41 Figure recognition 2.03
Motor functions: Changing left-hand position .22 Token test .11
Changing right-hand position .21 WCST
Alternating movements .39 Correct responses .25
Opposite reactions .12 Categories .01
Memory: Words .35 Boston Naming test .13
Cueing .21 Rey-Osterrieth Complex Figure
Recognition .10 Copy .13
Semi-complex figure .42 Immediate recall .03
linguistic contexts other than the context for our understanding of current neuropsychological
which they were originally developed has become instruments.
evident. This effort has been observed in many It can be anticipated that, in the future, some
different countries worldwide. instruments may become relatively universal tests
Until recently, test ability to detect brain in neuropsychology. It can be also anticipated that
pathology was enough. Today, sophisticated some instruments may be usable only within
neuroimaging techniques are available, which certain cultural and/or linguistic contexts. This is
could potentially contribute to extending our an open question, and much more research is
understanding of cognitive testing procedures. required.
Furthermore, today we have extended databases
for many neuropsychological instruments, which
can be used to have a better understanding of the
variables affecting test performance and the score REFERENCES
dispersions that can be found.
Abrahams, S., Goldstein, L. H., Simmons, A.,
We are proposing that neuropsychological tests
Brammer, M. J., Williams, S. C., Giampietro, V.
should fulfil four different criteria: to have a large P., Andrew, C. M., & Leigh, P. N. (2003). Functional
enough normative database; to know the effects of magnetic resonance imaging of verbal fluency and
different types of brain damage on this test; to confrontation naming using compressed image acqui-
know how the brain is activated when the test is sition to permit overt responses. Human Brain
Mapping, 20, 2940.
performed; and to know how this test correlates Abrahams, S., Leigh, P. N., Harvey, A., Vythelgum,
with other cognitive tests). This means that the G. N., Grise, D., & Goldstein, L. H. (2000). Verbal
test: Fluency and executive dysfunction in amyotrophic
lateral sclerosis (ALS). Neuropsychologia, 38,
1. Has been administered to a diversity of 734747.
populations, and diverse normative data are Ardila, A. (1999). A neuropsychological approach to
available. The test has been demonstrated to be intelligence. Neuropsychology Review, 9, 117136.
appropriate to evaluate different populations Ardila, A., Galeano, L. M., & Rosselli, M. (1998).
Toward a model of neuropsychological activity.
(normative criterion). Simpler tests with simple Neuropsychology Review, 8, 171190.
instructions have a better chance of becoming Ardila, A., Ostrosky-Solis, F., & Bernal, B. (2005). The
widely used tests. effect of culture, language and education on
2. Has been demonstrated to be clinically Semantic Verbal Fluency (ANIMALS). Archives of
useful (clinical criterion). This means that Clinical Neuropsychology. Manuscript submitted for
publication.
specific patterns of performance and error types Ardila, A., Ostrosky-Solis, F., Rosselli, M., & Gomez, C.
on the test are known to be associated with specific (2000). Age related cognitive decline during normal
brain pathologies. aging: The complex effect of education. Archives of
3. Has been used experimentally in normal Clinical Neuropsychology, 15, 495514.
populations, and the pattern of brain activation Ardila, A., Rosselli, M., & Bateman, J. R. (1994).
Factorial structure of cognitive activity. Behavioural
correlated with the performance in this test is well Neurology, 7, 4958.
known (experimental criterion). Bradley, N. A., Torner, R., Fisher, T., & Aharon-
4. Correlates with other assessment tests Peretz, J. (2001). Preliminary analyses of Hebrew
(psychometric criterion). Validity of the test verbal fluency measures. Applied Neuropsychology, 8,
248250.
would be established not only based on the
Bruyer, R., & Tuyumbu, B. (1980). Fluence verbale et
intercorrelation with other tests, but also from lesions du cortex cerebrale: performances et types
the specific pattern of performance in the test derreurs. Encephale, 6, 287197.
found in individuals with different types of Chen, R. Y. L., Chen, E. Y. H., Chan, C. K. Y., Lam,
pathologies. L. C. W., & Lie-Mak, F. (2000). Verbal Fluency in
schizophrenia: Reduction in semantic store.
We have taken the example of Semantic Verbal Australian and New Zealand Journal of Psychiatry,
34, 4348.
Fluency using the category ANIMALS. But, Crowe, S. F. (1992). Dissociation of two frontal lobe
potentially, the same type of analysis could be syndromes by a test of verbal fluency. Journal of
carried out with other basic neuropsychological Clinical and Experimental Neuropsychology, 14,
tests; for example, the Token Test, the Wisconsin 327339.
Card Sorting Test (using a simpler version), or the Cooper, D. B., Epker, M., Lacritz, L., Weine, M.,
Rosenberg, R. N., Honig, L., & Cullum, C. M.
naming tests (assuming that it is possible to (2001). Effects of practice on category fluency in
develop a good cross-cultural and cross-linguistic Alzheimers disease. The Clinical Neuropsychologist,
naming test). This type of analysis may advance 15, 125128.
Curtis, V. A., Bullmore, E. T., Bramer, M. J., Wright, I. Nell, V. (2000). Cross-cultural neuropsychological assess-
C., Williams, S. C., Morris, R. G., Sharma, T. S., ment: Theory and practice. Hillsdale, NJ: Lawrence
Murray, R., & McGuire, P. K. (1998). Attenuated Erlbaum Associates Inc.
frontal activation during a verbal fluency task in Okada, G., Okamoto, Y., Morinobu, S., Yamawaki, S.,
patients with schizophrenia. The American Journal of & Yokota, N. (2003). Attenuated left prefrontal
Psychiatry, 155, 10561063. activation during a Verbal Fluency Task in patients
Donovan, K., Siegert, R., & McDowall, J. (1999). with depression. Neuropsychobiology, 47, 2126.
Clustering and switching in verbal fluency in Ostrosky-Sols, F., Ardila, A., & Rosselli, M. (1999).
Parkinsons disease. New Zealand Journal of Neuropsi: A brief neuropsychological test battery
Psychology, 28, 6166. in Spanish with norms by age and educational level.
Fama, R., Sullivan, E. V., Shear, P. K., Cahn-Weiner, Journal of the International Neuropsychological
D. A., Marsh, L., Lim, K. O., Yesavage, J. A., Society, 5, 413433.
Tinklenberg, J. R., & Pfefferbaum, A. (2000). Ostrosky-Sols, F., Ardila, A., Rosselli, M., Lopez, G.,
Structural brain correlates of verbal and nonverbal & Mendoza, V. (1998). Neuropsychological test
fluency measures in Alzheimers disease. performance in illiterates. Archives of Clinical
Neuropsychology, 14, 2940. Neuropsychology, 13, 645660.
Gaillard, W. D., Sachs, B. C., Whitnah, J. R., Ostrosky-Sols, F., Ramirez, M., Lozano, A.,
Ahmad, Z., Balsamo, L. M., Petrella, J. R., Picasso, H., & Velez, A. (2004). Culture or education:
Braniecki, S. H., McKinney, C. M., Hunter, K., A study with indigenous Maya population.
Xu, B., & Grandin, C. B. (2003). Developmental International Journal of Psychology, 39, 3646.
aspects of language processing: fMRI of verbal Parks, R. W., Loewenstein, D. A., Dodrill, K. L.,
fluency in children and adults. Human Brain Barker, W. M., Yoshii, F., Chang, J. Y., Emran, A.,
Mapping, 18, 176185. Apicella, A., Sheramata, W. A., & Duara, R. (1988).
Goodglass, H., & Kaplan, E. (1972). The assessment of Cerebral metabolic effects of verbal fluency tests: A
aphasia and related disorders. Philadelphia: Lea & PET scan study. Journal of Clinical and Experimental
Febiger. Neuropsychology, 10, 565575.
Herrmann, M. J., Ehlis, A. C., & Fallgatter, A. J. (2003). Perret, E. (1974). The left frontal lobe of man and the
Frontal activation during a Verbal Fluency task as suppression of habitual responses in verbal cat-
measured by near-infrared spectroscopy. Brain egorical behavior. Neuropsychologia, 12, 323330.
Research Bulletin, 61, 5156. Pihlajamaki, M., Tanila, H., Hanninen, T.,
Ho, A. K., Sahakian, B. J., Robbins, T. W., Barker, Kononen, M., Laakso, M., Partanen, K.,
R. A., Rosser, A. E., & Hodges, J. R. (2002). Verbal Soininen, H., & Aronen, H. J. (2000). Verbal fluency
fluency in Huntingtons disease: A longitudinal activates the left medial temporal lobe: A functional
analysis of phonemic and semantic clustering and magnetic resonance imaging study. Annals of
switching. Neuropsychologia, 40, 12771284. Neurology, 47, 470476.
Hugdahl, K., Lundervold, A., Ersland, L., Smievoll, Radvin, L. D., Katzen, H. L., Agraval, P., & Relkin, N.
A. I., Sundberg, H., Barndon, R., & Roscher, B. E. R. (2003). Letter and semantic fluency in older
(1999). Left frontal activation during a semantic adults: Effects of mild depressive symptoms and
categorization task: An fMRI-study. International age stratified normative data. The Clinical
Journal Neuroscience, 99, 4958. Neuropsychologist, 17, 195202.
Goodglass, H., & Kaplan, E. (1972). The assessment of Ramier, A. M., & Hecaen, H. (1970). Role respectif des
aphasia and related disorders. Philadelphia: Lea & attaintes frontales et la lateralisation lesionelle dans
Febiger. le deficits de la fluence verbale. Revue
Lezak, M. D. (2004) Neuropsychological assessment (4th Neurologique, 123, 1722.
ed.). New York: Oxford University Press. Ravnkilde, B., Videbech, P., Rosenberg, R., Gjedde, A.,
Lopera, F. (n.d.). Fluidez verbal en diferentes categoras & Gade, A. (2002). Putative test of frontal lobe
semanticas [Verbal fluency in different semantic function: A PET-study of brain activation during a
categories]. Unpublished data. Stroops test and Verbal Fluency. Journal of Clinical
Lurito, J. T., Kareken, D. A., Lowe, M. J., Chen, S. H., and Experimental Neuropsychology, 24, 534547.
& Mathews, V. P. (2000). Comparison of rhyming Rosselli, M., Ardila, A., & Rosas, P. (1990).
and word generation with fMRI. Human Brain Neuropsychological assessment in illiterates II:
Mapping, 10, 99106. Language and praxic abilities. Brain and Cognition,
Matute, E., Rosselli, M., & Ardila, A. (2004). Verbal 12, 281296.
and nonverbal fluency in 6- to 15-year-old Spanish- Ruff, R. M., Allen, C. C., Farrow, C. E., Niemann, H.,
speaking children. Developmental Neuropsychology, & Wylie, T. (1994). Figural fluency: Differential
26, 647660. impairment in patients with left versus right frontal
Mitrushina, M. N., Boone, K. B., & DElia, L. F. lobe lesions. Archives of Clinical Neuropsychology, 9,
(1999). Handbook of normative data for neuropsycho- 4155.
logical assessment. New York: Oxford University Schlosser, R., Hutchinson, M., Joseffer, S., Rusinek, H.,
Press. Saarimaki, A., Stevenson, J., Dewey, S. L., &
Morris, J. C., Heyman, A., Mohs, R. C., Hughes, J. P., Brodie, J. D. (1998). Functional magnetic resonance
Van Belle, G., Fillenbaum, G., Mellits, E. D., imaging of human brain activity in a verbal fluency
Clark, C., the CERAD investigators, (1989). The task. Journal of Neurology, Neurosurgery and
Consortium to Establish a Registry for Alzheimers Psychiatry, 64, 492498.
Disease (CERAD). Part I. Clinical and neuropsy- Spreen, O., & Strauss, E. (1998). A compendium of
chological assessment of Alzheimers disease. neuropsychological tests (2nd ed.). New York: Oxford
Neurology, 39, 11591165. University Press.
Suhr, J. A., & Jones, R. D. (1998). Letter and semantic Tucha, O. W., Smely, C. W., & Lange, K. W. (1999).
fluency in Alzheimer, Huntington and Parkinsons Verbal and figural fluency in patients with mass
disease. Archives of Clinical Neuropsychology, 13, lesions of the left or right frontal lobes. Journal of
447454. Clinical and Experimental Neuropsychology, 21,
Testa, J. A., Troster, A. I., Fields, J. A., Gleason, A. C., 229236.
Salmon, D., & Beatty, W. W. (1998). Semantic Vingerhoets, G., Van Borsel, J., Tesink, C., Van den
fluency performance of patients with cortical and Noort, M., Deblaere, K., Seurinck, R.,
subcortical neurodegenerative diseases. Aging, Vandemaele, P., & Achten, E. (2003).
Neuropsychology and Cognition, 5, 203214. Multilingualism: An fMRI study. Neuroimage, 20,
Troster, A. I., Fields, J. A., Testa, J. A., Paul, R. H., 21812196.
Blanco, C. R., Hames, K. A., Salmon, D. P., & Warburton, E., Wise, R. J. S., Price, C. J., Weiller, C.,
Beatty, W. W. (1998). Cortical and subcortical Harad, U., Ramsay, S., & Frackowiak, R. S. J.
influences on clustering and switching in the perfor- (1996). Noun and verb retrieval in normal subjects
mance of verbal fluency tasks. Neuropsychologia, 36, studies with PET. Brain, 119, 159179.
295304. Wechsler, D. (1997). WAIS-III: Administration and
Troyer, A. K., Moscovitch, M., & Winocur, G. (1998a). scoring manual. San Antonio, TX: The
Clustering and switching on verbal fluency tests in Psychological Corporation.
Alzheimers and Parkinsons disease. Journal of the Weiss, E. M., Siedentopf, C., Hofer, A.,
International Neuropsychological Society, 4, 137143. Deisenhammer, E. A., Hoptman, M. J., Kremser, C.,
Troyer, A. K., Moscovitch, M., Winocur, G., Golaszewski, S., Felber, S., Fleischhacker, W. W., &
Alexander, M. P., & Stuss, D. (1998b). Clustering Delazer, M. (2003). Brain activation pattern during a
and switching on verbal fluency: The effects of focal verbal fluency test in healthy male and female
frontal- and temporal-lobe lesions. Neuropsychologia, volunteers: A functional magnetic resonance imaging
25, 388493. study. Neuroscience Letters 14.
Literature Review
Illiteracy: The Neuropsychology of Cognition Without Reading
Alfredo Ardila 1, *, Paulo H. Bertolucci 2, Lucia W. Braga 3, Alexander Castro-Caldas 4, Tedd Judd 5,
Mary H. Kosmidis 6, Esmeralda Matute 7, Ricardo Nitrini 8, Feggy Ostrosky-Solis 9, Monica Rosselli 10
1
2
Department of Neurology and Neurosurgery, Sao Paulo Universidade Federal de Sao Paulo, Sao Paulo, Brazil
3
SARAH Network, Brasilia, Brazil
4
Portuguese Catholic University, Lisbon, Portugal
5
Seattle Pacific University, Seattle, WA, USA
6
School of Psychology, Aristotle University of Thessaloniki, Thessaloniki, Greece
Downloaded from acn.oxfordjournals.org by guest on November 12, 2010

7
Universidad de Guadalajara, Guadalajara, Mexico
8
University of San Paulo, Sao Paulo, Brazil
9
National Autonomous University of Mexico, Mexico DF, Mexico
10
Psychology, Florida Atlantic University, Davie, FL, USA
*Authors are presented in alphabetical order pointing to a similar contribution, except that the project was initiated and coordinated by the first author.
Corresponding author at: Department of Communication Sciences and Disorders, Florida International University,
AHC3-431B, Miami, FL 33199, USA. Tel.: +1-305-551-7975.
E-mail address: ardilaa@fiu.edu (A. Ardila).
Accepted 17 September 2010
Abstract
Illiterates represent a significant proportion of the worlds population. Written language not only plays a role in mediating cognition, but
also extends our knowledge of the world. Two major reasons for illiteracy can be distinguished, social (e.g., absence of schools), and personal
(e.g., learning difficulties). Without written language, our knowledge of the external world is partially limited by immediate sensory infor-
mation and concrete environmental conditions. Literacy is significantly associated with virtually all neuropsychological measures, even
though the correlation between education and neuropsychological test scores depends on the specific test. The impact of literacy is reflected
in different spheres of cognitive functioning. Learning to read reinforces and modifies certain fundamental abilities, such as verbal and visual
memory, phonological awareness, and visuospatial and visuomotor skills. Functional imaging studies are now demonstrating that literacy and
education influence the pathways used by the brain for problem-solving. The existence of partially specific neuronal networks as a probable
consequence of the literacy level supports the hypothesis that education impacts not only the individuals day-to-day strategies, but also the
brain networks. A review of the issues related to dementia in illiterates is presented, emphasizing that the association between the education
level and age-related cognitive changes and education remains controversial. The analysis of the impact of illiteracy on neuropsychological
test performance represents a crucial approach to understanding human cognition and its brain organization under normal and abnormal
conditions.
Keywords: cognitive abilities; dementia; education; illiteracy; neuropsychology tests; schooling
Introduction
Illiterates represent a significant proportion of the worlds population. In 2000 2004, close to onefifth of the worlds popu-
lation was illiterate: 13% of men over 15 years of age and 23% of women (UNESCO, 2005). Only a few centuries ago, reading
and writing abilities were simply uncommon among the general population. Writing has only a 6,000-year history; for about
95% of our history, Homo sapiens were preliterate. We hypothesize that the acquisition of literacy skills has somehow changed
the brain organization of cognition.
# The Author 2010. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org.
doi:10.1093/arclin/acq079
690 A. Ardila et al. / Archives of Clinical Neuropsychology 25 (2010) 689712
Table 1. World-wide illiteracy rate during 19702015 (projected) (UNESCO, 2005)
Year Illiteracy rate (%)
1970 37
1980 30
1990 25
2000 20
2005 18
2010 16
2015 15
When studying illiteracy, the most obvious factor is the generational effect found in most countries (Table 1). Elders in most
societies generally have a lower-educational level than their children and grandchildren. Moving back in history, the duration
of education is progressively shorter on average.
Reasons for Illiteracy

Two major classes of reasons account for illiteracy:
(a) Social reasons. Illiteracy may be due to the absence of schools, social disapproval of literacy, child labor, and/or
poverty.
(b) Personal reasons. For a subgroup of illiterates, learning difficulties, mental retardation, significant motor and/or
sensory problems, early central nervous system pathologies, and similar conditions may account for their failure
to learn to read despite adequate exposure to education.
There is a significant diversity in literacy rates among countries, from 99% (e.g., Japan, Canada, and Finland) to below 20%
(e.g., Mali, Niger, and Burkina Faso; International Adult Literacy Survey, 1998). Illiteracy due to personal reasons (learning
difficulties, mental retardation, significant motor and/or sensory problems, early central nervous system pathologies, and
similar conditions) likely accounts for only a small percentage of illiteracy in industrialized countries and potentially somewhat
higher in poor countries with increased prevalence of early central nervous system pathologies.
These two main classes of reasons for illiteracy present potential confounds for research. Populations that are illiterate for
social reasons are likely to differ in important ways other than literacy from the schooled, literate populations that have been the
subjects of most research in cognitive psychology (prototypically U.S. college students). In general, illiterates are more likely
to be of a lower socioeconomic class, they may have a variety of health problems, they may have less exposure to media of
communication, and they may have limited experience with testing and, thus, distrust of researchers, among other differences.
Those who are illiterate due to personal reasons are quite likely to have atypically functioning brains and cognitive systems.
One way around this problem is to compare individuals to themselves rather than to other populations. Studies of adults
before and after they acquire literacy can help to control for these factors and can furthermore control, to a large extent, for
the effects of development. An alternative is the sample used in a series of Portuguese studies (e.g., Reis & Castro-Caldas,
1997) comparing sisters, one of whom is literate and the other illiterate due to social demands placed on the eldest daughter
to take on the responsibility of child care of younger siblings.
Why Is It Important to Study Illiteracy?
Different fundamental and practical reasons to study illiteracy could be distinguished (Olson & Torrance, 2009). There are,
however, at least four important reasons to study the neuropsychology of illiteracy:
(a) (Neuro)cognitive Theory. The study of illiteracy can contribute to a broader understanding of the organization of
cognition.
(b) Clinical Neuropsychology. The analysis of illiteracy can help to discern the influence of both literacy and schooling
on cognitive test performance. Educational variables significantly influence performance on a diversity of neurop-
sychological tests (Lezak, Howieson, & Loring, 2004).
A. Ardila et al. / Archives of Clinical Neuropsychology 25 (2010) 689712 691
(c) Teaching Literacy. Understanding the basic brain mechanisms, subserving cognition in illiterates may contribute to
improving reading and writing teaching methods for illiterate adults (e.g., Ardila, Ostrosky & Mendoza, 2000).
(d) Literacy and Neuropsychological Functioning. Exploring the effects of education on literacy and neuropsychologi-
cal functioning.
This article will focus primarily on those aspects of the study of illiteracy most pertinent to the neurocognitive theory, and
clinical neuropsychology, as well as the analysis of the effects of education on neuropsychological functioning.
Conceptual Issues in the Study of Illiteracy
A fundamental conceptual presumption of neuropsychology is that the potential for basic cognitive abilities, and corre-
spondingly, their brain mechanisms, are universal and inherent for any human being with normal brain development, regardless
of the specific language spoken and ones environmental conditions. Literacy (i.e., extending spoken language to a symbolic
visual representation) plays a major role in mediating cognitive processes. Luria (1931, 1933, 1966, 1973) and Vygotsky (1934/
1978) developed the concept of extracortical organization of higher mental functions to account for the interaction of bio-
logical and cultural factors in the development of human cognition (Kotik-Friedgut & Ardila, 2004).

Historical Development of Writing and Schools
Historically, writing is thought to represent an evolution of painting, whereas painting is the evolution of representation
capabilities, intentional gestures, and visually guided movements (No author, 1993). Wall paintings appeared in Europe
during the Paleolithic era, some 30,000 35,000 years ago (Childe, 1936) representing mainly animals, but also people, instru-
ments, and environmental conditions. Prewriting evolved with the gradual standardization of paintings to represent specific
elements (i.e., a standard bird reflected the general category bird). This suggests an emerging human ability to imitate
complex behavioral patterns (Blackmore, 1999). Thus, writing began with concrete pictograms that reflected realities acces-
sible to the senses, particularly to vision (Lecours, Pena-Casanova, & Ardila, 1998). These pictograms further evolved and
became abstract, progressively diverging from concrete representation toward representation of the phonological structure
of language. Todays simple and straightforward alphabetic systems in fact are the outcome of a very long and complex
process that took many centuries to reach its present form (Haarmann, 1991). The earliest generally accepted beginning of
writing in human history was in Sumer (contemporary Iraq) about 53 centuries ago (Schmandt-Besserat, 1996; but see
Haarmann, 2008, regarding earlier Danube script; and Li, 2003, regarding earlier Chinese). Symbols (graphemes) reflected
the meaning of words, so these original writing systems are regarded as logographic. Graphemes representing sounds (sylla-
bles) appeared about 4,000 years ago in Phoenicia (Sampson, 1985), and graphemes representing phonemes appeared even
later in Greece. Over the course of this transition, several languages developed with a coexistence of ideograms and phonolo-
gical representations, such as cuneiform writing and the Egyptian hieroglyphic system (Haarmann, 1991). Writing systems
evolved independently in pre-Columbian civilizations in Central America, most notably Mayan writing (Coe & Vanstone,
2005). The sequence of the evolution of writing in consequence was:
Drawings. pictograms. logograms. syllabic graphemes. phonemic graphemes
Several existing writing systems use a mixture of logograms, syllabic graphemes, phonemic graphemes, and extraphonemic
markers such as capitalization, punctuation, and spatial layout. The relative emphasis on each level of representation varies
from system to system. For example, the Chinese system is predominantly based on logograms with some phonemic gra-
phemes, and extraphonetic markers, and with an alternate phonemic grapheme system available. Likewise, the above sequence
is not meant to imply that any one system is more efficient or superior.
Thus, initial writing (or rather, prewriting) was a visuoconstructive ability (i.e., representing external elements visually), and
only later did it become an ideomotor praxis ability (i.e., making specific learned and fixed sequences of movements with the
hand to create a pictograma standardized representation of external elements). Still later, writing became a linguistic ability
(i.e., associating the pictogram with a word, and further analyzing the word into its constituent sounds; Ardila, 2004). It is not
surprising that three major disorders in writing can be observed as a result of brain pathology: Visuoconstructive (spatial or
visuospatial agraphia: Impairment in letter, word, and text formation due to disturbed visuospatial skills), ideomotor
(apraxic agraphia: Selective disorder of letter formation caused by ideomotor apraxia), and linguistic (aphasic agraphia:
Writing impairment due to a language disturbance; Table 2). The evolution of writing has continued with the development
Table 2. Representational forms of writing and corresponding cognitive skills
Representational form Historical example Neurocognitive skill Form of agraphia
Drawings pictograms Cave paintings Egyptian hieroglyphics Visuoconstructive ability Visuospatial

Logograms Sumerian Ideomotor praxis Apraxic
Syllabic graphemes Japanese Kana Linguistic ability Aphasic
Phonemic graphemes Spanish
of different technical instruments for writing: The stylus for etching in clay, feathers, the pencil, the typewriter, and the com-
puter, as well as the development of different writing styles (e.g., handwriting, script, uppercase, lowercase) and different
media (e.g., clay, papyrus, paper, computer screens). Moreover, images first and writing later allowed the possibility of rep-
resentation; thus, they enable humans to think and act beyond their immediate needs (Suddendorf & Corballis, 1997) and
promote abstract thinking.
Defining Literacy

The definition of literacy is not definitive, nor is there an overt cut-off point; consequently, intermediate stages, such as
semiliteracy and functional illiteracy, are often recognized. UNESCO (http://www.uis.unesco.org/i_pages/indspec/
TecSpe_literacy.htm) defines adult illiteracy as the percentage of the population aged 15 years and over who cannot both
read and write a comprehensible short simple statement on their everyday life. Literate does not necessarily mean schooled,
even though literacy is usually highly associated with formal schooling. Reading can be transmitted from parents or tutors
to children without formal school attendance (Berry & Bennett, 1992; Scribner & Cole, 1981). Under normal conditions,
reading is not learned by simple exposure, even though illiterates can develop some ability to holistically recognize a few
highly frequent words, such as the name of their country, and the name of some popular commercial products (Goldblum
& Matute, 1986). Extremely infrequently, the ability to read can be acquired without any deliberate training (so-called hyper-
lexia usually associated with autism, Martos-Perez & Ayuda-Pascual, 2003; Nation, 1999).
The degree to which contemporary illiterates are confined to access to immediate sensory information probably also reflects
the degree to which they are confined in their access to various media of information and communication and, possibly, also to
formal education. Da Silva, Petersson, Fasca, Ingvar, and Reis (2004) demonstrated in Portugal that whereas literates and illit-
erates perform equally well on a fluency task asking for elements known through direct sensory information (items that can be
found in a supermarket), the two groups differed on a semantic fluency task requiring naming animals. In contrast, literate
people can name dinosaurs, camels, or kangaroos, evidently unknown in Portugal, but learned by reading; whereas illiterates
can only name cats, dogs, horses, and other animals existing in the immediate surrounding environment and known through
direct sensory experience. When the information is related to real life and direct experience, it can be considerably easier
to understand. Thus, for the illiterate person, it is easier to repeat a word than to repeat a pseudoword (Castro-Caldas,
Petersson, Reis, Stone-Elander, et al., 1998; Kosmidis, Tsapkini, & Folia, 2006) and to remember objects presented to
them visually than word lists (Folia & Kosmidis, 2003).
The impact of literacy is reflected in all spheres of cognitive functioning. Learning to read reinforces certain fundamental
abilities, such as verbal and visual memory, phonological awareness, executive functioning, and visuospatial and visuomotor
skills (Bramao et al., 2007; Matute, Leal, Zarabozo, Robles, & Cedillo, 2000; Petersson, Reis, Askelof, Castro-Caldas, &
Ingvar, 2000; Petersson et al., 2001). It is not surprising that illiterate people underachieve on cognitive tests that assess
these particular abilities. Furthermore, attending school also reinforces certain attitudes and values that may speed the learning
process, such as the attitude that memorizing information is important or that knowledge is highly valuable; that learning is a
step-by-step process moving from the simple to the more complex. It has been emphasized that schooling improves an indi-
viduals ability to explain the basis of performance in cognitive tasks (Laboratory of Comparative Human Cognition, 1983).
Whether a fundamental aim of education, or a by-product, the fact remains that all schools and schooling, regardless of their
location, have a common influence on those who go through the system, namely, they train students and reinforce certain
values. Hence, school may be seen as a culture unto itself, a transnational culture, and the culture of school.
School not only teaches, but also helps in developing certain strategies and attitudes that will be useful for future learning.
Ciborowski (1979) observed that schooled and nonschooled children can learn a new rule equally well, but once acquired,
schooled children tend to apply it more frequently in subsequent similar cases. School attendance, however, does not mean
that educated people simply possess certain abilities that less-educated individuals lack. It does not mean that highly educated
people have the same abilities as less-educated individuals, plus something else (Ardila, Ostrosky & Mendoza, 2000). The
individual with no formal education most likely has acquired knowledge, skills, and values that educated people have not.
Nonetheless, formal cognitive testing evaluates those abilities in which educated people were trained; therefore, it is not sur-
prising that they will outperform those with no formal education. In fact, one study demonstrated that once the effects of edu-
cation and literacy were separated from each other, some types of language processing were enhanced by any level of literacy,
whereas others continued to improve with an increasing level of education (Kosmidis et al., 2006). It is noteworthy that the
educational level has a substantial relationship with performance on some cognitive tests, but is not systematically related
to everyday problem-solving (functional criterion of intelligence; Cornelious & Caspi, 1987). As a matter of fact, when func-
tionally oriented tests, such as the Rivermead Behavioural Memory Test (Wilson, Cockburn, & Baddeley, 1985) and the Fuld
Object Memory Evaluation (Fuld, 1980), are administered to people with low education, no significant differences with high-
educated people have been observed (Yassuda et al., 2009).
Although schooling represents, in some respects, a transnational culture, that culture is not uniform. Literacy and numeracy
are almost universal as the major focus of primary education, but beyond that the qualities of the educational experience vary
considerably. Wealthy educational systems are likely to have low student:teacher ratios, abundant teaching materials, extensive
communication outside the school (Internet, field trips, visiting instructors), ancillary services (psychologists, therapists, parent
services), and instruction in a wide variety of subjects and skills. Schools with less funding often lack these resources. Manly
(2006) has demonstrated a relationship between school expenditure and neuropsychological test performance.
School qualities depend on much more than funding. Schools vary greatly in their formal curricula and also in unwritten

characteristics. There is much variability in schools behavioral expectations and institutional cultures, as well as in their
emphasis on bilingualism, visual-spatial skills, abstract reasoning, problem-solving, speeded performance, social skills, indi-
vidualism, collectivism, moral values, and religious instruction, among other themes. These varying characteristics can reason-
ably be expected to have an impact on neuropsychological test performance, and this impact is not likely to be entirely linear
nor unifactorial, but these relationships have not yet been systematically explored.
Contemporary illiteracy is not the same as historical preliteracy. Literacy facilitates a number of cognitive technologies that
may have replaced preliterate cognitive skills. Those preliterate cognitive skills may require intact preliterate societies and may
be extinct or vestigial in the contemporary world and in contemporary marginalized illiterates. A functioning illiterate or pre-
literate society may be required to foster the full flowering of such skills. For example, there are many cultural traditions of
bards who memorize long poems and pass them on through generations (Lord, 1995). We tend to think of illiterates as concrete
thinkers, yet the fables, proverbs, myths, idioms, and even the metaphors built into the very structure of our languages that have
come to us from largely preliterate societies from millennia ago are often quite abstract.
Literacy Versus Schooling
Very important cognitive consequences of learning to read and write have been suggested: Changes in visual perception,
logical reasoning (i.e., executive functions), remembering strategies (Laboratory of Comparative Human Cognition, 1983),
as well as access to lexical storage and explicit phonological processing (Kosmidis et al., 2006) and improved working
memory. Schooling appears to influence formal operational thinking (Laurendeau-Bendavid, 1977) and functional brain organ-
ization (Castro-Caldas, Petersson, Reis, Stone-Elander, et al., 1998; Ostrosky-Solis, Ramirez, Lozano, Picasso, & Velez, 2004)
and to reinforce particular attitudes and values related to learning. Conversely, training these abilities may make it easier to
learn to read and write. Interestingly, adults who received training in those abilities in which illiterates significantly underscore
(e.g., verbal memory, phonological abstraction, etc.) showed a significant improvement in neuropsychological test scores and a
facilitation in their learning-to-read process (Ostrosky-Sols et al., 2003).
Literacy without schooling: The Vai. Scribner and Cole (1981) attempted to separate the effects of literacy from the effects of
formal schooling by studying Vai people in Liberia who were literate in the Vai script, but who had not attended school. Vai
people have their own script. The script is taught at home, rather than school, allowing the researchers to separate school-based
education from literacy. Indeed, there are three educational systems in the Vai culture: (a) Traditional socializationthe bush
school, taught by men for boys, and by women for girls; (b) English schoolingmuch like American schooling, and (c)
Quranic schoolingconducted in Arabic. They found that there were no general effects of literacy on a battery of cognitive
tests, but performance on some tests was related to particular features of the Vai script and literacy practices. Scribner and Cole
proposed that there are definite cognitive skills associated with literacy, but not necessarily with classroom learning. And these
cognitive skills are dictated by each culture and situation. Berry and Bennet (1989) carried out a partial replication of this study
among the Cree of Northern Ontario.
Comparison of groups with low and high education
Kosmidis, Tsapkini, Folia, Vlahou, and Kiosseoglou (2004) and Kosmidis et al. (2006) attempted to disentangle the effect of
education from literacy in a sample of urban dwellers (illiterates, low-education literates, and high-education literates). They
found that education influenced lexical decision-making, concluding that the greater the amount of education, the more effec-
tive access to, and, perhaps, the greater the size of the lexical store. In contrast, literacy influenced the capacity of the phono-
logical loop: Low- and high-education literates performed equally well on pseudoword repetition, and better than illiterates.
Therefore, they concluded that literacy influences the effectiveness of the direct route from the auditory analysis system to
the phoneme level, whereas education influences lexical processing. Similarly, Kosmidis and colleagues (2004) found that
semantic processing was affected by the level of education attained, whereas phonological processing depended on whether
the individual had attained symbolic representation per se through learning graphemephoneme correspondence or had
been exposed to formal schooling.
Neither literacy nor school attendance is a simple and linear variable. Reading ability is not only correlated with training
time, but also with the idiosyncrasies of the reading system. For example, learning to read is easier in a transparent writing
system (regular spelling, e.g., Spanish, German, Greek) than in an opaque writing system (irregular spelling, e.g., English,
French; Paulesu et al., 2001). Furthermore, school attendance goes far beyond learning to read, write, and calculate. School
provides background knowledge in many areas (geography, history, language, etc.), but also contributes to developing learning

skills. In school, children are exposed to information that is neither part of their immediate environment nor part of their direct
experience. They learn how to establish taxonomies of their newly learned concepts, they learn to use paper and pencil, and
they use a language different from that used in their everyday life (Montiel & Matute 2006). School develops and reinforces
certain cultural and socialization values by creating alternative parental and authority images; certain forms of group behavior;
attitudes toward government, church, or other institutions sponsoring their education; and the process of being tested (Serpell,
1993). Children learn from their teachers, their peers, the group activities at school, and the new context.
The Illiterate Brain
Two different approaches can be used to understand the potential differences between the literate and the illiterate brain:
Functional and anatomic measures.
Traditional Functional Measures: Dichotic Listening
Several studies have used dichotic listening paradigms to study the potential influence of literacy on cerebral lateralization
for language (Castro & Morais, 1987; Damasio, Damasio, Castro-Caldas, & Ferro, 1979; Karavatos et al., 1984; Kosmidis
et al., 2004; Tzavaras, Kaprinis, & Gatzoyas, 1981) yielding conflicting results. Some investigators reported decreased later-
ality for language among illiterates relative to literates (Joanette et al., 1983), others no laterality differences (Damasio,
Castro-Caldas, Grosso, & Ferro, 1976; Kosmidis et al., 2004), and yet others stronger lateralization among illiterates relative
to literates (Tzavaras et al., 1981). Potential reasons for the conflicting results may relate to characteristics of the stimuli used in
each investigation, such as the nature of the stimuli (i.e., words, pseudowords, numbers, syllables), the phonetic and the acous-
tic characteristics of the sounds (i.e., stop, nasal, fricative and liquid consonants, or consonants with abrupt onset), and differ-
ences in onset and intensity of stimuli in a pair (Ahonniska et al., 1993; Castro & Morais, 1987). More recent investigations of
the dichotic listening process (among literate individuals) have highlighted the role of attentional resource activity in addition
to cerebral dominance for language (Reinvang et al., 1994; van Ettinger-Veenstra et al., 2010; Westerhausen & Hugdahl,
2008). Unfortunately, none of the studies of dichotic listening in illiteracy took this parameter into account. Therefore, although
not informative with regard to the question of hemispheric dominance for language processing in illiterates in general, past
studies on dichotic listening among illiterates highlight the need for breaking down language processing into more specific
cognitive mechanisms.
Contemporary Functional Measures
Recent studies using different functional measures have demonstrated that literacy and education influence the pathways
used by the brain for problem-solving. Schooled literates and illiterates show equivalent left hemisphere attenuation of cortical
event-related potentials during a verbal memory task (for instance, memorizing words), but intrahemispheric differences at
parietotemporal areas, suggesting that learning literacy produces intrahemispheric specialization (Ostrosky-Solis et al., 2004).
Similarly, several studies using positron emission tomography (PET) and statistical parametric mapping or functional mag-
netic resonance imaging (fMRI) found additional differences between schooled literates and illiterates in the brain areas acti-
vated during language-based tests (i.e., left hemisphere perisylvian area). Interestingly, in one study using repetition of words
and repetition of pseudowords, these differences were much more relevant during pseudoword repetition suggesting that the
inability to process phonological segmentation was the inability to activate the brain regions responsible for doing it; the only
activation in words versus pseudowords that was greater in the literate than in the illiterate group was the more prominent left-
sided posterior parietal activation. In particular, when masking with the activation pattern defined by the words pseudowords
contrast in literates, there was increased activation of a left inferior parietal region (BA 40) in literates compared with illiterates
in words versus pseudowords. During real-word repetition, the difference was confined to a left parietal region that can be
related to the process of writing (Castro-Caldas, Petersson, Reis, Stone-Elander, et al., 1998; Fig. 1). This differential activation
of neural structures to real versus nonreal words suggested fundamental differences in cognitive processing based on phono-
logical information. A subsequent replication of the aforementioned PET findings (Petersson et al., 2000) traced differences in
the neuronal networks used by illiterates versus schooled literates in pseudoword repetition to the interaction between Brocas
area and the inferior parietal lobe, as well as the posterior-midinsula bridge between Wernickes and Brocas areas. Finally,
education-related differences were reported in an fMRI study of the effects of education on neural activation in healthy right-
handed Chinese subjects (Li et al., 2006). These differences in activation patterns emerged in the left inferior/middle frontal
gyrus and both sides of the superior temporal gyrus for a silent word recognition task and in the bilateral inferior/middle frontal

gyrus and limbic cingulate gyrus for a silent word and picture-naming task. It was suggested that these results indicate that the
patterns of neural activation associated with language tasks are strongly influenced by education. Education appears to have
enhanced cognitive processing efficiency in language tasks.
Even when their cognitive performance is equivalent, schooled literates and illiterates appear to use different brain acti-
vation patterns to perform the same task. Deloche, Souza, Braga, and Dellatolas (1999) studied healthy right-handed
Brazilians (unschooled illiterates and college graduates) on a task judging the magnitude of numbers within a familiar
context (e.g., 10 people in a standard-sized car, 9 students in a school: Is this a lot, average, or too little?). Both groups per-
formed similarly well with regard to the correctness of their judgments. Functional MRI scans, however, demonstrated differ-
ent brain activation patterns: The college graduates used the left hemisphere exclusively, specifically, the left frontal (BA 47),
inferior parietal (BA 40), and temporal lobes (BA 42, 21, and 22). In contrast, the illiterates activated both hemispheres during
the task, specifically, the temporal (BA 39, 20, and 22) and the occipital lobes (BA 19 and 37) bilaterally. Activation of the left
temporal lobe (BA 21 and 22) was the only finding common to both groups in the number processing task. In post-fMRI inter-
views inquiring about the strategies that the participants used to perform the judgment task, 95% of the illiterates reported
having used visual images, which may explain the activation of the occipital lobe in this group. The college graduates predo-
minantly used abstraction to solve the problems, which corresponds to their left hemisphere activation.
Carreiras and colleagues (2009) compared structural brain scans from those who learned to read as adults (late literates) with
matched illiterates. It was observed that late literates had more white matter in the splenium of the corpus callosum and more
grey matter in bilateral angular, dorsal occipital, middle temporal, and left supramarginal and superior temporal gyri. It was
Fig. 1. Maximum intensity projections of all significant activations thresholded. Results of the interaction analysis (A) word repetition (literateilliterate) and
(B) pseudo-word repetition (literate illiterate; adapted from Castro-Caldas, Petersson, Reis, Stone-Elander, et al., 1998).
also disclosed that the anatomical connections linking the left and right angular and dorsal occipital gyri through the area of the
corpus callosum white matter was larger in late literates than in illiterates. Results were interpreted to suggest that reading
increased the interhemispheric functional connectivity between the left and right angular gyri.
In sum, the data produced by functional imaging studies on brain activation during cognitive tasks support the contention
that learning literacy impacts not only the individuals day-to-day strategies, but also influences the functional architecture of
the adult human brain.
Anatomical Interhemispheric Comparisons
Comparing schooled literates and illiterates, Castro-Caldas and colleagues (1999) observed a slightly thinner corpus callo-
sum in illiterates, specifically the region where parietal fibers are thought to cross. The morphology of the corpus callosum
appears sensitive to the quality of the information content of the brain. The corpus callosum is larger for instance, in musicians
presumably because their performance with musical instruments requires the involvement of both hemispheres (Norton et al.,
2005). In the aforementioned PET scan study (Castro-Caldas, Petersson, Reis, Stone-Elander, et al., 1998), differences between
groups indicated a dissociation between the superior and the inferior parts of the angular-supramarginal regions, that is, the
superior parts being more active on the left than on the right in illiterates compared with literates, whereas the reverse was

the case for the inferior parts and the precuneus. This suggests differential recruitment of right and left regions of the parietal
lobe related to literacy and is in accordance with the finding reported above for magnitude judgments.
Some authors have postulated that hemispheric specialization is dependent on the nature of the task rather than the nature of
the stimulus; for instance, Stephan and colleagues (2003) used fMRI in a group of normal subjects during letter and visuos-
patial decision tasks with identical word stimuli to address two unresolved problems. It was noted that hemispheric specializ-
ation depended on the nature of the task rather than on the nature of the stimulus. Furthermore, the increased coupling between
left anterior cingulate cortex and left inferior frontal gyrus during letter decisions was observed, whereas right anterior cingu-
late cortex showed enhanced coupling with right parietal areas during visuospatial decisions; these authors suggested that cog-
nitive control is thus localized in the same hemisphere as task execution. We would propose an additional point based on the
aforementioned findings: Hemispheric specialization may also be related to the strategy used to solve the problem (verbal,
spatial, etc.), which is dependent on the individuals skills acquired by learning.
Neuropsychological Test Performance in Illiterates
General Cognitive Functioning
Mini-Mental State Exam. The Mini-Mental State Examination (MMSE) is a brief exam widely used by physicians primarily to
screen for the level of delirium or dementia. It consists of items of orientation, memory, attention, drawing, reading, writing,
repetition, naming, and following directions, scored on a 30-point scale (Folstein, Folstein, & Mchugh, 1975). The total MMSE
scores for illiterate samples average around 15 20 points (Table 3), in the range of severe cognitive alterations according to the
23 of 24 point cut-off suggested by Folstein and colleagues (1975).
Given its widespread use as a screening instrument for cognitive dysfunction, potential effects of literacy on performance
may be biased against those who are illiterate (Table 3). In fact, Bertolucci, Brucki, Campacci, and Juliano (1994) found that
illiterates differed from schooled literates on the MMSE not only on items related to writing and arithmetic (calculation,
reading a command, writing a sentence, and copying a line drawing), but also on orientation to time. Other studies have
found even more extensive differences; in one study, 16 of 19 questions differentiated schooled literates from illiterates at a
level of significance of P , .001, where only repetition, recall, and naming were unrelated to literacy (Rosselli et al.,
Table 3. Scores for illiterates on the MMSE in different normative studies

Author Age M SD
Ostrosky and colleagues (2000) 1650 19.7 3.7

5165 16.7 4.4
6689 16.4 4.2
Brucki and colleagues (2003) 6592 18.8 2.9
Laks and colleagues (2003) 6584 17.2 4.4
.84 14.3 3.9
Note: MMSE Mini Mental State Exam.
2000), with differences in performance being observed even between illiterates and people with as little as 1 year of education
at a level of significance of P , .0001 (Brucki, Nitrini, Caramelli, Bertolucci, & Okamoto, 2003).
Motor Functions
Writing is a fine movement activity; therefore, illiterate people may be less proficient with complex motor tests when com-
pared with people able to write. In fact, several studies have supported this hypothesis. Rosselli, Ardila, and Rosas (1990) found
statistically significant differences between illiterates and university-educated individuals in clinical tests of apraxia (buccofa-
cial and ideomotor praxis, finger alternating movements, meaningless movements, coordinated movements with both hands,
and motor impersistence tasks). When mimicking movements, illiterates frequently used the hand as instrument
(body-part-as-object). Illiterates also tended to replace the movement for the corresponding verbalization. Nitrini,
Caramelli, Herrera, Charchat-Fichman, and Porto (2005) administered Lurias fist-edge-palm test to 745 individuals with
different educational levels, including 238 illiterates. Logistic regression showed that illiteracy was associated with failure,
whereas gender and age were not. The proportion of individuals failing was inversely related to years of schooling. For
those able to reproduce the sequence, the number of demonstrations for successful reproduction was also inversely related
to years of schooling. Finally, Ostrosky, Ardila, and Rosselli (1999) investigated performance on four motor testschanging
left hand position, changing right hand position, alternating movements with both hands, and Lurias opposite reactions

testacross different educational levels. On all four motor tests, performance was higher in the participants with a higher edu-
cational level. The effect of level of education was robust on the first three tests, but minimal on the fourth.
Bramao and colleagues (2007) corroborated the association between reading skills and performance on visuomotor tasks.
They instructed illiterate and literate (Portuguese) participants to use the right or the left index finger to touch a randomly pre-
sented target on the right or left side of a touch screen. The literate subjects were significantly faster in detecting and touching
targets on the left compared with the right side of the screen. Presentation side did not affect performance of the illiterate group.
In conclusion, illiterates demonstrate poorer performance in a diversity of motor tests, including reproducing movements
and sequences of movements, alternating movements with both hands, and imitating meaningless movements.
Calculation and Number Processing
The assessment of calculation and number processing in illiterates and semiliterates after a brain injury is often a complex
task because these processes are closely linked to education. On the other hand, calculation is present in myriad daily life situ-
ations, such as in the use of money, means of transportation (bus routes, subways, trains), telephone numbers, and mental
calculations.
An international group of neuropsychologists created a battery of tests (EC 301) for evaluating calculation and number pro-
cessing in literate brain-damaged adults (Deloche et al., 1994). This instrument was adapted into several languages (English,
Finnish, French, German, Greek, Italian, Portuguese, and Spanish), thus providing a common tool for structurally equivalent
evaluations of examinees abilities. Nevertheless, the use of this tool in illiterates and semiliterates was shown to be ineffective.
A standardized, validated version of the original EC 301 Battery in Portuguese was developed in Brazil. Ten simple tasks asses-
sing counting, number processing, elementary calculation, and quantity estimation were presented to 122 normal Brazilian
adults, aged between 18 and 58 years with 0, 1, 2, 3, or 4 years of education (Deloche et al., 1999). Tasks such as counting
the number of elements in small sets were almost perfectly mastered by these illiterates and semiliterates; however, a sizeable
proportion of the sample made errors on other tasks (e.g., those assessing knowledge of the correspondence between numbers
and banknotes).
Language
Language abilities have been strongly correlated with a socioeducational level. Low socioeconomic parents use more non-
verbal than verbal strategies with children (Robinson, 1974). Along the same lines, the language used by people with a low
socioeconomic level is less fluent and has a simpler grammatical structure; it relies much more on emotional than on
logical strategies (Bernstein, 1974). Similarly, rural, unschooled children may lack symbolic representation skills because
their linguistic ability is tied to the immediate context of the referent (Bruner, Oliver, & Greenfield, 1966). Thus, it appears
that formal education facilitates the development of language into a fully symbolic tool. At least one study (Lantz, 1979),
however, showed that rural unschooled children performed better than schooled Indian or American children in coding and
decoding culturally relevant objects, such as grain, and seeds. Thus, children without formal schooling were able to separate
language symbols from the physical referent and to use those symbols for communicating accurately, but display of this ability
depended on the cultural relevance of the stimuli used (Laboratory of Comparative Human Cognition, 1983). In fact, the sig-
nificance of schooling lies not just in the acquisition of new knowledge, but in the creation of new motives and formal modes of
discursive verbal and logical thinking divorced from immediate practical experience (Luria, 1976).
Verbal fluency tests have been administered to illiterate populations in several countries. Table 4 presents the results of the
available normative studies for the category animals. Most frequently, the number of animals found in 1 min is about 10 12.
With increasing education, the number of items progressively increases.
Performance on phonemic or letter fluency tests is extremely difficult for illiterates. Available data suggest that letter fluency
in illiterates may be about 3 4 words in 1 min, at least for Spanish and Greek, although this could vary by language (e.g.,
Kosmidis et al., 2004; Montiel & Matute, 2006; Ostrosky et al., 1999, 2004).
Language repetition ability in illiterates is equivalent to that of schooled literates if real, high-frequency words are pre-
sented. However, when using pseudowords, difficulties are observed (Kosmidis et al., 2004; Petersson et al., 2000; Reis &
Castro-Caldas, 1997). Similarly, illiterates repeat high-frequency, but not low-frequency, words normally (Rosselli et al.,
1990). Of course, low-frequency words may be equivalent to pseudowords for a person with a limited vocabulary. Reis and
Castro-Caldas (1997) proposed that illiterate individuals (a) have difficulties in repeating pseudowords, (b) are worse at mem-
orizing pairs of phonologically related words compared with pairs of semantically related words, and (c) have difficulties in
generating words according to a formal criterion. Illiterate persons use strategies that are good for semantic processing, but
inadequate for explicit phonological analysis, while literate individuals are able to use several parallel running strategies.

It can be conjectured that vocabulary size is significantly correlated with educational level, considering that current edu-
cational systems are especially directed toward reinforcing verbal abilities and verbal knowledge. Reis, Guerreiro, and
Castro-Caldas (1994) and Reis, Petersson, Castro-Caldas, and Ingvar (2001) compared schooled literates and illiterates on
the task of naming real objects, naming photographs of these objects, and naming drawings of them. Results showed that
although both groups performed similarly at naming real objects, illiterates performed more poorly on naming photographs
and even more poorly on naming drawings. The authors suggested that introducing color information was useful in improving
the performance of illiterates in naming drawings (Reis et al., 2001). Therefore, increasing the amount of information contrib-
uted to better access to the name.
Metalinguistic Awareness
It has been observed that illiterate versus literate people do not differ in categorical perception of phonemes, but illiterate
people display a less precise categorical boundary and a stronger lexical bias (Serniclaes et al., 2005; Ventura, Kolinsky,
Querido, Fernandes, & Morais, 2007). Literacy, however, has been related to metalinguistic awareness (which can be regarded
as an executive function), since it is by means of the latter that it is possible to turn aspects of language into objects of reflec-
tion. Specifically, studies associated with phonological awareness suggest that learning to read leads children to dissect
language into small, nonmeaningful units. Illiterate adults find it difficult to consider words and nonwords as sequences of
phonemes (Morais, Cary, Alegria, & Bertelson, 1979) and, as a consequence, they underperform on tasks that require thinking
about the words phonemic characteristics (e.g., phonemic fluency tasks; Kosmidis et al., 2004; Reis & Castro Caldas, 1997).
Table 4. Animal fluency in illiterates

Author Place n Mean age SD Animals SD
Ardila and Moreno (2001) Aruaco Indians (Colombia) 5 26.2 2.9 9.8 1.5
Brucki, Malheiros, Okamoto, and Bertolucci (1997) Brazil 11.9
Brucki and Rocha (2004) Brazil 12.1
Kosmidis and colleagues (2004) Greece 19 72.0 7.6 11.8 3.2
Manly and colleagues (1999) USA 43 74.8 5.7 11.5 3.1
Montiel and Matute (2006) Mexico 10 38.2 10.7 13.5 3.0
Matute (1986) Mexico 25 46 10.4
Ostrosky and colleagues (1999) Mexico 50 21.3 3.3 13.2 3.7
Mexico 50 39.6 6.7 13.7 4.5
Mexico 49 58.8 4.1 12.7 5.0
Mexico 50 71.2 4.0 13.1 7.1
Ostrosky and colleagues (2004) Maya Indians (Mexico) 7 58.4 8.8 11.0 2.5
Ostrosky, Ramirez, and Ardila (2004) Pame Indians 7 53.6 6.1 11.8 4.0
Reis & Castro Caldas (1997) Portugal 23 F 62.0 7.1 10.8 3.7
Rosselli and colleagues (1990) Colombia 100 1665 10.6 3.4
However, illiterate performance is better, although still inferior to that of schooled literates, when the critical unit is the
syllable rather than a phoneme, as well as on rhyme detection (Morais, Bertelson, Cary, & Alegria, 1987). When comparing
illiterate to literate nonschooled groups on explicit phonological awareness tasks, Montiel and Matute (2006) found equivalent
performance between groups when dealing with syllables, but differences on the onset detection task and on the initial
phoneme detection test. Nonetheless, both groups diverge from the schooled literate group on more complex phonemic aware-
ness tasks, such as phonemic segmentation, phoneme blending, and initial phoneme deletion. Illiterate children had difficulty in
identifying the number of words in an oral sentence, suggesting that illiterate children find it difficult to think about language as
a string of words.
The aforementioned findings suggest that the influence of literacy extends beyond mere explicit phonemic awareness to
affect metalinguistic awareness, as well.
Visuoperceptual and Spatial Abilities
One visual-spatial effect of reading is the training of a specific direction of visual scanning (left to right, for most
European-origin writing systems). Using nonlinguistic stimuli, Ostrosky-Sols, Efron, and Yund (1991) studied the scanning
mechanism through a computer target detection paradigm in 60 illiterate and 60 literate individuals matched on a socioeco-
nomic level. Although target detection accuracy was identical in the two groups, there were significant differences between

the dectectability gradients of the literate and illiterate groups, suggesting that learning to read trains the scanning mechanisms
toward more consistent scan paths. Padakannaya and colleagues (2002) administered a picture array naming and recall task to
three groups of child readersunidirectional right-to-left readers of Arabic, unidirectional left-to-right readers of Kannada,
and bidirectional readers of Urdu and Englishand one group of Urdu illiterate adults. The results showed a right-to-left direc-
tion of visual scanning in the Arabic and Urdu readers. In the latter group, the strength of the scanning effect decreased with
more schooling in English. No right-to-left effect was observed in Kannada readers or Urdu illiterates. These results suggest
that the reading direction can affect scan habits in nonlinguistic tasks. The same effect has been documented in drawing.
Vaid, Singh, Sakhuja, and Gupta (2002) analyzed the influence of reading/writing direction and handedness on the direction
of stroke movement in free-hand figure drawing. Adult readers of scripts with opposing directionality (Hindi vs. Urdu) and
illiterate controls were observed while drawing a tree, a hand, a house, an arrow, a pencil, and a fish. Right-handers (including
right-handed illiterates) and left-to-right readers drew most figures from left to right, whereas left-handers (including left
handed illiterates) and right-to-left readers more often drew the figures from right to left.
Various neuropsychological studies have shown significant differences between schooled literates and illiterates in perform-
ing spatial and visuoperceptual tests. Ardila, Rosselli, and Rosas (1989) administered a basic neuropsychological test battery to
two extreme educational groups: Illiterate and professional adults, matched by sex and age. All of the visuospatial tasks (copying
a cube, a house, and the Rey Osterrieth Complex Figure; telling the time on a clock; recognizing superimposed figures; recog-
nizing the national map; and drawing the plan of the room) differentiated the two extreme educational groups. Matute and col-
leagues (2000) used four stick construction tasks (a test with a higher ecological validity for illiterates than copying figures) in
illiterate, semiliterate, and literate participants. Although illiterates generally made more errors than semiliterates and semilite-
rates more than schooled literates, only some of these differences were statistically significant. Significant differences were found
for lack of global fidelity and disarticulation errors when all four figures were considered together. Dansilio and Charamelo
(2005) used both figure copying and constructional abilities in 15 illiterates and a matched group with about 6 7 years of
school. The most frequent and relevant findings in illiterates in figure copying were an inability to reproduce the perspective
(13 of 15), unfolding (4 of 15), and unstructured copying (3 of 15). No errors were observed on the stick construction task.
Goncalves (2004) emphasized the difficulty illiterates had in copying drawings. As can be seen in Fig. 2 (reproduced from
Castro-Caldas, 2004), an illiterate copied one of the Bender drawings following a sequence that did not respect the idea of two
separate but connected elements, which is the way literate individuals tend to read the drawing. This resembles the concept of
integrative agnosia proposed by Riddoch and Humphreys (1987). Visually guided hand motor behavior may also be biased by
literacy. In a test paradigm in which literate and illiterate individuals were asked to direct the cursor toward a target on the
screen of a computer using the mouse, illiterate individuals were slower then literate controls, in particular when the right
hand had to move the cursor to the left side of the screen. This crossed condition is apparently trained by writing (Reis &
Castro-Caldas, 1997).
Memory
Culture and literacy can affect strategies used to recall information. Barltlett (1932) proposed that illiterates more frequently
use rote learning whereas literate people use more active information integration procedures (metamemory). Cole and
Fig. 2. Sequence of copying (right) of a Bender drawing (left).
Scribner (1974) observed that when memorizing information, schooled literates and illiterates make use of their own groupings
to structure their recall; for instance, high-school students rely mainly on taxonomic categories, whereas illiterate bush farmers
make little use of this principle. The authors argue that cultural differences in memorizing do not consist in the presence or the
absence of mnemonic techniques in general, but in the utilization of a specific technique: Reorganization of the

to-be-remembered material. This particular strategy for recall could be tied to school learning experiences.
Illiterates generally perform more poorly than schooled literates on conventional neuropsychological memory measures
such as wordlist learning and recall (Ardila et al., 1989; Cole, Frankel, & Sharp, 1971; Cole, Gay, Glick, & Sharp, 1971;
Folia & Kosmidis, 2003; Montiel & Matute, 2006; Nitrini et al., 2004), story learning and recall (Montiel & Matute, 2006),
verbal paired associates (Montiel & Matute, 2006; Reis & Castro-Caldas, 1997), digits backwards (Montiel & Matute,
2006), number-months (an adaptation of the Wechsler Memory Scale III Letter-Number Sequencing task), and complex
figure drawing (Ardila et al., 1989). However, the performance of illiterates seems to approach that of literates on object
memory (Folia & Kosmidis, 2003; Nitrini, Caramelli, Herrera, Porto, et al., 2004) and wordlist recognition memory (Ardila
et al., 1989). Illiterates low performance in memory tests may be specific to some of the psychometric procedures generally
used in testing memory.
The discrepancy observed between poor free recall and good recognition of illiterate individuals on object learning tasks
suggests inefficient encoding and retrieval strategies or poor organization of the material to be learned (Eslinger & Grattan,
1993). Recall of learned information generally occurs without cues or external support, requiring considerable self-initiated
activity and executive skills in addition to memory abilities (see Parkin & Leng, 1993). Among illiterates, these skills may
be adequate for a relatively passive cognitive process such as efficient recognition of stimuli learned through repetition, yet
inadequate to support a relatively active and effortful cognitive process such as free recall. This contention is supported by
the findings of an improvement in cognitive functions, including verbal memory, not specifically targeted in an adult
learning-to-read program; presumably, learning to read enhanced metacognitive abilities such as using analytic strategies, plan-
ning, and organizing output sequences, in this sample (Ardila, Ostrosky & Mendoza, 2000). Whether this improvement is a
direct result of literacy acquisition or of schooling is not clear.
A PET scan study found no group differences between schooled literates and illiterates on encoding and retrieving paired
associate words (Petersson et al., 2000). Both groups showed a positive correlation between cued-recall success and the acti-
vation of the left inferior prefrontal cortex and the medial temporal lobes during encoding, suggesting that literacy and school-
ing do not alter the basic neuroanatomy of encoding verbal material.
Educational Effects in Test Performance
Test-Wiseness
Test-wiseness refers to the examinees knowledge of how tests are designed, the expectations behind tests, and strategies for
taking tests, independent of the content the tests are designed to measure Thorndike (1951). It is seen as a nuisance variable that
gives its possessors better scores on tests which may not necessarily reflect better abilities on the capacity being measured.
Much of the literature on test-wiseness has investigated students well-entrenched in an educational career and examined rela-
tively sophisticated test strategies in group administered tests, suggesting that effects are modest (Green & Stewart, 1984).
However, what few studies are available for very young children and for those from developing countries suggest that the
effect is more substantial for these populations. Therefore, an important part of the differences between illiterates and literates
on cognitive tests may have to do not so much with cognitive abilities per se, as with test-taking abilities and familiarity. Test
taking and, consequently, test-wiseness are most likely skills learned primarily in school (Nell, 2000).
The Effect of Increasing Schooling Time
Even 1 or 2 years of school can make a significant difference in performance on some neuropsychological tests. Ostrosky,
Ardila, Rosselli, Lopez-Arango, and Uriel-Mendoza (1998) examined 64 illiterates, 64 adults with 1 2 years of school, and 64
adults with 3 4 years of school with the NEUROPSI. Statistically significant differences were found between the illiterates
versus 1 2 years of schoolfavoring the latteron the Language Comprehension, Phonological Verbal Fluency, and
Similarities subtests. Significant differences were observed between illiterates versus 3 4 years of schoolfavoring the
latterin the following subtests: Visual Detection, 20 3, Copy of a Semicomplex Figure, Calculation Abilities,
Sequences, Alternating movements, Recall of words: Cueing, and Recall of a Semicomplex Figure (Table 5).
Nonetheless, when comparing the scores in the same NEUROPSI test battery of participants with 10 12, 13 17, and 18 24
years of school, only a few significant differences were found. These observations indicate that the educational effect on neu-
ropsychological test performance is not linear. Differences between zero and 3 years of education are usually highly signifi-
cant; differences between 3 and 6 years of education can be lower; between 6 and 9 are even lower; and so forth, with virtually
no differences between, for example, 12 and 15 years of education. This means that on neuropsychological tests, education

effects represent a negatively accelerated curve, tending to plateau. This is not surprising, considering that, in general, neurop-
sychological tests have a low ceiling. Another consequence of this finding is that neuropsychological tests that show no edu-
cation effect between, for example, North Americans with fewer and greater than 12 years of education (but with no fewer than
8 years of education) cannot be safely extrapolated to assume no educational effect for fewer than 8 or as low as 0 years of
education.
The length, qualities, and content of the school day and year vary considerably from country to country and even from
school to school. More refined measures of length and qualities of schooling are needed in future studies.
Studies in Illiterate Children
In almost all occidental cultures, formal reading acquisition starts as soon as aged 5 6. Thus, the process to become literate
starts early in childhood; however, little attention has been paid to illiterate condition in children. A pioneer study in illiterate
children was performed in Mexico, with the aim to investigate whether or not the effects of literacy on neuropsychological test
performance are already evident during childhood, as has been reported previously for illiterate adult populations. Seeking that
purpose, the performance of 21 illiterate children and 22 literate children aged 6 13 was compared on 13 cognitive domains of
the Evaluacion Neuropsicologica Infantil (ENI; Child Neuropsychological Assessment): Attention, constructional abil-
ities, verbal memory coding and delayed recall, visual memory coding and delayed recall, tactile perception, visual perception,
auditory perception, oral language, metalinguistic awareness, calculation, and spatial abilities. Demographic variables were
controlled to avoid effect of socioeconomic characteristics. For the illiterate group, school nonattendance was due to social-
family reasons, thus controlling learning disabilities effects. Results showed that the illiterate group significantly
Table 5. Mean (SD) performance and comparisons of groups by the level of education on NEUROPSI neuropsychological tests (adapted from Ostrosky et al.,
1998)
Tests Score range Years of schooling Significant differences
0 1 2 3 4
Visual detection 0 16 9.9 (4.5) 11.2 (4.1) 12.5 (3.2) G3 versus G1

20 minus 3 0 5 3.1 (1.9) 3.1 (1.7) 4.1 (1.2) G3 versus G1,G2
Semicomplex figure (copy) 0 12 7.5 (2.0) 8.8 (2.4) 9.4 (1.9) G3 versus G1,G2
Language: Comprehension 0 6 3.7 (1.2) 4.4 (0.8) 4.6 (1.0) G1 versus G2,G3
Phonological fluency 0 15 3.1 (3.7) 6.5 (4.0) 7.0 (4.1) G1 versus G2,G3
Similarities 0 6 2.1 (2.2) 3.5 (2.1) 3.9 (1.9) G1 versus G2,G3
Calculation abilities 0 3 0.9 (1.0) 1.5 (1.1) 1.6 (1.1) G3 versus G1,G2
Sequences 0 1 0.1 (0.3) 0.2 (0.4) 0.4 (0.5) G3 versus G1,G2
Alternating movements 0 2 0.8 (0.7) 1.1 (0.7) 1.3 (0.7) G3 versus G1
Recall: Words Cueing 0 6 4.1 (1.4) 4.3 (1.4) 4.7 (1.4) G3 versus G1
Semicomplex figure 0 12 6.3 (2.2) 7.0 (2.4) 8.4 (2.3) G3 versus G1,G2
Notes: G1 illiterates; G2 1 2 years of school; G3 3 4 years of school.
underperformed the literate group in mostly all measures except tactile perception. Moreover, age was a significant covariant,
where higher scores were related to older ages in both groups. However, when analyzing the cognitive domains that it is known
that are related to schooling (metalinguistic awareness and calculation), it was found that metalinguistic awareness task per-
formance improve with age in literate children but it was not the case for the illiterate group; at the same time for the calculation
abilities, an effect of age was evident in both groups suggesting that math learning is school and environment-dependent
(Matute et al., Submitted).
Executive Function was also investigated in these two groups (Matute et al., unpublished) four domains were analyzed:
Verbal fluency, mental flexibility, planning and concept formation, and reasoning where 8 measures were compared:
Verbal fluency, semantic verbal fluency, phonemic verbal fluency, categories achieved in the card sorting test (mental flexi-
bility test), number of correct designs done with the minimum number of movements in the Pyramid of Mexico Test (a plan-
ning and organization test), similarities, and matrices and arithmetic problems. With the exception of the matrices test, the
illiterate children scored significantly below the literate children suggesting that differences in the development of executive
functions between these two populations are already evident in childhood. The absence of difference in matrices test is related
to low score in both groups; thus, it could be possible that those skills underlying this type of test are acquired at a later stage,
and intergroup differences would not be evident in childhood.
Learning to Read in Children Versus Learning to Read in Adults

Several studies were designed to compare brain activation of individuals who learned to read and write in childhood at the
usual age and those who learned in adulthood. In one of these studies, participants were asked to identify written words during
magnetoencephalography (MEG). Results showed that late literates had more late sources of activity in right temporoparietal
areas than their controls, who, on the other hand, had more early activation of left inferior frontal areas (Castro-Caldas,
Peterson, Reis, Askelof, et al., 1998; Castro-Caldas, Petersson, Reis, Stone-Elander, et al., 1998). This suggests that the mech-
anisms related to reading are different in late literates despite similar performance on the task for both groups. Castro-Caldas
and colleagues (2009) in a further study compared late literates to controls to subjects that learned at school at the usual age.
MEG was done while subjects were reading words. It was found that although the reading performance was the same in both
groups while performing the task, the pattern of source distribution was different between groups. There were more late sources
in the right temporoparietal areas of late literates compared with controls and more late sources in the left inferior frontal cortex
in control subjects. It is concluded that learning to read in adulthood is a process supported by different brain structures from
the ones used when learning occurs at the usual age. Castro-Caldas, et al., (2009) compared seven women who learned how to
read and write after the age of 50 (ex-illiterates) and five women with 10 years of regular schooling (controls) in a language
recognition test while brain activity was being recorded using MEG. It was found that both groups performed similarly on the
task of identifying target words. Analysis of the number of sources of activity in the left and right hemispheres revealed sig-
nificant differences between the two groups, showing that ex-illiterate subjects exhibited less brain functional asymmetry
during the language task. These findings were interpreted as reinforcing the concept that poorly educated subjects tend to
use the brain for information processing in a different way to subjects with a high-educational level or who were schooled
at the regular time.
The Gender Effect
Gender differences in cognitive abilities have been widely analyzed during the last decades (e.g., Buffery & Gray, 1972;
Caplan, Crawford, Hyde, & Richardson, 1997; Hedges & Nowell, 1995; Maccoby & Jacklin, 1974). Three major differences
between men and women in cognitive abilities have usually been reported: (a) verbal abilities, favoring women; (b) spatial
abilities, favoring men; and (c) arithmetical abilities, favoring men. Differences in calculation abilities have sometimes
been interpreted as a result of mens superior spatial abilities (Benbow, Libinski, Shea, & Eftekhari-Sanjani, 2000; Geary,
1996).
The origin of these differences is not clear, even though some neurological (e.g., Blanch Brennan, Condon, Santosh, &
Hadley, 2004; Rilea, Roskos-Ewoldsen, & Boles, 2004) and environmental (e.g., Quaiser-Pohl & Lehmann, 2002) factors
have been proposed. Gender differences have also been associated with hormonal influences (e.g., Aleman, Bronk, Kessels,
Koppeschaar, & van Honk, 2004).
The collaborative Mexican Colombian research program on the effects of education on neuropsychological test perform-
ance has consistently found over the last 30 years that education significantly interacted with gender (e.g., Ardila et al., 1989,
1992; Ostrosky et al., 1985, 1986, 1998; Rosselli et al., 1990). On most neuropsychological tests, among illiterates or people
with limited education, men outperformed women, a gender difference which decreased as the level of education increased and
Fig. 3. Performance in the Rey Osterrieth Complex Figure in normal adults with different levels of education (age 2175 years; N 824) (adapted from
Ardila et al., 1989; Rosselli & Ardila, 2003).
virtually disappeared at about the level of a high-school education. To the best of our knowledge, such an interaction between
gender and education on cognitive testing has not been reported elsewhere. As an illustration, Fig. 3 presents the scores
obtained from individuals with different levels of education in the ReyOsterrieth Complex Figure. As clearly observed,
scores are about 20% higher in illiterate men than in illiterate women. When increasing the education level, differences

become progressively smaller and after about 12 years of school, performance is virtually identical.
We may assume that this gender difference can be partially due to the characteristics of the population from which the
sample was recruited. In Latin American countries, it is not unusual to find that low-educated women remain at home and
are solely responsible for child care and house work, particularly for the historical age cohorts in which these studies were
carried out. Men go to the factory (or elsewhere) to work and, thus, acquire a more complex understanding of their environment
(navigating around the city, taking buses, and similar); they maintain a higher level of social interaction, they handle money,
and they are much better informed about national and international events. Simply speaking, they have a higher level of stimu-
lation and use a greater amount of information than women, at least regarding skills needed for the tests used. Studies in other
populations are required in order to support or reject this interpretation. This effect suggests that, at least for this population,
formal education may have an equalizing gender effect on some cognitive skills, rather than being responsible for gender
differences. In any case, these findings suggest that education does have an impact on gender differences in cognitive tests.
Age-Related Cognitive Decline in Illiterates
Capitani, Barbarotto, and Laicana (1996) proposed three possible patterns of association between age related decline and
education: (a) Parallelism: Age related decline runs the same course in different educational groups, that is, no interaction; (b)
Protection: Age related decline is attenuated in well educated participants; and (c) Confluence: The initial advantage of
well educated groups in middle age is reduced in later life. The researchers explored these hypotheses in a group of 307
Italians aged 40 85 and mean education levels of 6 versus 13 years; their data supported the parallelism theory for verbal
fluency, spatial memory, and Ravens Progressive Matrices, but the protection theory for visual attention and verbal
memory. Confluence was not observed on any of their five tests. They concluded that the protective effect of education is
not ubiquitous, but depends on the specific cognitive ability measured.
With a wider age and education range on the NEUROPSI, Ardila and colleagues (2000) found parallelism, protection, and
two different subtypes of confluence (upward and downward; Table 6). For some subtests, a specific pattern was not evident.
They, too, concluded that different patterns of age related cognitive decline and education may be found, depending on the
specific cognitive domain. These results agree with Capitani and colleagues for verbal memory, but not for verbal fluency, and
other domains (and the participant samples) are not comparable. Furthermore, these are cross-sectional studies and subject to
possible cohort effects.
Dementia in Illiterates
Decreased prevalence and incidence of dementia among persons with a higher level of education have been reported by
some population-based studies (Qiu et al., 2001; Zhang et al., 1990), findings that have been explained by the cognitive
reserve hypothesis, which holds that the individual differences in how tasks are processed provide differential resistance
against brain pathology or age-related changes (Katzman et al., 1989; Stern, 2006). According to this hypothesis, highly edu-
cated individuals may either have greater neural reserve, that is, brain networks that are less susceptible to disruption, or they
may have more neural compensation, which renders them more capable of developing efficient strategies to cope with the
Table 6. Examples of different patterns of cognitive decline across age ranges; mean scores and standard deviations (in parentheses) are presented
Age (years)
16 30 31 50 5165 6685
Parallelism: Copy of a Figure

Over 10 years 11.8 (0.5); 100% 11.7 (0.5); 99% 11.3 (1.2); 95% 10.9 (1.4); 92%
Illiterates 8.1 (2.1); 100% 7.9 (1.8); 97% 7.6 (2.2); 94% 7.2 (2.7); 89%
Protection: Recall of Words
Over 10 years 5.3 (0.9); 100% 4.9 (1.2); 92% 4.6 (1.5); 87% 3.7 (1.8); 70%
Illiterates 4.3 (1.6); 100% 3.6 (2.2); 84% 2.4 (2.4); 56% 2.1 (2.3); 49%
Confluence upwards: Digits Backwards
Over 10 years 4.3 (1.0); 100% 4.4 (1.0); 102% 4.0 (1.0); 93% 3.9 (1.1); 90%
Illiterates 2.2 (1.1); 100% 2.8 (1.1); 127% 2.9 (1.0); 132% 2.7 (0.9); 123%
Confluence downwards: Semantic Verbal Fluency
Over 10 years 21.6 (5.4); 100% 22.3 (5.0); 103% 20.1 (5.1); 93% 18.4 (4.8); 85%
Illiterates 13.2 (3.7); 100% 13.7 (4.5); 104% 12.7 (5.0); 96% 13.1 (7.1); 99%
Note: In each case, performance at the 1630 years is taken as 100% (adapted from Ardila, Ostrosky & Mendoza, 2000).

Table 7. Prevalence of dementia in relation to educational level (adapted from Herrera et al., 2002)
Schooling N Mean age (SD) Dementia (N) Dementia (%) Prevalence ratio
Illiterate 567 74.6 (7.1) 69 12.2 3.5

1 3 years 590 72.2 (5.9) 26 4.4 1.2
4 7 years 356 72.1 (6.4) 18 5.0 1.4
8 years or more 143 72.5 (6.7) 5 3.5 1a
a
The prevalence in this age group was taken as reference.
negative effects of disease and ageing (Stern, 2006). In a postmortem study of a large number of elderly individuals, between
12% 19% of the individuals fulfilled neuropathological criteria for Alzheimers disease (AD), although they had not been
considered demented during life. When the authors investigated this phenomenon, they found that as the number of years
of formal education increases, the odds of a clinical dementia diagnosis decreases by approximately 0.82 0.87 with each
additional year of education, thus supporting the cognitive reserve hypothesis (Roe, Xiong, Miller, & Morris, 2007).
A positive association between AD and low education has been found in many studies (e.g., Callahan et al., 1996; Friedland,
1993; Katzman, 1993; Letenneur, Commenges, Dartigues, & Barberger-Gateau, 1994; Liu et al., 1994; Mortimer & Graves,
1993; Rocca et al., 1990; Yu et al., 1989; Zhang et al., 1990) carried out in different countries: Brazil (Caramelli et al., 1997),
China (Hill et al., 1993; Yu et al., 1989), Finland (Sulkava, Wikstrom, & Aromaa, 1985), France (Dartigues, Gagnon, &
Michel, 1991), Italy (Bonaiuto, Rocca, & Lippi, 1990; Rocca et al., 1990), Israel (Korczyn, Kahan, & Gulper, 1991),
Sweden (Fratiglioni et al., 1991), and the USA (Stern et al., 1994). In a Brazilian population study (Herrera, Caramelli,
Silveira, & Nitrini, 2002), the prevalence ratio of dementia in illiterates was 3.5 times higher than in those with 8 or more
years of schooling (Table 7). Given that the low-educational level is often coupled with the low socioeconomic level, the
authors performed a multivariate analysis, which confirmed that the low-educational level was associated with the higher
prevalence of dementia (Herrera et al., 2002).
In a recent analysis of six Latin American studies, which included 26,199 elderly, dementia was two times more frequent in
illiterate than in literate individuals, a relevant finding because the pooled data show that the rate of illiteracy among the elderly
was approximately 10%. Significant differences of prevalence between illiterate and literate individuals were observed in seven
out of the eight population studies (Nitrini et al., 2009)
Negative results, however, have also been reported (e.g., Christensen & Henderson, 1991; Knoefel et al., 1991; OConnor,
Pollitt, & Treasure, 1991; Rocca et al., 1990). For instance, the Framingham study found no association between the edu-
cational levels and the risk of AD (Knoefel et al., 1991) in a predominantly literate population. Furthermore, a pooled analysis
of European data (Launer, Dinkgreve, Jonker, Hooijer, & Lindeboom, 1993) showed that the deleterious effect of a lower edu-
cation was more marked in women than in men. Neither gender nor education was associated with a positive family history of
dementia or the presence of the APOe4 allele (Duara et al., 1996). Finally, a study of a predominantly illiterate population in
the rural areas of India also failed to detect an influence of education on the prevalence of dementia (Chandra et al., 1998).
The higher prevalence of dementia among illiterates could be explained by possible overdiagnosis in this group due to a lack
of test-wiseness. However, a majority of studies adjusted cut-off scores or used tests more appropriate for populations with
educational heterogeneity (Zhang et al., 1990), as well as informant questionnaires (Herrera et al., 2002; Nitrini et al., 2004),
decreasing the likelihood of this possibility.
In an investigation of the factors affecting age of onset and rate of progression of AD, the researchers concluded that AD
progresses steadily, regardless of education, occupation, and other factors tested (Bowler, Munoz, Merskey, & Hachinski,
1998); however, the age of referral to a memory clinic, which was correlated with the age of onset, differed according to
the educational level. Those with a high level of education tended to be younger upon referral than those with a low level
of education. However, the scores in the extended scale for dementia on initial assessment were similar between patients
with a low and those with a high level of education. The authors further investigated the influence of education and socioeco-
nomic status in the autopsy-confirmed cases of AD (Munoz et al., 2000). They compared two groups of autopsied patients, 115
confirmed AD and 142 patients 65 years or older without dementia who died in the hospital without neurodegenerative disease.
They found no substantial differences in education, occupation, and socioeconomic or income levels between groups.
Hendrie and colleagues (2001) and Ogunniyi and colleagues (2000) compared the prevalence rate of dementia of 2,494
elderly poorly educated Nigerian residents in Nigeria to the prevalence rate of 2,212 African Americans with similar education,
living in the USA. The rates of the Nigerians (2.29%) were much lower than the rates for the U.S. group (8.24%), which
suggests that education is not the relevant factor for dementia. Furthermore they report different risk factors in these two
groups. In Nigeria, old age and female gender were revealed to be significant risk factors, whereas among African
Americans old age, rural living below the age of 19 years, low education, and family history of dementia were the risk

factors. In addition, the frequencies of the vascular risk factors investigated were lower in Nigerians, which may suggest invol-
vement of environmental factors in disease processes.
In conclusion, the issue of the influence of the educational level on the clinical diagnosis, prevalence, progression, or sever-
ity of AD is still far from fully understood. Indeed, we may find some differences between the well-selected groups of partici-
pants. These differences may be a reflection of cognitive strategies used based on different levels of school achievement. More
prospective and well-designed studies are needed to further clarify this topic.
Brain Damage and Illiteracy
Two opposing points of view have emerged in the neuropsychological literature regarding the influence of education on
brain organization of language. Cameron, Currier, and Haerer (1971) reported a lower frequency of aphasias associated
with injuries of the left hemisphere among right handed illiterate patients than among educated ones. The authors concluded
that language is more bilaterally represented in the illiterate group. In contrast, Damasio and colleagues (1976) claimed that
there is no qualitative or quantitative difference between the aphasias of educated and illiterate patients. The aphasia of
schooled literates or illiterates did not differ in the prevalence rate, distribution of clinical types, or semiological structure.
Matutes research (1988) supports the Damasio and colleagues (1976) conclusions. She compared three groups of right
handed Mexican adults: Brain damaged illiterates, braindamaged literates, and normal illiterates. An aphasia test was
given to all three groups in the context of a more extensive neuropsychological assessment. No differences between the
two brain damaged groups were evident in regard to language measures related to repetition, oral expression, and comprehen-
sion. However, when analyzing incidence of aphasia due to unilateral brain lesions, all left hemisphere damaged illiterates
presented aphasia, although the lesion locus for some of them were out of the perisylvian area, whereas none presented
aphasia after right hemisphere damage. Thus, the data obtained in this study suggest a less intrahemispheric specialization
for language on the left hemisphere in illiterates.
The hypothesis that illiterates might have a different cerebral organization for language and, therefore, deviate from literates
in their clinical profile; the severity of their aphasia or their prognosis was evident as early as 1867 with Scoresby Jacksons
observation of an illiterate patient. This author presented a case of an illiterate patient with a severe motor aphasia. After the
postmortem exam, the author found that although the lesion was very extended in the left hemisphere, the frontal lobe was
attainted only in the posterior part of the third frontal circumvolution, where Broca had located the spoken language. Since
it was only the posterior part of the third frontal circumvolution that was attainted, Scoresby Jackson suggested that a
bigger part of this circumvolution would participate in language only with a grater language acquisition (given by reading
acquisition). Thus, in the illiterate people, it would be only the posterior extreme of the third frontal circumvolution that
will be active in language expression, whereas in a literate person, all the circumvolution will be employed. Finally, the
author stated that language and writing will be learned with the posterior part of the left third frontal circumvolution and
with the experience given by the exercise in the art of speaking and writing a larger part of this circumvolution will be
employed. Critchley (1956) was probably the first author to suggest that hemispheric functional asymmetry could be influenced
by schooling. He suggested that low-educated persons tended to have less left hemispheric lateralization for language
functions. This suggestion was supported by Wechslers (1976) interpretation of the crossed aphasia in his illiterate patient
(Castro-Caldas et al., 1987).
Fonseca and Castro-Caldas (2002) compared the recovery process of literate and illiterate aphasics. They studied 24 illit-
erates and compared those with 42 schooled literates matched for age, gender, and type of aphasia. Generally, all scores
obtained in subtests of the Aphasia battery were lower in illiterates than they were in literate controls. Patients were tested
in the first month of their disease and 6 months later. The global scores of aphasia improved similarly in both groups;
however, the correlation between the test scores suggested that the process of recovery was different for each group.
Lecours and colleagues (1987a, 1987b, 1988) studied the relationship between brain damage and schooling with regard to
aphasic impairments of language. On the basis of their findings, the authors concluded that: (a) There was a greater right hemi-
sphere language involvement in illiterates than in the well educated patients; and (b) left stroke school educated patients
seemed to be sicker, as it were, than their illiterate counterparts, that is: (i) The classical symptoms of aphasia (suppression
stereotype, jargon aphasia) were more apparent among left stroke schooled literates than among left stroke illiterates; and (ii)
auditory comprehension was more frequently impaired among the left stroke literate patients.
Lecours and colleagues (1987b) also studied the influence of education on unilateral neglect syndrome. They analyzed a
large sample of right handed unilingual brain damaged individuals: Illiterates (left stroke and right stroke) and schooled lit-
erates (left stroke and right stroke). Evidence of a unilateral neglect syndrome was found in both left- and right-brain-damaged
schooled literates and illiterates. Their results provided no indication that tropisms were globally stronger depending on the side

of the lesion or on the educational level of the patients. Rosselli, Rosselli, Vergara, and Ardila (1985), however, reported a
higher frequency of right hemispatial neglect in low educated patients.
In summary, studies of brain damaged illiterates, when compared with brain damaged literates, have indicated that: (a)
Literacy does not change the dominance of the left hemisphere for language, illiterates as well as literates present aphasia
most often after left brain damage, and not after right brain damage; and (b) the right hemisphere appears to have a dispropor-
tionate involvement in language in illiterate when compared with literate individuals. This is based on the following evidence:
Left-damaged literates present a larger number of errors in aphasia tests than left-damaged illiterates (Lecours et al., 1988;
Matute, 1988) and right damaged illiterates more frequently present poorer performance in aphasia tests than right-damaged
literates (Lecours et al., 1987a, 1987b).
Some Clinical Implications
Brain injury, either due to stroke or TBI, may have various clinical manifestations, depending on the brain areas affected.
The individual may develop isolated or associated disorders of language, reading, writing, calculation, memory, attention,
vision, visuoconstruction, behavior, and/or movement. These problems need to be identified in order to establish an adequate
rehabilitation program. However, identifying many of these disorders is sometimes difficult because there are still very few
neuropsychological tools created for, standardized, and normed on illiterates. Furthermore, many existing tests depend on
content or processes related to education. It is often challenging to distinguish whether test difficulties experienced by an illit-
erate individual are due to brain injury or if they are preexistent, because the individual never learned the skills being measured.
Tests that have already been standardized on illiterates in different countries, include the MMSE, NEUROPSI, NEUROPSI
Attention and Memory, Rey Osterrieth Complex Figure, Lurias fist-edge palm test, motor tests, verbal fluency tests, visuo-
perceptual and spatial tests, memory tests, executive function tests, calculation, number processing, and others (Ardila et al.,
1989, 1992, 2000, 2003; Bertolucci et al., 1994; Brucki et al., 2004; Dansilio et al., 2005; Deloche et al., 1999; Folia &
Kosmidis, 2003; Kosmidis et al., 2004; Loureiro et al., 2004; Nitrini, Caramelli, Herrera, Porto, et al., 2004; Nitrini et al.,
2004, 2005; Ostrosky-Solis et al., 1985, 1986, 1998, 1999, 2004; Rosselli et al., 1990). Additional research is needed to deter-
mine if these norms can be generalized to other illiterate populations and to determine their sensitivity to brain impairment and
their prediction of functional abilities. More tools are also needed. However, tests not normed on these populations may be
helpful in demonstrating preserved abilities and strengths.
Neuropsychological assessment of illiterates should emphasize history taking, understanding their functioning within their
social and cultural context, and exploration of the ways in which that functioning differs from that of their peers and/or from
their own premorbid functioning. That history should also include distinguishing whether their illiteracy is due to lack of
opportunity or failure to learn.
The neurorehabilitation of illiterate people after brain injury or illness presents challenges to rehabilitation programs
oriented toward literates. Neuropsychological rehabilitation yields much more effective results when it is adapted to the indi-
viduals specific context (Lave, 1996; Rogoff, 1990). Consequently, any such program must take into consideration the
persons work environment, family life, and cultural context to maximize success. Current cognitive rehabilitation techniques
for memory, and, to some extent, executive functions, tend to favor written compensatory techniques such as memory books
and written directions. Such techniques need to be adapted to serve the illiterate patient (Judd & DeBoard, 2007).
Conclusions Regarding Human Cognition Through the Study of Illiteracy
Literacy is strongly reflected in the performance of tasks used in psychological, as well as neuropsychological evaluations.
Acquiring literacy appears to influence visual perception, logical reasoning, remembering strategies (Laboratory of
Comparative Human Cognition, 1983), and formal operational thinking (Laurendeau-Bendavid, 1977). It is a mistake to
assume that the inability to perform simple cognitive tasks, such as those frequently incorporated in current neuropsychological
test batteries, necessarily means abnormal brain function. The degree of literacy can often represent the crucial variable.
The influence of literacy seems to go further: Literacy may somehow change the brain organization of cognition. Studies of
the consequences of brain damage in illiterate populations evidence a more bilateral representation among illiterates than lit-
erates not only for linguistic, but probably also for visuospatial abilities. Apparently, literacy does not change the direction of
laterality in the brain organization of cognition, but the degree of this lateralization.
School attendance does not mean that educated people simply possess certain abilities that less-educated individuals lack.
Reading ability rather seems to represent an additional instrument to extend cognition. Illiteracy does not mean either that
highly educated people have the same abilities as less-educated individuals, plus something else. Nonetheless, formal cognitive

testing evaluates those abilities in which the educated child was trained. Therefore, it is not surprising that he or she will out-
perform the child with no formal education. It must be emphasized that the educational level has a substantial relationship with
performance on some cognitive tests, but is not systematically related to everyday problem-solving (functional criterion of
intelligence; Cornelious & Caspi, 1987). It is not totally accurate to assume that people with low levels of education are
somehow deprived. It may be more accurate to assume that they have developed different types of learning, more procedural,
pragmatic and sensory oriented.
Therefore, the effects of literacy and schooling on cognition include the following:
(1) Literacy provides an additional instrument for acquiring information. Information can be obtained not only using
oral tradition, but also from books, journals, and so forth.
(2) The process of learning to read and write may train specific additional abilities, such as explicit phonological aware-
ness, spatial perception, and fine movements.
(3) Reading and writing implies an additional instrument of conceptualization, interpretation, and mediation of the world
(Luria, 1966; Vygotsky, 1934/1978). Memory not only means being able to recall something, but also recording the
information in a notebook (auxiliary hippocampus); social communication not only means speaking and listening,
but also reading and writing. An explanation about how to improve agriculture can be replaced by a written pamphlet.
The instruction about how to go to a particular place can be found through reading a paper or using written street names.
(4) Attending school also has certain consequences on cognition, independent of reading and writing. It contributes to the
development of specific attitudes toward knowing, understanding, and thinking. School provides training in reading,
but also in mathematics, geography, drawing, history, and natural sciences. Furthermore, school represents a special
type of setting centered on specific values: Learning, memorizing, understanding, and achieving. This is not necess-
arily valid to the same extent in every type of school (e.g., in a school of arts, a trade school or a religious school).
Consequently, many of the effects of literacy are not only due to learning to read, but also to attending school, and, thus,
developing a way of interpreting the worlda cosmovision, so to speakthe cosmovision of learning and understanding.
It also appears that lateralized fine motor skills develop in the process of learning to write. Habits in directions of visual
scanning also appear to develop with the acquisition of literacy.
It is fairly clear that illiterates generally perform more poorly than literates on a variety of cognitive and neuropsychological
tests. The reasons for this are not obvious, but candidate explanations are:
(a) Acquiring literacy develops cognitive skills that are measured by the tests.
(b) Cognitive instruction that is part of schooling but not specifically part of learning literacy develops some of the cog-
nitive skills measured by the tests.
(c) Schooling develops an understanding of what is expected in testing and how to go about taking tests effectively.
(d) Schooling develops a predisposition toward achievement in testing.
The relative weight of each of these factors, or perhaps even other, unknown factors, on the neuropsychological performance of
illiterates is not yet known. Future studies in this area are needed to determine the factors that may contribute to differential
neuropsychological performance between literates and illiterates, as well as the relative role of the particular cognitive func-
tions being tested, the educational system and philosophy to which examinees were exposed, the particular language, writing
system, and/or population from which they have been influenced.
The process of elucidating cognitive mechanisms particular to illiteracy, especially as they contrast with those involved in
literacy, may help to expand our understanding of human cognition and functional brain organization as they relate to both
normal and pathological conditions.
Conflict of Interest
None declared.
References

Aleman, A., Bronk, E., Kessels, R. P., Koppeschaar, H. P., & van Honk, J. (2004). A single administration of testosterone improves visuospatial ability in young
women. Psychoneuroendocrinology, 29, 612617.
Ahonniska, J., Cantell, M., Tolvanen, A., & Lyytinen, H. (1993). Speech perception and brain laterality: The effect of ear advantage on auditory event-related
potentials. Brain and Language, 45, 127 146.
Ardila, A. (2004). There is not any specific brain area for writing: From cave painting to computers. International Journal of Psychology, 39, 6167.
Ardila, A., & Moreno, S. (2001). Neuropsychological evaluation in Aruaco Indians: An exploratory study. Journal of the International Neuropsychological
Society, 7, 510 515.
Ardila, A., Ostrosky, F., & Mendoza, V. (2000). Learning to read is much more than learning to read: A neuropsychologically-based learning to read method.
Journal of the International Neuropsychological Society, 6, 789 801.
Ardila, A., Ostrosky-Solis, F., Rosselli, M., & Gomez, C. (2000). Age related cognitive decline during normal aging: The complex effect of education. Archives
of Clinical Neuropsychology, 15, 495514.
Ardila, A., & Rosselli, M. (2003). Educational effects on the ROCF performance. In J. Knight & E. Kaplan (Eds.), Rey-Osterrieth complex figure handbook (pp.
271281). Lutz, FL: Psychological Assessment Resources.
Ardila, A., Rosselli, M., & Ostrosky, F. (1992). Socioeducational factors in neuropsychological assessment. In A. E. Puente & R. J. McCaffrey (Eds.),
Handbook of neuropsychological assessment: A biopsychosocial perspective (pp. 181 192). McCaffrey: Plenum Press.
Ardila, A., Rosselli, M., & Rosas, P. (1989). Neuropsychological assessment in illiterates: Visuospatial and memory abilities. Brain and Cognition, 11, 147
166.
Barltlett, F. C. (1932). Remembering. Cambridge: Cambridge University Press.
Benbow, C. P., Libinski, D., Shea, D., & Eftekhari-Sanjani, H. (2000). Sex differences in mathematical reasoning ability at age 13: Their status 20 years later.
Psychological Science, 11, 169232.
Bernstein, B. (1974). Language and roles. In R. Huxley & E. Ingram (Eds.), Language acquisition: Models and methods (pp. 220256). New York: Academic
Press.
Berry, J. W., & Bennet, J. (1989). Syllabic literacy and cognitive performance among the Cree. International Journal of Psychology, 24, 429450.
Berry, J. W., & Bennett, J. A. (1992). Cree conceptions of cognitive competence. International Journal of Psychology, 27, 73 88.
Bertolucci, P. H. F., Brucki, S. M. D., Campacci, S., & Juliano, Y. (1994). O Mini-Exame do Estado Mental em uma populacao geral: impacto da escolaridade.
Arquivos de Neuropsiquiatria, 52, 1 7.
Blackmore, S. (1999). The Meme Machine. New York: Oxford University Press.
Blanch, R. J., Brennan, D., Condon, B., Santosh, C., & Hadley, D. (2004). Are there gender-specific neural substrates of route learning from different perspec-
tives? Cerebral Cortex, 14, 1207 1213.
Bonaiuto, R., Rocca, E., & Lippi, A. (1990). Impact of education and occupation on the prevalence of Alzheimers disease (AD) and multi infarct dementia in
Appignano, Macerata Province, Italy [abstract]. Neurology, 40, 346.
Bowler, J. V., Munoz, D. G., Merskey, H., & Hachinski, V. (1998). Factors affecting the age of onset and rate of progression of Alzheimers disease. Journal of
Neurology, Neurosurgery and Psychiatry, 65, 184190.
Bramao, I., Mendonca, A., Fasca, L., Ingvar, M., Petersson, K. M., & Reis, A. (2007). The impact of reading and writing on visuo-motor integration task: A
comparison between illiterate and literate subjects. Journal of the International Neuropsychological Society, 13, 359 364.
Brucki, S. M., Malheiros, S. M., Okamoto, I. H., & Bertolucci, P. H. (1997). Dados normativos para o teste de fluencia verbal categoria animais em nosso meio.
Arquivos de Neuropsiquiatria, 55, 5661.
Brucki, S. M., & Rocha, M. S. (2004). Category fluency test: Effects of age, gender and education on total scores, clustering and switching in Brazilian
Portuguese-speaking subjects. Brazilian Journal of Medical and Biological Research, 37, 17711777.
Brucki, S. M. D., Nitrini, R., Caramelli, P., Bertolucci, P. H. F., & Okamoto, I. H. (2003). Sugestoes para o uso do Mini-Exame do Estado Mental no Brasil.
Arquivos de Neuropsiquiatria, 61, 777 781.
Bruner, J., Oliver, R., & Greenfield, P. (1966). Studies in cognitive growth. New York: Wiley.
Buffery, A., & Gray, J. (1972). Sex differences in the development of spatial and linguistic skills. In C. Ounsted & D. C. Taylor (Eds.), Gender differences, their
ontogeny and significance. New York: Churchill Livingstone.
Callahan, C. M., Hall, K. S., Hui, S. L., Beverly, S. M., Unverzagt, F. W., & Hendrie, H. C. (1996). Relationship of age, education, and occupation with demen-
tia among a community-based sample of African Americans. Neurologic Reviews, 53, 134140.
Cameron, R. F., Currier, R. D., & Haerer, A. F. (1971). Aphasia and literacy. British Journal of Disorders of Communication, 6, 161163.
Capitani, E., Barbarotto, R., & Laicana, M. (1996). Does education influence age related cognitive decline? A further inquiry. Developmental
Caplan, P. J., Crawford, M., Hyde, J. S., & Richardson, J. T. E. (1997). Gender differences in human cognition. New York: Oxford University Press.
Caramelli, P., Poissant, A., Gauthier, P., Bellavance, A., Gauvreau, D., Lecours, A. R., et al. (1997). Educational level and neuropsychological heterogeneity in
dementia of the Alzheimer type. Alzheimer Disease and Associated Disorders, 11, 9 15.
Carreiras, M., Seghier, M. L., Baquero, S., Estevez, A., Lozano, A., Devlin, J. T., et al. (2009). An anatomical signature for literacy. Nature, 461 (7266), 983
986.
Castro, S. L., & Morais, J. (1987). Ear differences in illiterates. Neuropsychologia, 25, 409417.
Castro-Caldas, A. (2004). Targeting regions of interest for the study of the illiterate brain. International Journal of Psychology, 39 (1), 5 17.
Castro-Caldas, A., Miranda, P. C., Carmo, I., Reis, A., Leote, F., Ribeiro, C., et al. (1999). Influence of learning to read and write on the morphology of the
corpus callosum. European Journal of Neurology, 6, 23 28.
Castro-Caldas, A., Nunes, M. V., Maestu, F., Ortiz, T., Simoes, R., Fernandes, R., et al. (2009). Learning orthography in adulthood: A magnetoencephalo-
graphic study. Journal of Neuropsychology, 3, 17 30.
Castro-Caldas, A., Peterson, K. M., Reis, A., Askelof, S., & Ingvar, M. (1998). Differences in inter-hemispheric interactions related to literacy, assessed by
PET. Neurology, 50, A43.
Castro-Caldas, A., Petersson, K. M., Reis, A., Stone-Elander, S., & Ingvar, M. (1998). The illiterate brain: Learning to read and write during childhood influ-
ences the functional organization of the adult brain. Brain, 121, 10531064.

Castro-Caldas, A., Confraria, A., Paiva, T., & Trindade, A. (1987). Contrecoup injury in the misdiagnosis of crossed aphasia. Journal of Clinical and
Experimental Neuropsychology, 8, 697701.
Chandra, V., Ganguli, M., Ratcliff, G., Pandav, R., Sharma, S., Belle, S., et al. (1998). Practical issues in cognitive screening of elderly illiterate populations in
developing countries. The Indo-US Cross-National Dementia Epidemiology Study. Aging (Milano), 10, 349357.
Childe, V. G. (1936). Man makes himself. London: Pitman Publishing.
Christensen, H., & Henderson, A. S. (1991). In age kinder to the initially more able? A study of eminent scientists and academics. Psychological Medicine, 21,
935946.
Ciborowski, I. J. (1979). Cross-cultural aspects of cognitive functioning: Culture and knowledge. In A. J. Marsella, R. G. Tharp, & I. J. Ciborowski (Eds.),
Perspectives in cross-cultural psychology (pp. 101 116). New York: Academic Press.
Coe, M., & Vanstone, M. (2005). Reading the Maya Glyphs. London: Thames & Hudson.
Cole, M., Frankel, F., & Sharp, D. W. (1971). The developmental of free recall learning in children. Developmental Psychology, 4, 109 123.
Cole, M., Gay, J., Glick, J., & Sharp, D. W. (1971). The cultural context of learning and thinking. Basic Books.
Cole, M., & Scribner, S. (1974). Culture and thought. New York: Wiley.
Cornelious, S. W., & Caspi, A. (1987). Everyday problem solving in adulthood and old age. Psychology of Aging, 2, 144153.
Critchley, M. (1956). Premorbid literacy and the pattern of subsequent aphasia. Proceedings of the Society of Medicine, 49, 335 336.
Damasio, A. R., Castro-Caldas, A., Grosso, J. T., & Ferro, J. M. (1976). Brain specialization for language does not depend on literacy. Archives of Neurology,
33, 300 301.
Damasio, H., Damasio, A. R., Castro-Caldas, A., & Ferro, J. M. (1979). Reversal for ear advantage for phonetically similar words in illiterates. Journal of
Clinical Neuropsychology, 1, 331 338.
Dansilio, S., & Charamelo, A. (2005). Constructional functions and figure copying in illiterates or low-schooled Hispanics. Archives of Clinical
Neuropsychology, 20, 1105 1112.
Dartigues, J. F., Gagnon, M., & Michel, P. (1991). Le programme de recherche PAQUID su lepidemiologie de la demence methodes et resultats initiaux. Revue
Neurologique, 145, 225 230.
Da Silva, C. G., Petersson, K. M., Fasca, L., Ingvar, M., & Reis, A. (2004). The effects of literacy and education on the quantitative and qualitative aspects of
semantic verbal fluency. Journal of Clinical and Experimental Neuropsychology, 26, 266277.
Deloche, G., Seron, X., Larroque, C., et al. (1994). Calculation and number processing: Assessment battery; role of demographic factors. Journal of Clinical
and Experimental Neuropsychology, 16, 195208.
Deloche, G., Souza, L., Braga, L. W., & Dellatolas, G. (1999). A calculation and number processing battery for clinical application in illiterates and semi-
literates. Cortex, 35, 503 521.
Duara, R., Barker, W. W., Lopez-Alberola, R., Loewenstein, D. A., Grau, L. B., Gilchrist, D., et al. (1996). Alzheimers disease: Interaction of apolipoprotein E
genotype, family history of dementia, gender, education, ethnicity, and age of onset. Neurology, 46, 15751579.
Eslinger, P. J., & Grattan, L. M. (1993). Frontal lobe and frontal-striatal substrates for different forms of human cognitive flexibility. Neuropsychologia, 31, 17
28.
Folia, V., & Kosmidis, M. H. (2003). Assessment of memory skills in illiterates: Strategy differences or test artifact? The Clinical Neuropsychologist, 17, 143
152.
Folstein, M. F., Folstein, S., & Mchugh, P. R. (1975). Mini-Mental State: A practical method for grading the cognitive state of patients for the clinicians.
Journal of Psychological Research, 12, 189198.
Fonseca, J., & Castro-Caldas, A. (2002). The knowledge of reading and writing rules influences aphasia outcome. Journal of the International
Fratiglioni, L., Grut, M., Forsell, Y., et al. (1991). Prevalence of Alzheimers disease and other dementias in an elderly urban population: Relationship with age,
sex, and education. Neurology, 41, 18861892.
Friedland, R. O. (1993). Epidemiology, education, and the ecology of Alzheimers disease. Neurology, 43, 246249.
Fuld, P. A. (1980). Fuld Object-Memory Evaluation. Wood Dale, Illinois: Stoelting.
Geary, D. C. (1996). Sexual selection and sex differences in mathematical abilities. Behavioral and Brain Sciences, 19, 229284.
Goldblum, M. C., & Matute, E. (1986). Are illiterate people deep dyslexics? Journal of Neurolinguistics, 2, 103 114.
Goncalves, P. M. (2004). Parents, presence, and education. Servir, 52, 2426.
Green, D. S., & Stewart, O. (1984). Test wiseness: The concept has no clothes. College Student Journal, 18, 416424.
Haarmann, H. (1991). Universalgeschichte der Schrift (2nd ed.). Frankfurt (Main)/New York: Campus.
Haarmann, H. (2008). Retrieved March 26, 2009, from, http://www.archaeomythology.org/journal/read_article.php?a=0108_2_haarmann.pdf.
Hedges, L. V., & Nowell, A. (1995). Sex differences in mental test scores, variability, and numbers of high-scoring individuals. Science, 269, 41 45.
Hendrie, H. C., Ogunniyi, A., Hall, K. S., et al. (2001). Incidence of dementia and Alzheimer disease in 2 communities: Yoruba residing in Ibadan, Nigeria, and
African Americans residing in Indianapolis, Indiana. Journal of the American Medical Association, 285, 739747.
Herrera, E., Caramelli, P., Silveira, A. S. B., & Nitrini, R. (2002). Epidemiological survey of dementia in a community-dwelling Brazilian population.
Alzheimer Disease and Associated Disorders, 16, 103108.
Hill, L. R., Klauber, M. R., Salmon, D. P., Yu, E. S., Zhang, M., & Katzman, R. (1993). Functional status, education, and the diagnosis of dementia in the
Shanghai survey. Neurology, 43, 138 145.
International Adult literacy Survey. (1998). Organization for Economic Co-operation and Development (OECD) and Statistics Canada. Retrieved Jun 26, 2007,
from http://www.statcan.ca/english/Dli/Data/Ftp/ials.htm.
Joanette, Y., Lecours, A. R., Lepage, Y., & Lamoureux, M. (1983). Language in right-handers with right-hemisphere lesions: A preliminary study including
anatomical, genetic and social factors. Brain and Language, 20, 217 248.
Judd, T., & DeBoard, R. (2007). Natural recovery: An ecological approach to neuropsychological recuperation. In B. Uzzell, M. Ponton, & A. Ardila (Eds.),
International handbook of cross-cultural neuropsychology. New York: Lawrence Erlbaum Associates.
Karavatos, A., Kaprinis, G., & Tzavaras, A. (1984). Hemispheric specialization for language in the congenitally blind: The influence of the Braille system.
Neuropsychologia, 22, 521 525.

Katzman, R. (1993). Education and the prevalence of dementia and Alzheimers disease. Neurology, 43, 1320.
Katzman, R., Aronson, M., Fuld, P., et al. (1989). Development of dementing illnesses in a 80-year-old volunteer cohort. Annals of Neurology, 25, 317 324.
Knoefel, J. E., Wolf, P. A., Linn, R. T., et al. (1991). Education has not effect on incidence of dementia and Alzheimers disease in the Framingham study.
Neurology, 41, 322 323.
Korczyn, A. D., Kahan, F., & Gulper, Y. (1991). Epidemiology of dementia in Ashkelom, Israel. Neuroepidemiology, 10, 100 120.
Kosmidis, M. H., Tsapkini, K., & Folia, V. (2006). Lexical processing in illiteracy: Effect of literacy or education? Cortex, 42, 1021 1027.
Kosmidis, M. H., Tsapkini, K., Folia, V., Vlahou, C. H., & Kiosseoglou, G. (2004). Semantic and phonological processing in illiteracy. Journal of the
International Neuropsychological Society, 10, 818 827.
Kotik-Friedgut, B., & Ardila, A. (2004). Systemic-dynamic Lurian theory and contemporary cross-cultural neuropsychology. In T. Akhutina, L. Moskovich,
& T. Dorothy (Eds.), Luria and contemporary psychology (pp. 5561). New York: Nova.
Laboratory of Comparative Human Cognition. (1983). Culture and cognitive development. In P. Mussen (Ed.), Handbook of child psychology: History, theories
and methods. New York: Wiley.
Lantz, D. (1979). A cross cultural comparison of communication abilities: Some effects of age, schooling and culture. International Journal of Psychology,
14, 171 183.
Launer, L. J., Dinkgreve, M. A. H. M., Jonker, C., Hooijer, C., & Lindeboom, J. (1993). Are age and education independent correlates of the mini-mental state
exam performance of community-dwelling elderly? Journal of Gerontology, 48, 271 277.
Laurendeau-Bendavid, M. (1977). Culture, schooling and cognitive development: A comparative study of children in French, Canada and Rwanda. In P. R.
Dasen (Ed.), Piagetian psychology: Cross cultural contributions. Cincinnati: Gardner.
Lave, J. (1996). The practice of learning. In S. Chaiklin, & J. Lave (Eds.), Understanding practice: Perspectives on activity and context. New York: Cambridge
University Press.
Lecours, A. R., Mehler, J., Parente, M. A., et al. (1987a). Illiteracy and brain damage I: Aphasia testing in culturally contrasted populations (control subjects).
Lecours, A. R., Mehler, J., Parente, M. A., et al. (1987b). Illiteracy and brain damage 2, manifestations of unilateral neglect in testing auditory comprehension
with iconographic material. Brain and Cognition, 6, 243265.
Lecours, A. R., Mehler, J., Parente, M. A., et al. (1988). Illiteracy and brain damage 3, a contribution to the study of speech and language disorders in illiterates
with unilateral brain damage (initial testing). Neuropsychologia, 26, 575589.
Lecours, A. R., Pena-Casanova, J., & Ardila, A. (1998). Orgenes y evolucion de la escritura. In A. R. Lecours, J. Pena -Casanova, & F. Dieguez-Vide (Eds.),
Dislexia y disgrafia: Teora, formas clnicas y exploracion. Barcelona, Spain: Masson.
Letenneur, L., Commenges, D., Dartigues, J. F., & Barberger-Gateau, P. (1994). Incidence of dementia and Alzheimers disease in elderly community residents
of south-westen France. International Journal of Epidemiology, 23, 1256 1261.
Lezak, M. D., Howieson, D. B., & Loring, D. W. (2004). Neuropsychological Assessment. New York: Oxford University Press.
Li, G., Cheung, R. T., Gao, J. H., Lee, T. M., Tan, L. H., & Fox, P. T. (2006). Cognitive processing in Chinese literate and illiterate subjects: An fMRI study.
Human Brain Mapping, 27, 144152.
Li, X. (2003). The earliest writing? Sign use in the seventh millennium BC at Jiahu, Henan Province, China. Antiquity, 77 (295), 3144.
Liu, H-C., Teng, E. L., Lin, K-N., Hsu, T. C., Guo, N-W., Chou, P., et al. (1994). Performance on a dementia screening test in relation to demographic variables.
Archives of Neurology, 51, 910 915.
Lord, A. B. (1995). The Singer Resumes the Tale. Ithaca: Cornell University Press.
Loureiro, C. S., Braga, L. W., Souza, L. N., Nunes Filho, G., Queiroz, E., & Dellatolas, G. (2004). Degree of illiteracy and phonological and metaphonological
skills in unschooled adults. Brain and Language, 89, 499 502.
Luria, A. R. (1931). Psychological expedition to central Asia. Science, 74, 383384.
Luria, A. R. (1933). The second psychological expedition to central Asia. Science, 78, 191 192.
Luria, A. R. (1966). Higher Cortical Functions in Man. New York: Basic Books.
Luria, A.R. (1973). The Working Brain. London: Penguin books.
Luria, A. R. (1976). Cognitive development. Harvard University Press.
Maccoby, E. E., & Jacklin, C. N. (1974). The psychology of sex differences. Palo Alto, CA: Stanford University Press.
Manly, J. J. (2006). Deconstructing race and ethnicity: Implications for measurement of health outcomes. Medical Care, 44 (11 Suppl. 3), S10S16.
Manly, J. J., Jacobs, D. M., Sano, M., Bell, K., Merchant, C. A., Small, S., et al. (1999). Effect of literacy on neuropsychological test performance in nonde-
mented, education-matched elders. Journal of the International Neuropsychological Society, 5, 191202.
Martos-Perez, J., & Ayuda-Pascual, R. (2003). Autismo e hiperlexia. Revista de Neurologia, 36, S57S60.
Matute, E. (1986). Aphasia in illiterates. Journal of Neurolinguistics, 2, 115130.
Matute, E. (1988). El aprendizaje de la lectoescritura y la especializacion hemisferica para el lenguaje. In A. Ardila & F. Ostrosky (Eds.), Lenguaje oral y
escrito. Mexico, DF: Trillas.
Matute, E., Leal, F., Zarabozo, D., Robles, A., & Cedillo, C. (2000). Does literacy have an effect on stick construction tasks? Journal of the International
Matute, E., Montiel, T., Pinto, N., Rosselli, M., Zarabozo, D., & Ardila, A. Neuropsychological test performance on healthy illiterate children. Manuscript
submitted for publication.
Montiel, T., & Matute, E. (2006). La relacion entre alfabetizacion y la escolarizacion con el desempeno en tareas verbales. In E. Matute (Ed.), Lectura y diver-
sidad cultural. Guadalajara, Mexico: Universidad de Guadalajara.
Morais, J., Bertelson, P., Cary, L., & Alegria, J. (1987). Literacy training and speech segmentation. In P. Bertelson (Ed.), The onset of literacy. Cognitive
process in reading acquisition. Boston: MIT Press.
Morais, J., Cary, L., Alegria, J., & Bertelson, P. (1979). Does awareness of speech as a sequence of phones arise spontaneously? Cognition, 7, 323 331.
Mortiner, J. A., & Graves, A. B. (1993). Education and other socioeconomic determinants of dementia and Alzheimers disease. Neurology, 43, 3944.
Nation, K. (1999). Reading skills in hyperlexia: A developmental perspective. Psychological Bulletin, 125, 338355.
Nell, V. (2000). Cross-cultural neuropsychological assessment: Theory and practice. New York: Lawrence Erlbaum Associates.

Nitrini, R., Bottino, C. M. C., Albala, C., Custodio, N. S., Ketzoian, C., Llibre, J. J., et al. (2009). Prevalence of dementia in Latin America: A collaborative
study of population-based cohorts. International Psychogeriatrics, 21, 622 630.
Nitrini, R., Caramelli, P., Herrera, E., et al. (2004a). Incidence of dementia in a community-dwelling Brazilian population. Alzheimer Disease and Associated
Disorders, 18, 241 246.
Nitrini, R., Caramelli, P., Herrera, E., Jr, Charchat-Fichman, H., & Porto, C. S. (2005). Performance in Lurias fist-edge-palm test according to educational
level. Cognitive Behavioral Neurology, 18, 211 214.
Nitrini, R., Caramelli, P., Herrera, E., Porto, C. S., Charchat-Fichman, H., Carthery, M. T., et al. (2004). Performance of illiterate and literate nondemented
elderly subjects in two tests of long-term memory. Journal of the International Neuropsychological Society, 10, 634 638.
No author. (1993). La naissance de lecriture. Lhistoire du Monde N82. Paris: Larousse.
Norton, A., Winner, E., Cronin, K., Overy, K., Lee, D. J., & Schanlg, G. (2005). Are there pre-existing neural, cognitive, or motoric markers for musical ability?
OConnor, D. W., Pollitt, P. A., & Treasure, F. P. (1991). The influence of education and social class on the diagnosis of dementia in a community population.
Psychological Medicine, 21, 219 234.
Ogunniyi, A., Baiyewu, O., Gureje, O., et al. (2000). Epidemiology of dementia in Nigeria: Results from the Indianapolis-Ibadan study. European Journal of
Neurology, 7, 485490.
Olson, D. R., & Torrance, N. (Eds.). (2009). The Cambridge Handbook of Literacy. Cambridge: Cambridge University Press.
Ostrosky, F., Ardila, A., & Rosselli, M. (1999). NEUROPSI: A brief neuropsychological test battery in Spanish. Journal of the International
Ostrosky, F., Ardila, A., Rosselli, M., Lopez-Arango, G., & Uriel-Mendoza, V. (1998). Neuropsychological test performance in illiterates. Archives of Clinical
Ostrosky-Sols, F., Arellano, M., & Perez, M. (2004). Can learning to read and write change the brain organization? An electrophysiological Study.
International Journal of Psychology, 39, 2735.
Ostrosky, F., Quintanar, L., Canseco, E., Meneses, S., Navarro, E., & Ardila, A. (1986). Habilidadis cognoscitivas y nivel sociocultural. Revista de
Investigacion Clinica, 38, 3742.
Ostrosky-Solis, F., Canseco, E., Quintanar, L., Navarro, E., Meneses, S., & Ardila, A. (1985). Sociocultural effects in neuropsychological assessment.
International Journal of Neuroscience, 27, 53 66.
Ostrosky-Sols, F., Efron, R., & Yund, E. (1991). Visual detectability gradients: Effects of illiteracy. Brain and Cognition, 17, 4251.
Ostrosky-Sols, F., Lozano, A., Ramrez, M., Picasso, H., Gomez, E., Velez, A., et al. (2003). Estudio Neuropsicologico de Poblacion Mexicana en Proceso de
Alfabetizacion. Revista Mexicana de Psicologa, 20, 5 17.
Ostrosky-Solis, F., Ramirez, M., Lozano, A., Picasso, H., & Velez, A. (2004). Culture or education: A study with indigenous Maya population. International
Journal of Psychology, 39, 36 46.
Padakannaya, P., Devi, M. L., Zaveria, B., Chengappa, S. K., & Vaid, J. (2002). Directional scanning effect and strength of reading habit in picture naming and
recall. Brain and Cognition, 48, 484 490.
Parkin, A. J., & Leng, N. R. C. (1993). Neuropsychology of the amnesic syndrome. Hove, UK: Erlbaum.
Paulesu, E., Demonet, J. F., Fazio, F., McCrory, E., Chanoine, V., Brunswick, N., et al. (2001). Dyslexia: Cultural diversity and biological unity. Science, 291,
2165 2167.
Petersson, K. M., Reis, A., Askelof, S., Castro-Caldas, A., & Ingvar, M. (2000). Language processing modulated by literacy: A network analysis of verbal
repetition in literate and illiterate subjects. Journal of Cognitive Neurosciences, 12, 364 382.
Petersson, K.M., Reis, A., & Ingvar, M. (2001). Cognitive processing in literate and illiterate subjects: A review of some recent behavioral and functional
neuroimaging data. Scandinavian Journal of Psychology, 42, 251267.
Qiu, C., Blackman, L., Winblad, B., et al. (2001). The influence of education on clinically diagnosed dementia incidence and mortality data from the
Kungsholmen Project. Archives of Neurology, 58, 20342039.
Quaiser-Pohl, C., & Lehmann, W. (2002). Girls spatial abilities: Charting the contributions of experiences and attitudes in different academic groups. The
British Journal of Educational Psychology, 72, 245 260.
Reis, A., & Castro-Caldas, A. (1997). Illiteracy: A bias for cognitive development. Journal of the International Neuropsychological Society, 3, 444 450.
Reis, A., Guerreiro, M., & Castro-Caldas, A. (1994). Influence of educational level of non brain-damaged subjects on visual naming capacities. Journal of
Clinical and Experimental Neuropsychology, 16, 939 942.
Reis, A., Petersson, K. M., Castro-Caldas, A., & Ingvar, M. (2001). Formal schooling influences two- but not three-dimensional naming skills. Brain and
Cognition, 47, 397411.
Reinvang, I., Bakke, S. J., Hugdahl, K., Karlsen, N. R., & Sundet, K. (1994). Dichotic listening performance in relation to callosal area on the MRI scan.
Riddoch, M. J., & Humphreys, G. W. (1987). A case of integrative visual agnosia. Brain, 110, 1431 1462.
Rilea, S. L., Roskos-Ewoldsen, B., & Boles, D. (2004). Sex differences in spatial ability: A lateralization of function approach. Brain and Cognition, 56, 332
343.
Robinson, W. P. (1974). Social factors and language development in primary school children. In R. Huxley & E. Ingram (Eds.), Language acquisition: Models
and methods. New York: Academic Press.
Rocca, W. A., Bonaiuto, S., Lippi, A., Luciano, P., Turtu, F., Cavarzeran, F., et al. (1990). Prevalence of clinically diagnosed Alzheimers disease and other
dementing disorders: A door-to-door survey in Appignano, Macerata Province, Italy. Neurology, 40, 626631.
Roe, C. M., Xiong, C., Miller, J. P., & Morris, J. C. (2007). Education and Alzheimer disease without dementia: Support for the cognitive reserve hypothesis.
Neurology, 68, 223 228.
Rogoff, B. (1990). Apprenticeship in thinking: Cognitive development in social context. Oxford: Oxford University Press.
Rosselli, D., Ardila, A., Pradilla, G., Morillo, L., Bautista, L., Rey, O., et al. (2000). El examen mental abreviado (Mini-Mental State Examination) como
prueba de seleccion para el diagnostico de demencia: estudio poblacional colombiano GENECO. Revista de Neurologia, 30, 428 432.
Rosselli, M., & Ardila, A. (2003). The impact of culture and education on nonverbal neuropsychological measurements: A critical review. Brain and Cognition,

52, 226 233.
Rosselli, M., Ardila, A., & Rosas, P. (1990). Neuropsychological assessment in illiterates II: Language and praxis abilities. Brain and Cognition, 12, 281 296.
Rosselli, M., Rosselli, A., Vergara, I., & Ardila, A. (1985). The topography of the hemi-inattention syndrome. International Journal of Neuroscience, 20, 153
160.
Sampson, G. (1985). Writing systems. Palo Alto, CA: Stanford University Press.
Schmandt-Besserat, D. (1996). How writing came about. Houston: University of Texas.
Scribner, S., & Cole, M. (1981). The psychology of literacy. Boston: Harvard University Press.
Serniclaes, W., Ventura, P., Morais, J., & Kolinsky, R. (2005). Categorical perception of speech sounds in illiterate adults. Cognition, 98 (2), B35B44.
Serpell, R. (1993). The significance of schooling: Life-journeys in an African society. Cambridge: Cambridge University Press.
Stephan, K., Marshall, J., Friston, K., Rowe, J., Ritzl, A., Killes, K., et al. (2003). Lateralized cognitive processes and lateralized task control in the human
brain. Science, 301, 384 386.
Stern, Y. (2006). Cognitive reserve and Alzheimer disease. Alzheimer Disease Associated Disorders, 20, S69 S74.
Stern, Y., Gurland, B., Tatemichi, T. K., Tang, M. X., Wilder, D., & Mayeux, R. (1994). Influence of education and occupation on the incidence of Alzheimers
disease. Journal of the American Medical Association, 271, 1004 1101.
Suddendorf, T., & Corballis, M. C. (1997). Mental time travel and the evolution of the human mind. Genetic, Social and General Psychology Monographs, 123,
133167.
Sulkava, R., Wikstrom, J., Aromaa, A., et al. (1985). Prevalence of severe dementia in Finland. Neurology, 35, 10251029.
Thorndike, R. L. (1951). In E. F. Lindquist (Ed.), Reliability in educational measurement. Washington, DC: American Council on Education.
Tzavaras, A., Kaprinis, G., & Gatzoyas, A. (1981). Literacy and hemispheric specialization for language: Digit dichotic listening in illiterates.
UNESCO (2005). Literacy. Retrieved March 27, 2009, from http://portal.unesco.org/education/en/ev.php-URL_ID=40341&URL_DO=DO_TOPIC&URL_
SECTION=201.html.
Vaid, J., Singh, M., Sakhuja, T., & Gupta, G. C. (2002). Stroke direction asymmetry in figure drawing: Influence of handedness and reading/writing habits.
van Ettinger-Veenstra, H. M., Ragnehed, M., Hallgren, M., Karlsson, T., Landtblom, A. M., Lundberg, P., & Engstrom, M. (2010). Right hemispheric brain
activation correlates to language performance. Neuroimage, 49, 34813488.
Ventura, P., Kolinsky, R., Querido, J. L., Fernandes, S., & Morais, J. (2007). Is phonological encoding in naming influenced by literacy? Journal of
Psycholinguistic Research, 36 (5), 341360.
Vygotsky, L. S. (1934/1978) Mind in society. Boston: Harvard University Press.
Wechsler, A. F. (1976). Crossed aphasia in an illiterate dextral. Brain and Language, 3, 164172.
Westerhausen, R., & Hugdahl, K. (2008). The corpus callosum in dichotic listening studies of hemispheric asymmetry: A review of clinical and experimental
evidence. Neuroscience and Biobehavioral Review, 32, 1044 1054.
Wilson, B. A., Cockburn, J., & Baddeley, A. (1985). Rivermead Behavioural Memory Test. San Antonio: Psychological Corporation.
Yassuda, M. S., Diniz, B. S., Flaks, M. K., Viola, L. F., Pereira, F. S., Nunes, P. V., et al. (2009). Neuropsychological profile of Brazilian older adults with
heterogeneous educational backgrounds. Archives of Clinical Neuropsychology, 24 (1), 7179.
Yu, E., Liu, W. T., Levy, P., et al. (1989). Cognitive impairment among elderly adults in Shanghai, China. Journal of Gerontology: Social Sciences, 44, S97
S106.
Zhang, M., Katzman, R., Salmon, D., Jin, H., Goujun, C., Wang, Z., et al. (1990). The prevalence of dementia and Alzheimers disease in Shanghai, China.
Annals of Neurology, 27, 428437.
Neuropsychology Review, Vol. 15, No. 4, December 2005 (
C 2005)
DOI: 10.1007/s11065-005-9180-y
Cultural Values Underlying Psychometric Cognitive Testing
Alfredo Ardila1,2
It is proposed that culture (values, beliefs, styles of behavior) can affect neuropsychological testing.
Cognitive testing represents a social situation thatas any social situation, it is one governed by
implicit cultural rules. At least eight different culture-dependent values underlie cognitive testing:
(1) One-to-one relationship, (2) Background authority, (3) Best performance, (4) Isolated environment
(5) Special type of communication, (6) Speed, (7) Internal or subjective issues, and (8) The use of
specific testing elements and strategies. In addition, it is proposed that the distance (e.g., gender,
age, ethnicity) between the examiner and the examinee may potentially impact the testing situation. A
special analysis regarding the function of instruction in cognitive tests is also presented emphasizing
that test instruction interpretation is also culture-dependent. Some potential avenues of research are
finally proposed.
KEY WORDS: cognitive testing; cultural values; culture differences; cross-cultural neuropsychology.
. . . a very limited kind of neuropsychology, appropriate tion of culture, including thinking, feeling, knowledge,
to only a fraction of the worlds population, is presented values, attitudes, and beliefs. (2) The behavioral dimen-
to the rest of the world as if there could be no other
sion, including the ways to relate with others, ways of
kind of neuropsychology, and as if the education and
cultural assumptions on which neuropsychology is based
behaving in different contexts and circumstances, festivi-
were obviously universals that applied everywhere in the ties and meeting, patterns of associations, etc. (3) Cultural
world. (Matthews, 1992, p. 421). elements: the physical elements characteristic of that hu-
man group such as symbolic elements, clothes, ornaments,
houses, instruments, weapons, etc. (Geertz, 2000; Harris,
WHAT IS CULTURE? 1983; Hofstede, 1997).
Culture represents a particular way to adapt to and
Culture refers to the set of learned traditions and live in a specific context. Cultural differences are strongly
living styles, shared by the members of a society. It in- related with environmental differences (Geertz, 2000;
cludes the ways of thinking, feeling and behaving (Harris, Peoples and Bailey, 2003). Arctic and Amazonian jungle
1983). Traditionally, culture has been interpreted as that cultural differences are to a significant extent due to the
complex whole which includes knowledge, belief, art, law, geographical, ecological and environmental differences
morals, custom, and any other capabilities and habits ac- between the Arctic and the Amazonian jungle regions.
quired by man (humans) as a member of society (Tylor, Cultures, however, are usually in some contact with each
1871). The minimal definition of culture could simply be, other and a significant degree of cultural diffusion is gen-
culture is the specific way of living of a human group erally observed. Cultural evolution and cultural changes
(Ardila, in press). are found throughout human history, depending upon,
Different dimensions of culture can be distinguished: (a) new environmental conditions, (b) contact with other
(1) The internal, subjective or psychological representa- cultures, and (c) internal cultural evolution (MacDonald,
1998; Linton, 1936; Sahlins and Service, 1960). For ex-
1 Department of Communication Sciences and Disorders, Florida ample, Gypsies in Russia, Spain, and the United States
International University, Miami, Florida.
2 To whom correspondence should be addressed at Department not only have many cultural commonalties, but also dif-
of Communication Sciences and Disorders, Florida International ferences due to the amount of time as well as degree of
University, HLS144, Miami, Florida 33199; e-mail: ardilaa@fiu.edu. interaction with and acceptance from their host societies.
185
1040-7308/05/1200-0185/0
C 2005 Springer Science+Business Media, Inc.
186 Ardila
Certain cultural elements have been particularly suc- attitudes (beliefs and feelings) and interpersonal behavior
cessful and have tended to strongly diffuse across cul- styles, may potentially affect cognitive testing. Cultures
tures (Linton, 1936). For instance, science and technology can be identified because of distinctive styles of behav-
have been extremely successful in solving different hu- iors which are based upon different patterns of beliefs and
man problems and have, in consequence, tended to spread these in turn are due to patterns of values.
throughout virtually all-existing world-wide societies. In
this regard, contemporary humanity has tended to become
more technologically homogeneous and to share the cul- Values
ture of science and technology. As communication speed
and access to shared information increases exponentially Culture supposes a specific value system. Kluckhohn
over its historic archive of record, cultural diffusion and and Strodbeck (1961) proposed a conceptual map which
its impact has become a considerable variable of research tried to include the complete range of values which it is
interest. possible for human beings to hold in relation to five key
Formal education and school have played a crucial issues about which all human beings hold opinions. The
role in the diffusion of science and technology, and in the five key issues are (1) human nature (how it is: good,
contemporary trend toward cultural homogenization. In bad, neutral, mixed), (2) mannature relationship (sub-
this regard, school can be considered as a subculture, the jugation to nature, harmony with nature or mastery over
subculture of school (Ardila et al., 1989, 2000). School not nature), (3) time orientation (past, present, future), (4) ac-
only provides some common knowledge but also trains tivity (being, being in becoming, doing), and (5) relational
some abilities and develops certain attitudes. Cognitive (lineality, collaterality, individuality). The prevalent value
testing is obviously based on the assumptions as well orientations of an individual, and indeed of the culture to
as the values of scientific and technologically oriented which s/he belongs, can act as a barrier to intercultural
societies. Schooled children usually share more scien- communication in that, what passes for common sense in
tific and technologic values and attitudes than their illit- one culture might appear deviant in another (Condon and
erate parents, and schooled subjects significantly outper- Yousef, 1981).
form illiterate individuals in cognitive testing (e.g., Ardila
et al., 1989; Goldblum and Matute, 1986; Lecours et al.,
1987a, 1987b, 1988; Manly et al., 1999; Matute et al., Attitudes and Beliefs
2000; Ostrosky et al., 1998; Reis and Castro-Caldas, 1997;
Rosselli et al., 1990). Usually refers to positive or negative evaluations of
School attendance, however, does not mean that edu- people, objects, or situations that often predispose people
cated people simply possess certain abilities that less edu- to feel and behave positively or negatively toward them.
cated individuals are lacking. It does not mean that highly According to Greif (1994) cultural beliefs are the ideas
educated people have the same abilities as less educated and thoughts common to several individuals that govern
individuals, plus something else (Ardila et al., 2000). The interactionbetween these people, and between them,
individual with no formal education obviously have cer- their gods, and other groupsand differ from knowledge
tain learnings that educated people do not. Nonetheless, in that they are not empirically discovered or analytically
formal cognitive testing evaluates those abilities that the proved. In general, cultural beliefs become identical and
educated people was trained in, and is not surprising that commonly known through the socialization process by
s/he will outperform the subjects with no formal edu- which culture is unified, maintained, and communicated.
cation. Noteworthy, educational level has a substantial Cultural beliefs serve not only to guide, but also to insti-
relationship with performance on some cognitive tests but gate action, that is, they possess emotional content. During
is not systematically related to everyday problem solv- childhood, people acquire beliefs and values from others
ing (functional criterion of intelligence; Cornelious and by teaching, imitation, and other forms of social learning.
Caspi, 1987). Luria (1976), for instance, observed that il- This process is known as cultural transmission. Many
literate individuals can have more rich and differentiated kinds of cultural transmission are possible.
perception, but they lack skills in the categorization and
generalization processes trained at school.
To operationalize culture is not simple. As mentioned Interpersonal Behavior
above, culture refers to the living styles (thinking, feel-
ing and behaving) shared by the members of a society For a long time, anthropology and cross-cultural
(Harris, 1983). Values (ways of thinking and interpreting), psychology have been interested in analyzing the ways
Cultural Values Underlying Psychometric Cognitive Testing 187
people interpret interpersonal behavior (Adamopoulos, As a matter of fact, neuropsychological testing uses
2002). The core question is, how people understand two different sources in assessing intellectual abilities:
the social behavior they experience in their environ- psychometric and neurological tradition. Both have been
ment. It has been proposed that there exist at least developed within the Western cultures. This does not
three universal dimensions used to interpret interper- mean, however, that every Western culture individual
sonal behavior: (1) Association-Dissociation (Affiliation); is equally familiar with neurological and psychometric
(2) Superordination-Subordination (Dominance); and testing procedures. Psychometric testing procedures have
(3) Intimacy-Formality. Naturally, this does not mean that particularly flourished in some specific countries. It is
other, culture-specific dimensions, do not exist (Triandis, understandable that some cultures and individuals do not
1978, 1994). It has been further proposed that around have many good models or roles for the undertaking of
the world, regardless of cultural, ethnic, and linguistic one person directly examining the intellectual abilities of
background, people understand social behavior as com- another. The idea that the examiner is able to determine the
municating primarily the presence or absence of affiliative characteristics of cognitive abilities using some specific
motives, the desire to dominate another or be submissive testing procedures may not be familiar to many people
to anothers authority, and the need for interpersonal close- worldwide. It may seem strange to some examinees to
ness (or its absence; Adamopoulos, 1988, 1991). have the examiner asking questions to which the examiner
An increasing emphasis in cultural variables, poten- obviously knows the answer (Nell, 2000).
tially affecting cognitive test performance has been evi-
dent during recent years (e.g., Ardila, 1995; Ferraro, 2002;
Fletcher-Janzen et al., 2000; Nell, 2000; Perez-Arce and HOW CULTURE AFFECTS COGNITIVE
Puente, 1996; Ponton and Leon-Carrion, 2001; Puente TEST PERFORMANCE
and Perez-Garcia, 2000; Puente and Salazar, 1998; Puente
and McCaffrey, 1992; Rosselli and Ardila, 2003; Samuda Luria (1966, 1973) and Vygotsky (1934/1978) ana-
et al., 1998; Toomela, 2003; Geisinger, 1992; Uzzell et al., lyzed the interaction between biological and cultural fac-
in press). Progressively, it has become evident that cog- tors in the development of human cognition. The main
nitive testing represents a culture-dependent activity and purpose of Lurias expedition to Uzbekistan during the
test scores may be affected by different culture conditions 1930s (Luria, 1931, 1933) was to investigate the influ-
and idiosyncrasies (Ardila, 2003). Neuropsychology has ence of culture, and particularly, its most important in-
become more and more aware of the cultural dimension stitution, education, on the development of higher mental
in behavior and cognition. functions. According to Luria (1973), mental functions
Furthermore, cognitive testing itself is a cultural con- are . . . social in origin and complex and hierarchical in
cept and was developed within a specific cultural context: their structure and they all are based on a complex system
Western societies (Puente and Agranovich, 2003). The of methods and means . . . (p. 30). An intrinsic factor
attempt to measure cognitive abilities represents one of in systemic organization of higher mental functions is the
the major endeavors in twentieth-century psychology. A engagement of external artifacts (objects, symbols, signs),
tremendous amount of research has been directed to un- which have an independent history of development within
derstanding the organization of intellectual activity and culture. It is this principle of construction of functional
to discuss the procedures appropriate to its measurement. systems of the human brain that Vygotsky (1934/1978)
As known, the history begins in France at the turn of called the principle of extracortical organization of com-
the century. In 1904 the Ministry of Education in France plex mental functions, implying that all types of human
commissioned Alfred Binet and Theophile Simon to de- cognitive processes are always formed with the support
velop a practical procedure to distinguish between men- of cultural elements.
tally retarded and normal children at school. To fulfill Greenfield (1997), in a magisterial paper published in
this purpose, they developed kind of developmental scale the journal American Psychologist, pointed out that there
describing the types of abilities that were normally ex- are three different reasons to account for why ability as-
pected at different ages (Binet, 1905, 1908). The concept sessments do not cross cultures: (1) Values and meanings,
of mental age was introduced to refer to the level of (2) modes of knowing, (3) and conventions of communi-
development expected at each age. The Binet-Simon tests cation.
were rapidly adopted in England, United States, and other Values and meanings refer to a lack of general agree-
Western countries. In the United States Terman (1916) ment on the value or merit of particular responses to par-
adapted and standardized the scales presented by Binet ticular questions. For example, some people may consider
and further developed the concept of IQ. that in the Ravens Progressive Matrices test it is a better
188 Ardila
response to focus on following an aesthetic principle (i.e., In contrast, parents who believe that children grow up
the figure that looks better in that position) rather than the tend to make fewer attempts at dialogue with their young
one sequenced according to a conceptual principle (i.e., child, and are less likely to prompt their child to recount
the figure that continues the sequence; Nell, 2000). Heath events to practice verbal communication.
(1989) found that many African American children do Finally, in many societies it can be frankly inappro-
not believe that authority figures (parents, teachers, test priate to question a stranger in an impersonal manner. In
designers, etc.) reward obvious answers, but rather that Latin societies, for example, it is expected that before test-
they do reward creative or expansive answers. If this is ing, some interpersonal relationship will be established
true, then they may prefer to select no obvious answers. between the examiner and the examinee (Dingfelder,
Furthermore, the same items do not necessarily have the 2005). Talking and interchanging ideas for a while before
same meaning in different cultures, regardless of how ap- beginning testing can be a prerequisite for a successful
propriate and accurate the translation is. An item referring testing. Without this previous personal contact, testing
to the protection of animals may have a rather different can be aversive and culturally disconcerting.
meaning in Europe than in a hunting society.
Knowing may be a collective endeavor, not an in-
dividual task. Many members of collective societies are CULTURAL VALUES IN PSYCHOMETRIC
distressed by testing situations that require individual re- COGNITIVE TESTING
sponses without the participation of the social group;
usual activities, such as farming, building houses, etc., Culture dictates what is and what is not relevant and
are carried out collectively. If most activities are carried significant in a particular context or situation (Berry, 1979;
out in a collective way, why should answering a test be Berry et al., 1992; Irvine and Berry, 1988). What is rele-
the exception? Many cultures, on the other hand, do not vant and worth learning or doing for an Eskimo does not
make a distinction between the process of knowing and necessarily coincide with what is relevant and worth learn-
the object of knowing. In consequence, questions such as ing or doing for an inhabitant of the Amazonian jungle.
why do you think?, or Why do you consider? may A culture provides specific models for ways of thinking,
be incomprehensible. The point is not what I think or I acting and feeling, and culture variations in cognitive test
consider; the point is how it is. The personal interpretation scores are evident (Anastasi, 1988; Heath, 1989, 1997;
is not the point; How do you consider that this sequence Wong et al., 2002).
should be continued? may be a question that does not Current neuropsychological testing uses specific
make too much sense. The real question is How does this conditions and strategies that may be not only unfamiliar
sequence continue?, and obviously others participation to many people, but may also violate some accepted cul-
may be required to find out the correct answer. tural norms. As mentioned above, interpersonal relations
Conventions of communication are highly culture- can be interpreted according to three dimensions: affilia-
dependent. Test questions assume that a questioner who tion, dominance and intimacy-formality (Triandis, 1978,
already has a given piece of information can sensibly ask 1994). Cultural variations in the characteristics of each
a listener for the same information. To ask or to answer one can be anticipated.
questions can be highly variable among cultures. Ameri- At least the following cultural values underlie psy-
can children, for example, learn that they should not talk chometrically oriented cognitive testing:
to strangers, but they also learn that they should answer
questions to the doctor, regardless of the doctor being a One-to-One Relationship
stranger. Furthermore, relevant information is not always
the same in every culture. Many types of questions can It refers to a specific way to relate with other peo-
be difficult to understand. To copy nonsense figures (e.g., ple (association dimension) (Triandis, 1978, 1994). There
Rey-Osterrieth Complex Figure) can be a cause for suspi- is an examiner and there is an examinee. Hence, it is
cion for many people. Heath (1989) proposed a distinction a one-to-one relationship between two people that very
between societies in which children are thought to grow likely never met before, are aliens, and will not meet ever
up and those in which children are raised or brought again in the future. In many cultures, activities are car-
up. She noted that parents who believe that children must ried out in a societal way. This style of behavior can be
be raised engage in a distinct set of verbalizations with perceived as contradicting the usual cultural way of liv-
their children involving highly specific verbal commu- ing (i.e., different people collaborate in diverse activities).
nication about events, requests for children to recount This is particularly true in communities relying heavily in
step-by-step features of their own actions, and so forth. social collaboration for the different daily life activities.
In individual-oriented (versus social-group-oriented) requesting best performance in testing supposes that the
societies, this condition may be easier to accept (Johnson- examinee knows the theoretical foundation of psychomet-
Powell et al., 1997). Heath (1983) suggested that each ric testing. Quite frequently, obviously this is not the case.
community has specific ways of socializing members and
helping them function in the community. There are several
features in social and linguistic environments which vary Isolated Environment
strikingly from one community to the other, including
the boundaries of the physical and social communities It also refers to a specific way to relate with other peo-
in which communication is possible; and the limits and ple (association dimension; Triandis, 1978, 1994). Test-
features of the situations in which talk occurs. ing is often done in an isolated room. Doors are often
closed and even locked. Usually, no one else is allowed
to be present, and in this regard it is a private and in-
Background Authority timate situation. Private appointments with aliens may
be quite inappropriate in many cultures. The examinee
It implies a subordinate relationship (dominance di- has to understand and accept this type of unusual social
mension; Triandis, 1978, 1994). Background or situa- relationship.
tional authority refers to the authority associated with a
particular role (Berry et al., 1992). Thus, the bank clerk
has the background authority to ask for our ID; the den- Special Type of Communication
tist has the background authority to request opening and
closing our mouth (but not to ask for our ID); and the It refers to the intimacy-formality dimension in inter-
barber has the authority to place a razor on our neck (but personal behavior (Triandis, 1978, 1994). Examiner and
not to ask for our ID or to ask for opening and closing the examinee do not maintain a normal everyday life con-
mouth!). In a standard testing situation, the examinee has versation or social rapport. Examiner uses stereotyped
to follow (obey) the instructions given by the examiner, utterances in a rather formal language. Examinees are not
and hence, the examiner is supposed to have a background allowed to talk about themselves in an informal way. This
or situational authority. It may be not so easy, however, is a type of formal communication relationship that can
to understand by whom and why this authority was con- be different from any previous type of relationship in the
ferred. Why should the examinee obey the examiner? This participants past experience. Heath (1983) pointed out
reluctance to obey may be especially evident if the exam- that the different ways children learned to use language
iner has certain personal characteristics; for example, age, were dependent on the ways in which members of each
gender, ethnicity, class or caste relative to the examinee. community structured their families, defined the roles that
community members could assume, and played out their
concepts of childhood that guided child socialization. She
Best Performance found that for children to get along with people and to
accomplish social goals, they need to learn their commu-
To do ones best may be most significant in a cul- nitys ways of language use, and they also acquire those
ture highly valuing competition, but not in a less compet- ways of using language through experiences in various
itive society. Psychometrically oriented cognitive testing community activities and interactions.
has flourished in highly competitive societies, often to the
disadvantage of those who do not value or understand
the process and outcome applications as is the case with Speed
current high stakes testing in American educational set-
tings. The examinee will perform at their optimal level. Time is understood differently across different cul-
Performance at best is only done in those endeavors tures (Adam, 1990; Hughes and Trautmann, 1995). Some
that are perceived and regarded as extremely important anthropological studied have approached time perception
and significant. It is assumed that the examinee has to in different cultural contexts (Adam, 1990; Munn, 1992;
perceive the testing as a most important and significant Hughes and Trautmann, 1995). Time has frequently been
endeavor. It may not be clear why it is so important and interpreted as a social and cultural construct rather than a
relevant to repeat a series of nonsense digits or to draw natural phenomenon (Goody, 1991). The main elements
an absurd figure. Why can it be so important to do well of time occur in all human societies, yet with different
in a memory test recalling every detail of the silly story emphases: sequence and duration, cyclical and linear pat-
of a truck driver who had an accident? In a certain sense, terns, and systems of time reckoning Perception of time,
190 Ardila
its conceptualization and the representation of concepts the culture where the test was developed usually obtain
across cultures are, however, culture-dependent, reflect- the highest scores.
ing specific cultural experience ( Lebedko, 2001). Speed Many of the presumptions implicitly assumed by cur-
supposes a particular type of time orientation (Kluckhohn rent cognitive tests that were mentioned above (e.g., best
and Strodbeck, 1961). For many cultural groups speed performance, background authority, etc) simply mean
tests are frankly inappropriate. Speed and quality may be that the testee has to have certain background knowledge
contradictory, and good products are the results of a slow about psychometric testing instruments. To some extent,
and careful process. Significant differences in attitudes s/he has to share the values of the culture where these
to timed procedures are found across different cultures instruments were developed.
(Puente and Agranovich, 2003; Perez-Arce and Puente,
1996). For instance, American children are used to be
tested since elementary school using timed tests, and in DISTANCE BETWEEN THE EXAMINER
general, American school reinforces the value that faster AND THE EXAMINEE
the performance, the better the result (Nell, 2000).
The demographic characteristics of the examiner
may play a crucial role in neuropsychological assessment.
Internal or Subjective Issues
Variables such as the examiners age, gender, and ethnicity
may be easily overlooked.
It refers to the Intimacy-Formality dimension of in-
In many societies worldwide age is associated with
terpersonal behavior (Triandis, 1978, 1994). What does
social status. Elders are supposed to be wiser and to have
privacy mean, and hence, what does it mean to disre-
significant background knowledge. This cultural assump-
spect ones privacy, is a question that has to be considered
tion has a simple consequence: older examiners deserve
within the culture framework. Dingfelder (2005) suggests
confidence; younger ones, not as much or to the same
that when dealing with Latin clients Therapists might
extent. It is easier to accept some authority and follow
consider sharing some minor details of their lives with
instructions given by older examiners; there may be more
these clients, to make the clients feel more comfortable
suspicion of younger examiners. On the other hand, in a
and welcome (p. 59). This type of intimacy is considered
society emphasizing progress and technology,such as
to be culturally appropriate. But asking the patient Are
the American societyto be older does not necessarily
you feeling depressed? may be regarded as an inappro-
mean to be wiser. The opposite pattern can even be true:
priate and invasive question. Questions about cognitive
younger members may have a better knowledge of sci-
issues (e.g., How is your memory?) may be questions
ence and technological issues. Age differences between
about internal subjective representations, the most per-
the tester and the testee do not seem to be a significant
sonal private sphere in some cultures.
barrier for Americans. But, it may be in many worldwide
cultures.
The impact of gender in neuropsychological testing
Use of Specific Testing Elements can be twofold: (1) the gender of the examiner; (2) the
and Testing Strategies gender match or mismatch between the examinee and ex-
aminer. Doctors may more frequently be males in some
Physical elements (figure, blocks, pictures, etc.) are cultures but there are also cultures in which doctors
culture-dependent elements (Geertz, 2000; Harris, 1983; are more frequently females. For instance, the ratio of
Hofstede, 1997). In a standard neuropsychological evalu- men/women for medical doctors in the United States
ation, the examiner uses figures, blocks, pictures, etc., as favors males (positive) but in Russia it favors females. By
part of the test materials. These physical elements may be the same token, the ratio men/woman psychologists is
unfamiliar to the client, or at least, not equally familiar for variable across different countries. It may be assumed that
clients with different cultural backgrounds. when attending cognitive testing, the examinee might have
In summary, the rationale and the procedures used in certain expectations about the gender of the examiner. It
cognitive testing rely on a whole array of cultural values is not yet clear if and how these expectations may affect
cannot be regarded as universal values. When examiners the testing situation.
use tests developed in their own culture to test members Gender matching or mismatching between the exam-
of a different culture, examinees often do not share the iner and the examinee has not been frequently analyzed.
presumptions implicitly assumed by the test (Greenfield, We simply do not know if it has any significance or for
1997, p. 1115). It is not surprising that the members of whom, as findings in certain areas of psychology (e.g.,
family therapy, sex therapy, etc.) indicate, it may be sig- tions in the first way, while most Latin American (and
nificant. Politeness (and willing to collaborate) may be European?) neuropsychologists interpret test instructions
higher when there is a gender mismatch between examiner in the second way. This is an issue that is rarely overtly an-
and examinee. For certain specific populations (e.g., peo- alyzed in testing. If the first interpretation is correct, obvi-
ple with specific sex attitudes) gender match or mismatch ously there is a language problem. Language usage differs
potentially may have some impact. according to cultural (and subcultural) background and
Ethnic matching may be a powerful variable in cog- strongly correlates with the subjects educational level.
nitive testing. By definition, people from other cultures are Sometimes, test instructions (and in general, the language
aliens (exogroup). People sharing the same culture are en- used in testing) are given in a formal languagenot in an
dogroup. Exogroup people raise suspicions to endogroup everyday normal language, which may be very difficult for
people; endogroup people deserve confidence (Harris, individuals with limited education to understand. Formal
1983). Communication flows easily with someone else language represents a sort of academic language, most
who shares in the same culture values and attitudes. Com- often found in a written form that many people neither
munication appears artificial and distancing with someone use nor completely understand. If the second interpreta-
belonging to a different cultural group. It can be antici- tion is correct (instructions are just a guideline), there
pated that the cultural mismatch between the examinee is an issue with regard to the examiners verbal ability to
and the examiner may result in a decreased cognitive test communicate the test instructions in a clear manner. The
performance. Different studies support this assumption examiner also has a cultural background, biasing his/her
(e.g., Miller and Rotatori, 1990; Mishra, 1980; Torrell understanding and interpretation of the testing situation
and Torrell, 1983). By the same token, acculturation and instructions. The interpretation of the instructions
(usually understood in United States as assimilation of may be biased depending on the examiners cultural back-
the American culture values) has been observed to result ground. As an example, the WAIS-III (Wechsler, 1997) is
in increased scores in cognitive testing (e.g., Manly et al., frequently regarded as the most reliable elaborated cog-
1998). nitive test. The tests Administration and Scoring Manual
(p. 63) states
THE FUNCTION OF INSTRUCTIONS
Before beginning the first subtest, introduce the WAIS-III
IN COGNITIVE TESTS
by saying I will be asking you to do a number of
things . . .
Regardless of the long tradition in psychometric test-
ing, literature gives the impression that in cognitive testing The following paragraph continues this instruction
there is a lack of agreement about what the function of with
test instructions and what impact they have on the exam-
inee. Or rather, there are implicit cultural values in the Examinees will differ in the amount of explanation that
they need before testing begins. Use your judgment to
interpretation of test instructions. decide if elaboration is needed . . .
Test instructions can be interpreted in two different
ways: Up to this point, it seems that the correct interpreta-
tion of the test instructions in the WAIS-III is the second
1. A standard set of statements that have to be pre-
one (i.e., the function of the instructions is to understand
sented to the examinee every time in exactly the
what the test is about and what the examinee must do).
same way. The examiner must tell the instructions
Nonetheless, when the description of the different
verbatim as they are written; instructions may or
subtests is presented, no indication to using judgment to
may not be repeated according to the test manual.
decide if elaboration is needed is readily found. Does
Otherwise, the testing would not be standard.
it mean that the correct interpretation of instructions is,
2. The function of the instructions is to understand
beginning at this point, the first one? (i.e., a standard set
what the test is about; what it is that the examinee
of statements that has to be presented to the examinee all
must do. Hence, test instructions are just a guide-
the times in exactly the same way)?
line for the examiner. Instructions can be worded
For instance, in describing the first subtest of the
and adapted according to the specific situation.
WAIS-III (Picture Completion), the Manual (p. 64) states
Further, you can add emphasis, pauses, specific
Before presenting any items, say I am going to show
prosody, etc.
you . . .. No indication to using judgment to decide if
In my clinical experience I have found that most elaboration is needed is mentioned. It is not even men-
neuropsychologists in the United States interpret instruc- tioned if the statements made by the examiner can be
192 Ardila
or not repeated. Or, what to do if the examinee asks for of close interpersonal relationships; Dingfelder,
additional explanation or makes some comments. Does 2005). These values are obvious if the examiner
the guideline presented at the beginning hold or not (Ex- belongs to the same culture that the examinee. But,
aminees will differ in the amount of explanation that they they may seem strange when there is a mismatch
need before testing begins. Use your judgment to decide in culture. The consequence is obviously that the
if elaboration is needed. . .)? It is not clear. Obviously, examiner and examinee should share the same
there are some implicit assumptions. My point is that there cultural background. Or the examiner should be
is an ambiguity. It is usually assumed that the Adminis- familiar enough with the cultural background of
tration and Scoring Manual is perfectly understandable the examinee.
for any psychologist. As it is written, however, interpreta- 3. Pinpointing cultural bias in neuropsychological
tion may depend on the examiners cultural background. tests. Selected statistical procedures can be used
There are implicit cultural values in understanding the test in detecting cultural bias. Some attempts in this
instructions. direction are currently available (e.g. Reynolds,
Does this difference in interpretation of the test in- 2000).
structions impact test performance? We cannot be sure, but 4. Re-developing cognitive tests according to the
it may be conjectured that it will impact the test perfor- cultural conditions. Simply speaking, making
mance in two different ways: (1) additional instructions tests more culturally and ecologically relevant.
may further help to understand the test; examinees not For instance, Da Silva et al. (2004) investigated
receiving additional instructions and explanations may be semantic verbal fluency tasks in illiterates and
penalized. (2) A rigid and stereotyped language may be literate participants. The performance on a food
acceptable for a person sufficiently familiar with psycho- criterion (supermarket fluency task), considered
metric testing. But it can be distancing, artificial, and even more ecologically relevant for the two literacy
distasteful for a person not sufficiently familiar with psy- groups, and an animal criterion (animal fluency
chometric testing. Ultimately, the function of language in task) were compared. The quantitative analysis
the everyday life is to talk. This is a basic understanding indicated that the two literacy groups performed
of language that most people have. equally well on the supermarket fluency task. The
overall results suggested that there is not a sub-
stantial difference between literate and illiterate
WHAT COULD BE DONE?
subjects related to the fundamental workings of
semantic memory. However, there is indication
There is not a simple answer to this question. Several
that the content of semantic memory reflects dif-
potential answers could be proposed:
ferences in shared cultural backgroundin other
1. Psychometric cognitive testing is only appropri- words, formal educationas indicated by the sig-
ate is those societies with a solid psychometric nificant interaction between level of literacy and
tradition, basically, Western societies. In other so- semantic criterion.
cieties, behavioral scales and more qualitative ap- 5. Obtaining appropriate norms for different cultural
proaches should be preferred. groups. A significant effort during the last years
2. Culturally most appropriate strategies should be has been observed in this direction. Just as an
selected in every society. Interpretations, values, example, different normative data are currently
and behaviors present variations according to available for verbal fluency tests using the cate-
the cultural context. These interpretations, val- gory ANIMALS (for a review, see Lezak, 2004;
ues and behaviors should be integrated in test- Mitrushina et al., 1999; Spreen and Strauss, 1998).
ing, and testing performed considering cultural Cross-cultural studies have found validity for ver-
idiosyncrasies. For instance, if has been pointed bal fluency tests in speakers of languages other
out that the following cultural values can af- than English; for example, Spanish (Ardila and
fect a psychologistclient relationship in Hispanic Rosselli, 1994), Finnish (Klenberg et al., 2001),
patients: familismo (family may have a signifi- Hebrew (Axelrod et al., 2001), Greek (Kosmidis
cant role in Latin cultures), simpata (Latino cul- et al., 2004), and Chinese (Chan and Poon,
tures tend to place greater value on interpersonal 1999). Results are also available in Ameridian
harmony), respeto (Latino cultures tend to give groups (Aruaco, Pame, and Maya Indians; Ardila
greater respect to individuals in position of author- and Moreno, 2001; Ostrosky-Solis et al., 2004a,
ity), personalismo (Latinos adhere to the value 2004b).
6. New test interpretation strategiesin addi- Adamopoulos, J. (1988). Interpersonal behavior: Cross-cultural and his-
tion to/instead of, the traditional psychometric torical perspectives. In: Bond, M. H. (ed.), The Cross-Cultural
Challenge to Social Psychology, Sage, Newbury Park, CA, pp. 196
strategies, could be developed. For example, 207.
Ardila (A cross-linguistic naming test, unpub- Adamopoulos, J. (1991). The emergence of interpersonal behavior: Di-
lished data) suggested to use the word frequency achronic and cross-cultural processes in the evolution of intimacy.
In: Ting-Toomey, S., and Korzenny, F. (eds.), International and In-
as a scoring criterion in naming tests. tercultural Communication Annual: Vol. 15. Cross-Cultural Inter-
7. Some tests are strongly culture-dependent (e.g., personal Communication, Sage, Newbury Park, CA, pp. 155170.
interpreting a proverb), whereas other tests may be Adamopoulos, J. (2002). The perception of interpersonal behaviors
across cultures. In: Lonner, W. J., Dinnel, D. L., Hayes, S. A., and
more cross-cultural (e.g., telling animal names). Sattler, D. N. (eds.), Online Readings in Psychology and Culture
Hence, some tests can be regarded as more uni- (Unit 15, Chapter 2), (http://www.wwu.edu/culture), Center
versal (e.g., mimicking everyday life movements for Cross-Cultural Research, Western Washington University,
Bellingham, Washington.
in testing for apraxia) whereas others can be use- Anastasi, A. (1988). Psychological Testing, Macmillan, New York.
ful just in some specific cultural contexts (e.g., Ardila, A. (1995). Directions of research in cross-cultural neuropsy-
spelling words aloudthat is frequent in lan- chology. J. Clin. Exp. Neuropsychol. 17: 143150.
Ardila, A. (2003). Culture in our brains: Cross-cultural differences in
guages with irregular writing systems, but unusual the brain-behavior relationships. In: Toomela, A. (ed.), Cultural
in other languages). This simply means the test Guidance in the Development of the Human Mind, Ablex,
should be selected considering the specific cul- Westport, CT, pp. 6386.
Ardila, A. (in press). The impact of culture on neuropsychological test
tural idiosyncrasies. performance. In: Uzzell, B., Ponton, M., and Ardila, A. (eds.),
International Handbook of Cross-Cultural Neuropsychology,
There is not a perfect solution, but all the points Erlbaum, Mahwah, NJ.
mentioned above may contribute to reach a better and Ardila, A., and Moreno, S. (2001). Neuropsychological evaluation in
Aruaco Indians: An exploratory study. J. Int. Neuropsychol. Soc.
fairer cognitive testing in different cultural contexts. 7: 510515.
Ardila, A., Ostrosky-Solis, F., Rosselli, M., and Gomez, C. (2000). Age
related cognitive decline during normal aging: The complex effect
of education. Arch. Clin. Neuropsychol. 15: 495514.
CONCLUSION Ardila, A., and Rosselli, M. (1994). Development of language, memory
and visuospatial abilities in 5 -to 12-year-old children using a
Cognitive testing represents a social situation that neuropsychological battery. Dev. Neuropsychol. 10: 97120.
Ardila, A., Rosselli, M., and Rosas, P. (1989). Neuropsychological
as any social situation, it is one governed by implicit assessment in illiterates: Visuospatial and memory abilities. Brain
cultural rules. The relationship between the examiner and Cogn. 11: 147166.
examinee, the type of environment, the style of commu- Axelrod, B., Tomer, R., Fisher, T., and Aharon-Peretz, J. (2001).
Preliminary analysis of the Hebrew Verbal Fluency measures.
nication that is maintained, and the activities carried out Appl. Neuropsychol. 8: 248251.
in a standard testing situation are embedded in a cul- Berry, J. W. (1979). Culture and cognition style. In: Mrsella, A., Tharp,
tural context. Psychometric tests have been developed in R. G., and Ciborowski, T. J. (eds.), Perspectives in Cross-Cultural
Psychology, Academic Press, New York, pp. 117135.
some specific cultures, and are strongly biased by the Berry, J. W., Poortinga, Y., Segall, M. H., and Dasen, P. (1992). Cross-
modal culture values observed in these cultures. Nonethe- Cultural Psychology, Cambridge University Press, Cambridge, UK.
less, these specific values and behavioral styles do not Binet, A. (1905). Analysis of C. E. Spearman, the proof and measurement
of association between two things and general intelligence objec-
necessarily represent universal values and behaviors. Un- tively determined and measured. LAnne Psychol. 11: 623624.
derstanding these cultural assumptions, and developing Binet, A. (1908). Le developpement de lintelligence chez les enfants.
assessment procedures correctly tailored to other cul- LAnne Psychol 14: 194.
Chan, A. S., and Poon, M. W. (1999). Performance of 7- to 95-year-old
tures, represents a major endeavor for twenty-first century individuals in a Chinese version of the category fluency test. J. Int.
neuropsychology. Neuropsychol. Soc. 5: 525533.
Condon, J. C., and Yousef, F. (1981). An Introduction to Intercultural
Understanding, Bobbs-Merrill, New York.
ACKNOWLEDGMENTS Cornelious, S. W., and Caspi, A. (1987). Everyday problem solving in
adulthood and old age. Psychol. Aging 2: 144153.
da Silva, C. G., Petersson, K. M., Faisca, L., Ingvar, M., and Reis, A.
My sincere gratitude goes to Dr Kevin Keating and (2004). The effects of literacy and education on the quantitative
the anonymous reviewers of this paper for their most valu- and qualitative aspects of semantic verbal fluency. J. Clin. Exp.
able suggestions. Neuropsychol. 26: 266277.
Dingfelder, S. F. (2005). Closing the gap for Latino patients. Monitor
Psychol. 36: 5861
Ferraro, F. R. (ed.) (2002). Minority and Cross-Cultural Aspects of
REFERENCES Neuropsychological Assessment, Swets and Zeitlinger, Lisse.
Fletcher-Janzen, E., Strickland, T. L., and Reynolds, C. R. (eds.) (2000).
Adam, B. (1990). Time in Social Theory, Temple University Press, Handbook of Cross-Cultural Neuropsychology. New York: Kluwer
Philadelphia. Academic/Plenum Publishers.
194 Ardila
Geertz, C. (2000). Interpretation of Cultures, Basic Books, New York. Linton, R. (1936). The Study of Man, Appleton-Century-Crofts,
Geisinger, K. (1992). Psychological Testing of Hispanics, American New York.
Psychological Association, Washington, DC. Luria, A. R. (1931). Psychological expedition to central Asia. Science
Goody, J. (1991). Time: Social organization. In: Sills, D. L. (ed.), 74: 383384.
International Encyclopedia of Social Sciences, Vol. 16, Macmillan, Luria, A. R. (1933). The second psychological expedition to central
New York, pp. 3042. Asia. Science 78: 191192.
Goldblum, M. C., and Matute, E. (1986). Are illiterate people deep Luria, A. R. (1966). Higher Cortical Functions in Man, Basic Books
dyslexics? J. Neurolinguist. 2: 103114. New York.
Greenfield, P. M. (1997). You cant take it with you: Why ability Luria, A. R. (1973). The Working Brain, Penguin Books, London.
assessments dont cross cultures. Am. Psychol. 52: 11151124. Luria, A. R. (1976). Cognitive Development, Harvard University Press,
Greif, A. (1994). Cultural beliefs and the organization of society: A Cambridge, MA.
historical and theoretical reflection on collectivist and individualist MacDonald, K. (1998). Evolution, culture and the five-factor model. J.
societies. J. Pol. Econom. 102: 912950. Cross Cultural Psychol. 29: 119149.
Harris, M. (1983). Culture, People, Nature: An Introduction to General Manly, J. J., Miller, S. W., Heaton, R. K., Byrd, D., Reilly, J., Velasquez,
Anthropology, 3rd edn., Harper and Row, New York. R., and Saccuzzo, D. (1998). The effect of African-American ac-
Heath, S. B. (1983). Ways With Words: Language, Life, and Work culturation on neuropsychological test performance in normal and
in Communities and Classrooms, Cambridge University Press, HIV-positive individuals. J. Int. Neuropsychol. Soc. 4: 291302.
Cambridge, UK. Manly, J. J., Jacobs, D. M., Sano, M., Bell, K., Merchant, C. A., Small,
Heath, S. B. (1989). The learner as cultural member. In: Rice, M. L., S. A., and Stern, Y. (1999). Effect of literacy on neuropsychological
and Schiefelbusch, R. L. (eds.), The Teachability of Language, test performance in non-demented, education- matched elders. J.
Paul H. Brookes, Baltimore, MD, pp. 333350. Int. Neuropsychol. Soc. 5: 191202.
Heath, S. B. (1997). Culture: Contested realm in research on children Matthews, C. G. (1992). Truth in labeling: Are we really an international
and youth. Appl. Dev. Sci. 1: 113123. society? J. Clin. Exp. Neuropsychol. 14: 418426.
Hofstede, G. (1997). Cultures and Organizations: Software of the Mind, Matute, E., Leal, F., Zarabozo, D., Robles, A., and Cedillo, C. (2000).
McGraw Hill, New York. Does literacy have an effect on stick construction tasks? J. Int.
Hughes, D. O., and Trautmann, T. R. (eds.) (1995). Time. Histories and Neuropsychol. Soc. 6: 668672.
Ethnologies. University of Michigan Press, Ann Arbor, MI. Miller, J., and Rotatori, A. F. (1990). Bias and the assessment of
Irvine, S. H., and Berry, J. W. (1988). Human Abilities in Cultural culturally different students. In: Rotatori, A. F., and Fox, R.
Context, Cambridge University Press, Cambridge, UK. A. (eds.), Comprehensive Assessment in Special Education:
Johnson-Powell, G., Yamamoto, J., Wyatt, G. E., and Arroyo, W. (1997). Approaches, Procedures and Concerns, Charles C. Thomas,
Transcultural Child Development. Psychological Assessment and Springfield, IL, pp. 492514.
Treatment, Wiley, New York. Mishra, S. P. (1980). The influence of examiners ethnic attributes on
Klenberg, L., Korkman, M., and Lahti-Nuuttila, P. (2001). Differential intelligence test scores. Psychol. School 17: 117122.
development of attention and executive functions in 3- to Mitrushina, M. N., Boone, K. B., and DElia, L. F. (1999). Handbook
12-year-old Finnish children. Dev. Neuropsychol. 20: 407428. of Normative Data for Neuropsychological Assessment, Oxford
Kluckhohn, F. R., and Strodbeck, F. L. (1961). Variations in Value University Press, New York.
Orientations, Row Petersen, San Diego. Munn, N. D. (1992). The cultural anthropology of time: A critical essay.
Kosmidis, M. H., Vlahou, C. H., Panagiotaki, P., and Kiosseoglou, G. Ann. Rev. Anthropol. 21: 931923.
(2004). The verbal fluency task in the Greek population: Normative Nell, V. (2000). Cross-Cultural Neuropsychological Assessment:
data, and clustering and switching strategies. J. Int. Neuropsychol. Theory and Practice, Erlbaum, Mahwah, NJ.
Soc. 10: 164172. Ostrosky, F., Ardila, A., Rosselli, M., Lopez-Arango, G., and
Lebedko, M. (2001). Time perception across Russian and American Uriel-Mendoza, V. (1998). Neuropsychological test performance
cultures. PAC3 at JALT2001 conference, Kitakyushu, Japan, in illiterates. Arch. Clin. Neuropsychol. 13: 645660.
November. Ostrosky-Solis, F., Ramirez, M., Lozano, A., Picasso, H., and Velez,
Lecours, R. L., Mehler, J., Parente, M. A., Caldeira, A., Cary, L., Castro, A. (2004a). Culture or education: A study with indigenous Maya
M. J., Dehaout, F. Delgado, R., Gurd, J., Karmann, D., Jakubovitz, population. Int. J. Psychol. 39: 3646.
R., Osorio, Z., Cabral, L. S., and Junqueira, M. (1987a). Illiteracy Ostrosky-Solis, F., Ramirez, M., and Ardila, A. (2004b). Effects
and brain damage I: Aphasia testing in culturally contrasted of culture and education on neuropsychological testing: A
populations (control subjects). Neuropsychologia 25: 231 preliminary study with indigenous and nonindigenous population.
245. Appl. Neuropsychol. 11: 186193.
Lecours, R. L., Mehler, J., Parente, M. A., Caldeira, A., Cary, L., Castro, Peoples, J., and Bailey, G. (2003). Humanity: An Introduction to
M. J., Dehaout, F., Delgado, R., Gurd, J., Karmann, D., Jakubovitz, Cultural Anthropology, 6th edn., Wadsworth, New York.
R., Osorio, Z., Cabral, L. S., and Junqueira, M. (1987b). Illiteracy Perez-Arce, P., and Puente, A. E. (1996). Neuropsychological
and brain damage 2: Manifestations of unilateral neglect in testing assessment of ethnic-minorities: The case of assessing Hispanics
auditory comprehension with iconographic material. Brain living in North America. In: Sbordone, R. J., and Long, C. J. (eds.),
Cogn. 6: 243265. Ecological Validity of Neuropsychological Testing, Gr Press/St
Lecours, A. R., Mehler, J., Parente, M. A., Beltrami, M. C., Canossa de Lucie, Delray Beach, FL, pp. 283300.
Tolipan, L., Castro, M. J., Carrono, V., Chagastelles, L., Dehaut, Ponton, M. O., and Leon-Carrion, J. (eds.) (2001). Neuropsychology
F., Delgado, R., Evangelista, A., Fajgenbaum, S., Fontoura, C., and the Hispanic Patient. Erlbaum, Mahwah, NJ.
de Fraga Karmann, D., Gurd, J., Hierro Torne, C., Jakubovicz, R., Puente, A. E., and Agranovich, A. V. (2003). The cultural in cross-
Kac, R., Lefevre, B., Lima, C., Maciel, J., Mansur, L., Martinez, cultural neuropsychology. In: Hersen, M., Goldstein, G., and
R., Nobrega, M. C., Osorio, Z., Paciornik, J., Papaterra, F., Beers, S. R. (eds.), The Handbook of Psychological Assessment,
Jourdan Penedo, M. A., Saboya, B., Scheuer, C., Batista da Silva, Vol. 1, Wiley, New York, pp. 321332.
A., Spinardi, M., and Texeira, M. (1988). Illiteracy and brain Puente, A. E., and McCaffrey, R. J. (eds.) (1992). Handbook of
damage 3: A contribution to the study of speech and language Neuropsychological Assessment: A Biopsychosocial Perspective,
disorders in illiterates with unilateral brain damage (initial testing). Plenum, New York.
Neuropsychologia 26: 575589. Puente, A. E., and Perez-Garcia, M. (2000). Neuropsychological
Lezak, M. D. (2004). Neuropsychological Assessment, 4th edn., Oxford assessment of ethnic minorities: Clinical issues. In: Cuellar, I.,
University Press, New York. Paniagua, F. A., and Freddy, A. (eds.), Handbook of Multicultural
Mental Health, Academic Press, San Diego, CA, pp. 419 Terman, L. M. (1916). The Measure of Intelligence, Houghton-Mifflin,
435. Boston.
Puente, A. E., and Salazar, G. D. (1998). Assessment of minority and cul- Toomela (ed.) (2003). Cultural Guidance in the Development of the
turally diverse children. In: Prifitera, A., and Saklofske, D. H. (eds.), Human Mind, Ablex Publishing, Westport, CT.
WISC-III Clinical Use and Interpretation: Scientist-Practitioner Torrell, F., and Torrell, S. (1983). The relationship between race of
Perspectives, Academic Press, San Diego, CA, pp. 227 examiner, cultural mistrust, and the intelligence test performance
248. of Black children. Psychol. Schools 20: 367369.
Reis, A., and Castro-Caldas, A. (1997). Illiteracy: A cause for biased Triandis, H. C. (1978). Some universals of social behavior. Pers. Soc.
cognitive development. J. Int. Neuropsychol. Soc. 5: 444 Psychol. Bull. 4: 16.
450. Triandis, H. C. (1994). Culture and Social Behavior, McGraw-Hill,
Reynolds, C. R. (2000). Methods for detecting and evaluating cultural New York.
bias in neuropsychological tests. In: Fletcher-Janzen, E., Strickland, Tylor, E. B. (1871). Primitive Culture: Researches Into the Development
T. L., and Reynolds, C. R. (eds.), The Handbook of Cross-Cultural of Mythology, Philosophy, Religion, Art, and Custom, John Murray,
Neuropsychology, Plenum, New York, pp. 249285. London.
Rosselli, M., and Ardila, A. (2003). The impact of culture and education Uzzell, B., Ponton, M., and Ardila A. (eds.) (in press). International
on nonverbal neuropsychological measurements: A critical review. Handbook of Cross-Cultural Neuropsychology. Lawrence Erlbaum
Brain Cogn. 52: 326333. Associates, Mahwah, NJ.
Rosselli, M., Ardila, A., and Rosas, P. (1990). Neuropsychological Vygotsky, L. S. (1934/1978). Mind in Society, Harvard University
assessment in illiterates II: Language and praxic abilities. Brain Press, Cambridge, MA.
Cogn. 12: 281296. Wechsler, D. (1997). WAIS-III: Administration and Scoring Manual,
Sahlins, M., and Service, E. R. (1960). Evolution and culture. University The Psychological Corporation, San Antonio, TX.
of Michigan Press. Wong, T. M., Strickland, T. L., Fletcher-Janzen, E., Ardila, A., and
Samuda, R. J., Feuerstein, R., Kaufman, A. S., Lewis, J. E., and Reynolds, C. R. (2002). Theoretical and practical issues in the
Stenberg, R. J. (1998). Advances in Cross-Cultural Assessment, neuropsychological assessment of culturally dissimilar patients. In:
Sage, Thousand Oaks, CA. Fletcher-Janzen, E., Strickland, T. L., and Reynolds, C. R. (eds.),
Spreen, O., and Strauss, E. (1998). A Compendium of Neuropsycholog- The Handbook of Cross-Cultural Neuropsychology, Plenum, New
ical Tests, 2nd edn., Oxford University Press, New York, York, pp. 318.
Neuropsychologic symptoms in the migraine syndrome
Alfredo Ardila, Erick Sanchez
CEPHALALGIA Ardila A, Sanchez E. Neuropsychologic symptoms in the migraine syndrome.

Cephalalgia 1988;8:67-70. Oslo. ISSN 0333-1024
Twenty patients with complex neuropsychologic symptoms associated with classic migraine
were selected from a group of 200 patients with vascular headache. Twenty types of
symptoms were found and grouped into six categories (language, visual symptoms,
cognitive-dysmnesic symptoms, olfactory-gustatory hallucinations, automatisms, and
somatosensory symptoms). Some of the symptoms found have apparently not been reported
previously. The importance of the analysis of this complex neuropsychologic set of symptoms
and of the possible consequences of complicated, frequent vascular headache is
emphasized. Classic migraine, neuropsychologic symptoms
Alfredo Ardila, Erick Sanchez, Instituto Neurologico de Colombia, Bogot, Colombia;

Correspondence to Alfredo Ardila, Ph.D., Apartado Aereo 17021, Bogot Colombia; Accepted
4 January 1988
In addition to the most frequently observed symptoms of classic migraine, such as sparkling scotomas,
paresthesia, motor deficits, hemianopsia, and so forth, other complex disorders can appear, such as aphasia,
alexia, visual hallucinations, micropsia, macropsia, achromatopsias, movement illusions, amnesia, macro- or
micro-somatognosia, confusional states, and so forth (1-7). In general, sensory, motor, or cognitive
dysfunction is transitory and usually ends after some 10 or 15 min. However, the analysis of such a complex
neuropsychologic set of symptoms can be important not only for understanding the possible mechanism
involved in migraine but also for the differential diagnosis of migraine and other nosologic entities.
A further analysis of the neuropsychologic symptoms associated with classic migraine was therefore
considered relevant.
Methods and results
Over the past 3 years 200 patients diagnosed as having vascular headache were studied in the Neurologic
Institute of Colombia. Of these, 20 patients were found (14 women, 6 men, average age, 35 years; age
range, 20-69 years) who showed some neuropsychologic symptoms as a prodrome (from about 5 min to
several hours) of classic migraine. Since some of the patients had more than one phenomenon, the total
number of symptoms recorded was 40. Table 1 shows the phenomena found and their frequency in the
sample analyzed.
Table 1. Complex neuropsychologic phenomena

associated with migraine and the number of subjects who
experienced them
Anomia (forgetting the name of things) 7
Difficulty in speaking 5
Depersonalization (feeling as if another person) 3
Seeing the world as strange 3
Macropsia (apparent increased object size 2
Micropsia (apparent decrease object size) 2
Simultaneous agnosia (only the object or part of
the object being looked at is recognized) 2
Automatic behavior 2
Inability to understand language 2
Olfactory hallucinations 2
Achromatopsia (disappearance of colors) 1
Chromatopsia (modification of object colors) 1
Palinopsia (visual perseveration) 1
Pelopsia (objects seem to approach the subject
and become larger) 1
Gustatory hallucinations 1
Alexia (inability to read) 1
Acalculia (calculation disturbance) 1
Teleopsia (objects seem small and far away) 1
Transient global amnesia 1
Hemiasomatognosia (unilateral misperception
of one's own body) 1
Table 2. Types of neuropsychologic phenomena associated with
migraine; frequency and percentage in the sample analyzed
Type Frequency Percentage
Language 16 40.0
Visual disturbances 11 27.5
Cognitive-dysmnesic symptoms 7 17.5
Olfactory-gustatory hallucinations 3 7.5
Automatism 2 5.0
Somatosensory symptoms 1 2.5
The 20 types of complex neuropsychologic phenomena found in the sample were grouped into 6
categories. Language and visual phenomena were found to represent approximately two-thirds (Table 2).
Complex somatosensory phenomena appeared in only one case.
The most frequently found phenomena were anomia ( forgetting the name of things, sometimes with the
presence of semantic paraphasia) and difficulties in speaking (which were usually associated with right-sided
paresthesia). In the first case, the territory involved corresponded theoretically to the temporo-occipital region
of the left hemisphere, and, in the second, to the anterior parietal region. In the case of anomia, the deficit
was associated with defects in the right visual field and with scotomas in this field. The difficulties in
understanding language were associated with anomia and also with defects in the right visual field.
Patient E.T., a 27-year-old woman, had a 14-year history of vascular headache. Initially, she presented
with right visual field hemianopia and difficulties in remembering the names of objects. This would last for
about 15-20 min; during this period she never tried to read. Afterwards, vascular headache appeared. She
had to lie down in bed. Headache lasted several hours. Findings on neurologic examination, computed
tomography (CT) scan, and EEG were normal.
Alexia (inability to read) appeared in only one patient as the inability to join letters to form words and was
associated with defects in the right visual field and anomia; it was therefore a verbal alexia, and, as a result,
we would expect it to be an alexia without agraphia (inability to write). Acalculia (calculation disturbance) was
evident in a female patient who performed simple, routine calculations in her work; "I didn't know what I had
to do to add up; it was as if I had never learned to add up". This was associated with anomia.
The visual phenomena took the form of macropsia (apparent increased object size), micropsia
(decreased size), achromatopsia (disappearance of colors), chromatopsia (modification of object colors; for
example, "everything looked yellow"), pelopsia (objects seem to approach the subject and become larger:
"things came to me"), teleopsia (objects seem small and far away), or palinopsia (visual perseveration; that
is, one patient reported that he kept seeing repeated figures or images). This was associated with left
homonymous hemianopsia and agnosia (on looking at an object, the edge of it was lost; that is, only the
object or part of the object being looked at is recognized); the latter two patients also had defects in their right
visual field.
Patient E.S., a 60-year-old woman, has since childhood presented with vascular headache. She reported
sparkling scotomas, paresthesia, and left hemianopia. On occasion, she reported macropsia, micropsia, and
repeated perception of figures (palinopsia). Neurologic examination and CT scan were normal.
The cognitive-dysmnesic type of phenomena (experimental and mnesic changes) took the form of
strangeness (feeling different, as if another person; that is, depersonalization); perceiving the world as if it
were different (two of the patients said they saw things "as if on the television" and the third mentioned that
everything seemed unknown-jamais-vu). The patient with transient global amnesia (with a history of migraine
of 49 years' duration) on one occasion while shaving himself observed: "I could only see one half on my face,
and I felt as if I were far away". While she talked to him, his wife realized that he could not understand what
was happening to him nor remember what she said to him. This lasted for 2h.
Patient R.M., a 26-year-old woman, has for about 8 years had episodes of paresthesia in the right arm
and face lasting about 30 min, whereafter she had a severe headache for about 2 h. Twice she had
presented with the "feeling of being another person" and with anguish, followed by right-sided paresthesia
and vascular headache. Neurologic examination and EEG were normal.
One patient reported "a kind of sweet taste in my mouth", simultaneously with paresthesia in his lips and
tongue. Another patient who experienced phenomena of strangeness referred to "a smell like acid", and still
another patient mentioned "a smell like sewage".
One patient (with an almost 30-year history of vascular headache) has over the past months experienced
episodes of cephalea associated with automatic behavior and subsequent amnesia. Similarly, the patient with
transient global amnesia experienced episodes of automatisms with sucking, which makes it difficult to decide
whether it is a question of two different phenomena (transient global amnesia and automatisms) or, on the
contrary, of just one phenomenon. It nevertheless points to an involvement of the mesial temporal territory.
The only case of complex somatosensory phenomena (hemiasomatognosia: unilateral misperception of

own's own body) is that of a 60-year-old woman with a history of vascular headache since childhood. On one
occasion, she experienced a feeling of strangeness in her left hand ("as if it weren't mine; it didn't belong to
me; it moved by itself without my controlling it") followed by left hemi-body dysesthesia.
Discussion
Above all, it is noteworthy that neuro-psychologic phenomena associated with migraine tend to appear in
patients with a long history of vascular headache. Even though some of our patients had a 50-year history of
headache, the appearance of complex symptoms went back only a few years at most and in most cases had
started appearing shortly before examination.
Although most of the symptoms found in our sample have already been described in the literature (8), it is
worth emphasizing some points.
Language disorders of various types occur most frequently in our material. However, according to
Frederiks (8), they are not frequent in migraine, and when they appear, they are usually of the expressive
kind. These disorders occurred frequently in our material owing to the high frequency of anomia. We suppose
that anomia is a symptom that runs the risk of being ignored if the patient is not asked directly about it. In
routine neurologic examinations only errors of the motor type tend to be included as language defects. Given
the high incidence of visual symptoms in migraine, it seems logical to assume that anomia is a frequent
symptom, considering that, topographically, anomia corresponds to the left temporo-occipital region (9),
which is next to the region that would produce visual disturbances. Although alexia, even without agraphia,
has been described in cases of migraine (1, 5), to our knowledge acalculia has not been reported as a
phenomenon connected with migraine.
Transient global amnesia has been reported associated with migraine (2, 3), but, in general,
cognitive-dysmnesic phenomena do not seem to be frequent in vascular headaches. Three of our patients
experienced depersonalization phenomena; two said they felt the world to be unreal ("as if I were watching
television"), although it is difficult to be sure that it was a question of purely cognitive-dysmnesic phenomenon
and not simply a perceptual one; and one experienced a jamais-vu phenomenon.
Although olfactory phenomena have been shown to occur in migraine (10), to our
knowledge no gustatory phenomena have been previously reported. In our patients it was found together with
paresthesia of the lips and tongue.
Phenomena related to changes in the body scheme include macrosomatognosia, micro-somatognosia,

and autoscopia (8). However, one of our patients had a clear hemiasomatognosia with a "feeling of a strange
hand".
The above-mentioned observations emphasize the diversity and richness of the neuropsychologic
symptoms that can be found in the case of migraine.
Finally, and bearing in mind the similarity between the mechanisms underlying the migraine type of
headache and transient cerebral ischemia, it is important to consider the possibility of permanent
neuro-psychologic sequelae in cases of migraine. This possibility has been analyzed by different authors
(11-13), who found that, in some cases at least, it is possible to find permanent cognitive deficits in frequent,
complicated vascular headache.
References
1. Bigley GK, Sharp FR. Reversible alexia without agraphia due to migraine. Arch Neurol 1983;40:114-5
2. Caplan L, Chedru F, Lhermitte F, Mayman C. Transient global amnesia and migraine. Neurology
1981;31: 1167-70
3. Crowell F, Stump DA, Biller J, McHenry C, Toole JF. The transient global amnesia-migraine
connection. Neurology 1984;41:75-9
4. Delaney RC, Wallace JD, Egelko S. Transient cerebral ischemic attacks and neuropsychological
deficits. J Clin Neuropsych 1980;2:107-14
5. Fleishman JA, Segall JD, Judge FP. Isolated transient alexia. Arch Neurol 1983;40:11-6
6. Hachinski VC, Porchawka J, Stecle JC. Visual symptoms in the migraine syndrome. Neurology 1973
;23:570-9
7. Klee A, Willanger R. Disturbances of visual perception in migraine. Acta Neurol Scand 1966;
42:400-14
8. Frederiks JAM. Paroxysmal neuropsychological disorders. In: Frederiks JAM ed Handbook of clinical
neurology. Vol 45, Clinical neuropsychology. Amsterdam: Elsevier 1985;507-14
9. Benson DF. Aphasia, alexia and agraphia. New York: Churchill Livingstone Inc 1979
10. Wakefield G. Clinical features of migraine, with special reference to the migraine aura. In: Pearce J ed
Modern topics in migraine. London: W. Heinemann Medical Books 1975;22-9
11. Bartleson JD. Transient and persistent neurological manifestations of migraine. Stroke 1984;15:383-6
12. Hooker W. Neuropsychological impairments in classic and common migraine. Paper presented at the
XII International Neuropsychological Society meeting. Houston, 1984
13. Zeitlin C, Oddy M. Cognitive impairment in patient with severe migraine. Br Med J Clin Psychol
1984;23:27-35
CORRESPONDENCIA
conocida o silente en el HI, control ipsilateral 7. Semenza C, Bertella L, Grana A, Mori I, cientemente, los extractos de Datura o la es-
de la mano dominante, representacin hemis- Conti F. Numbers and calculation in crossed copolamina pura empezaron a mezclarse con
frica bilateral del lenguaje o alteraciones du- aphasia. Brain Lang 2003; 87: 131-2. tranquilizantes (benzodiacepinas y fenotiaci-
rante el desarrollo de la lateralizacin de las 8. Bakar M, Kirshner HS, Wertz RT. Crossed nas), mezcla que usualmente se conoce como
funciones lingsticas [4]. aphasia. Functional brain imaging with PET nueva burundanga. Los tranquilizantes se
Bakar et al [8] realizaron estudios con to- or SPECT. Arch Neurol 1996; 53: 1026-32. aaden para evitar la agitacin psicomotora
mografa por emisin de positrones (PET) o 9. Marien P, Paquier P, Engelborghs S, De Deyn de la vctima. Sin embargo, la intoxicacin
tomografa computarizada por emisin de fo- PP. Acquired crossed aphasia in dextral chil- con Datura con fines rituales se ha comunica-
tn nico (SPECT) en pacientes con AC, y en- dren revisited. Brain Lang 2001; 79: 426-43. do durante siglos y se encuentra en diferentes
contraron reas de hipometabolismo ms ex- 10. Alexander MP, Annett M. Crossed aphasia pases latinoamericanos [7].
tensas que las lesiones encontradas en tomo- and related anomalies of cerebral organiza- El gnero Datura pertenece al orden Sola-
grafa computarizada (TC) o RM, as como tion: case reports and a genetic hypothesis. naceae, junto con la mandrgora (Mandrago-
depresin funcional de las reas del lenguaje Brain Lang 1996; 55: 213-39. ra), la henbana (Hyoscyamus) y la belladona
del HI. 11. Paghera B, Marien P, Vignolo LA. Crossed (Atropa belladonna). Cada uno de estos agen-
Alexander y Annett [10] plantean un origen aphasia with left spatial neglect and visual tes botnicos contiene cantidades variables de
gentico para la AC; por ello, la ausencia de imperception: a case report. Neurol Sci 2003; atropina y otros alcaloides tropanos relaciona-
un gen hara que las funciones del lenguaje se 23: 317-22. dos (hiosciamina, norhiosciamina y escopola-
distribuyeran aleatoriamente a los hemisferios 12. Marien P, Engelborghs S, Vignolo LA, De mina); todos ellos pueden tener efectos aluci-
cerebrales, y ocasionaran una organizacin Deyn PP. The many faces of crossed aphasia ngenos [7]. Por esta razn, se han utilizado
anmala de las funciones cognitivas en peque- in dextrals: report of nine cases and review of con fines rituales. La escopolamina es una
os segmentos de la poblacin. the literature. Eur J Neurol 2001; 8: 643-58. droga anticolinrgica; bloquea los receptores
Paparounas et al [3] consideran que la ma- 13. Carr MS, Jacobson T, Boller F. Crossed apha- colinrgicos muscarnicos, lo cual puede ex-
yora de casos seran compatibles con la hip- sia: analysis of four cases. Brain Lang 1981; plicar sus efectos sobre la memoria; la grave-
tesis de la imagen en espejo, por la cual este 14: 190-202. dad de su efecto depende de la dosis utilizada
sndrome resultara de una lesin hemisfrica 14. Castro-Caldas A, Confrara A. Age and type [8]. La escopolamina cruza fcilmente la ba-
derecha en espejo de lo que habra pasado si of crossed aphasia in dextrals due to stroke. rrera hematoenceflica, y por esto afecta direc-
la lesin hubiera estado en el HI, mientras que Brain Lang 1984; 23: 126-33. tamente el sistema nervioso central (SNC).
la teora de Alexander y Annett explicara los 15. Joanette Y, Puel M, Nespoulous JL, Rascol A, Una vez ingerida oralmente, se absorbe rpi-
casos de AC anmala. Lecours AR. Crossed aphasia in right-handed damente en el tracto digestivo y se distribuye
Si bien no existe un acuerdo en los meca- patients. I. Review of the literature. Rev Neu- ampliamente en el cuerpo. La mayora de la
nismos involucrados en la AC, nuestro caso rol (Paris) 1982; 138: 575-86. droga se excreta a travs de la orina durante las
provee evidencia para enriquecer la discusin 16. Urbain E, Seron X, Remits A, Cobben A, Van primeras 12 horas. Esta excrecin rpida ex-
en la lateralizacin y organizacin intrahemis- der Linden M, Mouchette R. Crossed aphasia plica la dificultad para obtener anlisis toxico-
frica de las funciones del lenguaje. in a right-handed patient: a report on one case. lgicos positivos [9,10], lo cual representa una
Rev Neurol (Paris) 1978; 134: 751-9. limitacin para demostrar que alguien ha reci-
L. Torres a, N. Mori-Quispe b, M. Vlez a, 17. Castro-Caldas A, Confrara A, Poppe P. Non- bido escopolamina.
A. Delgado-Salinas a, A. Anicama-Hernndez c, verbal disturbances in crossed aphasia. Apha- La intoxicacin por escopolamina se asocia
C. Cosentino-Esquerre a siology 1987; 1: 403-13. con un cuadro clnico de midriasis, taquicar-
dia, sequedad bucal, visin borrosa, retencin
Aceptado tras revisin externa: 17.11.05. urinaria, desorientacin, incoordinacin en los
a
Instituto Especializado en Ciencias Neurolgicas. Intoxicacin por escopolamina movimientos y amnesia grave [8,9]. Estos son
b
Universidad Nacional Mayor de San Marcos. Li- los efectos generales observados en casos de
ma. c Universidad Nacional San Luis Gonzaga. Ica, (burundanga): prdida de la administracin de burundanga. La intoxica-
Per.
capacidad de tomar decisiones cin con otros agentes anticolinrgicos produ-
Correspondencia: Dr. Luis Torres. Instituto Especia- ce un cuadro clnico similar [11,12]. Existe en-
lizado en Ciencias Neurolgicas. Departamento de
Enfermedades Neurodegenerativas. Jr. Ancash, 1271.
Desde hace muchos aos se conocen los efec- tonces un sndrome anticolinrgico central
Lima 01, Per. E-mail: torresl@terra.com.pe tos de la intoxicacin con extractos derivados (psicosis atropnica o delirio txico). El deli-
de plantas pertenecientes al gnero Datura o rio por escopolamina y el delirio causado por
Brugmansia [1-4]. Todas las especies son na- otros agentes anticolinrgicos, sin embargo,
BIBLIOGRAFA tivas de Suramrica, pero algunas de ellas se podran ser diferentes.
cultivan extensamente alrededor del mundo. Algunos informes mencionan la psicosis
1. Damasio A. Aphasia. N Engl J Med 1992; 326: Actualmente se encuentran en diferentes pa- potencialmente asociada con escopolamina
531-40. ses. Los escritos de la poca colonial indican [13-19]. Se menciona intranquilidad, excita-
2. Zarranz JJ. Neurologa. 1 ed. Madrid: Har- que los paseos por las calles plantadas con cin, alucinaciones, euforia, desorientacin, y
court-Brace; 1998. p. 210-5. cacao sabanero (Brugmansia candida o Da- tambin estupor, coma, depresin respiratoria
3. Paparounas K, Eftaxias D, Akritidis N. Dis- tura candida) representan un excelente trata- [14,15] y tambin muerte [20,21]. Algunos
sociated crossed aphasia: a challenging lan- miento para el insomnio [5]. informes sealan tambin que la herona oca-
guage representation disorder. Neurology En Colombia, la administracin criminal sionalmente se adultera con escopolamina
2002; 59: 441-2. con extractos de Datura apareci durante los [22,23]. Se observa en estos casos una sinto-
4. Rodrguez-Campello A, Pascual-Calvet J, aos cincuenta, pero muy especialmente du- matologa anticolinrgica de herona mezclada
Munteis E, Gomis M, Serra A, Pou A. Cros- rante los ochenta. Desde entonces, estos ex- con escopolamina. En casos de aplicacin con
sed aphasia: description of a case. Neurolo- tractos se conocen popularmente como bu- burundanga, algunos pacientes pueden presen-
gia 2000; 15: 250-2. rundanga. La palabra burundanga tiene un tar, algn tiempo despus de la intoxicacin
5. Alexander MP, Fischette MR, Fischer RS. origen africano. Aparece en una cancin afro- aguda, una especie de psicosis que puede in-
Crossed aphasias can be mirror image or cubana de Celia Cruz, muy popular durante cluir alucinaciones relacionadas con el mo-
anomalous. Case reports, review and hypo- los aos 50 [6]. Esta palabra no se ha incorpo- mento de la intoxicacin. En ocasiones, la
thesis. Brain 1989; 112: 953-73. rado al espaol, ni siquiera en el rea de Cari- familia del paciente comunica cambios de per-
6. Coppens P, Hungerford S, Yamaguchi S, Ya- be. No es clara la razn por la cual se selec- sonalidad (apata, indiferencia y similares).
madori A. Crossed aphasia: an analysis of the cion esta palabra en particular para designar Sin embargo, aparentemente, poco se mencio-
symptoms, their frequency, and a comparison algunos extractos de Datura. na la conducta de sumisin, falta de crtica y
with left-hemisphere aphasia symptomato- A comienzos de los aos ochenta se empe- obediencia asociadas con la intoxicacin por
logy. Brain Lang 2002; 83: 425-63. z a utilizar la escopolamina pura. Ms re- escopolamina.
REV NEUROL 2006; 42 (2) 125

CORRESPONDENCIA
Durante los ltimos aos se ha observado cin por escopolamina. Este grupo inclua 54 muy cerca de la cafetera, donde ella viva,
cierto inters por conocer las consecuencias hombres (88,5%) y siete mujeres (11,5%) [32]. trayndole y entregndole al hombre su dine-
de la intoxicacin criminal por burundanga. Ardila et al [32] desarrollaron un cuestiona- ro y sus pertenecas, mientras l permaneca
Actualmente existen decenas de sitios en in- rio dirigido a calcular la frecuencia de intoxi- en la cafetera y simplemente la esperaba. Al
ternet que mencionan este fenmeno [24-28]. cacin criminal por burundanga y las caracte- regresar, el hombre le pregunt si tena algo
Sin embargo, este fenmeno no se conoce su- rsticas generales de las vctimas en Bogot. ms, y ella le cont todas las cosas que tena
ficientemente en neurologa y neuropsiquia- Se tomaron seis muestras en diferentes estra- en su apartamento. El hombre le pidi que re-
tra. Existen pocas publicaciones cientficas tos socioeconmicos (373 hombres, 404 mu- gresara y trajera algunas cosas ms, y ella as
especialmente dedicadas a la intoxicacin con jeres; intervalo de edad: 18-55 aos). El cues- lo hizo. El hombre simplemente tom todas
burundanga. Por ejemplo, el texto reciente de tionario inclua la pregunta de si el participan- las cosas y se fue.
toxicologa publicado por la Asociacin Co- te se haba intoxicado con burundanga, y en
lombiana de Neurologa incluye el captulo tal caso, las principales caractersticas clni- Caso 3. Estudiante de 20 aos, que entr a
Neurotoxicidad por escopolamina y la nueva cas observadas. 16 participantes (2,06% de la una cafetera cerca del edificio de su universi-
burundanga [10]. muestra) afirmaron que en algn momento de dad, en Bogot, y pidi un caf con leche y
Ardila et al [1] enfatizaron las similitudes su vida se haban intoxicado con burundanga. algunas galletas. En la cafetera haba mucha
entre la intoxicacin por escopolamina y la La frecuencia de intoxicacin por escopola- gente. Alguien se le acerc, comenz a ha-
amnesia global transitoria. En ambas se en- mina fue mayor en los estratos socioeconmi- blarle y a bromear. El estudiante afirma creer
cuentra un dficit abrupto en la memoria de cos altos, y tambin mayor en los hombres que, posteriormente, alguien lo condujo hasta
los eventos recientes (amnesia antergrada). que en las mujeres. En la mayora de los casos un camin estacionado frente a la cafetera.
Durante el episodio amnsico, sin embargo, el el robo fue el mvil principal, pero en las No existen memorias claras posteriores a este
paciente mantiene una identidad personal y mujeres, en algunas ocasiones, se mencion el evento, pero cree recordar que estuvo traba-
puede ejecutar adecuadamente sus actividades abuso sexual. La amnesia retrgrada fue mni- jando en una plantacin y haba personas
normales de la vida diaria. Ardila et al propu- ma, pero la amnesia antergrada se extendi armadas alrededor. Tambin afirma que en
sieron que la intoxicacin por burundanga tie- entre 1 y 72 horas. Los individuos con amne- cierto momento alguien le dijo: No lo necesi-
ne dos caractersticas principales: sias prolongadas tambin comunicaron cam- tamos ms. Se puede ir o lo matamos. Usted
Una amnesia antergrada grave, probable- bios de personalidad y trastornos neuropsico- decide. El estudiante comenz entonces a ca-
mente relacionada con su accin sobre el lgicos, tales como defectos de atencin y di- minar. Estaba en la selva amaznica y supone
hipocampo. ficultades en la memoria. Slo el 50% de que estuvo caminando quizs durante dos das
Una incapacidad para reaccionar a la agre- estos casos se llev a un hospital, y menos del y finalmente lleg a Puerto Inrida (Puerto
sin externa (conducta de sumisin), pro- 20% de los casos se comunicaron a la polica. Inrida es una pequea poblacin en medio de
bablemente asociada con alguna accin so- Estos resultados sugieren que probablemente la selva amaznica; la nica forma de llegar es
bre la amgdala. varios cientos de personas podran intoxicarse por avin, y a su alrededor se encuentran mu-
con burundanga cada mes en Bogot. chas plantaciones ilegales). Al llegar pregunt
La intoxicacin con escopolamina podra con- A continuacin se presentan tres casos ilus- dnde se encontraba y alguien le dijo que en
siderarse como un modelo de amnesia global trativos. Los dos primeros se seleccionaron Puerto Inrida. Fue a la iglesia y el sacerdote
transitoria, segn Ardila et al. simplemente porque representan casos tpi- le dio comida, lo alberg y le permiti llamar
Es difcil obtener una informacin precisa cos. Pueden considerarse como casos proba- a su familia a Bogot. Su familia estaba extre-
sobre la incidencia de intoxicacin con burun- bles. El ltimo caso se describe porque inclu- madamente preocupada, porque haba desapa-
danga, ya que: ye una situacin en la que la vctima permane- recido desde hacia aproximadamente dos me-
No todos los casos llegan a centros hospi- ci en un estado confusional durante unos dos ses, y le envi dinero para tomar un avin has-
talarios. meses. Representa un caso extremadamente ta Bogot. Posteriormente, su familia comuni-
Los resultados toxicolgicos positivos son inusual. Puede interpretarse como un caso de ca algunos cambios de personalidad, como
improbables, ya que la escopolamina se ex- posible intoxicacin por escopolamina. apata, fallos en la memoria, dificultades para
creta rpidamente. concentrarse, y pobre rendimiento acadmico.
Caso 1. Mujer divorciada, de 55 aos, que
Pea et al [29] comunicaron 98 casos durante asisti a una fiesta con algunos amigos. Du- El primero y el ltimo caso ilustran los cam-
el perodo 1980-1981 en el hospital Ramn rante la fiesta conoci a un amable caballero, bios tpicos de personalidad hallados algunas
Gonzlez Valencia (Bucaramanga, Colombia). con quien bail y convers animadamente. veces con la intoxicacin por burundanga. El
Durante los primeros cinco meses de 1984 se Una vez terminada la fiesta, decidi invitarlo segundo caso es particularmente importante,
recibieron 25 casos de intoxicacin por bu- a su apartamento. La mujer afirma que recuer- ya que la memoria se encuentra slo parcial-
rundanga en la Clnica San Pedro Claver de da cuando llegaron al apartamento, comenza- mente alterada, pero existe una clara conducta
Bogot [30]. El Instituto Nacional de Medici- ron a hablar, y ella prepar alguna bebida. Su de sumisin. Es decir, la conducta de sumi-
na Legal de Bogot informa de 122 exmenes siguiente memoria corresponde al da siguien- sin se observa antes de la aparicin de la
toxicolgicos positivos para escopolamina pu- te, cuando se despert y se encontr sola. To- amnesia. Existe alguna posibilidad remota de
ra o mezclada durante el perodo junio 1988- do su dinero y sus joyas, al igual que el ama- que otras causas txicas y farmacolgicas, co-
junio 1989 [31]. ble caballero, haban desaparecido. Su familia mo uso de cocana, drogas de diseo, frma-
Como se menciona con anterioridad, la es- comunica algunos cambios de conducta des- cos, etc., podran haber provocado los sndro-
copolamina se elimina rpidamente y es infre- pus de este evento. Se volvi ms pasiva y mes clnicos descritos. Consideramos la posi-
cuente obtener pruebas positivas de laborato- aislada. Previamente, era una mujer activa y en- bilidad extremadamente baja, dadas las mani-
rio [10]. En el estudio de Pea et al [29], sola- tusiasta, con una gran diversidad de intereses festaciones clnicas de los sujetos. Ninguno
mente el 50% de los casos analizados durante personales. Su entusiasmo y sus intereses per- de los tres pacientes, adems, tena anteceden-
las 10 primeras horas fueron positivos. En el sonales disminuyeron. Tras dos aos, todava tes de uso de drogas, alcohol y otros estimu-
Instituto Nacional de Medicina Legal de Bo- se observan estos cambios de personalidad. lantes del SNC, que podran haber provocado
got, solamente el 22% de los casos analiza- un cuadro clnico semejante. Dadas las cir-
dos fueron positivos [31]. Usualmente, cuan- Caso 2. Estudiante universitaria, de 25 aos, cunstancias asociadas con los tres casos ante-
do se practican los exmenes de laboratorio, que fue a tomar un refresco en una cafetera riores, fue imposible obtener un examen cog-
han transcurrido ms de 12 horas desde la ad- cercana a su universidad. Un hombre de unos nitivo, neurolgico y neuropsiquitrico con
ministracin de la burundanga. 30 aos se sent en la misma mesa e inici detenimiento de estos pacientes.
Entre enero y octubre de 2000 se recibieron alguna conversacin. La estudiante comunica El fenmeno ms intrigante que se encuen-
en el Hospital Universitario de San Ignacio tener algunas memorias fragmentadas e im- tra en el caso de intoxicacin con burundanga
(Bogota) 216 casos de intoxicacin exgena; precisas de los eventos que sucedieron a con- es la conducta sumisa y obediente que presen-
61 (28%) correspondan a probable intoxica- tinuacin. Recuerda yendo al apartamento, ta la vctima. Aparentemente, esta conducta
126 REV NEUROL 2006; 42 (2)

CORRESPONDENCIA
sumisa no se ha comunicado en los casos de denadas por amenazas potenciales, y conduc- pinas endgenas [46]. Como anteriormente se
amnesia global transitoria. El paciente intoxi- tas desencadenadas por peligros reales, espe- seal, la intoxicacin con escopolamina po-
cado con burundanga sigue cualquier orden, cialmente predadores. Los mecanismos ner- dra interpretarse como una amnesia global
no presenta ninguna resistencia e incluso ofre- viosos responsables del sistema de defensa transitoria [1]. El alcohol tambin podra con-
ce su dinero y sus pertenencias. La vctima constituiran una red jerrquica con la amg- tribuir a la amnesia. La observacin clnica su-
regresa donde est el victimario sin hacer nin- dala como punto de convergencia de los va- giere que frecuentemente pero no siempre
gn intento por escapar (caso 2). Se observa rios estmulos amenazantes. El ncleo central la burundanga se les administra a las vctimas
que la vctima frecuentemente ofrece entregar- de la amgdala proyecta a la sustancia gris potenciales mientras consumen alcohol (caso
le cosas adicionales a su victimario. El abuso periacueductal del cerebro medio, el hipotla- 1). Todava ms, la respuesta observada des-
sexual se observa muy rara vez en casos de mo y el tronco cerebral, el cual coordina los pus de la intoxicacin con burundanga tiene
intoxicacin con burundanga, probablemente diferentes sistemas de defensa, tales como la algunas similitudes con la respuesta hallada
corresponde a menos del 5% de los casos [32]. lucha defensiva, las respuestas de evitacin, en casos de amnesia por alcohol: culpa, ansie-
Una susceptibilidad aumentada y una propen- las posturas de sumisin, y la activacin aut- dad, preocupacin por lo que haya podido su-
sin a aceptar o responder a afirmaciones es- noma. Se sabe que el hipotlamo y la sustan- ceder durante el perodo amnsico, etc.
pecficas se han sealado tambin con relacin cia gris peroacueductal del cerebro medio de- Sorprendentemente, la intoxicacin con es-
a otros alucingenos del gnero Datura [33, sempean un papel fundamental en la expre- copolamina (burundanga) ha atrado una aten-
34]. Esta conducta de sumisin probablemente sin de las conductas de defensa [43]. cin apenas modesta de la neurologa y la
sea el resultado de la accin de la escopolami- Shaik et al [44] observaron que tres regiones neuropsiquiatra. Aunque el fenmeno se en-
na sobre las estructuras cerebrales que normal- en la amgdala modulan la conducta defensiva. cuentra bastante restringido a Colombia, hay
mente controlan la conducta emocional, parti- stas incluyen el ncleo medial, el complejo en la bibliografa informes de formas inusua-
cularmente la conducta defensiva. basal y el ncleo central de la amgdala. La les de amnesia que se ajustan muy bien al cua-
Adems de su accin sobre el hipocampo, estimulacin simultnea de un ncleo amigda- dro clnico hallado con la burundanga. Por
que es de naturaleza colinrgica, se podra su- lino y de sitos del hipotlamo medio o la sus- ejemplo, Brizer et al [47] informaron de dos
poner una accin selectiva sobre otras reas tancia periacueductal del cerebro medio en los casos de amnesia ocurridos en Nueva York
del sistema lmbico temporal (muy probable- cuales se desencadena conductas defensivas, con algunas caractersticas clnicas prctica-
mente la amgdala) [35]. La intoxicacin por demostraron que dos de las regiones facilitan mente idnticas a las observadas en casos de
burundanga no solamente implica amnesia, la conducta defensiva: los ncleos mediales y intoxicacin por burundanga: sequedad bucal,
sino tambin una incapacidad para responder el complejo basal, y una tercera el ncleo visin borrosa, taquicardia y amnesia anter-
a un atraco monetario o asalto sexual. Se pue- central suprime la conducta de defensa. Los grada evidente. Probablemente, muchos casos
de conjeturar un trastorno en la habilidad para mecanismos que modulan cada una estas re- de intoxicacin con burundanga han permane-
reaccionar normalmente a los estmulos ame- giones son diferentes. La amgdala medial que cido sin diagnstico. La posibilidad de into-
nazantes y aversivos. facilita la conducta de defensa incluye una va xicacin con burundanga debe considerarse
La amgdala contiene varios ncleos que (la estra terminal) que proyecta directamente siempre que existe amnesia asociada con robo
pueden dividirse en dos grupos principales: al hipotlamo medio y utiliza la sustancia P y/o abuso sexual. Los exmenes de laborato-
corticomedial y basolateral. La amgdala corti- como neurotransmisor. La facilitacin basal rio pueden contribuir al diagnstico, pero los
comedial desempea un papel central en el amigdalina utiliza una va que se dirige a la resultados negativos no descartan la posibi-
control de la agresin [36] y en la conducta sustancia gris periacueductal, en la cual algu- lidad de intoxicacin con escopolamina. Las
agonista [37]. La amgdala basolateral presenta nos aminocidos excitatorios actan como neu- caractersticas clnicas y la situacin general
una influencia excitatoria en el ataque afectivo: rotransmisores. El ncleo central tambin pro- relacionada con la amnesia pueden ser ms
su estimulacin produce un ataque afectivo, en yecta a la sustancia gris periacueductal. importantes que los resultados de los exme-
tanto que su lesin disminuye el ataque afecti- Consecuentemente, se puede suponer una nes de laboratorio.
vo normalmente desencadenado por un estmu- accin restringida de los extractos de Datura El anlisis de la accin de la intoxicacin
lo nocivo [38]. La amgdala desempea un sobre el sistema lmbico, particularmente el con burundanga sobre el cerebro tiene conse-
papel muy especial en el aprendizaje de evita- hipocampo y la amgdala. cuencias significativas para comprender cmo
cin de los estmulos peligrosos; es decir, es Durante siglos las plantas del gnero Datu- se toman normalmente las decisiones. La toma
responsable de la aparicin de respuestas de ra se han utilizado ampliamente con fines ri- de decisiones no slo es una habilidad frontal
defensa en caso de ataque. En caso de lesiones tuales, en chamanismo y en brujera [5,7]. En [48-51], como fcilmente podra suponerse, si-
amigdalinas, se espera razonablemente una Amrica, varias plantas de este gnero han no que es tambin el resultado de un tipo parti-
conducta de sumisin y cierta incapacidad para servido para preparar extractos y ungentos cular de estado emocional. El sistema emocio-
responder a la agresin. Neuronas colinrgicas con propsitos rituales [4]. En diferentes pa- nal del cerebro puede contribuir significativa-
median la mayora de los ncleos amigdalinos; ses se encuentran informes que apoyan este mente al proceso de toma de decisiones. La in-
adems, la amgdala tiene conexiones impor- tipo de utilidad. Por ejemplo, los habitantes toxicacin con burundanga puede representar
tantes con el ncleo basal de Meynert [39]. precolombinos Chibchas que habitaban el te- un excelente modelo para comprender cabal-
Plotnik et al [40] analizaron la accin de la rritorio donde actualmente se encuentra Co- mente el proceso de toma de decisiones.
escopolamina sobre la conducta de sumisin lombia administraban un extracto de Datura
en monos. Observaron que cuando se limita el a las viudas y sirvientes de los jefes reciente- A. Ardila-Ardila a, C.B. Moreno b,
espacio y, en consecuencia, se maximiza la mente muertos, para enterrarlos vivos junto S.E. Ardila-Gmez c
posibilidad de interaccin social, la escopola- con el cuerpo de su jefe [5]. En varios pases
mina aumentaba consistentemente la agresivi- de Amrica Central, la especie Datura se ha Aceptado tras revisin externa: 17.11.05.
a
dad de los monos dominantes y disminua la utilizado extensamente para producir ungen- Florida International University. Miami, Florida,
agresividad en monos no dominantes. Cuando tos con propiedades anestsicas y para crear EE. UU. b Grupo de Neurociencias NEUROS. Uni-
se aument el espacio y se minimiz, en con- sentimientos de levitacin en ceremonias ri- versidad del Rosario.Bogot. c Hospital Universita-
rio San Ignacio. Bogot, Colombia.
secuencia, la posibilidad de interaccin social, tuales [3]. Se ha propuesto que la Datura stra-
la escopolamina produjo una reduccin de la monium se utiliza en el proceso de creacin de Correspondencia: Dr. Alfredo Ardila. 12230 NW 8th
Street. Miami, Florida 33182. USA. E-mail: alfredo-
agresin en todos los monos. Los estudios de los llamados zombis en Hait [45]. ardila@cs.com
ratones con conducta de sumisin adquirida en El cuadro clnico observado en casos de in-
un modelo de contacto sensorial han demos- toxicacin con burundanga, sin embargo, pue-
trado un aumento en los niveles de serotonina de resultar no slo de la accin de la escopola- BIBLIOGRAFA
(5-HT) en el complejo amigdalino, el hipo- mina, sino tambin de las benzodiacepinas y,
campo, el ncleo caudado y el hipotlamo [41]. frecuentemente, del alcohol que se combinan. 1. Ardila A, Moreno C. Scopolamine intoxica-
Misslin [42] propuso una distincin entre Por ejemplo, se ha sugerido una asociacin tion as a model of transient global amnesia.
conductas de defensa anticipatorias desenca- entre amnesia global transitoria y benzodiace- Brain Cogn 1991; 15: 236-45.
REV NEUROL 2006; 42 (2) 127

CORRESPONDENCIA
2. Castaneda C. The teaching of Don Juan: A 20. Sticht G, Kaferstein H, Staak M. Results of 36. Verges M. Control amygdalien de comporte-
Yaqui way of knowledge. Berkeley, CA: Uni- toxicological investigations of poisonings ments dagression chez le rats. Physiol Behav
versity of California Press; 1968. with atropine and scopolamine. Acta Med 1976; 17: 429-44.
3. Furst PT. Hallucination and culture. San Fran- Leg Soc (Liege) 1989; 39: 441-7. 37. Bolhuis JJ, Fitzgerald RD, Dijk DJ, Koolhaas
cisco: Chandler & Sharp Publishing; 1976. 21. Urich RW, Bowerman DL, Levisky JA, Pflug JM. The corticomedial amygdale and learn-
4. Lockwood TE. The ethnobotany of Brug- JL. Datura stramonium: a fatal poisoning. J ing in an agonistic situation in the rat. Physi-
mansia. J Ethnopharmacol 1979; 1: 147-64. Forensic Sci 1982; 27: 948-54. ol Behav 1984; 32: 575-9.
5. Arvalo C. Aspectos toxicolgicos de la bu- 22. Hamilton RJ, Perrone J, Hoffman R, Hen- 38. Hilton SM, Zbrozyna AW. Amygdaloid re-
rundanga. Medicina 1983; 7: 22-3. retig FM, Karkevandian EH, Marcus S, et al. gion for defense reactions and its afferent path-
6. URL: http://new.blackvoices.com/news/fea- A descriptive study of an epidemic of poi- ways to the brain stem. J Physiol 1963; 165:
tures/ny-celia4,0,7214581.realaudio?coll= soning caused by heroin adulterated with 160-73.
bv-news-archive. Fecha ltima consulta: scopolamine. J Toxicol Clin Toxicol 2000; 39. Carlson NR. Physiology of behavior. 2 ed.
14.11.2004 38: 597-608. Boston: Allyn & Bacon; 1986.
7. Harner MJ. Hallucinogens and shamanism. 23. Centers for Disease Control and Prevention. 40. Plotnik R, Mollenauer S, Gore W, Popov A.
New York: Oxford University Press; 1973. Scopolamine poisoning among heroin users Comparing the effects of scopolamine on
8. Goodman GA, Goddman LS, Gilman A. The New York City, Newark, Philadelphia, and operant and aggressive responses in squirrel
pharmacological basis of therapeutics. 6 ed. Baltimore, 1995 and 1996. JAMA 1996; 276: monkeys. Pharmacol Biochem Behav 1975;
New York: Macmillan Publishing; 1980. 92-3. 3: 739-48.
9. Britzer DA, Manning DW. Delirium induced 24. URL: http://earthops.org/burundanga.html. Fe- 41. Devoino LV, Alperina EL, Podgornaya EK,
by poisoning with anticholinergic agents. Am cha ltima consulta: 12.12.2004. Polyakov OV, Idova GV, Ilyuchenok RY.
J Psychiatry 1982; 139: 1343-4. 25. URL: http://www.fpnotebook.com/ER106.htm. Nature of the distribution of serotonin and a
10. Uribe MG, Uribe C. Neurotoxicidad por esco- Fecha ltima consulta: 12.12.2004. serotonin metabolite in brain structures and
polamina y nueva burundanga. In Uribe MG, 26. URL: http://www.corpus-delicti.com/smp/ the development of immunosuppression in
ed. Neurotoxicologa. Bogot: Asociacin Pittel_Burundanga.html. Fecha ltima consul- submissive mice. Neurosci Behav Physiol
Colombiana de Neurologa; 2001. p. 89-96. ta: 12.12.2004. 2003; 33: 473-7.
11. Hall RCW, Pfefferbaum P, Gardner ER, Stick- 27. URL: http://earthops.net/klaatu/sc-palamine1. 42. Misslin R. The defense system of fear: be-
ney SK, Perl M. Intoxication with angels html. Fecha ltima consulta: 12.12.2004. havior and neurocircuitry. Neurophysiol Clin
trumpet: anticholinergic delirium and hallu- 28. URL: http://sec-global.com/advisories/SA 2003; 33: 55-66.
cinocis. J Psychodelic Drugs 1978; 10: 251-3. 0002270400.html. Fecha ltima consulta: 43. Gregg TR, Siegel A. Brain structures and
12. Baraka A, Harim S. Reversal of central anti- 12.12.2004. neurotransmitters regulating aggression in
cholinergic syndrome by galanthamine. JA- 29. Pea LB, Puente F, Arenas LA. Caractersti- cats: implications for human aggression. Prog
MA 1977; 238: 2293-4. cas de la intoxicacin con escopolamina (vu- Neuropsychopharmacol Biol Psychiatry 2001;
13. Rubner O, Kummerhoff PW, Haase H. An rundanga) y su tratamiento en el Hospital 25: 91-140.
unusual case of psychosis caused by long- Universitario Ramn Gonzlez Valencia. Re- 44. Shaikh MB, Siegel A. Neuroanatomical and
term administration of a scopolamine mem- vista UIS 1983; 11: 31-46. neurochemical mechanisms underlying
brane patch. Paranoid hallucinogenic and 30. Moreno C, Rodrguez R, Jannaut MJ, Orozco amygdaloid control of defensive rage behav-
delusional symptoms. Nervenarzt 1997; 68: M, Romero CA, Ruiz LM, et al. Efectos de la ior in the cat. Braz J Med Biol Res 1994; 27:
77-9. intoxicacin por escopolamina (burundanga) 2759-79.
14. Muller J, Wanke K. Toxic psychoses from sobre la memoria [no publicado]. 45. Davis W. Passage to the darkness: the ethno-
atropine and scopolamine. Fortschr Neurol 31. Moreno C, Yomayusa N, Vargas VA, Mora- biology of the Haitian Zombies. Chapel Hill,
Psychiatr 1998; 66: 289-95. Izquierdo R. Intoxicacin por escopolamina: NC: University of North Caroline Press; 1988.
15. Ziegler B, Tonjes W. Scopolamine poisoning casos del Instituto de Medicina Legal. II World 46. Danek A, Uttner I, Straube A. Is transient
as a cause of acute paranoid hallucinatory Congress of Legal Medicine. Bogot, Co- global amnesia related to endogenous benzo-
psychoses. Psychiatr Prax 1991; 18: 21-4. lombia; 1990. diazepines? J Neurol 2002; 249: 628.
16. Mego DM, Omori JM, Hanley JF. Transder- 32. Ardila A, Ardila SE. Intoxicacin por es- 47. Brizer DA, Manning DW. Delirium induced
mal scopolamine as a cause of transient psy- copolamina (burundanga). Caractersticas so- with anticholinergic agents. J Psychiatry 1981;
chosis in two elderly patients. South Med J ciales. Neuropsicologa, Neuropsiquiatra y 139: 1343-4.
1988; 81: 394-5. Neurociencias 2002: 4; 161-74. 48. Ardila A, Surloff C. Executive dysfunction.
17. Sennhauser FH, Schwarz HP. Toxic psycho- 33. Tart C. Altered states of consciousness. New Ann Arbor, NJ: Neurobase; 2003.
sis from transdermal scopolamine in a child. York: John Wiley & Sons; 1969. 49. Goldberg E. The executive brain. New York:
Lancet 1986; 2: 1033. 34. Dobkin de Ros M. Psychodelic healing in Oxford University Press; 2001.
18. Cairncross JG. Scopolamine psychosis revis- the Peruvian Amazon. San Francisco: Chand- 50. Luria AR. Frontal lobe syndromes. In Vinken
ited. Ann Neurol 1983; 13: 582. ler Publishing; 1972. PJ, Bruyn GW, eds. Handbook of clinical neu-
19. Cheng SW, Hu WH, Hung DZ, Yang DY. 35. Sah P, Faber ESL, Lpez de Armentia M, rology. Vol 2. Amsterdam, North Holland;
Anticholinergic poisoning from a large dose Power J. The amygdaloid complex: anato- 1969. p. 725-57.
of Scopolia extract. Vet Hum Toxicol 2002; my and physiology. Physiol Rev 2003; 83: 51. Stuss DT, Knight RT. Principles of the frontal
44: 222-3. 803-34. lobe function. New York: Oxford; 2002.
128 REV NEUROL 2006; 42 (2)

Neuropsychology Review, Vol. 9, No. 3,1999
A Neuropsychological Approach to Intelligence
Alfredo Ardila1,2
This paper proposes that current psychometric intelligence tests are limited in evaluating cognitive
activity. From a neuropsychological perspective, they fail to measure some fundamental cognitive
abilities such as executive functions, memory, and visuospatial abilities. The analysis of the Wechsler
Intelligence Scale presented shows that the original rationale for selecting the specific subtests in-
cluded in the WAIS was unclear. The concept of a g factor in cognition is also analyzed, with the
conclusion that the g factor continues to be controversial. The value of intelligence tests in predicting
school performance is also criticized. It is proposed that the psychometric concept of general intelli-
gence should be deleted from cognitive and neurological sciences. Finally, it is proposed that, in the
future, neuropsychological instruments sensitive to more specific cognitive abilities replace current
psychometric intelligence tests.
KEY WORDS: Intelligence; IQ; neuropsychological assessment; cognitive abilities.
INTRODUCTION brain-damaged populations. Neither argument is accurate.

Psychometric intelligence tests are frequently included in
Curiously, intelligence tests do not appraise intelligence. the neuropsychological evaluations of brain-damaged in-
(Anonymous) dividuals. Even a Wechsler Intelligence Scale adapted for
neuropsychological purposes has been developed (WAIS-
The idea that is presented in this paper is very simple: From
R-NI; Kaplan et al., 1991). Neuropsychological instru-
a neuropsychological perspective, current psychometric
ments can be and are frequently used with normal popula-
intelligence tests (e.g., Wechsler Adult Intelligence Scale;
tions. Initially, neuropsychological tests are administered
WAIS) are limited in evaluating cognitive abilities. Fur-
to normal subject populations (norming studies) before
thermore, when using compound scores (e.g., IQ), it is not
being used with abnormal subject populations. It could
sufficiently apparent what these general scores measure.
be further argued that neuropsychological tests are fre-
It is suggested that in the future, current testing methods
quently easy and often have a low ceiling. Indeed, neu-
be replaced by neuropsychological cognitive assessment
ropsychological tests target pathological people and, in
instruments.
neurologically normal people, the ceiling frequently is
This idea, developed throughout the paper, emerges
rapidly reached. The ceiling effect varies depending on the
from the following points. (1) There are two different
specific test. However, this is not an intrinsic limitation,
sets of instruments directed to the appraisal of cognitive
and the ceiling can be raised. (2) From a neuropsycholog-
abilities: psychometric intelligence tests (e.g., the WAIS
ical point of view, intelligence tests do not evaluate some
in its different versions) and neuropsychological assess-
abilities that should be included as "fundamental cogni-
ment tests (e.g., Luria-Nebraska Neuropsychological Bat-
tive abilities" (i.e., "intelligence"). An analysis of execu-
tery, NEUROPSI). No evident reason seems to exist to
tive functions (i.e., "frontal lobe" abilities), memory, and
maintain this duality. It may be argued that psychomet-
visuospatial abilities is presented in this paper. It is em-
ric intelligence tests are directed to normal populations,
phasized that according to contemporary neuropsychol-
whereas neuropsychological instruments are directed to
ogy, these abilities represent some of the most important
cognitive abilities. They are inappropriately tested using
1 Instituto Colombiano de Neuropsicologia, Bogota, Colombia.
current psychometric instruments. (3) There was not suf-
2 Correspondence regarding this article should be addressed to the author
at 12230 NW 8 Street, Miami, Florida 33182. ficient scientific rationale for selecting the set of subtests
117
1040-7308/99/0900-01 17$16.00/0 C 1999 Plenum Publishing Corporation
118 Ardila
included in current intelligence batteries. Most of our cur- including culture, ecological demands, primary language,
rent knowledge about the brain organization of cognition and educational level. Test scores are associated, there-
has been obtained during the past 50 years. The specific fore, not only with the subject's learning opportunities, but
WAIS subtests currently in use were selected prior to ac- also with those variables that a culture dictates worthy of
quiring this knowledgenearly three quarters of a cen- cognitive amplification (Ardila, 1995a). Different cultural
tury ago. environmental contexts will result in the development of
The analysis of intelligence testing presented in this different patterns of abilities (Berry, 1971,1979). Further,
paper will center on the Wechsler Intelligence Scales when tests are used with members of a different culture,
(WIS). There are several reasons for this. First, it repre- testees often do not share the presumptions about val-
sents the most widely used intelligence scale, not only ues, knowledge, and communication implicitly assumed
in the United States but also in many other countries. by the test (Greenfield, 1997). (2) The specific tasks used
Second, it is the best studied and analyzed intelligence to tap intelligence are inappropriate. The first point has
scale. Many research studies using the WIS in different been extensively analyzed, particularly in anthropology
areas are easily available. Third, there are different ver- (e.g., Irvine and Berry, 1988) and cross-cultural psychol-
sions of the test (WAIS-R, WISC-III, WAIS-III, etc.) that ogy (e.g., Berry et al., 1992). In this paper, the major
have attempted to overcome the shortcomings that existed emphasis will be placed on the second point (i.e., that
in previous versions. As a result, they can be considered the the specific tasks used to tap intelligence are inappro-
best designed intelligence scales to date; at the least, very priate).
significant amounts of time and effort have been devoted A general conclusion of this paper is that the con-
to their design, redesign, and use. And fourth, they are the cept of general intelligence should be abandoned. But
most frequently used intelligence test battery components abandoning the concept of intelligence may seem too ex-
in neuropsychology. Researchers have even adapted the treme. Intelligence has become a fundamental cornerstone
WAIS to the neuropsychological perspective (WAIS-R- of contemporary psychology. Of course, there are two dif-
NI; Kaplan etal., 1991). ferent issues in this regard: the word intelligence and the
It has to be emphasized that not all the concerns existence of a general factor (g) in cognition.
presented with regard to the WIS are applicable to other It is proposed that the word intelligence be replaced
intelligence test batteries. For example, the second most with either cognitive abilities or simply cognition. What is
popular intelligence scale, the Stanford-Binet Intelligence the difference? The answer is simple: intelligence is con-
Scale (Thorndike et al., 1986) uses a somewhat differ- fusing and difficult to operationalize. Furthermore, it has
ent approach to intelligence. Stanford-Binet postulates a been frequently equated with psychometric intelligence
three-level hierarchical model of intelligence. The first tests (IQ).
level is represented by a general intelligence factor (g). It is also pointed out that the existence of a g fac-
There are three second level factors (Crystallized Abilities, tor in cognitive testing is questionable. An examination
Fluid-Analytic Abilities, and Short-Term Memory). Crys- of the history of the g factor in intelligence, and the dif-
tallized Abilities include Verbal Reasoning and Quanti- ferent factor-analytic studies carried out during the last
tative Reasoning. The third level refers to the subtests decades, leads to the conclusion that the assumption of a
included in this intelligence scale: Vocabulary, Compre- g factor is difficult to sustain. Many researchers consider
hension, Absurdities, and Verbal Relations to evaluate Ver- that g-based factor hierarchy confusing and misleading
bal Reasoning; Quantitative, Number Series, and Equation (e.g., Ceci, 1990; Lezak, 1995).
Building to assess Quantitative Reasoning; Pattern Anal- Finally, it has to be emphasized that intelligence is
ysis, Copying, Matrices, and Paper Folding and Cutting obviously a construct, not a physical entity. Frequently,
to appraise Fluid-Analytic Abilities (Abstract-Visual Rea- however, the term intelligence has been used as if it were
soning); and finally Bead Memory, Memory for Sentences, a physical entity, a reification that can be easily and objec-
Memory for Digits, and Memory for Objects to evaluate tively measured. In this paper, it is argued that the construct
Short-Term Memory. It is evident that the Stanford-Binet of intelligence is no longer tenable, and thus, measures of
overtly recognizes that reasoning and memory represent general intelligence are inappropriate and misleading.
fundamental elements of cognition. Many words from popular language have been in-
The concept of intelligence can be criticized from corporated into psychology. Terms such as will, mind, and
two different points of view. (1) Cognitive abilities mea- consciousness are just a few examples that have been elim-
sured by intelligence psychological tests represent, at least inated with the evolution of psychology because their pre-
in their contents, culturally learned abilities. Performance cise meaning is vague. For the same reason, it is proposed
is influenced by a vast array of moderating variables, that the term intelligence should also be removed.
Intelligence and Neuropsychology 119
INTELLIGENCE TESTING: A BRIEF to emphasize the importance of a common factor in intel-

HISTORICAL OVERVIEW ligence" (Brody, 1992, p. 13).
The attempt to measure intelligence represents one

of the major endeavors in 20th century psychology. A Multiple-Factor Approaches
tremendous amount of research has been directed to un-
derstanding the organization of intellectual activity and The most distinguished representant of the second
discussing the procedures appropriate to its measurement. point of view was L. L. Thurnstone (1938, 1947), who
In 1904 the Ministry of Education in Prance com- further developed factor analysis, attempting to obtain the
missioned Alfred Binet and Theophile Simon to develop most parsimonious solutions of the data to which they were
a practical procedure to distinguish between mentally re- applied. He introduced new concepts and more sophisti-
tarded and normal children at school. To fulfill this pur- cated procedures in factor analysis, such as oblique-factor
pose, they developed a kind of developmental scale de- structure and centroid methods. He proposed a relatively
scribing the types of abilities that were normally expected limited number of factors that would correspond to the fun-
at different ages (Binet, 1905,1908). The concept of "men- damental or primary mental abilities: Space, Verbal Com-
tal age" was introduced to refer to the level of development prehension, Word Fluency, Induction, Perceptual Speed,
expected at each age. Later, Stern (1912) introduced the Deduction, Rote Learning, and Reasoning. He supposed
concept of IQ. The Binet-Simon tests were rapidly adopted that each factor should correspond to certain specific ner-
in England, the United States, and other countries. In the vous system activity. Further studies (e.g., Kaiser, 1960)
United States, Terman (1916), at the University of Stand- have significantly supported most of the original primary
ford, adapted and standardized the scales presented by factors proposed by Thurstone.
Binet and named them the Stanford Revision of the Binet Vernon (1950) developed a kind of hierarchical model
Scale, or simply Stanford-Binet. Terman also further de- to describe the organization of cognitive abilities. He as-
veloped the concept of IQ. Nonetheless, how to understand sumed two major or second-order factors to group primary
cognitive abilities measured in intellectual tests remained abilities or primary factors: Verbal-Educational (v:ed) and
significantly controversial. Spatial-Mechanical (k:m) abilities. Thus, there was a kind
Two different interpretations of cognitive abilities of hierarchy in intelligence from the most general factor
rapidly became evident: (1) There is a general intelligence g, to the major factors v:ed and k:m, to the minor factors
factor that potentially may be measured and even quanti- or primary abilities, and finally to those factors specific to
fied; and (2) there are different cognitive abilities, not a each test.
single one. That is, single compound scores are not ac- Guilford (1967,1968; Guilford and Hoepfner, 1971)
ceptable. took a somewhat different approach. He proposed a
three-dimensional classification of intelligence including
contents (letters, numbers, words, and behavioral descrip-
The "g" Intelligence Factor tions); operations (memory, evaluation, convergent think-
ing, and divergent thinking); and products (units, classes,
Spearman (1904,1923) may be considered the most relations, systems, transformations, and implications).
important representative of the first point of view. He hy- Consequently, according to Guildford, 120 different in-
pothesized a two-factor theory of intelligence. He sup- tellectual abilities could be distinguished. He supposed
posed that any test measures a g factor common to all that empirical data would support the existence of this
other cognitive tests; and a specific factor (s) unique to high number of intellectual abilities.
that particular test. The relation between g and s compo- Cattell (1971) proposed that more than one second-
nents may be variable, but g is always included in any order analysis factor could be found. He distinguished be-
cognitive tests. Tests without the g factor may be tests of tween "Fluid Intelligence" (corresponding to and reflect-
sensory or motor abilities, but they do not represent cog- ing a pattern of neurophysiological and incidental learning
nitive tests. The existence of this g factor constitutes the influences) and "Crystalized Intelligence" (highly sensi-
theoretical basis to accept that intelligence can be quanti- tive to each person's unique cultural, educational, and en-
tatively measured using a simple score (IQ). Spearman's vironmental experiences). This distinction rapidly became
theory was subjected to diverse fundamental criticism on quite popular. Cattell's distinction between two different
empirical grounds. However, "while Spearman was aware types of intelligence is similar to the two major intellec-
that his theory had been empirically refuted, he continued tual factors proposed by Hebb (1942): Intelligence A and
120 Ardila
Intelligence B. Intelligence A represents the basic biolog- abilities, and contemporary factor analytic studies of neu-
ical ability to acquire knowledge. Intelligence B reflects ropsychological tests, is remarkable.
the influence or expression of acculturation, education,
and personal experiences.
Sternberg's Triarchic Theory
The Idea of Multiple Intelligences Sternberg (1988) defined intelligence as "the men-
tal activity underlying purposive adaptation to, shaping
Gardner's multiple intelligence approach might be of, and selection of real-world environments relevant to
interpreted as a return to Thurnstone. Gardner (1983) pro- one's life" (p. 69). The triarchic theory of intelligence con-
posed the existence of different and independent types of sists of three closely interrelated subtheories: a contextual
intelligence. In developing his model of intelligence, he subtheory, a componential subtheory, and an experiential
began with several observations: (1) Damage in differ- subtheory (Sternberg, 1985).
ent neural structures may result in impairing certain abil- The contextual subtheory limits intelligence to men-
ities while sparing other abilities; that is, isolated defects tal activity underlying environments relevant to one's life.
in some types of cognition can be observed in cases of In consequence, intelligence should be conceptualized
brain pathology (a procedure known in neuropsychology considering the real conditions existing in the immediate
as "double dissociation"). (2) Individuals such as idiot- environment. Intelligence represents adaptation to one's
savants demonstrate a significant dissociation in different environment. Mental activity is directly inferable through
cognitive abilities. That is, in non-brain-damaged indi- techniques widely available to cognitive psychology.
viduals intellectual abilities may be dissociated, and even The componential subtheory states that the mental
extremely dissociated. (3) Every type of ability is iden- mechanisms are those that affect and are affected by con-
tified by a specific set of operations related to a neural text. Intelligence makes sense only within a particular con-
mechanism; in this regard, Gardner is attempting to rec- text. The basic "mental unit" of analysis in this subtheory
oncile the idea of several types of intelligence with current is the information processing component. This refers to the
research about brain organization of cognition. (4) A spe- process transforming sensory inputs into conceptual repre-
cific developmental history for each type of intelligence; sentations, transforming a conceptual representations into
that is, different cognitive abilities ("intelligences") de- another, or transforming conceptual representations into
velop independently in a child. (5) An evolutionary his- motor acts.
tory exists for each intelligence; that is, different intelli- The experiential subtheory states that tasks are par-
gences may have different origins in subhuman species ticularly relevant to the measurement of intelligence when
and may have evolved in different ways. (6) Experimen- they measure cognitive performance either when a task or
tal psychology supports the existence of different intel- situation is novel or when the task is in the process of
ligences. (7) Psychometric studies support the indepen- becoming automatized.
dence of different cognitive abilities; Gardner insists that Sternberg (1997) has attempted to apply his interpre-
psychometric research has not investigated widely enough tation of intelligence to testing in the field of intelligence
the diversity of intellectual abilities that are observed in and the understanding of lifelong learning. His interpre-
real contexts. (8) Susceptibility of different abilities to en- tation of intelligence allows significant cultural variations
coding in a symbolic system; he proposed that cognitive and emphasizes the understanding of the behavioral con-
abilities tend to be encoded in culturally different devised text.
symbolic systems.
Departing from these considerations, Gardner pro-
poses six different types of intelligence: Linguistic, mu- A Processing Speed Interpretation of Intelligence
sical, logic-mathematical, spatial, body-kinesthetic, and
personal. This group of intelligences may partially corre- It has also been proposed that intelligence depends
spond to Thurnstone's primary mental abilities. However, on what may be called "the neural efficiency of the brain"
Gardner is relying not simply on psychometric procedures (Eysenck, 1986). Several recent studies have demonstrated
but also on a broad array of contemporary research, includ- that the time required to perform some simple perceptual
ing contemporary neuropsychology. tests are significantly correlated with psychometric intel-
In brief, Gardner proposes a relatively limited num- ligence test scores. This means that intelligence may be
ber of basic and independent abilities or types of intelli- related to some characteristics of information processing
gence. The similarity with Thurnstone's primary mental in the central nervous system. Jensen (1987) observed a
correlation between choice reaction time and scores on and a supplement test (Mental Status). Each IQ (Fluid,
intelligence tests. These correlations, however, were not Crystalized, and Composite) has a mean of 100 and stan-
particularly impressive (about 0.20 to 0.30). It was dard deviation of 15.
observed that reaction time was inversely correlated with Naglieri and Das (1996) suggested that intelligence
IQ and measures thought to singly predict approximately should be seen as a cognitive construct, integrating neuro-
10-15% of the variance in IQ (Brody, 1992). Higher cor- physiological findings, cognitive processing research, and
relations on the order of -0.40 using more complex reac- sociocultural components of human performance. They
tion time techniques have been reported by Frearson and base their intelligence theory on Luria's interpretation
Eysenck (1986). about the three brain functional units (motivation-
Nettlebeck (1987) found a correlation on the order emotion, processing-storing information, and planning-
of 0.50 between inspection time and IQ. The technique controlling behavior). They assume that intelligence con-
used consisted of tachistoscopic presentations of two ad- sists of these three components: attentional processes that
jacent vertical lines followed by a masking stimulus. The provide focused cognitive activity, information processes
time of exposure varied and a psychophysical function was of two types (simultaneous and successive), and planning
obtained. The task was used to ascertain the minimal ex- processes that provide control of attention; the use of
posure time required to obtain a certain level of accuracy information processes, internal and external knowledge,
in recognizing which one of the two lines was longer. and cognitive tools; and self-regulation to achieve de-
Reed and Jensen (1992) have used visual evoked- sired goals (Naglieri, 1997). They refer to their theory as
potentials to assess what they call nerve conduction ve- the Planning, Attention, Successive, Simultaneous (PASS)
locity. They calculate this velocity by dividing the subtheory of intelligence (Das et al., 1994). They then devel-
ject's head length by the latency of an early visual evoked oped a Cognitive Assessment System (CAS) applicable to
potential component. Using this procedure, they report a children up to the age of 18. The CAS includes measures
correlation between nerve conduction velocity and intel- of attention (Expressive Attention, Number Detection, Re-
ligence on the order of 0.20 to 0.30. ceptive Attention), simultaneous processing (Matrices,
In brief, some significant correlations have been es- Figure Memory, Verbal-Spatial Relations), successive pro-
tablished between speed in information processing and cessing (Word Series, Sentence Repetition, Sentence
scores on psychometric intelligence tests. These measures, Question, Speech Rate), and planning (Number Match-
however, usually predict only a relatively modest percent- ing, Planned Codes, Planned Connection).
age of the variance. Both test batteries have at least three major common
points: (1) They relate intelligence with brain activity and
in this regard represent neuropsychologically oriented in-
Neuropsychologically Oriented Intelligence Tests telligence scales; (2) they are based on Luria's theory
about brain organization of cognition; and (3) they attempt
Some attempts have been made to approach the con- to include those cognitive abilities associated with pre-
cept of intelligence and to develop intelligence test bat- frontal functions (Luria's third functional unit; prefrontal
teries based on a neuropsychological perspective. Two of or "executive" functions). In this regard, they recognize
these attempts will be briefly examined: The Kaufman that executive functions must be regarded as crucial ele-
Adolescent and Adult Intelligence Test (KAIT; Kaufman ments of intelligent behavior.
and Kaufman, 1993,1997) and the Cognitive Assessment
System (CAS; Das etal., 1994; Naglieri and Das, 1996).
The KAIT provides three types of scores: Fluid, Crys- THE WECHSLER INTELLIGENCE SCALES
talized, and Composite IQS. It is applicable to people be-
tween the ages of 11 and 85. According to the authors, What is intelligence? Many definitions of intelligence
the tests were developed based on the models of Piaget's have been proposed (e.g., Binet, 1908; Jensen, 1980;
formal operations and Luria's planning ability in an at- Sternberg, 1985; Wechsler, 1944). In current literature,
tempt to include high-level decision making tasks (Luria's we still find a wide variety of definitions, many of which
third functional unit). The Crystalized Scale includes Def- make reference to the mental abilities. For the purpose
inition, Auditory Comprehension, Double Meaning, and of this analysis of intelligence tests, Wechsler's definition
Famous Faces subtests. The Fluid Scale includes Rebus will be used. Wechsler (1944) defined intelligence as "the
Learning, Logical Steps, Mystery Cards, and Memory for aggregate or global capacity of the individual to act pur-
Block Designs. The KAIT also include two additional sub- posefully, to think rationally and to deal effectively with
tests (Rebus Delayed Recall and Auditory Delayed Recall) his environment" (p. 3).
122 Ardila
This definition can be divided into four different ele- to evaluate the capacity of the individual to act purpose-
ments: (1) Intelligence is an aggregate or global capacity, fully, to think rationally, and to deal effectively with his
(2) to act purposefully, (3) to think rationally, and (4) to environment. Wechsler (1944) explains that:
deal effectively with the environment.
The first element of Wechsler's definition of intel- In arriving at our final choice of tests we used the following
procedure: (1) Careful analysis was made of the various stan-
ligence refers to a core issue: Is there such a thing as a dardized tests of intelligence now in use. These were studies
global or general intelligence, or rather, is intelligence an with special attention to the author's comments with reference
aggregate of abilities? In his defintion, Wechsler does not to the type of functions measured, the character of the popula-
take a definite position, but he assumes a theory of general tion on which the scales were originally standardized, and the
intelligence in developing his test battery (Full Scale IQ). evidence of the test's reliability. (2) An attempt was made to
evaluate each test's claim to validity as evidenced by its degree
Yet he recognizes at least two major types of intelligence: of correlation (a) with other recognized test and, (b) more im-
verbal and performance intelligence. This question about portantly still, with subjective ratings of intelligence. The later
one or several intelligences continues up to the present included teachers' estimates, rating by army officers (as in the
day (see Neisser et al., 1996). case of the Army Alpha and Beta), and estimates of business
The second element in Wechsler's definition of in- executives (as in the case of various tests which had been tried
out in industry). (3) An attempt was made to rate the tests on
telligence ("to act purposefully") could be understood as the basis both of our own clinical experience and of that of
the control, organization, and planning of behavior. Acting others. (4) Some two years were devoted to the preliminary
purposefully is evidently a frontal lobe function (executive experimental work of trying out various likely tests to on the
function) if taken from a neuropsychological perspective several groups of known intelligence level, (p. 76)
(e.g., Luria, 1980; Stuss and Benson, 1986).
The third element ("to think rationally") might be Unfortunately, Wechsler fails to explain exactly how
understood as either organization of cognition (metacog- these steps were taken.
nition) or problem-solving ability. In either case, the def-
inition deals with executive functions (Stuss and Benson,
1986). It means that intelligence, to a significant degree, DO INTELLIGENCE TESTS PREDICT
refers to executive functions. From the neuropsychologi- SCHOOL PERFORMANCE?
cal perspective, this approach may sound quite attractive
(i.e., intelligence means planning behavior and organiz- The most frequent argument of the defenders of in-
ing cognition). Unfortunately, it will be explained that telligence testing and IQ is that intelligence tests can in
WIS blatantly fails to evaluate executive functions. There a reliable way predict school performance (e.g., Jensen,
is an overt discrepancy between the defintion of intelli- 1980). As a matter of fact, this was the initial purpose of
gence presented by Wechsler and the testing included in intelligence tests.
the WIS. The influence of educational variables on intelligence
The final element ("to deal effectively with the envi- test performance represents a well established observation
ronment") refers to the functional criteria of intelligence. (e.g., Anastasi, 1988; Cronbach, 1990). Educational at-
Of course, intelligence may be understood not only from tainment significantly correlates with scores on standard
a psychometric perspective but also from a functional per- tests of intelligence. This correlation ranges from about
spective (Pirozzolo, 1985). Wechsler appropriately recog- 0.57 to 0.75 (Matarazzo, 1972). Correlations with verbal
nizes that intelligence has to be considered with regard intelligence subtests are usually higher (from about 0.66
to the specific environment. The physical and social en- to 0.75) than correlations with performance intelligence
vironment can be quite different between Seattle and the subtests (from about 0.57 to 0.61). But correlation does
Amazonian jungle, as well as between different subcul- not mean causality; it simply means association.
tures existing in a complex city such as New York. Colom- The crucial question is: Do intelligence tests really
bian street children can deal extremely well with their city predict school performance? Or, do schools train those
environment, even though from a psychometric point of abilities appraised in intelligence tests? To answer these
view they may score at the level of mental retardation. questions is not easy, even though frequently the interpre-
Unfortunately, it is not so easy to evaluate effectiveness in tation has been that IQ does predict school performance
dealing with the environment from the outside. This can (e.g., Hunter, 1986). Other researchers, however, consider
only be appropriately evaluated from inside the culture or that IQ scores are, to a significant extent, a measure of
subculture itself. In this regard, intelligence becomes an direct and indirect school learning (e.g., Ceci, 1990).
anthropological issue. Ceci (1991) presented an extensive and detailed re-
The question at this point is how Wechsler decided view of available data in this area. The general conclusion
that the WIS subtests he selected were the most appropriate is that school attendance accounts not only for a substantial
portion of variance in children's IQ but also apparently From a neuropsychological perspective, this represents an
some, though not all, of the cognitive processes that un- extremely important observation when constructing any
derpin successful performance in IQ tests. The magni- theory about the organization of cognitive abilities; inde-
tude of this influence ranges between 0.25 to 6 IQ points pendent cognitive abilities can be independently impaired
per year of school. As a result, the association between and can independently deteriorate during normal and ab-
IQ and education cannot be interpreted as indicating that normal aging. According to Lezak (1995), the concept of
IQ predicts school success. Intelligence and schooling intelligence has limited application. The concept of IQ,
have complex bidirectional relationships, with each one she notes, represents so many kinds of more or less con-
influencing variations in the other (Ceci and Williams, founded functions as to be conceptually meaningless. She
1997). concludes that the "IQ as a score is inherently meaning-
There are two additional observations that should be less and not infrequently misleading IQ as a catchword
emphasized. (1) The largest correlations between IQ and has outlived whatever usefulness it may once have had and
school performance are not found with Full Scale IQ but should be discharged" (p. 25).
with Verbal IQ, and particularly with some verbal subtests Heterogeneity of cognitive abilities is supported by
(e.g., Vocabulary). So, the question that might be raised is, empirical neuropsychological data. This is true not only
Why bother to administer the complete intelligence scale in abnormal but also normal populations. As an illustra-
if the verbal subscale (and even one single subtest) is suf- tion, Ardila et al. (1998) selected a homogenous sample of
ficient, and may even be better? And, (2) IQ may "predict" normal subjects (300-subject sample, aged 17-25 years;
performance in language, reading, writing, and arithmetic all of them right-handed, middle-class male university
but cannot predict performance in other areas, such as students). An extensive neuropsychological test battery
drawing or music. Simply speaking, if language tests are was administered including language, memory, perceptual
used as predictors, verbal performance can be predicted. abilities, concept formation, and praxis abilities. Forty-
That is quite obvious. Evidently, our current educational one different scores were calculated. Table I presents the
system is significantly biased in favor of verbal abilities. dispersion in scores observed on some of the most "classi-
This is not true in other cultures' educational systems, nor cal" psychological and neuropsychological tests (WAIS,
is it always true in all of our educational programs. It there- Wechsler Memory Scale, etc.) that were included in this
fore seems questionable that the WIS is a good predictor research. It is evident that a particularly high dispersion
of school performance in music and art schools. in scores is found in this completely normal and homoge-
Ceci (1991) emphasizes that there is a circularity in nous population. The ratio between the lowest and highest
the association between school performance and IQ. IQ scores in most tests was around 1:5-l 0. In some test scores
test items were initially selected from those items that ac- it was even higher. In the WAIS subtests, dispersion was
cording to teachers' opinions, poor learners had the most particularly high in Information, Arithmetic, Block de-
difficulty answering in class. Variants of these questions sign, Object assembly, and Digit-symbol. It was relatively
are still found in large number on contemporary IQ tests. lower in Digits, Picture completion, and Picture arrange-
So, it is obvious that IQ tests would predict school suc- ment. Even in young, normal, and highly educated indi-
cess, as they were composed of items that poor learners viduals there is a very significant dispersion in the per-
found most difficult. Consequently, school failure is both formance of usual psychological and neuropsychological
explained as a lack of intelligence and is itself the basis test scores. Significant intersubject and intrasubject dif-
for the definition of lack of intelligence. The circularity is ferences have to be taken into consideration in any theory
evident. about organization of cognition, and in the evaluation of
intellectual abilities.
A NEUROPSYCHOLOGICAL PERSPECTIVE OF
COGNITIVE ABILITIES AND THEIR MEASURE Luna's Interpretation of Brain Organization
of Cognition
In neuropsychology a significant opposition to the
use of compound IQ scores is frequently observed. Lezak No doubt, one of the most influential theoreticians
(1995), for example, stressed that neuropsychological ob- in contemporary neuropsychology has been A. R. Luria.
servations demonstrate that there are independent intel- Luria (1980) proposed that cognitive abilities represent
lectual functions; Brain damage can impair certain func- "functional systems." The concept of the functional sys-
tions while sparing others. In consequence, compound, tem is understood as a group of interconnected biological
global, or total scores can be artificial and meaningless. operations that produces a particular biological effect. The
124 Ardila
Table I. Performance of 300 Normal Subjects in "Classical" for the spatial organization of numbers, figures, drawings,
Psychological and Neuropsychological Tests etc. In all the functional systems in which the paricular
Test Mean SD Range Ratioa ability is included, the defect will be apparent. The brain
damage produces not the loss of a specific cognitive pro-
WAIS cess (functional system), but its disturbance at a specific
Information 16.7 4.4 4-27 :6.7 level. This implies that neuropsychological assessment
Similarities 18.7 3.6 6-26 :4.3
Arithmetic 11.8 7.7 3-18 :6.0
will be aimed at disclosing the fundamental defects un-
Vocabulary 53.5 9.2 18-76 :4.2 derlying the apparent deficits. For this purpose, it will be
Comprehension 17.6 4.5 6-27 :4.5 necessary to administer to the patients different types of
Digits 11.6 2.2 5-17 :3.4 tasks and to analyze how the particular difficulties in per-
Picture completion 16.8 2.8 8-21 1:2.6 forming each one of them are manifested.
Picture arrangement 25.3 5.5 10-36 1:3.6
Block design 39.3 7.4 0-48 Clinical-anatomical correlations were widely devel-
Object assembly 31.7 7.6 2-44 1:22.0 oped by Luria. As a matter of fact, he is a precursor of the
Digit-symbol 56.0 15.0 11-90 1:8.2 method of the superimposition of lesions to disclose criti-
WMS cal areas in a particular type of disorder. His study of 800
Information 5.8 0.4 3-6 1:2.0 patients to determine the critical brain area for phonemic
Orientation 4.9 0.2 4-5 1:1.2
Mental Control 7.3 1.9 0-9 discrimination deficits has become classic. This procedure
Logical Memory 15.1 3.2 5-21 1:4.2 of superimposing lesions to highlight critical areas re-
Visual Reproduction 12.1 1.9 6-19 1:3.1 sponsible for clinical syndromes is extensively used in the
Associative Learning 18.1 3.1 3-26 1:8.6 present-day neuropsychological research (e.g., Damasio
Verbal Fluency and Damasio, 1989; Kertesz, 1983). Luria strived to es-
Phonologic 11.6 2.5 3-17 1:5.6
Semantic 14.7 2.5 7-25 1:3.6
tablish correlations between brain pathology and distur-
The Rey-Osterrieth Complex Figure bances at specific levels of information processing (e.g.,
Copy 34.9 1.7 26-36 1:1.4 phonemic discrimination), not to correlate brain pathology
Immediate memory 28.2 5.0 9-36 1:4.0 with performance in specific tests. Tests may be changed,
Finger Tapping Test but since some specific level of information processing
Right hand 61.2 10.4 23-88 1:3.8
Left hand 53.0 8.7 21-79 1:3.7
would still be required, impairment will be manifested.
Reading speed (words/minute) 172.3 46.7 64-426 1:6.6 Because performance on even apparently very simple tests
WCST can require the participation of different brain systems,
Categories achieved 5.7 1.1 0-6 performance on such simple tests can be altered as a con-

Perseverative errors 9.6 9.3 0-77
sequence of very different brain pathology, although the
Nonperseverative errors 9.3 8.1 0-62 specific errors will be different. Many different types of
a
Ratio between the highest and lowest scores. When the lowest score brain pathology can alter, for instance, calculation abili-
is zero, the ratio cannot be calculated. From the mathematical point of ties; however, in each case the difficulty (and the errors)
view, it would be infinite. These conditions are indicated by a dash. will be the result of a disturbance at a different level. Pa-
tients with frontal lobe damage and patients with angular
functional system is based on a complex dynamic constel- gyros damage can both present with serious difficulties
lation of stages, situated at different levels of the nervous in performing simple calculation tests. However, the un-
system, which in performing an adaptative task, may be derlying impaired mechanism and the type of errors mani-
changed without the task itself being changed. To write, fested are quite different (Rosselli and Ardila, 1989). Con-
for instance, represents a complex psychological process sequently, the validity derived from correlating the site of
(functional system) that requires the participation of mul- the brain pathology with performance on a particular test
tiple areas of the brain; each of these areas makes its par- appears, in Luria's interpretation, as a very crude approx-
ticular contribution to the whole system. A focal lesion of imation.
the brain will disrupt the ability to write at a particular level For Luria, the information collected from the obser-
(the ability to perform the skilled movements required for vation of brain-damaged patients should be helpful for
writing, the spatial organization of writing, the selection developing a more accurate picture of brain organization
of words, the ability to sequence graphemes, etc.). How- of cognitive processes. If we knew well enough how the
ever, such particular focal damage will also disrupt all the brain works we should be able to accurately predict brain
functional systems for which that particular operation is pathology when analyzing in detail the performance of a
required. For instance, the patient will not only have dif- patient on a set of tests. The departure point in the neu-
ficulties for the spatial organization of writing but also ropsychological assessment is the knowledge about how
Table II. Factors Underlying Different Aphasia Syndromes, language area (Broca's area), and (6) the primary motor
According to Luria (1976a)a cortex controlling language articulation. Benson points
Aphasia Type Impaired Factor out that depending on the material used in reading, other
additional brain areas could also be involved in the read-
Acoustic-Agnostic Phoneme discrimination ing process. This whole array of brain areas would rep-
Acoustic-Amnesic Verbal memory resent the "brain system" (Ardila, 1995b) underlying the
Amnesic Semantic structure of words
Semantic Understanding logical-grammatical reading aloud process, and supporting the "functional sys-
(quasi-spatial) structures tem" for reading, according to a Lurian interpretation. In
Afferent Motor Articuleme discrimination case of damage in any of these written language process-
Efferent Motor Disturbances in speech kinetic structure ing levels, a deficit in reading will appear, even though
Dynamic Verbal initiative it would be different depending on the specific impaired
a
"Factor" in Luria's theory refers to the fundamental defect responsible area. Furthermore, other abilities also relying on one of
for a particular neuropsychological syndrome. these processing levels ("factors") would be also affected.
Contemporary neuroimaging and electrophysiologi-
the brain works, not the knowledge about how to apply a cal techniques have provided most valuable information
series of tests in standardized conditions. about brain activity during performance of different cog-
It is interesting to note that Luria extensively, but nitive tasks. Thus, departing from measures of focal brain
not systematically, used the term factor to refer to the metabolism, positron emission tomography allows one
deficit that can underlie an overt clinical disorder. At other to visualize levels of brain activity and focal involve-
times he simply referred to the basic deficit or underlying ment during different conditions. It has been observed that
defect affecting normal psychological performance. And, when performing complex intellectual tasks (e.g., reading
undoubtably, he did not use the term factor with a mathe- aloud, speaking, etc.) a complex matrix of activated ar-
matical meaning (i.e., factor analysis). eas is revealed (Posner et al., 1988). Different brain ar-
At this point, the question arises as to which "fac- eas participate, making specific contributions to the per-
tors" underlay performance in different neuropsycholog- formance during, for example, a reading task: occipital
ical tests. These factors would, in consequence, represent (visual perception), temporal (language decoding), and
the basic elements of cognition, Luria discussed this ques- frontal Broca's area (language control and production)
tion in some detail with regard to language. In his last (Pettersen et al., 1989). For each one, a somehow lim-
book Basic Problems of Neurolinguistics" (1976a), Luria ited region is fully activated, where some other areas can
analyzed the factors that can underlie the different apha- be only partially active (Price et al., 1994). While speak-
sic syndromes (Table II). As a matter of fact, these same ing, a specific activation of the left mouth area can be
factors had been previously pointed out by Luria years observed, as well as in the superior temporal lobe and the
before. However, it is not easy to deduce the impaired supplementary motor area.
factors in other neuropsychological syndromes (e.g., ag-
nostic or apraxic disorders). This "factorial theory" of cog-
nitive activity represents one of the most interesting and Factor Analysis in Neuropsychology
outstanding points in Luna's neuropsychological perspec-
tive. Unfortunately, Luria did not completely develop this In the neuropsychological domain, factor analysis
factorial theory of psychological activity. has been more frequently applied to some specific tests and
scales directed to measure single cognitive abilities. For
example, several factor analytic studies of the Wechsler
Further Developments Memory Scale are available to date (Ardila and Rosselli,
1994; Bornstein and Chelune, 1988; Elwood, 1991; Roid
Luria's interpretation of brain organization of cogni- etal., 1988; Wechsler, 1987). Nonetheless, factor analyses
tive activity has received support from contemporary re- of extensive neuropsychological battery tests are scarce.
searchers in the area. Thus, for Benson (1994), any com- Ponton etal. (1994) administered a neuropsycholog-
plex psychological activity requires the participation of ical test battery including 10 different tests to 300 nor-
different brain areas. As an example, according to Benson mal subjects. A factor analysis was used and five different
(1982), six different brain areas participate under normal factors were found: a Verbal Factor (measured basically
conditions in reading aloud: (1) primary visual cortex, through verbal fluency and naming), a Learning Factor
(2) association visual cortex, (3) angular gyrus, (4) tempo- (measured specially with an auditory verbal learning test),
ral areas involved in language recognition, (5) the frontal a factor related to the Speed in Processing Information
126 Ardila
(attention; measured with Digit-symbol subtest), a Visual the different scores of the Wisconsin Card Sorting Test.
Processing Factor (measured with the Rey-Osterrieth Factor IV (7.9% of the variance) was a Fine Movements
Complex Figure), and finally, a Psychomotor Speed Factor factor, and Factor V (6.7% of the variance) represented a
(measured with the Pin Test). Verbal Memory factor.
Ardila et al. (1994) administered a general neuro- Ostrosky etal. (1999) administered a short neuropsy-
psychological test battery to a 98-subject sample. Their chological test battery assessing a wide spectrum of cog-
battery included language, memory, spatial abilities, con- nitive functions including orientation, attention, memory,
cept formation, and praxis abilities tests. A factor analysis language, visuoperceptual abilities, and executive
with varimax rotation found nine different factors account- functions. Normative data in an 800-subject sample from
ing for about 70% of the variance. Factor I (Verbal Fac- 16 to 85 years of age, and from zero to 24 years of educa-
tor; accounting for 14.2% of the variance) was measured tion were obtained. A factor analysis with varimax rotation
by a Sequential Verbal Memory Test and Verbal Fluency disclosed seven different factors. Factor I (accounting for
subtests. Factor II (accounting for 12.9% of the variance) 28.6% of the variance scores) best correlated with Dig-
was measured by the WMS Visual Reproduction subtests its backwards, Copy of a semicomplex figure, Calculation
(Nonverbal Memory and Constructional Factor; imme- abilities, and Language Comprehension. Factor II (9.6%
diate and delayed reproduction) and the Rey-Osterrieth of the variance) highly correlated with the writing scores.
Complex Figure (copy and immediate reproduction). Fac- Factor III (accounting for 6.1% of the variance) best cor-
tor III (Verbal Memory Factor; accounting for 9.8% of the related with verbal fluency tests. Factor IV (accounting for
variance) was measured by the WMS Logical Memory 5.7% of the variance) was correlated with motor functions.
subtests (immediate and delayed). Factor IV (Fine Move- Factor V (accounting for 4.3% of the variance) correlated
ments Factor; accounting for 6.4% of the variance) was with all the recall scores. Factor VI (accounting for 3.9%
associated with fine movements (tapping subtests, right of the variance) correlated with Orientation in Space. Fac-
and left hand). Factor V (Verbal Knowledge; accounting tor VII (accounting for 3.6% of the variance) correlated
for 6.0% of the variance) was mainly measured by the with Orientation in Person.
Information subtest of the WMS and the Boston Naming In summary, several factor analyses of extensive neu-
Test. Factor VI (Praxis Ability Factor; accounting for 5.5% ropsychological test batteries have yielded quite similar
of the variance) represented ideomotor praxis tests. Fac- results: Some 5-10 factors are found, accounting for about
tor VII (Delay Associative Learning Factor; accounting two-thirds of the total variance. The first factor accounts
for 5.4% of the variance) was measured by the Delayed for some 15-30% of the total variance, and it is usually a
Associative Learning subtest, and Factor VIII (Arithmetic verbal factor. Some additional factors are also observed:
Factor; accounting for 5.0% of the variance) was measured spatial, memory, perceptual, fluency, motor skills, etc. The
by Digit Span. Factor IX (Mental Control Factor; account- damage in the "brain systems" supporting the intellectual
ing for 4.4% of the variance) was measured by the Mental activities corresponding to these factors (verbal abilities,
Control subtest of the WMS. Correlations between some spatial abilities, verbal fluency, etc.) would result in spe-
tests were negative (e.g., between Logical Memory from cific neuropsychological syndromes (aphasia, spatial ag-
the WMS and Rey-Osterrieth Complex Figure-Copy con- nosia, amnesia, etc.). These factors should be matchable
dition). Several correlations were around zero. This obser- with the neuropsychological syndromes found in cases
vation is particularly important from the point of view of of brain pathology (see Table III). Nonetheless, it does
the existence of a general factor in cognition (g factor). not seem realistic to suppose that the exact number of
Ardila et al. (1998) administered a comprehensive these factors can be determined. From a factor analysis
neuropsychological test battery: language, memory, per- perspective, the factors to be found obviously depend not
ceptual abilities, concept formation, and praxis abilities only on the types of tests that are included but also on some
tests to 300 normal subjects. Forty-one different scores additional variables (e.g., the subjects included in the anal-
were calculated. It was found that some of the tests pre- ysis, the type of factor analysis, etc.). From a neuropsycho-
sented a complex intercorrelation system, whereas other logical perspective, to exactly pinpoint the cognitive syn-
tests presented few or no significant correlations. A factor dromes observed in cases of brain pathology is not an easy
analysis with varimax rotation of the neuropsychologi- task, at least at the moment. As an illustration of this point,
cal battery tests was performed. Five different factors ac- we do not know well enough how to classify spatial dis-
counted for 63.6% of the total variance. Factor I (26.7% turbances associated with brain pathology (e.g., Benton,
of the variance) represented a clearly Verbal factor. Fac- 1989; De Renzi, 1982, 1985; Hecaen, 1962; Hecaen and
tor II was a perceptual or Nonverbal factor (12.5% of the Albert, 1978; Morrow and Ratcliff, 1988; Newcombe and
variance). Factor III (9.8% of the variance) correlated with Ratcliff, 1989). We do not know yet well enough either
Table HI. Factors Observed in Ardila etal. (1 994) Neuropsychological Test Battery, the Tests Most
Saturated by These Factors, and Probable Neuropsychological Syndromes that Might Be Associated with
Impairments in Those Factors
Factor Test Probable Neuropsychological Syndrome
I Verbal Production Verbal fluency Convexital left prefrontal syndrome

II Constructional- Visuospatial Rey-Osterrieth Figure Constructional apraxia
Visual memory WMS Spatial agnosia
III Verbal Memory Logical memory Verbal amnesia
Weroicke aphasia
IV Fine Movements Tapping test Premotor syndrome
Kinetic apraxia
V Verbal Knowledge Information Wernicke aphasia
Boston Naming Test Anomia
VI Praxis Ability Ideomotor apraxia test Ideomotor apraxia
VII Delayed Associative Learning Delayed Associative Hippocampal amnesia
Learning WMS
VIII Arithmetic Digits Acalculia
IX Attentional Mental control WMS Orbital prefrontal syndrome
Table IV. Some Relatively Constant Factors Found Across Different frequency across different factorial studies. This is ob-
Factor-Analytic Studies (Carroll, 1993) and the Neuropsychological served in different cognitive areas: reasoning, language,
Syndromes with Which They Might Be Associated
memory, visual perception abilities, etc. Table IV presents
Factor Neuropsychological Syndrome a summary of these relatively constant factors found across
different factor analytic studies. From a neuropsycholog-
Language ical perspective, these factors are expected to be impaired
Lexical Knowledge Wernicke aphasia
Grammatical Sensitivity Broca aphasia in cases of focal brain pathology. Some neuropsycholog-
Communication Ability Prefrontal syndrome ical syndromes are expected to be observed in cases of
Oral Production Verbal apraxia? disruption of these brain systems supporting these basic
Speech Sound Discrimination Word deafness cognitive factors.
Naming Facility Anomia
Expressional and Word Fluency Extransylvian motor aphasia
Reasoning
Sequential Prefrontal syndrome Frontal Lobes and Intelligence
Inductive Prefrontal syndrome
Quantitative Frontal acalculia Long ago it was noted that frontal damage did not
Visual Perception result in evident deficits in psychometric intelligence tests
Spatial Relations Spatial agnosia
Serial Perceptual Integration Topographic agnosia? (Hebb, 1939; Hebb and Penfield, 1940). This was true
Perceptual Speed Visual agnosia even in cases of bilateral frontal lobectomy. It was some-
Numerical how surprising to find that IQ in patients with frontal lobe
Number Facility Acalculia damage could be normal (Hebb, 1945). These initial obser-
Attention and Concentration vations carried out during the 1940s have been further doc-
Attention and Concentration Prefrontal syndrome
umented in neuropsychology (e.g., Brazzelli etal., 1994;
Damasio and Anderson, 1993). Milner (1963) reported
a mean loss of only 7.2 IQ points following dorsolateral
how exactly to classify language disturbances associated frontal lobectomies, with mean postoperative IQ scores re-
with brain damage (e.g., Benson and Ardila, 1996). And maining in the average range. This observation meant that
we do not know well enough the exact variants of the either frontal lobes do not have much to do with intelli-
prefrontal syndrome (Damasio and Anderson, 1993). gence or psychometric intelligence tests were not sensitive
Interestingly, some fundamental intellectual factors to frontal lobe deficits ("executive dysfunctions," accord-
can be found throughout different psychometric factor ing to contemporary terminology).
analytic studies. Carroll (1993) analyzed 461 factor-ana- Teuber (1972) carried out a rather extensive research
lytic studies presented in the literature up to date. He ob- project in order to pinpoint the deficits associated with
served that some factors tend to appear with a significant frontal pathology. He compared patients with right frontal,
128 Ardila
left frontal, and bilateral damage. The results demonstrated Full Scale IQ. In the WCST only Persevertive Errors neg-
that in general, patients with frontal lesions performed as atively correlated with Verbal IQ and Full Scale IQ. Only
well as other patients in a variety of intelligence tests. two correlations were found to be significant with regard to
Teuber, however, found deficits in some visuoperceptual the TMT: TMT Form B Errors negatively correlated with
tests, such as visual search tasks. These defects in visual WISC-R Vocabulary subtest; and TMT A Time negatively
search have also been pointed out by different authors correlated with Performance IQ. Results were interpreted
(e.g., Luria, 1980). Milner (1982) also found impaired as supporting the assumption that traditional intelligence
performance in same-different comparisons using clicks, tests are not fully evaluating executive functions.
flashes, and colors. Further, she pointed out that frontal In general, prefrontal lobe activity has been asso-
damage patients have difficulties indicating the recency ciated with self-regulation, control of cognition (meta-
of an item in a series using either figures or words. Signif- cognition), temporal organization of behavior, monitoring
icant difficulties in sequential or temporal memory have of behavior, selective inhibition of responses to immedi-
been observed in this group of patients, and it has even ate stimuli, planning behavior, and control of attention
been proposed that the temporality of behavior represents (Brown, 1985;Damasio and Anderson, 1993;Fuster, 1989;
a core defect in cases of frontal pathology (Fuster, 1989). Hecaen, 1964; Luria, 1966,1969,1973,1980; Perecman,
In traditional psychometric intelligence tests, perfor- 1987;Pribram, 1973;Stuss and Benson, 1983,1986,1987).
mance of frontal damage patients can be normal or near These are the abilities not tapped by psychometric intel-
normal. Black (1976) found a mean WAIS verbal IQ of ligence tests. The term executive function has been pro-
99.1 and a mean performance IQ of 99.5 in a group of posed to refer to the multi-operational system mediated by
44 Vietnam veterans who had sustained unilateral frontal prefrontal areas of the brain and their reciprocal cortical
lobe shrapnel injuries. Janowsky et al. (1989) described and subcortical connecting pathways (Stuss and Benson,
seven patients with various focal frontal lobe lesions who 1986). Executive dysfunction may be summarized in two
obtained a mean WAIS-R Full Scale IQ of 101. Damasio cardinal defects: in controlling behavior and in organizing
and Anderson (1993) analyzed 10 patients with frontal le- cognition.
sions (ventrolaeral and dorsolateral) caused by either vas- Evidently, traditional intelligence tests do not appro-
cular events or surgical resection for treatment of tumors. priately evaluate executive function disturbances. It has
The most notable feature of the WAIS-R testing in these to be concluded that either executive functions should not
patients was the consistent preservation of the cognitive be included as elements of "intelligent behavior" or psy-
abilities required to perform the various intellectual tasks chometric intelligence tests are insufficient in testing for
following frontal lobe damage. intelligence. They are not sensitive to the most impor-
By the same token, in normal subjects, low corre- tant elements of "intelligence": "To act purposefully" (i.e.,
lations between traditional intelligence test scores and controlling and planning behavior) and "to think ratio-
executive function measures have been reported. Welsh nally" (i.e., organizing and directing cognition) according
et al. (1991) observed in children that most of the execu- to Wechsler's (1944) definition of intelligence.
tive function tasks (Visual Search, Verbal Fluency, Motor The conclusion is evident: psychometric intelligence
Planning, Tower of Hanoi, Wisconsin Card Sorting Test tests do not appropriately appraise intelligence. Or at least,
(WCST), and Matching Familial Figures Test) were un- they are not appraising those abilities that from a neuropsy-
correlated with IQ. Visual Search, Verbal Fluency, WCST, chological perspective (and also from the point of view of
and Tower of Hanoi did not correlate with any IQ mea- the Wechsler's intelligence testing) should be understood
sure (Verbal, Quantitative, and Nonverbal) from the Iowa as the most important elements in intelligence.
Test of Basic Abilities. Using a 300-subject college stu-
dent sample, Ardila et al. (1998) observed that Verbal
Fluency tests presented a low but significant correlation Memory and Visuoperceptual Abilities
(about 0.20 to 0.25) with some WAIS verbal subtests, par-
ticularly Digits, Arithmetic, and Information. However, One of most basic functions of the cerebral cortex is
WCST scores did not correlate at all with the Verbal, Per- to encode and store new information (memory). One basic
formance, or Full Scale IQ. Ardila etal. (in press) analyzed area in assessing cognitive activity refers to memory eval-
the correlation between IQ and some executive function uation (Lezak, 1995; Spreen and Strauss, 1991). The WIS
measures (WCST, verbal fluency, and Trial Making Test does not appropriately measure memory. Wechsler him-
(TMT), Form A and Form B). Fifty 13- to 16-year-old self realized the significant shortcoming of his intelligence
normal children were selected. It was found that verbal scale and created a parallel scale directed specifically to
fluency tests correlated about 0.30 with Verbal IQ and measure memory (Wechsler Memory Scale; Wechsler,
1945). However, the WIS has remained within the domain (some 25% of the variance). (4) Like every trait, intelli-
of psychometric intellectual measures, whereas the gence is the joint product of genetic and environmental
Wechsler Memory Scale has been widely used in the neu- factors. (5) School affects intelligence in many different
ropsychology domain. ways: transmitting specific information, developing cer-
It could be argued that some WIS subtests are indeed tain skills and attitudes. Failure to attend school has neg-
evaluating memory: the Information subtest measures re- ative consequences in intelligence testing. (6) Some bi-
mote memory and Digits measures immediate memory. ological conditions have clear negative consequences on
It is easy to agree with this perspective. However, there intelligence. Examples are perinatal complications, expo-
are at least two significant shortcomings in WIS memory sure to environmental lead, and exposure to high blood
appraisal. (1) The memory process is not evaluated (i.e., levels of alcohol. (7) There is a steady rise across time
the ability to encode, store, and retrieve new information). in intelligence test scores known as the "Flynn effect"
This is the most critical type of memory test in neuropsy- (Flynn, 1984,1987). Mean IQ scores have increased more
chology (e.g., Lezak, 1995; Luria, 1976b). (2) At least than 15 points in the last 50 years. Some reasons may be
verbal and nonverbal memory testing should be included improved nutrition, cultural changes, experience with test-
in a cognitive evaluation. Evidently, memory assessment ing, shifts in schooling or child-rearing practices, or some
using the WIS is insufficient, even though WAIS-III has other unknown factors. (8) Ethnic differences in intelli-
attempted, at least partially, to overcome this significant gence reflect complex patterns. No overall generalization
shortcoming. about them is appropriate. (9) Many of the most critical
By the same token, WAIS subtests fail to appropri- questions about intelligence remain unanswered.
ately measure spatial and visuoperceptual abilities. In- Some brief comments may be presented to these
deed, some WIS subtests partially tap into spatial and selected conclusions: (1) Evidently, the concept and in-
visuoperceptual abilities (e.g., Picture completion). But terpretation of intelligence continue to be controversial.
it is difficult to accept that WIS subtests are good enough Neisser et al. (1996) recognized that there are different
to evaluate spatial and visuoperceptual abilities. There are ways to interpret intelligence. No single interpretation of
many better tests in the area (see Lezak, 1995). intelligence testing data is widely accepted, (2) Neisser
etal.(1996) refer to the bidirectional relationship between
school and IQ: intelligence predicts school achievement,
SOME CONTEMPORARY VIEWS ABOUT and school affects intelligence. (3) Many factors may be si-
INTELLIGENCE multaneously acting on the scores obtained in intelligence
tests: genetic factors, some early biological conditions, en-
Even though the issue of intelligence has been a vironmental factors, cultural values, etc. (4) Given the so-
"hot" topic for many years, the main questions remain called flynn effect, several factors may be simultaneously
unsettled. Recently, some integrative papers have been interacting to cause the recent rapid rise in test scores. For
published, attempting to distinguish what is really known a person coming from a nonpsychometrically oriented cul-
and what still remains controversial with regard to intelli- ture (as the author of this paper), however, it is evident that
gence. Neisser et al. 's (1996) paper "Intelligence: Knowns the most crucial factor may be the tremendous training in
and Unknowns" published in American Psychologist rep- testing abilities that Americans have been progressively
resents perhaps the most authoritative report. The paper exposed to. Although exposure to psychometric testing
was prepared by a task force specifically appointed by the has extended to other countries, it is markedly higher in
Board of Scientific Affairs of the American Psychologi- the United States than in most countries. School children
cal Association. Some of the main conclusions presented currently spend a significant amount of time developing
by Neisser and his 10 expert co-authors are (1) There are those strategies required in answering tests, and in prac-
many ways to be intelligent, and there are also many con- ticing tasks similar to those included in intelligence tests.
ceptualizations of intelligence. (2) Psychometry has been This was not observed one generation ago. (5) No gener-
able to measure a wide range of abilities that are distinct alization or general conclusion about ethnic differences in
from one another and yet intercorrelated. It is possible to intelligence is acceptable. There are many ways to be in-
describe the complex relationships between these abilities telligent in different cultural contexts. Good performance
in many different ways. Some authors have searched for a on psychometric intelligence tests is just one way to be
"general intelligence" (g) factor, whereas others have pre- intelligent in a quite specific cultural context.
ferred to refer to a set of independent factors. Still others Reactions to Neisser et al.'s (1996) paper rapidly ap-
have opted for a hierarchy of factors. (3) Intelligence is peared (see American Psychologist, 52(1), 1997). Reac-
correlated with school achievement at a level of about 0.50 tions were so mixed that the only conclusion that can be
130 Ardila
safely drawn is that intelligence continues as a very contro- 2.3. Lexical knowledge (naming, vocabulary, or
versial and, in many regards, a poorly understood topic. other similar tests)
If a conservative paper like that of Neisser el al. could 3. Calculation abilities
trigger so many different and opposite reactions, it must 3.1. Arithmetical operations
be concluded that the concept of intelligence is on very 3.2. Numerical problems
fragile ground.
4. Perceptual abilities
4.1. Visual recognition of figures under different
conditions (e.g., visual detection, to recognize
WHAT SHOULD BE INCLUDED WHEN embedded or unusually presented figures, to
TESTING FOR COGNITIVE ABILITIES? find similarities and differences between fig-
ures, etc.)
Since Thurnstone (1938,1947), there is the converg- 4.2. Recognition of sounds and music (verbal-pho-
ing consensus that some fundamental cognitive abilities nological discrimination; and nonverbal rhy-
may be distinguished. Researchers refer to a limited num- thms, melodies, music, etc.)
ber of domains, usually six to nine, frequently appearing
in factor analytic studies of psychological (Carroll, 1993) 5. Memory and learning
and neuropsychological test batteries (Ardila etal., 1994, 5.1. Verbal learning (Serial Verbal Learning, Cali-
1998; Pont6n et al., 1994). A similar idea is presented by fornia Verbal Learning Test, Rey Auditory
Gardner (1983) when he proposed different types of in- Verbal Learning test, Logical Memory, etc.)
telligence. Evidently, these are the cognitive domains that 5.2. Nonverbal learning (Benton Visual Retention
should be included when testing for intellectual abilities. test, immediate and delayed recall of figures)
There are no fixed tests to evaluated these domains, even 6. Visuoconstructive and visuospatial abilities
though some tests may be better, at least at a certain his- 6.1. Visuoconstructive (such as Rey-Osterrieth
torical moment. In the future, new and better tests can be Complex Figure)
developed to appraise these domains, and these domains 6.2. Tests for spatial abilities (such as line orienta-
may even be restated and rearranged. tion)
In neuropsychology there are several tests that have 7. Motor
become widely accepted and extensively used (see Lezak,
1995; Spreen and Strauss, 1998). They are considered re- 7.1. Fine movements (such as the Finger Tapping
liable, sensitive, and in general "good" tests. There is a Test or other fine movements test)
significant research body supporting their reliability and 7.2. Praxis ability tests
validity. An evaluation of cognitive abilities should in- 8. Executive function abilities
clude these widely accepted tests. As a matter of fact, 8.1. Abstraction (e.g., Similarities)
many of them have been taken from the intelligence test- 8.2. Reasoning (e.g., Raven Progressive Matrixes)
ing research, and in this regard, psychometric intelligence 8.3. Concept formation tests (the Category Test,
testing and neuropsychological testing may be comple- Wisconsin Card Sorting Test, etc.)
mentary rather than mutually exclusive. Of course, it is ex- 8.4. Some tests directed to "maintain instructions"
pected that in the future, superior testing instruments will (Stroop test, Trial Making Test Form B, Luria's
be developed, replacing the current tests that now are con- opposite reactions, etc.)
sidered the best available neuropsychological instruments.
Examples of these cognitive domains, and potentially Of course, these tests are not necessarily evaluating
useful tests are: a single cognitive domain. Attention is required for an
appropriate performance in any intellectual test. Calcula-
1. Attention tion abilities represent a rather complex and multifactorial
ability. Verbal memory depends on language understand-
1.1. Focused attention (e.g., digits backwards)
ing. For example, phonological verbal fluency can be in-
1.2. Sustained attention (e.g., serial subtractions
etc.) terpreted as an executive function test, whereas semantic
verbal fluency is closer to a lexical knowledge test. Fur-
2. Language thermore, all these tests are significantly influenced by
2.1, Verbal fluency (using semantic and phonolog- education, age, and cultural background. Norms for dif-
ical categories) ferent groups should be obtained. Although raw scores
2.2. Language comprehension (token test) can be nonequivalent in different educational, cultural,
and age groups, standard normalized scores are equiva- der) is added. In the Performance Scale Matrix, Reasoning
lent. Each group itself represents its own norm. Tests must (the examinee looks at a matrix from which a section is
be standardized and norms obtained not only for different missing and either identifies the number or points to one
age ranges but also for different educational and cultural of five response options that completes the matrix) and
groups. Otherwise, what is normal for one group might be Symbol Search (the examinee visually scans two groups
interpreted as pathological for another. When a particu- of symbols, a target group composed of two symbols and
lar group outscores another, this simply means that wrong a search group composed by five symbols, and indicates
norms have been used. whether either target symbol matches any of the symbols
in the search group) were added. The Digit-symbol subtest
is used under different conditions: Digit symbol-Coding
THE WAIS-III (as previously used), Digit symbol-Incidental Learning
(paring to recall the symbols matched with the numbers,
In the last 50 years, the same 11 subtests proposed by and free recall to recall the symbols used in coding sec-
Wechsler in the 1930s were repeated over and over again, tion), and Digit symbol-Copy (to copy the symbols that
with just minor changes. Not until 1997 were some fun- were used in the Digit symbol-Coding). These two last
damental changes introduced to the WIS. The WAIS-III conditions are optional and are not used in calculating
(Wechsler, 1997) represents a very significant improve- IQs. Table V presents the cognitive domains and areas
ment over previous versions. But changes are far from included in the WAIS-III subtests.
enough. Several major concerns arise: (1) Testing for cogni-
The WAIS-III includes some new subtests directed tive abilities is still insufficient (see above); (2) It is not
to overcome some of the limitations found in the previous easy to understand why if several factor analyses disclosed
WIS versions: Instead of 11 subtests, there are 14 (seven four different factors (Verbal Comprehension, Perceptual
verbal subtests and seven performance subtests) used to Organization, Working Memory, and Processing Speed),
calculate IQs. In the Verbal Scale, the Letter-Number Se- the WAIS-III insists on calculating only two compound
quencing subtest (combinations of numbers and letters are scores (Verbal and Performance) instead of four. In the
read and the examinee is required to recall the numbers different factor analytic studies that are presented in the
first in ascending order and then letters in alphabetical or- WAIS-III Technical Manual, the Object Assembly subtest
Table V. Cognitive Domains and Areas Included in the WAIS-III Subtests.

Cognitive Domain Area WAIS-OI Subtest Factora
1. Attention Focused attention Digit span Working Memory

Sustained attention Digit-symbol Processing Speed
2. Language Fluency Not included
Language comprehension Not included
Lexical knowledge Vocabulary Verbal Comprehension
3. Calculation abilities Arithmetical operations Not included
Numerical problems Arithmetics Working Memory
4. Perceptual abilities Visual recognition Picture completion Perceptual Organization
Symbol search Processing Speed
Auditory recognition Not included
5. Memory Verbal learning Not included
Nonverbal learning Digit-symboI-IL
6. Visuoconstructive Visuoconstructive Block Design Perceptual Organization
and visuospatial Object Assembly
Spatial abilities Not included
7. Motor Fine movements Not included
Praxis Not included
8. Executive functions Abstraction Similarities Verbal Comprehensionb
Reasoning Matrix reasoning Perceptual Organization
Concept formation Not included
"Maintain instructions" Letter-number Working Memory
a
Factors found in the WAIS-III (Wechsler, 1997).
b
Comprehension is a subtest difficult to interpret and includes two different types of subtests (proverb interpretation and
knowledge of social conventions).
132 Ardila
was not included. No reason for this exclusion is men- anthropology (e.g., Bernatzik, 1957), and contemporary
tioned. (3) Scores are corrected according to age but not humans speak over 4,000 different languages (Swadesh,
according to educational level, although the most impor- 1967). Evidently, an extended analysis of cognitive dis-
tant variable affecting psychological and neuropsycholog- turbances in different cultural and ecological contexts is
ical test performance is education, not age (e.g., Anastasi, necessary for us to understand and serve the neuropsycho-
1988; Cronbach, 1990; Ostrosky et al., 1998). Neverthe- logical needs of our constituency.
less, raw scores should be corrected by both education and Supposedly, comparable fundamental cognitive dis-
age. In the standardization sample used, about two thirds turbances are to be found in every human species mem-
of the subjects had 12 or more years of education. The ber, regardless of cultural background, educational level,
lowest educational group included in the normative sam- language, and ecological demands, as a consequence of
ple had "eight or less years of education," which is most brain lesions. There are some fundamental characteris-
likely an inappropriate education cut-off point (Ostrosky tics in the human brain, and in brain-behavior relation-
et al., 1998). The WAIS-III seems inappropriate to test ships that one would expect to observe in every human
people with low educational levels. (4) The WAIS-III re- subject. Basic cognitive processes are universal, and cul-
tains the total compound score (Full Scale IQ) despite tural differences in cognition reside more in the situations
the fact that the factor analytic studies presented in the to which particular cognitive processes are applied than
WAIS-III Technical Manual show that the four factors in the existence of the process in one cultural group and
obtained in the different factor analyses are rather inde- the absence in another. Culture prescribes what should be
pendent. learned and at what age. Consequently, different cultural
In brief, the WAIS-III is undoubtedly an improve- environments lead to the development of different patterns
ment over previous WIS versions. This is the first time of abilities. Cultural and ecological factors play a role in
that Wechsler's original testing schema has been, at least developing different cognitive styles (Berry, 1971,1979).
partially, abandoned, and new tests are included. After al- Furthermore, cultural variables can eventually influence
most 50 years of extensive use, it was finally accepted the brains' organization of cognition. For example, it has
that the 11 WIS subtests proposed by Wechsler were in- been reported that the degree (not the direction) of brain
sufficient to test cognitive abilities. Evidently, WAIS-III lateralization of language can depend on literacy, and in
represents an implicit recognition that at least executive general, on the verbal training histories (Lecours et al.,
functions and memory had been insufficiently tested in the 1987,1988; Matute, 1988).
WIS previous versions. Nonetheless, it does not mean that Language, memory, visuospatial abilities, praxis
new WAIS-III has overcome the major difficulties pointed skills, and cognitive abilities in general studied in psychol-
out above, and no research studies using the WAIS-III are ogy and neuropsychology are under cultural influence.
yet available. Cognitive disturbances associated with brain pathology
are related to the way those abilities have been trained.
Brain damage results in the disturbance in a specific level
COGNITIVE ABILITIES EN DIFFERENT of information processing, even though the actual manifes-
CULTURAL CONTEXTS tation depends on the specific pattern of cognitive abilities.
This specific pattern of cognitive abilities is culturally de-
Psychometric intelligence tests have been developed pendent. Reading and writing can illustrate the enormous
in a very specific cultural context. They rely on those abil- complexity of brain organization of any cognitive process.
ities that are significant in that particular cultural context Supposedly, reading is based on certain fundamental abil-
and on those approaches that are culturally most valuable. ities (e.g., complex shape perception, cross-modal learn-
In neuropsychology, cognitive disturbances associ- ings, etc.) already existing 5,000 years ago in preliterate
ated with brain pathology of a very limited subsample humans, and of course, existing in illiterate individuals.
of the human speciescontemporary Western, and most The human brain might be specialized not for reading or
often, urban middle-class, and literate brain-damaged writing per se, but for certain basic abilities (information
individualshave been relatively well analyzed. Our un- processing levels or cognition factors) required to read and
derstanding about the brain's organization of cognitive to write, albeit not only to read or to write. It does not seem
abilities, and their disturbances in cases of brain pathol- reasonable to assume that some brain areas are specialized
ogy, is therefore not only partial but, undoubtedly, cul- in reading or writing, just as it does not seem reasonable
turally biased. Cultural and linguistic diversity is an enor- to suppose that some brain areas are specialized in using
mous, but frequently overlooked, moderating variable. computers. Even though the fundamental defect may be
Several thousand different cultures have been described by the same, the actual manifestation of brain pathology may
be different depending on the training histories and the a neuropsychological perspective (and also according to
actual pattern of cognitive abilities. Wechsler himself) could be better interpreted as "intel-
If, despite some existing basic characteristics in its ligence." At the least, they fail in appraising some most
brain organization, oral and written language disturbances fundamental aspects.
are associated with language idiosyncracies (e.g., apha- 2. The concept of IQ might disappear. It is archaic,
sia is not completely equivalent in Chinese and Spanish; and theoretically remains a controversial concept. Subtests
alexia and agraphia can be different in English, Span- used to measure "intelligence" are inappropriate.
ish, and Japanese, etc.; Ardila etal., 1996; Sasanuma and 3. In the future, cognitive evaluation may rely on
Fujimura, 1971; Yamadori, 1975; Yu-Huan etal., 1990), neuropsychological instruments instead of using psycho-
other cognitive abilities such as spatial cognition may also metric intelligence tests. Neuropsychological tests have
depend on the specific ability learning history and the par- a clear and overt rationale from the point of view of the
ticular cognitive style. Supposedly, similar spatial cog- brain organization of cognitive activity. No clear rational
nition disturbances are to be found in a similar way in for the WIS subtests is easily found.
every human species member, regardless of the cultural 4. It would seem more appropriate to use standard
background and the ecological demands. Basic spatial scores (such as T, z, or percentiles) for the individual
cognition abilities, however, are applied in rather differ- tests and cognitive domains, than using global intellec-
ent ways depending on the specific cultural context and tual scores (such as IQ).
the ecological demands. Contemporary city individual's
spatial abilities are not necessarily inferior (or superior)
to the Eskimos' or Amazonian Indians' spatial abilities. ACKNOWLEDGMENTS
Spatial abilities may have evolved with new living and cul-
tural conditions. Spatial cognition abilities can be required My most sincere gratitude to Virginia Standish for her
in many contemporary, and historically recent skills, not most valuable support and help while preparing this paper.
found in every human group. The author of this paper Thanks to Dr. Kevin Keating, Dr. Mariano Alemagny, and
had the opportunity to study a university chemistry pro- Dr. Sarah Ransdell for their observations and suggestions
fessor who suffered a small right-parietal infarction with- on this paper. My gratitude to the anonymous reviewers
out any overt spatial disturbance. Although she had not for their most helpful, encouraging, and interesting obser-
any evident spatial difficulty in her everyday activities, vations.
she could not continue teaching chemistry, because she
was "unable to have a spatial representation of molecules
and all the time got confused." Mathematics (Ardila and REFERENCES
Rosselli, 1990; Luria, 1980), painting, chemistry, chess
(Chabris and Hamilton, 1992), reading and writing (Ardila Anastasi, A. (1988). Psychological Testing (6th ed.), MacMillan, New
and Rosselli, 1993; Benson and Ardila, 1996), mechanics York.
(Benton, 1989), and even music (Henson, 1985) all rep- Ardila, A. (1995a). Directions of research in cross-cultural neuropsy-
chology. Journal of Clinical and Experimental Neuropsychology
resent, at least partially, spatially based skills. They can 17:143-150.
be impaired in cases of right hemisphere damage of those Ardila, A. (1995b). Estructura de la actividad cognoscitiva: Hacia una
same areas that in an Eskimo or Amazonian Indian would teoria neuropsicologica [Structure of cognitive activity: Toward
a neuropsychological theory]. Neuropsychologia Latina 1(2): 21-
imply an impossibility to move around the snow or the 32.
jungle. Ardila, A., Galeano, L. M., and Rosselli, M. (1998). Toward a model of
The analysis of cognitive disturbances associated neuropsychological activity. Neuropsychology Review 8:171-190.
Ardila, A., Pineda, D., and Rosselli, M. (in press). Correlation between
with brain pathology in different cultural contexts, no intelligence test scores and executive function measures. Archives
doubt, represents a challenge for 21st century neuropsy- of Clinical Neuropsychology.
chology. This analysis may eventually allow researchers to Ardila, A., and Rosselli, M. (1990). Acalculias. Behavioural Neurology
3:39-48.
have a better understanding of the basic elements (factors) Ardila, A., and Rosselli, M. (1993). Spatial agraphia. Brain and Cogni-
of cognition. tion 22: 75-95.
Ardila, A., and Rosselli, M. (1994). Development of language, mem-
ory and visuospatial abilities in 5- to 12-year-old children using a
neuropsychological battery. Developmental Neuropsychology 10:
SOME TENTATIVE CONCLUSIONS 97-120.
Ardila, A., Rosselli, M., and Bateman, J. R. (1994). Factorial structure
of cognitive activity. Behavioral Neurology 7:49-58.
1. Psychometric intelligence tests (e.g., WIS in its Ardila, A., Rosselli, M., and Ostrosky, F. (1996). Agraphia in Spanish-
different versions) do not seem to measure what from language. Aphasiology 10: 723-739.
134 Ardila
Benson, D. F. (1982). The alexia: A guide to the neural basis of reading. J. A. M. (ed.), Handbook of Clinical Neurology, Vol. 45: Clinical
In Kirsher, H. S., and Freemon, F. (eds.), The Neurology of Aphasia, Neuropsychology, Elsevier, Amsterdam, pp. 405-422.
Swets, Amsterdam, pp. 139-161. Elwood, R. W. (1991). Factor structure of the Wechsler Memory
Benson, D. F. (1994). The Neurology of Thinking, Oxford University Scale-Revised (WMS-R) in a clinical sample: A methodological
Press, New York. reappraisal. The Clinical Neuropsychologist 5:329-337.
Benson, D. F., and Ardila, A. (1996). Aphasia: A Clinical Perspective, Eysenck, H. (1986). Inspection time and intelligence: A historical
Oxford University Press, New York. introduction. Personality and Individual Differences 7:603-607.
Benton, A. (1989). Constructional apraxia. In Goodglass, H., and Flynn, J. R. (1984). The mean IQ gains of Americans: Massive gains
Damasio, A. R. (eds.). Handbook of Clinical Neuropsychology 1932-1978. Psychological Bulletin 95:29-51.
(Vol. 2), Elsevier, Amsterdam, pp. 287-294. Flynn, J. R. (1987). Massive IQ gains in 14 nations: What IQ tests really
Bernatzik, H. A. (1957). Razas y pueblos del mundo [World races and measure. Psychological Bulletin 101:171-191.
people]. Vols. 1-3. Ediciones Ave, Barcelona. Frearson, W. M., and Eysenck, H. J. (1986). Intelligence, reaction time
Berry, J. W. (1971). Ecological and cultural factors in spatial perceptual [TR] and a new "odd-man-out" RT paradigm. Personality and
development. Canadian Journal of Behavioral Sciences 3: 324- Individual Differences7: 807-817.
336. Fuster, J. M. (1989). The Prefrontal Cortex (2nd ed.), Raven Press, New
Berry, J. W. (1979). Culture and cognition style. In Mrsella, A., Tharp, York.
R. G., and Ciborowski, T. J. (eds.). Perspectives in Cross-Cultural Gardner, H. (1983). Frames of Mind: The Theory of Multiple Intelli-
Psychology, Academic Press, New York, pp. 117-135. gences, Basic Books, New York.
Berry, J. W., Poortinga, Y. H., Segall, M. H., and Dasen, P. R. Greenfield, P. M. (1997). You can't take it with you: Why ability as-
(1992). Cross-Cultural Psychology, Cambridge University Press, sessments don't cross cultures. American Psychologist 52: 1115-
Cambridge. 1124.
Binet, A. (1905). Analysis of C.E. Spearman: The proof and measure- Guilford, J. P. (1967). The Nature of Human Intelligence, McGraw-Hill,
ment of association between two things and general intelligence New York.
objectively determined and measured. L'Anne Psychologique 11: Guilford, J. P. (1968). Intelligence has three facets. Science 160:
623-624. 615-620.
Binet, A. (1908). Le developpement de 1'intelligence chez les en- Guilford, J. P., and Hoepfner, R. (1971). The Analysis of Intelligence.
fants. [The development of intelligence in children.] L'Anne McGraw-Hill, New York.
Psychologique 14:1-94. Hebb, D. (1939). Intelligence in man after large removals of cerebral
Black, F. W. (1976). Cognitive deficits in patients with unilateral tissue: Report of four left frontal lobe cases. Journal of General
war-related frontal lobe lesions. Journal of Clinical Psychology Psychology 21:73-87
32: 366-372. Hebb, D. O. (1942). The effects of early and late brain injury upon test
Bornstein, R. A., and Chelune, G. J. (1988). Factor structure of the scores and the nature of normal adult intelligence. Proceedings of
Wechsler Memory Scale-Revised. The Clinical Neuropsychologist the American Phylosophical Society 85:275-292.
2:107-115. Hebb, D. O. (1945). Man's frontal lobe: A critical review. Archives of
Brazzelli, M., Colombo, N., Delia Salla, S., and Spinnler, H. (1994). Neurology and Psychiatry 54:421-438.
Spared and impaired abilities after bilateral frontal damage. Cortex Hebb, D., and Penfield, W. (1940). Human behavior after extensive
30: 27-51. bilateral removal from the frontal lobes. Archives of Neurology
Brody, N. (1992). Intelligence (2nd ed.). Academic Press, New York. and Psychiatry 44: 421-438.
Brown, J. E. (1985). Frontal lobe syndromes. In Frederiks J. A. Hecaen, H. (1962). Clinical symptomatology in right and left hemi-
M. (ed.), Handbook of Clinical Neurology, Vol. 45: Clinical sphere lesions. In Mountcastle, V. B. (ed.), Interhemispheric
Neuropsychology, Elsevier, Amsterdam, pp. 32-41. Relations and Cerebral Dominance, John Hopkins University,
Carroll, J. B. (1993). Human Cognitive Abilities: A Survey of Factor Baltimore, pp. 215-243.
Analytic Studies, Cambridge University Press, Cambridge. Hecaen, H. (1964). Mental symptoms associated with tumors of the
Cattell, R. B. (1971). Abilities: Their Structure, Growth and Action, frontal lobe. In Warren, J. M., and Akert, K. (eds.), The Frontal
Houghton-Mifflin, Boston. Granular Cortex and Behavior, McGraw-Hill, New York, pp.
Ceci, S. J. (1990). On Intelligence... More or Less: A Bioecological 335-352.
Treatise on Intellectual Development, Prentice Hall, Englewood, Hecaen, H., and Albert, M. L. (1978). Human Neuropsychology, Wiley,
NJ. New York.
Ceci, S. J. (1991). How much does schooling influence general intelli- Henson, R. A. (1985). Amusia. In Frederiks, J. A. M. (ed.), Handbook
gence and its cognitive components? A reassessment of evidence. of Clinical Neurology, Vol. 45: Clinical Neuropsychology, Elsevier,
Developmental Psychology 27:703-722. Amsterdam, pp. 483-490.
Ceci, S. J., and Williams, W. M. (1997). Schooling, intelligence and Hunter, J. (1986). Cognitive ability, cognitive aptitudes, job knowledge,
income. American Psychologist 52:1051-1058. and job performance. Journal of Vocational Behavior 29: 340-
Chabris, C. F., and Hamilton, S. E. (1992). Hemispheric special- 363.
ization for skilled perceptual organization by chessmasters. Irvine, S. H., and Berry, J. W. (eds.), (1988). Human Ability in Cultural
Neuropsychologia 30:47-57. Context, Cambridge University Press, Cambridge.
Cronbach, L. J. (1990). Essentials of Psychological Testing (5th ed.), Janowsky, J. S, Shimamura, A. P., Kritchevsky, M., and Squire, L. R.
Harper and Row, New York. (1989). Cognitive impairments following frontal lobe damage and
Damasio, A. R., and Anderson, S.W. (1993). The frontal lobes. In Heil- its relevance to human amnesia. Behavioral Neurosciences 103:
man, K. M., and Valenstein, E. (eds.), Clinical Neuropsychology, 548-560.
Oxford University Press, New York, pp. 409-460. 3rd edition. Jensen, A. R. (1980). Bias in mental testing. Free Press, New York.
Damasio, H., and Damasio, A. R. (1989). Lesion Analysis in Neuropsy- Jensen, A. R. (1987). Individual differences and the Hick paradigm. In
chology, Oxford University Press, New York. P. A. Vernon (ed.), Speed of Information Processing and Intelli-
Das, J. P., Naglieri, J. A., and Kirby, J. R. (1994). Assessment of gence, Ablex, Norwood, pp. 101-175.
Cognitive Processes: The PASS Theory of Intelligence, Allyn & Kaiser, H. F. (1960). The varimax solution for Primary Mental Abilities.
Bacon, Needham Heights, MA. Psychometrika 25:153-158.
De Renzi, E. (1982). Disorders of Space Exploration and Cognition, Kaplan, E., Fein, D., Morris, R., and Delis, D. (1991). WAIS-R as a
Wiley, New York. Neuropsychological Instrument, Psychological Corporation, San
De Renzi, E. (1985). Disorder of space exploration. In Frederiks, Antonio. TX.
Kaufman, A. S., and Kaufman, N. L. (1993). Manual for the Kaufman Nettlebeck, T. (1987). Inspection time and intelligence. In Vernon,
Adolescent and Adult Intelligence Test (KAIT), American Guidance P. A. (ed.), Speed of Information Processing and Intelligence,
Service, Circle Pines, MN. Ablex, Norwood, NJ, pp. 295-346.
Kaufman, A. S., and Kaufman, N. L. (1997). The Kaufman Adolescent Newcombe, P., and Ratcliff, G. (1989). Disorders of visuospatial anal-
and Adult Intelligence Test. In Flanagan, 0. P., Genshaft, J. D., ysis. In Goodglass, H., and Damasio, A. R. (eds.), Handbook of
and Harrison, P. L. (eds.), Contemporary Intellectual Assessment: Clinical Neuropsychology (Vol. 2), Elsevier, Amsterdam, pp. 333-
Theories, Tests and Issues, Guilford Press, New York, pp. 209-229. 356.
Kertesz, A. (1983). Localization in Neuropsychology, Academic Press, Ostrosky, F, Ardila, A., and Rosselli, M. (1998). Neuropsychological
New York. test performance in illiterates. Archives of Clinical Neuropsychol-
Lecours, A. R., Mehler, J., Parente, M. A., Beltrami, M. C, Canossa ogy 13:645-660.
de Tolipan, L., Castro M. J., Carrono, V., Chagastelles, L., Dehaut, Ostrosky, F., Ardila, A., and Rosselli, M. (1999). NEUROPSI: A brief
P., Delgado, R., Evangelista, A., Fajgenbaum, S., Fontoura, C., neuropsychological test battery in Spanish with norms by age and
de Fraga Karmann, D., Gurd, J., Hierro Tome, C., Jakubovicz, R., educational level. Journal of the International Neuropsychological
Kac, R., Lefevre, B., Lima, C., Maciel, J., Mansur, L., Martinez, Society 5:413-433.
R., Nobrega, M. C., Osorio, Z., Paciomik, J., Papaterra, P., Perecman, E. (1987). Consciousness and the meta-functions of the
Jourdan Penedo, M. A., Saboya, B., Scbeuer, C., Batista da Silva, frontal lobes. In Perecman, E. (ed.). The Frontal Lobes Revisited,
A., Spinardi, M., and Texeira, M. (1988). Illiteracy and brain IRBN Press, New York, pp. 1-10.
damage. III: A contribution to the study of speech and language Pettersen, S. E., Fox, P. T, Snyder, A. Z., and Raichle, M. E. (1989).
disorders in illiterates with unilateral brain damage (initial testing). Activation of extrastriate and frontal cortical areas by visual words
Neuropsychohgia 26:575-589. and word-like stimuli. Science 249:1041-1044.
Lecours, R. L., Mehler, J., Parente, M. A., Caldeira, A., Cary, L., Pirozzolo, F. J. (1985). Mental retardation. In Frederiks, J. A. M.
Castro, M. J., Dehaout, P., Delgado, R., Gurd, J., Karmann, D., (ed.) Handbook of Clinical Neurology, Vol. 46: Neurobehavioral
Jakubovitz, R., Osorio, Z., Cabral, L. S,, and Junqueira, M. (1987). Disorders, Elsevier, Amsterdam, pp. 1-40.
Illiteracy and brain damage, I: Aphasia testing in culturally con- Ponton, M. O., Satz, P., and Herrera, L. (1994). Factor analysis of a
trasted populations (control subjects). Neuropsychohgia 25: 231- neuropsychological screening battery for Hispanics (NeSBHIS).
245. XVII European Meeting of the International Neuropsychological
Lezak, M. D. (1995). Neuropsychological Assessment (2nd ed.), Oxford Society, June.
University Press, New York. Posner, M. I., Petersen, S. E., Fox, P. T., and Raichle, M. E. (1988).
Luria, A. R. (1966). Human Brain and Psychological Processes, Harper Localization of cognitive operations in the human brain. Science
and Row, New York. 240:1627-1631.
Luria, A. R. (1969). Frontal lobe syndromes. In Vinken, P. J., and Pribram, K. H. (1973). The primate frontal cortexExecutive of the
Bruyn, G. W. (eds.), Handbook of Clinical Neurology, Vol. 2, brain. In Pribram, K. H., and Luria, A. R. (eds). Psychophysiology
North Holland, Amsterdam, pp. 725-757. of the Frontal Lobes, Academic, New York, pp. 293-314.
Luria, A. R. (1973). The frontal lobes and the regulation of behavior. Price, C. J., Wise, R. J. S., Watson. J. D. G., Patterson, K., Howard, D.,
In Pribram, K. H., and Luria, A. R. (eds.), Psychophysiology of the and Frackowiak, R. S. J. (1994). Brain activation during reading:
Frontal Lobes, Academic Press, New York, pp. 3-26. The effects of exposure duration and task. Brain 117:1255-1269.
Luria, A. R. (1976a). Basic Problems ofNeurolinguistics, Mouton, The Reed, T. E., and Jensen, A. R. (1992). Conduction velocity in a brain
Hague, Netherlands. nerve pathway of normal adults correlates with intelligence level.
Luria, A. R. (1976b). The Neuropsychology of Memory, Halstead, Intelligence 16: 259-272.
Somerset, NJ. Roid, G. H., Prifitera, A., and Ledbetter, M. (1988). Confirmatory analy-
Luria, A. R. (1980). Higher Cortical Functions in Man (2nd ed.), Basic, sis of the factor structure of the Wechsler Memory Scale-Revised.
New York. Clinical Neuropsychology 2: 116-120.
Matarazzo, J. D. (1972). Wechsler's Measurement and Appraisal of Rosselli, M., and Ardila, A. (1989). Calculation deficits in patients with
Adult Intelligence, Oxford University Press, New York. right and left hemisphere damage. Neuropsychohgia 27: 607-
Matute, E. (1988). El aprendizaje de la lectoescritura y laespecializacion 617.
hemisferica para el lenguaje [Reading and writing learning, and Sasanuma, S., and Fujimura, O. (1971). Kanji versus Kana processing
hemispheric specialization for language]. In Ardila, A., and in alexia with transient agraphia: A case report. Cortex 7:1-18.
Ostrosky-Solis, F. (eds.), Lenguaje oral y escrito [Oral and written Spearman, C. (1904). "General intelligence" objectively determined
language]. Mexico D. F., Mexico: Editorial Trillas, pp. 310-358. and measured. American Journal of Psychology 15:201-293.
Milner, B. (1963). Effects of different brain lesions on card sorting. Spearman, C. (1923). The Nature of "Intelligence" and the Principles
Archives of Neurology 9:90-100. of Cognition, MacMillan, London.
Milner, B. (1982). Some cognitive effects of frontal-lobe lesions in Spreen, O., and Strauss, E. (1991). A Compendium of Neuropsycholog-
man. Philosophical Transactions of the Royal Society of London ical Tests, Oxford University Press, New York.
298:211-226. Spreen, O., and Strauss, E. (1998). A Compendium of Neuropsycho-
Morrow, L., and Ratcliff, G. (1988). The neuropsychology of spatial logical Tests: Administration, Norms and Commentary (2nd ed.),
cognition. In Stiles-Davis, J., Kritchevsky, M., and Bellugi, U. Oxford University Press, New York.
(eds.), Spatial Cognition: Brain Bases and Development, Erlbaum, Stern, W. (1912). Die psychologischen Methoden der Intelligenz pru-
Hillsdale, NI, pp. 5-32. fung. [The Psychological Method to Measure Intelligence.] Leipzig:
Naglieri, J. A. (1997). Planning, attention, simultaneous and succesive Barth.
theory and the cognitive assessment system: A new theory-based Sternberg, R. J. (1985). Beyond IQ: A Triarchic Theory of Human
measure of intelligence. In Flanagan, D. P., Genshaft, J. D., and Intelligence, Cambridge University Press, New York.
Harrison, P. L. (eds.), Contemporary Intellectual Assessment: Sternberg, R. J. (1988). A triarchic view of intelligence, In Irvine,
Theories, Tests and Issues, Guilford Press, New York, pp. 247-267. S. H., and Berry, J. W. (eds.), Human Ability in Cultural Context,
Naglieri, J. A., and Das, J. P. (1996). Das Naglieri Cognitive Assessment Cambridge University Press, New York, pp. 60-85.
System, Riverside, Chicago. Sternberg, R. J. (1997). The concept of intelligence and its role in life-
Neisser, U., Boodoo, G., Bouchard, T. J., Boykin, A. W., Brody, N., long learning and success. American Psychologist 52:1030-1037.
Ceci, S. J., Halpern, D. R, Loehlin, J. C, Perloff, R.. Stenberg, Stuss, K. H., and Benson, D. F. (1983). Frontal lobe lesions and
R. J., and Urbina, S. (1996). Intelligence: Knowns and unknowns. behavior. In Kertesz, A. (ed.), Localization in Neuropsychology,
American Psychologist 51:77-101. Academic Press, New York, pp. 429-454.
136 Ardila
Stuss, D. T, and Benson, D. F. (1986). The Frontal Lobes, Raven Press, Vernon, P. E. (1950). The Structure of Human Abilities, Wiley, New York.
New York. Wechsler, D. (1944), The Measurement of Adult Intelligence. Williams
Stuss, D. T., and Benson, D. P. (1987). The frontal lobes and control of and Wilkins, Baltimore.
cognition and memory. In Perecman, E. (ed.), The Frontal Lobes Wechsler, D. (1945). A standardized memory scale for clinical use.
Revisited, IRBN Press, New York, pp. 141-158. Journal of Psychology 19: 87-95.
Swadesh, M. (1967). El Lenguaje y la Vtda Humana [Language and Wechsler, D. (1987). Manual for the Wechsler Memory Scale-Revised,
Human Life], Fondo de Cultural Economica, Mexico. Psychological Corporation, San Antonio, TX.
Terman, L. M. (1916). The Measure of Intelligence, Houghton-Mifflin, Wechsler, D. (1997). WAIS-III: Administration and Scoring Manual,
Boston. Psychological Corporation, San Antonio, TX.
Teuber, H. L. (1972). Unity and diversity of frontal lobe function. Acta Welsh, M. C., Pennington, B. F., and Groisser, D. B. (1991). A normative-
Neurobiology Experimental 32:615-656. developmental study of executive function: A window on prefrontal
Thorndike, R. L., Hagen, E. P., and Sattler, J. M. (1986). Stanford- function in children. Developmental Neuropsychology 7: 131-
Binet Intelligence Scale: Guide for Administration and Scoring, 149.
Riverside, Chicago. Yamadori, A. (1975). Ideogram reading in alexia. Brain 98: 231-
Thurnstone, L. L. (1938). Primary Mental Abilities, University of 238.
Chicago Press, Chicago. Yu-Huan, H., Ying-Guan, Q., and Gui-Qing, Z. (1990). Crossed aphasia
Thurnstone, L. L. (1947). Multiple Factor Analysis, University of in Chinese: A clinical survey. Brain and Language 39: 347-
Chicago Press, Chicago. 356.
Brain and Cognition 68 (2008) 9299
Contents lists available at ScienceDirect
Brain and Cognition

journal homepage: www.elsevier.com/locate/b&c
On the evolutionary origins of executive functions

Alfredo Ardila *
Department of Communication Sciences and Disorders, Florida International University, 10900 SW 13 Street, HLS 139, Miami, FL 33199, USA
a r t i c l e i n f o a b s t r a c t
Article history: In this paper it is proposed that the prefrontal lobe participates in two closely related but different exec-
Accepted 3 March 2008 utive function abilities: (1) metacognitive executive functions: problem solving, planning, concept for-
Available online 7 April 2008 mation, strategy development and implementation, controlling attention, working memory, and the like;
that is, executive functions as they are usually understood in contemporary neuroscience; and (2) emo-
Keywords: tional/motivational executive functions: coordinating cognition and emotion/motivation (that is, fulfill-
Executive functions ing biological needs according to some existing conditions). The first one depends on the dorsolateral
Metacognition
prefrontal areas, whereas the second one is associated with orbitofrontal and medial frontal areas. Cur-
Language evolution
rent tests of executive functions basically tap the first ability (metacognitive). Solving everyday problems
(functional application of executive functions), however, mostly requires the second ability (emotional/
motivational); therefore, these tests have limited ecological validity. Contrary to the traditional points of
view, recent evidence suggests that the human prefrontal lobe is similar to other primates and hominids.
Other primates and hominids may possess the second (emotional executive functions) prefrontal ability,
-but not the first (metacognitive executive functions) one. It is argued that metacognitive executive func-
tions are significantly dependent on culture and cultural instruments. They probably are the result of the
development and evolution of some conceptualization instruments; language (and written language as
an extension of oral language) may represent the most important one. The second executive function
ability (emotional/motivational) probably is the result of a biological evolution shared by other primates.
! 2008 Elsevier Inc. All rights reserved.
goal-directed behaviors, and anticipate the consequences of ones

Nothing in biology makes sense except in the light of evolution
actions (Ardila & Surloff, 2007; Denckla, 1994, 1996; Goldberg,
(Theodosius Dobzhansky, 1973)
2001; Luria, 1969, 1980; Stuss & Benson, 1986; Stuss & Knight,
2002). The concept of morality, ethical behaviors, self-awareness,
1. What does executive functions mean? and the idea of the frontal lobes as manager and programmer of
the human psyche are also included (Anderson, Bechara, Damasio,
Executive functions is a relatively new term in the neurosci- Tranel, & Damasio, 1999; Damasio, 1994; Luria, 1980; Moll, Zahn,
ences. Luria (1966, 1973, 1980) is the direct antecessor of the conde Oliveira-Souza, Krueger, & Grafman, 2005).
cept of executive functions. He distinguished three functional units During the late 19th and early 20th centuries, clinical investiga-
in the brain: (1) arousal-motivation (limbic and reticular systems); tions documented diverse behavioral disorders in cases of frontal
(2) receiving, processing, and storing information (post-rolandic pathology. Frontal lobe syndrome was conceptualized by Feu-
cortical areas); and (3) programming, controlling, and verifying chtwanger (1923). He correlated frontal pathology to behaviors
activity, depending on the activity of the prefrontal cortex. Luria that were not related to overt speech, memory, or sensorimotor
mentions that this third unit has an executive role. Lezak (1983) deficits. He emphasized the personality changes in motivation,
referred to executive functioning to discriminate cognitive func- affective dysregulation, and the capacity to regulate and integrate
tions from the how or whether of human behaviors. Baddeley other behaviors. Goldstein (1944) expanded the capacity of frontal
(1986) grouped these behaviors into cognitive domains that in- lobe behaviors to include the abstract attitude, initiation, and
cluded problems in planning, organizing behaviors, disinhibition, mental flexibility. Luria (1966, 1969) related prefrontal lobe activ-
perseveration, reduced fluency, and initiation. Baddeley also ity with programming motor behavior, inhibiting immediate re-
coined the term dysexecutive syndrome. sponses, abstracting, problem solving, verbal regulation of
The definition of executive function includes the concept of behavior, reorienting behavior according to behavioral conse-
mental flexibility and also ability to filter interference, engage in quences, temporal integration of behavior, personality integrity,
and consciousness. During the 1970s, 1980s, and 1990s several
* Fax: +1 305 348 2750.
books exclusively devoted to the analysis of the prefrontal cortex
E-mail address: ardilaa@fiu.edu were published (e.g., Fuster, 1989; Levin, Eisenberg, & Benton,
0278-2626/$ - see front matter ! 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.bandc.2008.03.003
A. Ardila / Brain and Cognition 68 (2008) 9299 93
1991; Miller & Cummings, 1998; Perecman, 1987; Pribram & Luria, sion of the limbic system. It seems that no laboratory test for exec-
1973; Roberts, Robbins, & Weiskrantz, 1998; Stuss & Benson, utive function taps into the ability to coordinate cognition and
1986). These works usually assumed that frontal (prefrontal) emotion, and in that regard, no executive function test has signif-
syndrome was synonymous with executive dysfunction. icant ecological validity.
Progressively, it became apparent that prefrontal syndrome By coordinating cognition and emotion, the prefrontal lobe
and executive dysfunction are not synonymous. The prefrontal plays a major function: controlling the limbic system impulses;
cortex plays a key monitoring role in executive functions, but other that is, making limbic impulses socially acceptable (e.g., Beer,
brain areas are also involved (Elliott, 2003). Intact frontal pro- John, Scabini, & Knight, 2006; Blair, 2004; Lezak, Howieson, Loring,
cesses, although not synonymous with intact executive function- & Hannay, 2004). The inability to make basic biological needs so-
ing, are and integral part of it. Attempts to localize executive cially acceptable, as clearly described in Phineas Gages case, fre-
functioning to discrete frontal areas have been inconclusive. The quently represents a major disturbance in prefrontal patients. Of
emerging view is that executive function is mediated by dynamic course, we all would like to hit somebody when frustrated, to take
and flexible networks. Neuroimaging results have also implicated possession of anything we want, to stay at home instead of per-
posterior, cortical, and subcortical regions in executive functioning forming fatiguing work, and to approach any potential sexual part-
(Roberts, Robbins, & Weiskrantz, 2002). ner. That is exactly what many patients with prefrontal lobe
Historically, Phineas Gage has become the most classical exam- pathology frequently do.
ple for prefrontal lobe pathology, and disturbances in executive
functions (Harlow, 1868). Phineas Gage was a reliable foreman
2. Is there any fundamental core ability accounting for
for a railroad company who suffered a tragic accident in which a
executive functions?
tampering rod was projected through his frontal lobes. Miracu-
lously he survived, but after this accident, he was described as
Some disagreement exists around the question of unity or diver-
profane, irascible, and irresponsible. Profound personality
sity (non-unitary) of executive functions (e.g., De Frias, Dixon, &
changes were reported, and according to Harlow, he began to be-
Strauss, 2006; Duncan, Emslie, Williams, Johnson, & Freer, 1996;
have as an animal. The Phineas Gage case is usually cited as a typ-
Grafman, 2006; Kimberg, DEsposito, & Farah, 1997; Parkin & Java,
ical example of executive function disturbances. It is obvious
1999). It is not clear what the particular unitary factor saturating
however, that Phineas Gages impairments were mostly situated
the different executive function tests is. Behavior inhibition has
at an emotional/motivational level, not at a purely cognitive (or
been considered as a potential candidate, as the single factor respon-
metacognitive) level. Overt behavioral changes were observed
sible for successful performance in different executive tests (Bark-
as frequently found in frontal lobe pathology, but purely cognitive
ley, 1997) or in combination with working memory (Pennington &
impairments were ill-documented, partially due to the lack of
Ozonoff, 1996). Salthouse (1996, 2005) suggested that reasoning
appropriate cognitive assessment instruments.
and perceptual speed represent the underlying factors related to
Most frequently, executive functions are analyzed in experi-
all executive functions. Salthouse (2005) observed that performance
mental conditions using diverse research strategies, such as solv-
on two common tests of executive functioning, the Wisconsin Card
ing diverse problems, finding similarities between two words,
Sorting Test (Berg, 1948; Heaton, 1981) and the Controlled Oral
providing an answer that requires inhibiting another, etc. A para-
Word Association Test (Benton, Hamsher, & Sivan, 1994), were
digm is created and the subject is required to solve it. The brain
strongly correlated with reasoning ability and perceptual speed.
activity can be simultaneously recorded, using brain electrical
Other authors challenge the existence of such a unitary factor.
activity or recording the regional level of activation (e.g., Osaka
Godefroy, Cabaret, Petit-Chenal, Pruvo, and Rousseaux (1999)
et al., 2004). Alternatively, executive functions are analyzed in
emphasized that certain frontal lobe patients perform well on some
brain-damaged populations in order to find the contribution of
tests purported to assess executive abilities but not on others. Cor-
different brain systems (e.g., Jacobs, Harvey, & Anderson, 2007).
relations among different executive tests are frequently moderate
This last approach represents the classical neuropsychological
or low, and many times lacking statistical significance (Friedman
method. Executive functions, however, rarely are analyzed in nat-
et al., 2006; Lehto, 1996; Salthouse, Atkinson, & Berish, 2003).
ural ecological conditions. How, indeed, do people solve everyday
Miyake, Friedman, Emerson, Witzki, and Howerter (2000)
problems? This is obviously a crucial question in understanding
adopted an intermediate position. They studied three often-postu-
human behavior.
lated aspects of executive functions (shifting, updating, and inhibi-
Tests for executive function typically represent external tasks,
tion) and concluded that, although they are clearly distinguishable,
requiring the correct application of some intellectual abilities to
they do share some underlying commonality. Based on the results
be solved; for example, the Wisconsin Card Sorting Test (Berg,
of their study, the authors stated that executive functions are sep-
1948; Heaton, 1981), the Tower of Hanoi (Simon, 1975), or the
arable but moderately correlated constructs thus suggesting both
Stroop test (Stroop, 1935) represent unusual and unfamiliar tasks,
unitary and non-unitary components of the executive system. By
requiring the development of new strategies, planning, thought
the same token, several authors have suggested different subcom-
flexibility, etc. Nonetheless, they are emotionally neutral tasks.
ponents of executive functions (e.g., Anderson, 2001; Delis, Kaplan,
Although executive functions depend on extended networks
& Kramer, 2001; Denckla, 1994; Elliott, 2003; Hobson & Leeds,
including different brain areas, it is assumed that the prefrontal
2001; Lafleche & Albert, 1995; Lezak, 1983; Piguet et al., 2002).
cortex plays a major controlling and monitoring role. Most impor-
tant, prefrontal cortex does not only participate in those classically
recognized executive operations (sequencing, alternating, inhibit- 3. Two proposed fundamental executive functions
ing, etc.), but also plays a core role in coordinating cognition and
emotion (Mitchell & Phillips, 2007). Most of the disturbances re- It may be conjectured that there are two different, but closely
ported in Phineas Gage (and in many cases of prefrontal syn- related types of prefrontal lobe abilities (e.g., Fuster, 2001, 2002;
dromes) refer to behavioral/emotional disturbances; or more Happaney, Zelazo, & Stuss, 2004):
exactly, disturbances in coordinating cognition and emotion/moti-
vation. The prefrontal lobe has extensive connections to subcorti- (1) Metacognitive executive functions which include problem
cal and limbic system areas (Barbas, 2006; Damasio & Anderson, solving, abstracting, planning, strategy development and
1993) and even its orbital portion could be regarded as an exten- implementation, and working memory (the usual under-
94 A. Ardila / Brain and Cognition 68 (2008) 9299
standing of executive functions, generally measured in neu- tions of negative mood appear to have little effect on cognitive
ropsychology executive functions tests); these are abilities control processes, whereas positive mood impairs aspects of
mostly related with the dorsolateral area of the prefrontal updating, planning, and switching. This effect of emotion on cogni-
cortex (e.g., Stuss & Knight, 2002).The dorsolateral prefrontal tion has been supported for different executive function tasks, such
cortex has been observed to participate in diverse planning, as working memory (Spies, Hesse, & Hummitzsch, 1996) and plan-
abstracting, problem solving, and working memory tasks. ning (Oaksford, Morris, Grainger, & Williams, 1996)using the
Using fMRI dorsolateral prefrontal activation has been found Tower of London (Shallice, 1982) as a paradigm. In other words,
in tasks such as solving the Tower of Hanoi (Fincham, Carter, when emotion is involved, metacognitive executive function abil-
van Veen, Stenger, & Anderson, 2002), Controlled Word ity decreases. Mitchell and Phillips (2007) emphasize that current
Association Test (letter fluency) (Baldo, Schwartz, Wilkins, evidence on the effects of mood on regional brain activity during
& Dronkers, 2006), working memory (Yoon, Hoffman, & executive functions demonstrates that the prefrontal cortex repre-
DEsposito, 2007), and solving the Wisconsin Card Sorting sents an area of integration between mood and cognition.
Test (Lie, Specht, Marshall, & Fink, 2006). These two types of executive functions (metacognitive and
(2) Emotional/motivational executive functions, which is emotional/motivational) depend on relatively different prefron-
responsible for coordinating cognition and emotion. That tal areas, and as a matter of fact, two major variants in the prefron-
means, the ability to fulfill basic impulses following socially tal syndrome are frequently distinguished, one mostly impairing
acceptable strategies. In the last case, what is most impor- cognition (or rather, cognitive control, that is, metacognition);
tant does not necessarily include what the best conceptual the other one mostly impairing behavior:
and intellectual result is, but what is in accordance with per-
sonal impulses (e.g., Bechara, Damasio, & Damasio, 2000). In (1) Dorsolateral syndrome. Cummings (1993) indicated that the
that regard, the core function of the prefrontal lobe is to find dorsolateral circuit is the most important to executive func-
acceptable justifications for limbic impulses. Following tioning. The most noted deficit is an inability to organize a
socially acceptable strategies actually involves inhibition of behavioral response to novel or complex stimuli. Symptoms
selfish or unsociable basic impulses in the first place, but are on a continuum and reflect capacity to shift cognitive
not necessarily arriving at the best conceptual solution. sets, engage existing strategies, and organize information
The ventromedial areas of the prefrontal cortex are involved to meet changing environmental demands. Various
in the expression and control of emotional and instinctual researchers, including Luria (1969), have noted persevera-
behaviors (Fuster, 1997a, 1997b, 2002). This function is tion, stimulus-bound behavior, echopraxia, and echolalia.
related with so-called inhibitory control of behavior Lateralization has been noted in executive dysfunction
(Miller & Wang, 2006). Clinical evidence (e.g., Luria, 1969; (Goldberg, 2001). Ventral and dorsal portions of prefrontal
Stuss & Knight, 2002) as well as experimental research cortex are bilieved to interact in the maintenance of rational
(e.g., Leung & Cai, 2007; Medalla, Lera, Feinberg, & Barbas, and non-risky decision making (Manes et al., 2002).
2007) suggest that the neural substrate for this inhibitory According to Fuster (1997a, 1997b, 2002), the most general
function resides mainly in the medial and orbital portions executive function of the lateral prefrontal cortex is the tem-
of the prefrontal cortex. Fuster (2002) points out that The poral organization of goal-directed actions in the domains of
apparent physiological objective of inhibitory influences behavior, cognition, and language.
from orbitomedial cortex is the suppression of internal and (2) Orbitofrontal and medial frontal syndrome. Orbitofrontal dam-
external inputs that can interfere with whatever structure age has been associated with desinhibition, inappropriate
of behavior, speech, or cognition is about to be undertaken behaviors, personality changes, irritability, mood liability,
or currently underway (page 382). tactlessness, distractibility, and disregard of important events
(Stuss & Knight, 2002). These patients are unable to respond to
There is no question that if metacognitive executive functions social cues. Noteworthy, it was observed by Laiacona and col-
were indeed used in solving external problems without the involve- leagues (1989) that these patients have no difficulty with card
ment of limbic impulses, most world-wide problems would have sorting tasks. Eslinger and Damasio (1985) coined the term
been solved by man, because contemporary man has sufficient reacquired sociopathy to describe dysregulation that couples
sources to solve all the major social problems (such as poverty and both a lack of insight and remorse regarding these behaviors.
war). Human conflicts in general would also be significantly reduced. The orbitofrontal cortex appears to be linked predominantly
Direct observation suggests that everyday problems usually with limbic and basal forebrain sites. Medial frontal lobe dam-
have an emotional content: talking with a friend, a boss or a spouse; age causes apathy or abulia (a severe form of apathy). Acute
driving in the street; deciding how to approach somebody; spend- bilateral lesions in the medial frontal area can cause akinetic
ing money, etc., are not emotionally neutral activities, as are the mutism, in which the individual is awake and has self-aware-
Wisconsin Card Sorting Test (Berg, 1948; Heaton, 1981) and the ness, but does not initiate behaviors (Ross & Stewart, 1981).
Tower of Hanoi (Simon, 1975). When other people are involved, it According to Fuster (1997a, 1997b, 2002) the ventromedial
is not easy to remain emotionally neutral. Social issues, simply areas of the prefrontal cortex are involved in expression and
speaking, are not emotionally neutral, because power/submission, control of emotional and instinctual behaviors.
personal benefits, and diverse biological drives are potentially in- It is evident that the two prefrontal syndromes can have
volved (Smith, Bond, & Kagitcibas, 2006). Most likely, throughout rather different clinical expressions (metacognitive and emo-
evolution of mankind (i.e., during the last 150,000 years) fulfilling tional/motivational) depending upon the specific location of
these drives has been the major application of the prefrontal func- the damage.
tions: to get power, to have a dominant role, to take food and goods
for ourselves, to get a sexual partner, etc. That means that emo-
tional/motivational executive functions have played a crucial role 4. Metacognitive executive functions as an internalization of
in survival and reproduction (e.g., facilitating behaviors such as actions
acquiring dominant roles, obtaining sexual partners, etc.).
Mitchell and Phillips (2007) have argued that emotion can af- As pointed above, disagreement persists around the potential
fect executive function ability. They propose that mild manipula- unitary factor in executive functions. I will suggest that action
representation (i.e., internally representing movements) may con- while external, social speech is for others. Vygotsky considered
stitute at least one basic metacognitive executive function factor. I that thought development is determined by language.
will refer to action representation and time perception, derived Vygotsky (1987) separated two different types of concepts:
from it; both ultimately potentially depending upon one single spontaneous and scientific. Spontaneous concepts are developed
core ability (sequencing?). in a parallel way with language, whereas scientific concepts are
Two departing observations are important to support the concepts learned at school. Children progressively develop reflec-
involvement of prefrontal cortex in motor representation: tive consciousness through the development of scientific concepts.
School is intimately related with learning a new conceptual instru-
1. Anatomical observation. Prefrontal cortex represents an exten- ment: reading. Written language is an extension of oral language,
sion and further evolution of the frontal motor areas. It may and represents the most elaborated form of language.
be conjectured that the prefrontal lobe should participate in Luria further extended Vygotskys ideas and attempted to find
complex and elaborated motor (executive) activities. the neurological correlates for different components of complex
2. Clinical observation. A diversity of motor control disturbances cognitive processes. He clearly stated that mental functions
are observed in prefrontal pathology, such as perseveration, uti- are . . .social in origin and complex and hierarchical in their struc-
lization behavior, paratonia, primitive reflexes, etc. (e.g., Ardila ture and they all are based on a complex system of methods and
& Rosselli, 2007a; Victor & Ropper, 2001). means. . . (Luria, 1973, p. 30).
In brief, Vygotsky (1934/1962) argued that complex psycholog-
Several authors have argued that thought, reasoning, and other ical processes (metacognitive executive functions) derives from
forms of complex cognition (metacognition) depend on an interior- language internalization. Thinking relies in the development of
ization of actions. Vygotsky (1929, 1934/1962, 1934/1978), for in- an instrument (language or any other), that represents a cultural
stance, proposed that thought (and in general, complex cognitive product.
processes) is associated with some inner speech. Vygotsky repre- Lieberman (2002a, 2002b) refers specifically to the origins of
sents the most classical author suggesting this interpretation for language. He postulates that neural circuits linking activity in
complex cognition. More recently, Lieberman (2002a, 2002b) sug- anatomically segregated populations of neurons in subcortical
gested that language in particular and cognition in general arise structures and the neocortex throughout the human brain regu-
from complex sequences of motor activities. late complex behaviors such as walking, talking, and compre-
Vygotsky (1934/1978) developed the concept of extracortical hending the meaning of sentences. The neural substrate that
organization of higher mental functions to account for the interac- regulates motor control (basal ganglia, cerebellum, and frontal
tion of biological and cultural factors in the development of human cortex) in the common ancestor of apes and humans most likely
cognition. Vygotskys (1934/1962, 1934/1978) understanding of was modified to enhance cognitive and linguistic ability. Lieber-
higher mental functions is roughly equivalent to metacognitive man (2002a, 2002b) suggests that motor activity is the departing
executive functions. A major factor in systemic organization of point for cognition. Speech communication played a central role
higher cognitive processes, according to Vygotsky, is the engage- in this process. The neural bases of mankind linguistic ability
ment of external instruments (objects, symbols, signs), which have are complex, involving structures other than Brocas and Wer-
an independent history of development within culture. This princi- nickes areas. Many other cortical areas and subcortical structures
ple of construction of brain functional systems was called by form part of the neural circuits implicated in the lexicon, speech
Vygotsky the principle of extracortical organization of complex production and perception, and syntax. The subcortical basal gan-
mental functions, implying that all types of mankind conscious glia support the corticalstriatalcortical circuits that regulate
activity are formed with the support of external auxiliary tools or speech production, complex syntax, and the acquisition of the
aids. However, different mediators and means, or significantly dif- motor and cognitive pattern generators that underlie speech pro-
ferent details within them (e.g., the direction of writing and degree duction and syntax. They most likely are involved in learning the
of letter-sound correspondence, orientation by maps or by the semantic referents and sound patterns that are instantiated as
behavior of sea-birds, etc.) may develop, and in fact are developed words in the brains dictionary.
in different cultures. Therefore, the analysis of cognitive processes The cerebellum and prefrontal cortex are also involved in learn-
must necessarily take into account these cross-cultural differences ing motor acts (e.g., Matsumura et al., 2004; Hernandez-Mueller,
(Kotik-Friedgut & Ardila, 2004). Mulas, & Mattos, 2005). Lieberman (2002a, 2002b) proposes that
The central point in Vygotskys (1934/1962) idea is that higher the frontal regions of the cortex are implicated in virtually all cog-
forms of cognition (cognitive executive functions) depend on cer- nitive acts and the acquisition of cognitive criteria; posterior corti-
tain mediation (language, writing or any other); the instruments cal regions are clearly active elements of the brains dictionary. The
used for mediating these complex cognitive processes are cultur- anterior cingulate cortex plays a part in virtually all aspects of lan-
ally developed. According to Vygotsky (1934/1962), the invention guage and speech. Real-word knowledge appears to reflect stored
(or discovery) of these instruments, will result in a new type of conceptual knowledge in regions of the brain traditionally associ-
evolution (cultural evolution), not requiring any further biological ated with visual perception and motor control. Some aspects of hu-
changes. Thinking is interpreted as a covert motor activity (inner man linguistic ability, such as the basic conceptual structure of
speech). words and simple syntax, are phylogenetically primitive and most
Vygotsky (1929) assumes that thought and speech develop dif- likely were present in the earliest hominids. Lieberman (2002a,
ferently and independently having different genetic roots. Before 2002b) further suggests that speech production, complex syntax,
two years of age, the development of thought and speech are sep- and a large vocabulary developed in the course of hominid evolu-
arate. They converge and join at about the age of two years, and tion, and Homo erectus most likely talked, had large vocabularies,
thought from this point ahead becomes language mediated (verbal and commanded fairly complex syntax. Full human speech capa-
thought). Language in consequence becomes the primary instru- bility, enhancing the robustness of vocal communication, most
ment for conceptualization and thinking. According to Vygotsky likely is a characteristic of anatomically modern humans.
(1934/1962), speech develops first with external communicative/ These two authors (Vygotsky and Lieberman), although using
social speech, then egocentric speech, and finally inner speech. rather different approaches, have both postulated that the devel-
Vocalization becomes unnecessary because the child thinks the opment of language and complex cognition are related with some
words instead of pronouncing them. Inner speech is for oneself, motor programs, sequencing, internalizing actions, and the like.
The recent discovery of so-called mirror neurons represents a the entire cerebrum on each specimen. In relative terms, prefrontal
new element in understanding inner speech and action representa- white matter was found to have the largest difference between hu-
tion. A mirror neuron is a neuron which fires both when an animal man and non-human, whereas gray matter showed no significant
performs an action and also when the animal observes the same difference. Increased brain interconnectivity may represent a ma-
action performed by another animal. In humans, brain activity con- jor characteristic of the human brain. As a note of caution, it is
sistent with mirror neurons has been found in the premotor cortex important to keep in mind that subjects used in this study were
and the inferior parietal cortex (Rizzolatti & Craighero, 2004; Riz- contemporary people, living in city environments, with a high level
zolatti, Fadiga, Gallese, & Fogassi, 1996). In monkeys, the rostral of education, etc., not humans living in those conditions existing
part of ventral premotor cortex (area F5) contains neurons that dis- 150,000 years ago. Stimulation can be rather different not only
charge, both when the monkey grasps or manipulates objects and qualitatively but even probably quantitatively, potentially impact-
when it observes the experimenter performing similar actions. ing on brain interconnectivity.
These neurons (mirror neurons) appear to represent a system that It can be tentatively concluded that it is questionable that the
matches observed events to similar, internally generated actions. size of the prefrontal cortex can account for the human executive
Transcranial magnetic stimulation and positron emission functions. Some other factors have to be considered, such as con-
tomography (PET) experiments suggest that a mirror system for nectivity (increased stimulation?).
gesture recognition also exists in humans and includes Brocas area
(Rizzolatti & Arbib, 1998). The discovery of mirror neurons in the
Brocas area might have immense consequences for understanding 6. Executive functions in pre-historical man
the organization and evolution of mankind cognition (Arbib, 2006;
Craighero, Metta, Sandini, & Fadiga, 2007). An obvious implication Some recent studies have approached the question of the evolu-
of mirror neurons is that they can participate in the internal repre- tion of the prefrontal cortex and executive functions (Risberg,
sentation of actions. Neuroimaging data have showed that interac- 2006; Roth & Dicke, 2005; Winterer & Goldman, 2003). It is usually
tions involving Brocas area and other cortical areas are weakest accepted that Homo sapiens sapiens appeared about 150,000 years
when listening to spoken language accompanied by meaningful ago, and during this time, his brain evolution has been minimal
speech-associated gestures (hence, reducing semantic ambiguity), (Wood, 1992). It means, the humans existing since about 150,000
and strongest when spoken language is accompanied by self- years ago had basically the very same neurological organization
grooming hand movements or by no hand movements at all sug- of contemporary individuals, including the biological foundations
gesting that Brocas area may be involved in action recognition for the so-called executive functions.
(Skipper, Goldin-Meadow, Nusbaum, & Small, 2007). PET studies How were executive functions used by pre-historical man? Of
have associated the neural correlates of inner speech with activity course, we cannot be sure, but some few papers have approached
of the Brocas area (McGuire et al., 1996). this question (e.g., Bednarik, 1994, 2003; Coolidge & Wynn, 2001,
2005; Sugarman, 2002; Wayne, 2006).
Coolidge and Wynn (2001) raised the question of how executive
5. Is there anything special in the prefrontal cortex? functions (supposedly, one of the key evolutionary acquisitions
that led to the development of modern thinking) were demon-
For some time it has been assumed that the prefrontal cortex is strated by prehistoric people. Coolidge and Wynn assumed that
significantly larger in humans than in other primates (e.g., Blinkov it is possible to match many of the features of executive function
& Glezer, 1968). This difference in volume was assumed to repre- with activities reconstructed from archaeological evidence. The po-
sent a major reason to account for differences in complex forms tential application of several components of executive function
of cognition (executive functions) observed in humans. (such as sequential memory, task inhibition, and organization
Nonetheless, such an assumption turned out to be incorrect. and planning) is analyzed by the authors: (1) Sequential memory:
Measures of prefrontal lobe volumes have not found differences it can be speculated in the Palaeolithic Lithic reduction sequences,
between mankind and non-human primates. Semendeferi, Dama- but even sophisticated procedures like Levallois can be explained
sio, Frank, and Van Hoesen (1997, 2002) measured the volume of without resort to closely linked sequences of action. The produc-
the frontal lobe as a whole and of its main sectors (including cortex tion and use of barbed bone projectile points is another potential
and immediately underlying white matter) in humans, chimpan- marker. The final product depends much more closely on a set se-
zees, gorillas, orangutans, gibbons, and macaques using three- quence of actions. It is a true multi-step technology. (2) Tasks of
dimensional reconstructions of magnetic resonance (MR) scans of inhibition, in which immediate gratification and action are de-
the brain. Although the absolute volume of the brain and the fron- layed, are harder to identify archaeologically. Agriculture requires
tal lobe was found to be largest in humans, the relative size of the such inhibition. Facilities such as traps, that capture remotely,
frontal lobe was similar across hominoids: macaque (28.1%), gib- are technologies of inhibition and were probably present in the
bon (31.1%), orangutan (35.3%), gorilla (32.4%), chimpanzee European Mesolithic. Palaeolithic examples are less convincing.
(35.9%), and human (36.7%). Most significantly, it was found that Coolidge and Wynn (2001, 2005) consider that nothing of Middle
humans do not have a larger frontal lobe than expected from a pri- Palaeolithic foraging, however, would require tasks of inhibition
mate brain of mankind size. Furthermore, the relative size of the (indeed, they conclude that nothing in the archaeological record of
sectors of the frontal lobe (dorsal, mesial, and orbital) was similar Palaeolithic appears to require executive function). (3) Organization
across the primate species studied. Semendeferi and colleagues and planning is another basic executive function ability that likely
suggested that the special cognitive abilities attributed to a frontal was required for activities such as migration and colonization.
advantage may be due to differences in individual cortical areas Coolidge and Wynns (2001) review of the archaeological evi-
and to a richer interconnectivity, none of which required an in- dence finds no convincing demonstration for executive functions
crease in the overall relative size of the frontal lobe during hominid among the traces left by Neanderthals. The authors conclude that
evolution. the archaeological records support the hypothesis that executive
Schoenemann, Sheehan, and Glotzer (2005) found that a major function was a late and critical acquisition in human cognitive evo-
difference between humans and other primates was the white lution. In a very ingenious study, Stout and Chaminade (2007)
matter volume. Using MRI from 11 primate species, the authors using positron emission tomography (PET) recorded the brain
measured gray, white, and total volumes for both prefrontal and activity from six inexperienced subjects learning to make stone
tools of the kind found in the earliest archaeological records. The process taking thousands of years, but the most critical element
authors found that tool making is associated with the activation of human language is the use of grammar, likely appearing some
of diverse parieto-frontal perceptual-motor systems, but no activa- 10,000100,000 years ago (Ardila, 2006). Probably, H. neanderthal-
tion was observed in dorsolateral prefrontal cortex. They con- ensis did not have a grammatical language and according to arche-
cluded that human capacities for sensorimotor adaptation, rather ological evidence, did not use executive functions (Coolidge &
than abstract conceptualization and planning, were central factors Wynn, 2008). Language grammar likely developed from action
in the initial stages of human technological evolution, such as mak- internalization (Ardila, 2006).
ing stone tools. Nonetheless, complex cognitive processes (i.e., Written language represents an extension of oral language.
metacognitive executive functions) seem to be crucial for further Written language appeared only some 60008000 years ago and
development and survival of H. sapiens. The key factor for H. sapiens its diffusion has been so slow that even nowadays about 20% of
late evolution seems to be the mental ability to plan and strategize, the world population is illiterate (UNESCO, 2000). Performance in
which allowed them to find innovative solutions to the many psychometric executive function tests has been observed to be
changing environmental problems which they were exposed (Coo- very significantly correlated with subjects educational level (e.g.,
lidge & Wynn, 2008). This may be one reason to account why H. Ardila, Ostrosky-Solis, Rosselli, & Gomez, 2000; Ardila, Rosselli, &
sapiens survived while Homo neanderthalensis disappeared. Chang- Rosas, 1989; Ardila & Rosselli, 2007b; Ostrosky, Ardila, Rosselli, Lo-
ing environmental conditions (e.g., global climates changes) may pez-Arango, & Uriel-Mendoza, 1998; Reis & Castro-Caldas, 1997;
require flexible survival strategies. It has been conjectured that Rosselli, Ardila, & Rosas, 1990). For instance, Gomez-Perez and
during the past history changing physical environment conditions Ostrosky-Solis (2006) observed that whereas tests related to mem-
resulted in a selection that gave human ancestors adaptive versa- ory are sensitive to aging, those related to executive functioning
tility to endure increasing environmental instability (Bonnefille, are mostly sensitive to education. It can be argued that illiterates
Potts, Chali, Jolly, & Peyron, 2004; Potts, 1996, 2004). possess basic executive functions (e.g., ability to internally repre-
Mithen (1994, 1996) has proposed the accessibility of mental sent actions) but they lack an important instrument to organize
modules as the impetus for mankind culture at the time of the executive functions: written language.
Middle/Upper Palaeolithic transition, about 60,000 to about
30,000 years ago. He identified these mental modules as general
intelligence, social intelligence, natural history intelligence, techni- 8. Tentative conclusions
cal intelligence, and language. Probably, language was the most
important one, increasing communication, and facilitating the The analysis of executive functions represents one of the most
transmission of knowledge, potentially resulting in an increased intensively studied neuroscience questions during the last decade.
probability of survival and reproduction. The emphasis in reasoning, abstracting skills, behavioral control,
It could be speculated that at the beginning of human history, anticipating the consequences of behavior, and similar abilities,
transmitting knowledge from generation to generation was lim- has contributed to the frequently found false idea that human
ited. Although some forms of learning can be transmitted by mod- behavior is guided by rationality. Human history blatantly contra-
eling or imitation (vicarious learning or social learning or dicts this idea.
modeling) (e.g., Bandura, 1977) language development repre- This misinterpretation of mankind behavior is linked to the
sented a powerful instrument to accumulate and transmit knowl- assumptions that the human brain is unique and superior to
edge about the world. The crucial point in the origins of executive the brain of other species. We refer to our species as the wise
functions becomes the possibility to conceptualize the environ- man (H. sapiens). Analyzing executive functions, it may be con-
ment (concepts are represented in words) and to transmit and pro- cluded that two different types of executive functions could be
gressively accumulate this knowledge about the world. separated: metacognitive and emotional/motivational, depending
on different brain systems. It could be argued that only the first
one should be referred to as executive functions; usually, however,
7. Metacognitive executive functions as a cultural product they are considered together in most definitions of executive func-
tions, assuming a certain unity.
There is no convincing evidence that Paleolithic individuals Contemporary testing of executive functions has focused on
used executive functions (Coolidge & Wynn, 2001), understood as abstracting, problem solving, and similar metacognitive abilities.
the ability for planning. . .etc. (first interpretation of prefrontal These metacognitive abilities seem to be useful in solving external
abilities: metacognitive executive functions). For thousands of and emotionally neutral problems. When social situations and bio-
years, prefrontal abilities were in consequence exclusively used logical drives are involved, the ability to rationally solve problems
to fulfill basic impulses following socially acceptable strategies seems to decrease in a significant way. In this regard, contempo-
(e.g., hierarchy in the group) (second interpretation of prefrontal rary testing of executive functions has limited ecological validity.
abilities: emotional/motivational executive functions). Archeological analysis has discovered only someif any, evi-
Which were the milestones for cultural development and how dence of metacognitive executive functions in prehistorical man.
did metacognitive executive functions appear? It could be specu- We have to conclude that metacognitive abilities represent a re-
lated that some crucial inventions fueled the development of cul- cent acquisition, not obviously dependent on recent biological
tural evolution (Vygotsky, 1934/1962). For instance, kind of changes. The development of some cultural instruments, poten-
cognitive fluidity has been postulated as a basic requisite for exe- tially resulting in a new type of evolution has been suggested. Lan-
cuting complex human activities (Gardner, 1983). The most impor- guage as an instrument not only to conceptualize the immediate
tant candidate for this crucial invention that fueled the experience, but for its transmission of knowledge, has been pro-
development of cultural evolution is language. Language allows posed as the major cultural instrument for metacognition. Lan-
the transmission of knowledge and facilitates survival and repro- guage complexity has historically increased with the
duction. Without language, children can learn from parents by imi- development of written language. No question, some other cultural
tation, but imitation is limited to some elementary activities, such instruments have also contributed to the development of metacog-
as making a simple stone ax. Language represents a major instru- nitive abilities; for instance, mathematics, drawing, and technology
ment of internal representation of the world and thinking (Vygot- (from the wheel to computers). From the point of view of the brain,
sky, 1934/1978). Language development obviously was a slow metacognitive executive functions are not necessarily correlated
with a further brain development; increased neural interconnec- Damasio, A., & Anderson, S. W. (1993). The frontal lobes. In K. M. Heilman & E.
Valenstein (Eds.), Clinical neuropsychology (4th ed.). New York: Oxford
tivity may potentially support the increased complexization of
University Press.
executive functions found in contemporary H. sapiens. De Frias, C., Dixon, R., & Strauss, E. (2006). Structure of four executive functioning
tests in healthy older adults. Neuropsychology, 20, 206214.
Acknowledgments Delis, D., Kaplan, E., & Kramer, N. (2001). DelisKaplan executive function system.
Psychological Assessment Resources, Inc..
Denckla, M. B. (1994). Measurement of executive function. In G. R. Lyon (Ed.),
My gratitude to the anonymous reviewers for their support, Frames of reference for the assessment of learning disabilities: New views on
most important suggestions, and time devoted to review this man- measurement issues (pp. 117142). Baltimore, MD: Paul H. Brooks.
Denckla, M. B. (1996). A theory and model of executive function: A
uscript. My sincere recognition to Loren Overholt and Dr. Eliane neuropsychological perspective. In G. R. Lyon & N. A. Krasnegor (Eds.).
Ramos for their invaluable help in editing the manuscript. Attention, memory and executive function (pp. 263277). Baltimore, MD: Paul
H. Brooks.
Dobzhansky, T. (1973). Nothing in biology makes sense except in the light of
References evolution. The American Biology Teacher, 35, 125129.
Duncan, J., Emslie, H., Williams, P., Johnson, R., & Freer, C. (1996). Intelligence and
Arbib, M. A. (2006). Aphasia, apraxia and the evolution of the language-ready brain. the frontal lobes: The organization of goal-directed behavior. Cognitive
Aphasiology, 20, 11251155. Psychology, 30, 257303.
Anderson, V. (2001). Assessing executive functions in children: Biological, Elliott, R. (2003). Executive functions and their disorders. British Medical Bulletin, 65,
psychological, and developmental considerations. Developmental 4959.
Neurorehabilitation, 4, 119136. Eslinger, P. J., & Damasio, A. R. (1985). Severe disturbance of higher cognition after
Anderson, S. W., Bechara, A., Damasio, H., Tranel, D., & Damasio, A. R. (1999). bilateral frontal lobe ablation: Patient EVR. Neurology, 35, 17311741.
Impairment in social and moral behaviour related to early damage in human Feuchtwanger, E. (1923). Die Funktionen des Stirnhirns. Berlin: Springer.
prefrontal cortex. Nature Neuroscience, 2, 10321037. Fincham, J. M., Carter, C. S., van Veen, V., Stenger, V. A., & Anderson, J. R. (2002).
Ardila, A. (2006). Orgenes del lenguaje: un anlisis desde la perspectiva de las Neural mechanisms of planning: A computational analysis using event-related
afasias. Revista de Neurologa, 43, 690698. fMRI. Proceedings of the National Academy of Sciences of the United States of
Ardila, A., Ostrosky-Solis, F., Rosselli, M., & Gomez, C. (2000). Age related cognitive America, 99, 33463351.
decline during normal aging: The complex effect of education. Archives of Friedman, N. P., Miyake, A., Corley, R. P., Young, S. E., DeFries, J. C., & Hewitt, J. K.
Clinical Neuropsychology, 15, 495514. (2006). Not all executive functions are related to intelligence. Psychological
Ardila, A., Rosselli, M., & Rosas, P. (1989). Neuropsychological assessment in Science, 17, 172179.
illiterates: Visuospatial and memory abilities. Brain and Cognition, 11, Fuster, J. M. (1989). The prefrontal cortex. New York: Raven Press.
147166. Fuster, J. M. (1997a). The prefrontal cortexAnatomy physiology, and neuropsychology
Ardila, A., & Rosselli, M. (2007a). Neuropsicologia clinica. Mexico: Manual Moderno. of the frontal lobe. Philadelphia: Lippincott-Raven.
Ardila, A., & Rosselli, M. (2007b). Illiterates and cognition: The impact of education. Fuster, J. M. (1997b). The prefrontal cortex: Anatomy, physiology, and neuropsychology
In B. Uzzell, M. Pontn, & A. Ardila (Eds.), International handbook of cross-cultural of the frontal lobe (3rd ed.). New York: Lippincott, Williams & Wilkins.
neuropsychology (pp. 181198). Mahwah, NJ: Lawrence Erlbaum Associates. Fuster, J. M. (2001). The prefrontal cortexAn update: Time is of the essence.
Ardila, A., & Surloff, C. (2007). Dysexcutive syndromes. San Diego: Medlink: Neuron, 30, 319333.
Neurology. Fuster, J. M. (2002). Frontal lobe and cognitive development. Journal of
Baddeley, A. (1986). Working memory. Oxford: Oxford University Press. Neurocytology, 31, 373385.
Baldo, J. V., Schwartz, S., Wilkins, D., & Dronkers, N. F. (2006). Role of frontal versus Gardner, H. (1983). Frames of mind: The theory of multiple intelligences. New York:
temporal cortex in verbal fluency as revealed by voxel-based lesion symptom Basic Books.
mapping. Journal of the International Neuropsychological Society, 12, 896900. Godefroy, O., Cabaret, M., Petit-Chenal, V., Pruvo, J. P., & Rousseaux, M. (1999).
Bandura, A. (1977). Social learning theory. New Jersey: Prentice Hall. Control functions of the frontal lobes. Modularity of the central-supervisory
Barbas, H. (2006). Organization of the principal pathways of prefrontal lateral, system? Cortex, 35, 120.
medial, and orbitofrontal cortices primates and implications for their Goldberg, E. (2001). The executive brain. New York: Oxford University Press.
collaborative interaction in executive functions. In J. Risberg & J. Grafman Goldstein, K. (1944). The mental changes due to frontal lobe damage. Journal of
(Eds.). The frontal lobes. Development, function and pathology (pp. 2168). Psychology, 17, 187208.
Cambridge: Cambridge University Press. Gomez-Perez, E., & Ostrosky-Solis, F. (2006). Attention and memory evaluation
Barkley, R. A. (1997). ADHD and the nature of self-control. New York: Guilford Press. across the life span: Heterogeneous effects of age and education. Journal of
Bechara, A., Damasio, H., & Damasio, A. R. (2000). Emotion, decision making and the Clinical and Experimental Neuropsychology, 28, 477494.
orbitofrontal cortex. Cerebral Cortex, 10, 295307. Grafman, J. (2006). Human prefrontal cortex: Processes and representations. In J.
Bednarik, R. G. (1994). Art origins. Anthropos, 89, 169180. Risberg & J. Grafman (Eds.). The frontal lobes. Development, function and
Bednarik, R. G. (2003). The earliest evidence of palaeoart. Rock Art Research, 20, pathology (pp. 6991). Cambridge: Cambridge University Press.
89135. Happaney, K., Zelazo, P. D., & Stuss, D. T. (2004). Development of orbitofrontal
Beer, J. S., John, O. P., Scabini, D., & Knight, R. T. (2006). Orbitofrontal cortex and function: Current themes and future directions. Brain and Cognition, 55, 110.
social behavior: Integrating self-monitoring and emotion-cognition Harlow, J. M. (1868). Recovery from the passage of an iron bar through the head.
interactions. Journal of Cognitive Neuroscience, 18, 871879. Massachusetts Medical Society Publications, 2, 327346.
Benton, A. L., Hamsher, K., & Sivan, A. B. (1994). Multilingual aphasia examination Heaton, R. (1981). Wisconsin card sorting test: Manual. Odessa: Psychological
(3rd ed.). Iowa City, IA: AJA Associates. Assessment Resources, Inc..
Berg, E. A. (1948). A simple objective technique for measuring flexibility in thinking. Hernandez-Mueller, S., Mulas, F., & Mattos, L. (2005). The contribution of the
Journal of General Psychology, 39, 1522. cerebellum to cognitive processes. Revista de Neurologa, 40(Suppl. 1), S57S64.
Blair, R. J. (2004). The roles of orbital frontal cortex in the modulation of antisocial Hobson, P., & Leeds, L. (2001). Executive functioning in older people. Reviews in
behavior. Brain and Cognition, 55, 198208. Clinical Gerontology, 11, 361372.
Blinkov, S. M., & Glezer, I. I. (1968). Das Zentralnervensystem in Zahlen und Tabellen. Jacobs, R., Harvey, A. S., & Anderson, V. (2007). Executive function following
Jena: Fischer. focal frontal lobe lesions: Impact of timing of lesion on outcome. Cortex, 43,
Bonnefille, R., Potts, R., Chali, F., Jolly, D., & Peyron, O. (2004). High-resolution 792805.
vegetation and climate change associated with Pliocene Australopithecus Kimberg, D., DEsposito, M., & Farah, M. (1997). Cognitive functions in the prefrontal
afarensis. Proceedings of the National Academy of Sciences of the United States of cortexWorking memory and executive control. Current Directions in
America, 101, 1212512129. Psychological Science, 6, 185192.
Coolidge, F. L., & Wynn, T. (2001). Executive functions of the frontal lobes and the Kotik-Friedgut, B., & Ardila, A. (2004). Systemic-dynamic Lurian theory and
evolutionary ascendancy of Homo sapiens. Cambridge Archaeological Journal, 11, contemporary cross-cultural neuropsychology. In T. Akhutina, L. Moskovich, &
255260. T. Dorothy (Eds.), A.R. Luria and contemporary psychology. New York: Nova
Coolidge, F. L., & Wynn, T. (2005). Working memory, its executive functions, Science Publishers.
and the emergence of modern thinking. Cambridge Archaeological Journal, 15, Laiacona, M., De Santis, A., Barbarotto, R., Basso, A., Spagnoli, D., & Capitani, E.
526. (1989). Neuropsychological follow-up of patients operated for aneurysms of
Coolidge, F. L., & Wynn, T. (2008). A stone-age meeting of minds neandertals anterior communicating artery. Cortex, 25, 261273.
became extinct while Homo sapiens prospered. A marked contrast in mental Lafleche, G., & Albert, M. (1995). Executive function deficits in mild Alzheimers
capacities may account for these different fates. American Scientist Online, disease. Neuropsychology, 9, 313320.
96, <http://www.americanscientist.org/template/AssetDetail/assetid/56450; Lehto, J. (1996). Are executive function tests dependent on working memory
jsessionid=aaa8s-19FU6GQB/>. Accessed 6.02.08. capacity? Quarterly Journal of Experimental Psychology, 49, 2950.
Craighero, L., Metta, G., Sandini, G., & Fadiga, L. (2007). The mirror-neurons system: Leung, H. C., & Cai, W. (2007). Common and differential ventrolateral prefrontal
Data and models. Progress in Brain Research, 164, 3959. activity during inhibition of hand and eye movements. Journal of Neurosciences,
Cummings, J. L. (1993). Frontal-subcortical circuits and human behavior. Archives of 27, 98939900.
Neurology, 50, 873880. Levin, H. S., Eisenberg, H. M., & Benton, A. L. (1991). Frontal lobe function and
Damasio, A. R. (1994). Descartes error. New York: Putnam. dysfunction. New York: Oxford University Press.
Lie, C. H., Specht, K., Marshall, J. C., & Fink, G. R. (2006). fMRI to decompose the Risberg, J. (2006). Evolutionary aspects of the frontal lobes. In J. Risberg & J. Grafman
neural processes underlying the Wisconsin card sorting test. Neuroimage, 30, (Eds.). The frontal lobes. Development, function and pathology (pp. 120).
10381049. Cambridge: Cambridge University Press.
Lezak, M. D. (1983). Neuropsychological assessment (2nd ed.). New York: Oxford Rizzolatti, G., & Arbib, M. A. (1998). Language within our grasp. Trends in
University Press. Neurosciences, 21, 188194.
Lezak, M. D., Howieson, D. B., Loring, D. W., & Hannay, J. (2004). Neuropsychological Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. Annual Review of
assessment. New York: Oxford. Neuroscience, 27, 169192.
Lieberman, P. (2002a). Human language and our reptilian brain. Cambridge, MA: Rizzolatti, G., Fadiga, L., Gallese, V., & Fogassi, L. (1996). Premotor cortex and the
Harvard University Press. recognition of motor actions. Cognitive Brain Research, 3, 131141.
Lieberman, P. (2002b). On the nature and evolution of the neural bases of human Roberts, A. C., Robbins, T. W., & Weiskrantz, L. (1998). The prefrontal cortex: Executive
language. Yearbook of Physical Anthropology, 45, 3662. and cognitive functions. Oxford: Oxford University Press.
Luria, A. R. (1966). Human brain and psychological processes. New York: Harper & Roberts, A. C., Robbins, T. W., & Weiskrantz, L. (2002). The prefrontal cortex: Executive
Row. and cognitive functions. Oxford: Oxford University Press.
Luria, A. R. (1969). Frontal lobe syndromes. In P. J. Vinken & G. W. Bruyn (Eds.). Ross, E. D., & Stewart, R. M. (1981). Akinetic mutism from hypothalamic damage:
Handbook of clinical neurology (Vol. 2, pp. 725757). Amsterdam: North Holland. Successful treatment with dopamine agonists. Neurology, 31, 14351439.
Luria, A. R. (1973). Working brain. London: Penguin books. Rosselli, M., Ardila, A., & Rosas, P. (1990). Neuropsychological assessment in
Luria, A. R. (1980). Higher cortical functions in man (2nd ed.). New York: Basic. illiterates II: Language and praxic abilities. Brain and Cognition, 12,
Manes, F., Sahakian, B., Clark, L., Rogers, R., Antoun, N., Aitken, M., et al. (2002). 281296.
Decision-making processes following damage to the prefrontal cortex. Brain, Roth, G., & Dicke, U. (2005). Evolution of the brain and intelligence. Trends in
125, 624639. Cognitive Sciences, 9, 250257.
McGuire, P. K., Silbersweig, D. A., Murray, R. M., David, A. S., Frackowiak, R. S. J., & Salthouse, T. (1996). The processing-speed theory of adult age differences in
Frith, C. D. (1996). Functional anatomy of inner speech and auditory verbal cognition. Psychological Review, 103, 403428.
imagery. Psychological Medicine, 26, 3839. Salthouse, T. (2005). Relations between cognitive abilities and measures of
Matsumura, M., Sadato, N., Kochiyama, T., Nakamura, S., Naito, E., Matsunami, K., executive functioning. Neuropsychology, 19, 532545.
et al. (2004). Role of the cerebellum in implicit motor skill learning: A PET Salthouse, T., Atkinson, T., & Berish, D. (2003). Executive functioning as a potential
study. Brain Research Bulletin, 63, 471483. mediator of age-related cognitive decline in normal adults. Journal of
Medalla, M., Lera, P., Feinberg, M., & Barbas, H. (2007). Specificity in inhibitory Experimental Psychology: General, 132, 566594.
systems associated with prefrontal pathways to temporal cortex in primates. Schoenemann, P. T., Sheehan, M. J., & Glotzer, L. D. (2005). Prefrontal white matter
Cerebral Cortex, 17(Suppl. 1), i136i150. volume is disproportionately larger in humans than in other primates. Nature
Miller, B. L., & Cummings, J. L. (1998). The human frontal lobes: Functions and Neuroscience, 8, 242252.
disorders. New York: The Guilford Press. Semendeferi, K., Damasio, H., Frank, R., & Van Hoesen, G. W. (1997). evolution of the
Miller, P., & Wang, X. J. (2006). Inhibitory control by an integral feedback signal in frontal lobes: A volumetric analysis based on three-dimensional
prefrontal cortex: A model of discrimination between sequential stimuli. reconstructions of magnetic resonance scans of human and ape brains.
Proceedings of the National Academy of Sciences of the United States of America, Journal of Human Evolution, 32, 375388.
103, 201206. Semendeferi, K., Lu, A., Schenker, N., & Damasio, H. (2002). Humans and great apes
Mitchell, R. L., & Phillips, L. H. (2007). The psychological, neurochemical and share a large frontal cortex. Nature Neuroscience, 5, 272276.
functional neuroanatomical mediators of the effects of positive and negative Shallice, T. (1982). Specific impairments of planning. Philosophical Transactions of
mood on executive functions. Neuropsychologia, 45, 617629. the Royal Society of London Series B, 298, 199209.
Mithen, S. (1994). From domain specific to generalized intelligence: A cognitive Simon, H. A. (1975). The functional equivalence of problem solving skills. Cognitive
interpretation of the middle/upper palaeolithic transition. In C. Renfrew & E. B. Psychology, 7, 268288.
W. Zubrow (Eds.), The ancient mind: Elements of cognitive archaeology Skipper, J. I., Goldin-Meadow, S., Nusbaum, H. C., & Small, S. L. (2007). Speech-
(pp. 2939). Cambridge, UK: Cambridge University Press. associated gestures, Brocas area, and the human mirror system. Brain and
Mithen, S. (1996). The prehistory of the mind. London: Thames and Hudson. Language, 101, 260277.
Miyake, A., Friedman, N., Emerson, M., Witzki, A., & Howerter, A. (2000). The unity Smith, P. B., Bond, M. H., & Kagitcibas, C. (2006). Understanding social psychology
and diversity of executive functions and their contributions to complex Frontal across cultures: Living and working in a changing world (sage social psychology
Lobe tasks: A latent variable analysis. Cognitive Psychology, 41, 49100. program). New York: Sage Publishers.
Moll, J., Zahn, R., de Oliveira-Souza, R., Krueger, F., & Grafman, J. (2005). The neural Spies, K., Hesse, F. W., & Hummitzsch, C. (1996). Mood and capacity in Baddeleys
basis of human moral cognition. Nature Reviews Neuroscience, 6, 799809. model of human memory. Zeitschrift fur Psychologie, 204, 367381.
Oaksford, M., Morris, F., Grainger, B., & Williams, J. M. G. (1996). Mood, reasoning, Stout, D., & Chaminade, T. (2007). The evolutionary neuroscience of tool making.
and central executive processes. Journal of Experimental Psychology: Learning, Neuropsychologia, 45, 10911100.
Memory and Cognition, 22, 477493. Stroop, J. R. (1935). Studies of inference in serial verbal reactions. Journal of
Osaka, N., Osaka, M., Mondo, H., Morishita, M., Fukuyama, H., & Shibasaki, H. (2004). Experimental Psychology, 18, 643662.
The neural basis of executive function in working memory: An fMRI study based Stuss, D. T., & Benson, D. F. (1986). The frontal lobes. New York: Raven Press.
on individual differences. NeuroImage, 21, 623631. Stuss, D. T., & Knight, R. T. (2002). Principles of frontal lobe function. New York:
Ostrosky, F., Ardila, A., Rosselli, M., Lopez-Arango, G., & Uriel-Mendoza, V. (1998). Oxford University Press.
Neuropsychological test performance in illiterates. Archives of Clinical Sugarman, R. (2002). Evolution and executive functions: Why our toolboxes are
Neuropsychology, 13, 645660. empty? Revista Espaola de Neuropsicologa, 4, 351377.
Parkin, A., & Java, R. (1999). Deterioration of frontal lobe function in normal aging: Victor, M., & Ropper, A. H. (2001). Adams & Victors principles of neurology. New York:
Influences of fluid intelligence versus perceptual speed. Neuropsychology, 13, McGraw-Hill.
539545. Vygotsky, L. S. (1929). The problem of the cultural development of the child II.
Pennington, B. F., & Ozonoff, S. (1996). Executive functions and developmental Journal of Genetic Psychology, 36, 415432.
psychopathology. Journal of Child Psychology and Psychiatry, 37, 5187. Vygotsky, L. S. (1934/1962). Thought and language. Cambridge, MA: MIT Press.
Perecman, E. (Ed.). (1987). The frontal lobes revisited. New York: The IRBN Press. Vygotsky, L. S. (1934/1978). Mind in society. Cambridge, MA: Harvard University
Piguet, O., Grayson, G., Browe, A., Tate, H., Lye, T., Creasey, H., et al. (2002). Normal Press.
aging and executive functions in Old-old community dwellers: Poor Vygotsky, L. S. (1987). Lecture 2. Memory and its development in childhood. The
performance is not an inevitable outcome. International Psychogeriatric collected works of L.S. Vygotsky (Vol. 1, pp. 301310). New York: Kluwer
Association, 14, 139159. Academic/Plenum Press.
portal.unesco.org/education/en/ev.php- Wayne, C. (2006). The evolutionary origins of volition. In D. Spurrett, H. Kincaid, D.
URL_ID=12874&URL_DO=DO_TOPIC&URL_SECTION=201.html. Ross, & L. Stephens (Eds.), Distributed cognition and the will: Individual volition
Potts, R. (1996). Evolution and climate variability. Science, 273, 922923. and social context. Cambridge: MIT Press, MA.
Potts, R. (2004). Paleoenvironmental basis of cognitive evolution in great apes. Winterer, G., & Goldman, D. (2003). Genetics of human prefrontal function. Brain
American Journal of Primatology, 62, 209228. Research: Brain Research Review, 43, 134163.
Pribram, K. H., & Luria, A. R. (Eds.). (1973). Psychophysiology of the frontal lobes. New Wood, B. (1992). Origin and evolution of the genus Homo. Nature, 355, 783790.
York: Academic Press. Yoon, J. H., Hoffman, J. N., & DEsposito, M. (2007). Segregation of function in the
Reis, A., & Castro-Caldas, A. (1997). Illiteracy: A cause for biased cognitive lateral prefrontal cortex during visual object working memory. Brain Research,
development. Journal of the International Neuropsychological Society, 5, 444450. 1184, 217225.
CORRESPONDENCIA
algn tiempo; en ocasiones estuvo desemplea- dentes sociales de extrema pobreza y de abu-
Neuropsicologa de do, simplemente deambulando por la ciudad. sos por parte de su madre. No existen antece-
los asesinos en serie Se traslad a los Estados Unidos a los 18 aos. dentes psiquitricos o neurolgicos. Su con-
En este pas trabaj en un restaurante como la- ducta criminal se inici en una etapa tarda de
Desde el punto de vista criminolgico, cuan- vaplatos y, en ocasiones, se dedicaba a recoger su vida y slo asesina a mujeres ancianas que
do un asesino reincide en sus delitos un mni- basura y a labores de limpieza. viven solas, habitualmente tras un proceso de
mo de tres ocasiones y con un cierto intervalo Presenta una historia positiva de psicosis y acercamiento progresivo a la vctima [8].
de tiempo entre ellos, es conocido como asesi- conductas extraas, por lo cual ha sido hospi- Ha vivido en la Ciudad de Mxico desde
no en serie. Suele matar por contacto directo talizado en instituciones psiquitricas tanto en que tena 3 meses de edad. Su padre abandon
con la vctima: las apuala, las estrangula o Cuba como en los Estados Unidos. Afirma la familia despus de que naciera y su madre
las golpea; casi nunca usa armas de fuego. que oye voces que pertenecen a la Santera se fue a vivir con otro hombre. sta era em-
Con frecuencia, sus crmenes suponen una es- (conjunto de sistemas religiosos que funden pleada domstica, afectada de alcoholismo
pecie de ritual en el que el asesino en serie se creencias catlicas con la religin tradicional grave, y su padrastro era un obrero no cualifi-
excita mezclando las fantasas personales con yoruba, practicada por descendientes de afri- cado. Durante su infancia vivi en condiciones
la muerte [1]. canos, sobre todo en Cuba), que le ordenan de extrema pobreza y nunca asisti a la escue-
Generalmente se ha credo que los asesinos hacer cosas y controlan su cuerpo. En una de la. Refiere que su madre siempre la agredi f-
en serie presentan unas caractersticas sociales sus hospitalizaciones fue diagnosticado de sica y verbalmente. Cuando tena 12 aos la
y psicolgicas especficas, como una persona- esquizofrenia paranoide crnica, con ideas entreg a un hombre a cambio de tres cerve-
lidad antisocial [2], una historia de abusos du- delirantes, alucinaciones auditivas y visuales zas. Este hombre la at y la viol, y con l
rante la infancia [3] y, con frecuencia, rasgos e ideacin paranoide. En ocasiones presenta continu durante 15 meses una relacin de so-
de sadismo sexual [4]. Sin embargo, existe una alucinaciones auditivas que le piden que se metimiento, de la que tuvo un hijo. Ms ade-
variabilidad importante, no slo relacionada mate y delirios de posesin satnica. Afirma lante trabaj vendiendo dulces y en una zapa-
con los antecedentes personales y con los inte- que durante su infancia a veces senta que el tera. Tuvo varias parejas y fue madre de cuatro
reses sexuales, sino tambin con respecto a la diablo estaba sobre l. Refiere que presenta hijos. Para conseguir ms dinero, los fines de
edad, a la historia de abuso de alcohol o de eyaculaciones espontneas durante el da y la semana trabajaba en espectculos de lucha li-
otras sustancias, al nivel cultural, al sexo, a las noche, por lo que utiliza un papel de celofn bre. A la edad de 43 aos, dej de participar de
circunstancias especficas que rodean los cr- para evitar que se le humedezca el pantaln. forma directa en ellos, pero continuaba como
menes y a las particularidades de las vctimas. En una resonancia magntica se le detect una promotora de competiciones de lucha libre.
Aunque los asesinos en serie suelen ser perso- lesin de pequeo tamao en el lbulo frontal No se han encontrado antecedentes neuro-
nas jvenes, este tipo de criminales tambin se izquierdo cerca de la cisura de Silvio, as co- lgicos o psiquitricos importantes. No exis-
ha dado entre nios y adolescentes [5], y se mo anormalidades en la sustancia blanca alre- ten antecedentes personales de alcoholismo o
han comunicado pocos casos de asesinatos en dedor de los ventrculos cerebrales. de abuso de drogas.
serie durante la etapa adulta media. Se observa Se le acusa (y lo acepta, pero luego lo nie- Est acusada de matar, por lo menos, a 18
tambin una distribucin sexual extremada- ga) de la muerte de cuatro mujeres, todas ellas ancianas, y de intentar matar a otra; en 10 de
mente dispar; a pesar de que la inmensa mayo- prostitutas de etnia negra. Los asesinatos los los asesinatos existen pruebas dactilares. Al
ra de los asesinos en serie son hombres, este comete de forma similar: las conduce a un lu- principio acept su participacin en tres cr-
tipo de crimen tambin se ha referido, aunque gar solitario o a una casa abandonada, las gol- menes, pero luego slo en uno. El patrn en
con una frecuencia mucho menor, en mujeres pea en la cabeza con una barra y despus las todos los casos ha sido similar: se presenta
[6]. Tomados los datos expuestos en su con- roca con gasolina e intenta quemar su cad- inicialmente como una persona protectora que
junto, el asesino en serie tipo sera un hombre ver. Los asesinatos los comete durante sus es- trata de ayudar a las ancianas. Despus, cuan-
joven, con una edad comprendida habitual- tados alucinatorios por las rdenes que recibe do ya logra entrar en sus domicilios, les solici-
mente entre los 25 y los 35 aos [1,3,7]. de la Santera y no incluyen actividades se- ta trabajo, en ocasiones se enoja por la res-
Existen, pues, diferencias evidentes entre xuales con la vctima. Parece tener, como m- puesta que recibe y las estrangula.
los asesinos en serie, por lo que en ocasiones nimo, una memoria parcial de los crmenes. Considera que se ha comportado de una
resulta difcil hallar rasgos comunes entre to- Los resultados de su evaluacin neuropsi- forma inapropiada porque conoce bien las le-
dos ellos. colgica revelaron una inteligencia general yes y las reglas sociales, y porque sabe que
Se presentan dos casos muy distintos para dentro de rango fronterizo, con habilidades matar es inaceptable. No obstante, considera
ilustrar este aspecto. El primer caso es el de un verbales superiores a las no verbales. Su ren- que sus acciones fueron justificadas, ya que
hombre con una evidente enfermedad cerebral dimiento en pruebas verbales (fluidez verbal, realmente ella es una vctima y pierde la pa-
asociada con limitaciones intelectuales, quien denominacin) se encuentra moderadamente ciencia cuando la insultan.
cometa sus crmenes cumpliendo las rdenes deprimido, mientras que sus puntuaciones en La paciente fue sometida a una evaluacin
que reciba durante sus estados alucinatorios. El pruebas no verbales (figura compleja de Rey- neuropsicolgica, psicolgica y psiquitrica
segundo caso es el de una mujer sin anteceden- Osterrieth: copia), de memoria (figura com- extensa. Exceptuando puntuaciones significa-
tes patolgicos evidentes, pero con claros ante- pleja de Rey-Osterrieth: memoria; memoria tivamente disminuidas en las pruebas de valo-
cedentes personales de abusos, quien mataba lgica) y de funcin ejecutiva (prueba de ras- racin de funciones frontales (batera de fun-
slo a mujeres ancianas porque la insultaban. treo, clasificacin de tarjetas de Wisconsin) ciones frontales y ejecutivas [9], que incluye
corresponden a un rango anormal. 14 subpruebas diferentes: prueba de Stroop,
Caso 1. Hombre cubano de 45 aos con ante- No fue posible realizarle pruebas de perso- prueba de cartas de Iowa, laberintos, seala-
cedentes de haber sufrido un traumatismo cra- nalidad. Sin embargo, clnicamente no parece miento autodirigido, memoria de trabajo vi-
neoenceflico a la edad de 5 aos, con prdida deprimido o ansioso; no se muestra crtico suoespacial, memoria de trabajo verbal, prue-
de conciencia y hospitalizacin. Afirma que, con sus delitos (en algn momento incluso ex- ba de clasificacin de cartas, laberintos, torre
despus del traumatismo, presentaba descoor- clam tanto problema que han armado por la de Hanoi, resta consecutiva, generacin de
dinacin en la pierna izquierda y sensaciones muerte de esas cuatro cucarachas!). Su con- verbos, generacin de clasificaciones semn-
anormales. Asimismo, refiere cefaleas frecuen- ducta en la crcel siempre ha sido excelente y ticas, comprensin y seleccin de refranes, y
tes que le incapacitaban. En un EEG realizado se muestra como una persona muy dcil. Al curva de metamemoria), no se encuentra nin-
posteriormente se observaron ondas agudas margen del asesinato de estas cuatro mujeres, guna anormalidad.
temporales. El paciente recibi fenitona. Sus no tiene antecedentes delictivos. No existen
familiares afirman que despus de los 5 aos antecedentes personales de abuso de alcohol o Estos dos casos ilustran claramente la gran
desarroll crisis epilpticas generalizadas. Su de drogas psicoactivas. variabilidad existente en los aspectos persona-
escolarizacin fue limitada (2-3 aos), con un les, sociales, intelectuales y psiquitricos de
bajo rendimiento acadmico. Tras abandonar Caso 2. Mujer acusada de 18 crmenes y de un los asesinos en serie. No es fcil encontrar ele-
la escuela, trabaj en la construccin durante intento de homicidio. Presenta unos antece- mentos comunes entre ambos. De hecho, es-
162 REV NEUROL 2009; 48 (3)

CORRESPONDENCIA
tos dos casos podran considerarse como in- proponerse que la disfuncin ejecutiva supone que afecta fundamentalmente a embarazadas,
usuales en la categora de los asesinos en se- un requisito necesario, pero no suficiente, pa- neonatos y pacientes con alteracin de la inmu-
rie, ya que no se ajustan a las caractersticas ra este tipo de conducta, y que diferentes fac- nidad celular. L. monocytogenes supone una
supuestamente habituales en este tipo de cri- tores sociales y psicolgicos contribuyen en causa infrecuente de meningitis, absceso cere-
minales: hombres jvenes de edades compren- grado variable a las manifestaciones de las dis- bral e hidrocefalia. A pesar de que las pruebas
didas entre los 20 y los 35 aos, con una per- tintas conductas criminales, incluyendo las de de neuroimagen cada vez aportan una infor-
sonalidad psicoptica, con una historia delic- los asesinatos en serie. macin ms precisa, sigue siendo prioritario
tiva de una gravedad variable, y, en un porcen- considerar el contexto clnico del paciente y la
taje alto de los casos, con evidentes motivacio- A. Ardila a, F. Ostrosky-Sols b totalidad de la informacin suministrada por
nes sexuales. las pruebas complementarias para establecer
Si se comparan los perfiles intelectual y de Aceptado tras revisin externa: 04.11.08. un diagnstico correcto lo antes posible, ya
personalidad de los dos casos, el nico ele- a
Department of Communication Sciences and Dis- que la demora en el tratamiento adecuado pue-
mento en comn que parece existir es el bajo orders. Florida International University. Miami, de agravar decisivamente el pronstico.
rendimiento observado en las pruebas de fun- Florida, Estados Unidos. b Laboratorio de Neuro-
psicologa y Psicofisiologa. Universidad Nacional
cin ejecutiva. De hecho, las puntuaciones en Autnoma de Mxico. Mxico DF, Mxico.
Varn de 53 aos de edad, con antecedentes
dichas pruebas, en particular en las que eva- de hepatopata enlica con cirrosis heptica,
Correspondencia: Alfredo Ardila PhD. Department
lan el control inhibitorio, suelen encontrarse of Communication Sciences and Disorders. Florida
hipertensin arterial y diabetes mellitus tipo 2.
disminuidas en criminales de diferentes tipos International University. 12230 NW 8th Street. Consult por debilidad en el hemicuerpo de-
[10] y no parecen ser especficas de los asesi- 33182 Miami, Florida, USA. E-mail: ardilaa@fiu.edu recho de dos semanas de evolucin, con ma-
nos en serie. Por ejemplo, Blair et al [11] en- yor torpeza para la marcha. En la exploracin
contraron que los individuos con psicopatas inicial, el paciente se encontraba plenamente
presentaban fallos significativos en las prue- BIBLIOGRAFA consciente, orientado y colaborador, con len-
bas neuropsicolgicas que detectaban las le- guaje normal, y en el que nicamente destaca-
siones orbitofrontales. La disfuncin ejecuti- 1. Kraemer GW, Lord WD, Heilbrun K. Com- ba la presencia de facial central derecho leve,
va sera entonces una condicin necesaria, pe- paring single and serial homicide offenses. hemiparesia derecha 4/5 sin dficit sensitivo
ro no suficiente, para detectar a este tipo de Behav Sci Law 2004; 22: 325-43. asociado y con reflejos osteotendinosos hipo-
criminal. 2. Hare RD. Without conscience: the disturbing activos. No existan signos de irritacin me-
Adems de los defectos en las funciones world of the psychopaths among us. New nngea y el paciente se encontraba afebril. El
ejecutivas, en cada uno de los dos casos exis- York: Guilford; 1983. paciente ingres para estudio, y se realizaron
ten otros factores aadidos, aunque franca- 3. Beasley JO. Serial murder in America: case pruebas de neuroimagen tomografa axial
mente diferentes, que pueden haber contribui- studies of seven offenders. Behav Sci Law computarizada (TAC) craneal y resonancia
do a la conducta criminal. En el primer caso, 2004; 22: 395-414. magntica (RM) craneal cuyos resultados re-
resulta evidente una enfermedad del sistema 4. Warren JI, Hazelwood RR, Dietz PE. The flejaron una lesin subcortical frontal izquier-
nervioso: hay antecedentes de traumatismo cra- sexually sadistic serial killer. J Forensic Sci da de 2,5 cm de dimetro con edema perile-
neoenceflico, epilepsia, anormalidades mor- 1996; 41: 970-4. sional asociado, compatible con metstasis
folgicas en las imgenes cerebrales, sensa- 5. Myers WC. Serial murder by children and cerebral nica. En la TAC craneal, la tumora-
cin de debilidad y parestesias en la pierna iz- adolescents. Behav Sci Law 2004; 22: 357-74. cin mostraba reas de hiperdensidad con re-
quierda (es llamativa la asociacin existente 6. Wilson W, Hilton T. Modus operandi of female lacin a una probable hemorragia intratumo-
entre estas dos ltimas manifestaciones y las serial killers. Psychol Rep 1998; 82: 495-8. ral. En los estudios de RM cerebral desataca-
eyaculaciones espontneas que presenta el pa- 7. Godwin GM. Hunting serial predators: a ba la captacin heterognea de contraste, as
ciente, lo que indicara la existencia de una multivariate classification approach to profil- como la estrecha relacin con el asta ventricu-
afeccin frontal mesial). Se encuentran, ade- ing violent behavior. Boca Raton, FL: CRC lar frontal izquierda. El sistema ventricular
ms, limitaciones intelectuales importantes y Press; 2000. mostraba un tamao y una configuracin nor-
antecedentes de psicosis. En el segundo caso, 8. Ostrosky-Sols F, Vlez-Garca A, Santana- males (Fig. 1). A las 48 horas del ingreso, el
hay dos elementos que pueden haber contri- Vargas D, Prez M, Ardila A. A middle-aged paciente present fiebre mayor de 38 C aso-
buido potencialmente a su conducta delictiva: female serial killer. J Forensic Sci 2008; 53: ciada a desorientacin y somnolencia. Se to-
los antecedentes de graves abusos durante su 1223-30. maron hemocultivos en los que se inform del
infancia y su adolescencia, y los de luchadora 9. Flores-Lzaro JC, Ostrosky-Sols F, Lozano crecimiento de bacilos grampositivos, que se
durante cerca de 13 aos. Aunque no se han A. Batera de funciones frontales y ejecuti- consideraron como probables contaminantes.
referido accidentes o traumatismos importan- vas: presentacin. Revista de Neuropsicolo- El paciente mejor inicialmente tras iniciar el
tes durante este perodo de su vida, evidente- ga, Neuropsiquiatra y Neurociencias 2008; tratamiento antibitico con cefotaxima. Al da
mente hubo una exposicin permanente y pro- 8: 141-58. siguiente, el paciente experiment en horas un
logada a diversos traumatismos craneales. Es 10. Bergvall AH, Wessely H, Forsman A, Hansen profundo deterioro del nivel de consciencia
conocido que la prctica de la lucha libre pue- S. A deficit in attentional set-shifting of violent hasta el coma escala de coma de Glasgow
de asociarse con diversas anormalidades cere- offenders. Psychol Med 2001; 31: 1095-105. (GCS) 5, respuesta ocular 1, verbal 1, moto-
brales [12]. Podra establecerse la hiptesis de 11. Blair KS, Newman C, Mitchell DG, Richell ra 3, por lo que precis intubacin y soporte
que el efecto acumulado de 13 aos de trau- RA, Leonard A, Morton J, et al. Differentiat- vital. Se realiz una nueva TAC craneal (Fig.
matismos enceflicos menores pueda haber ing among prefrontal substrates in psychopa- 2) que evidenci hidrocefalia aguda tetraven-
dado como resultado alguna disfuncin fron- thy: neuropsychological test findings. Neuro- tricular, por lo que se le traslad a nuestro
tal, asociada con las conductas criminales de psychology 2006; 20: 153-65. centro, donde fue intervenido de urgencia pa-
la paciente. De hecho, las conductas crimina- 12. Toth C, McNeil S, Feasby T. Central nervous ra implantarle una derivacin ventriculoperi-
les slo se han observado en la ltima etapa de system injuries in sport and recreation: a sys- toneal de lquido cefalorraqudeo (LCR). El
su vida. tematic review. Sport Med 2005; 35: 685-715. anlisis microbiolgico del LCR tomado du-
rante la ciruga puso de manifiesto un creci-
En resumen, aunque se han comunicado perfi- miento de L. monocytogenes. A pesar de exte-
les habituales de asesinos en serie (p. ej., hom- riorizar el drenaje ventriculoperitoneal e ins-
bres jvenes con antecedentes delictivos), di- Ventriculitis por taurar terapia antibitica especfica con ampi-
chos perfiles no siempre se encuentran y, de cilina y gentamicina intravenosa, el paciente
hecho, existe una diferenciacin importante Listeria monocytogenes permaneci con bajo nivel de consciencia
en cuanto a las caractersticas personales, psi- (GCS 7), y falleci a los tres das por deterio-
colgicas y sociales de estos asesinos. Sobre Listeria monocytogenes es un bacilo grampo- ro gasomtrico en relacin con una probable
la base de los dos casos analizados, podra sitivo de replicacin intracelular y extracelular neumona aspirativa.
REV NEUROL 2009; 48 (3) 163

REVIEW ARTICLE
published: 23 June 2010
EVOLUTIONARY NEUROSCIENCE doi: 10.3389/fnevo.2010.00007
On the evolution of calculation abilities

Alfredo Ardila*
Edited by: Some numerical knowledge, such as the immediate recognition of small quantities, is observed
J. M. Davis, University of Georgia,
in animals. The development of arithmetical abilities found in mans evolution as well as in childs
USA; University of Macau, China
development represents a long process following different stages. Arithmetical abilities are
Reviewed by:
Anthony Dugbartey, University of relatively recent in human history and are clearly related with counting, i.e., saying aloud a series
Victoria, Canada of number words that correspond to a collection of objects. Counting probably began with
Tomone Takahashi, Shinshu University, finger sequencing, and that may explain the 10-base found in most numerical systems. From
Japan
a neuropsychological perspective, there is a strong relationship between numerical knowledge
Kelly Robinson, Georgia State
University, USA and finger recognition, and both are impaired in cases of left posterior parietal damage (angular
*Correspondence: or Gerstmanns syndrome). Writing numbers appeared earlier in human history than written
Alfredo Ardila, Department of language. Positional digit value is clearly evident in Babylonians, and around 1,000 BC the zero
Communication Sciences and was introduced. Contemporary neuroimaging techniques, specifically fMRI, have demonstrated
Disorders, NHS 431B, Florida
that the left parietal lobe, particularly the intraparietal sulcus, is systematically activated during
International University, Miami, FL
33182, USA. a diversity of tasks; other areas, particularly the frontal lobe, are also involved in processing
e-mail: ardilaa@fiu.edu numerical information and solving arithmetical problems. It can be conjectured that numerical
abilities continue evolving due to advances in mathematical knowledge and the introduction
of new technologies.
Keywords: calculation abilities, numerical knowledge, calculation origins, brain evolution
INTRODUCTION Mechner, 1958; Capaldi and Miller, 1988). It could be conjectured

Calculation abilities have followed a long process from the initial that pigeons and rats can count, at least up to a certain quantity;
quantification systems up to modern algebra, geometry, and phys- that is, they can recognize how many times a motor act to peck
ics. Some rudimentary numerical concepts are observed in animals, on a board or to press a lever has been repeated. Whether or not
and there is no question that pre-historic man used some quanti- this behavior can really be interpreted as counting is nonetheless
fication. However, the ability to represent quantities, the develop- questionable. However, this behavior is observed, at least after a
ment of a numerical system, and the use of arithmetical operations long and painstaking training. Nonetheless, these animal responses
are found only in old civilizations. (to peck or to press the lever) are not precise but just approximate.
This paper reviews the evolution of calculation abilities, includ- In other words, when the rat is required to press the lever seven
ing numerosity and counting in non-human animals, calculation times, the rat presses it about seven times (i.e., 5, 6, 7, 8 times).
abilities in primitive and modern humans, and links between lan- As Dehaene (1997) emphasizes, for an animal, 5 plus 5 does not
guage and number concepts throughout human history. In addi- make 10, but only about 10. According to him, such fuzziness in
tion, the paper reviews contemporary studies of the neurological the internal representation of numbers prevents the emergence of
substrates of numerical abilities and discusses the implications exact numerical arithmetical knowledge in animals. Using highly
of technological advances with regard to continued evolution of controlled and sophisticated designs, it has been pointed out that
these abilities. chimpanzees can even use and add simple numerical fractions (e.g.,
1/2 1/4 = 3/4) (Woodruff and Premack, 1981). These observa-
NUMERICAL CONCEPTS IN ANIMALS tions support the assumption that some quantity concepts can be
The origin of mathematical concepts can be traced to sub-human found in different animals.
species. Throughout recent history different reports have argued Counting (or rather, approximately counting) motor responses
that animals (horses, rats, dogs, chimpanzees, dolphins, and even is just a motor act as is walking or running. Counting lever press-
birds) can use numerical concepts and perform arithmetical opera- ings is not very different from estimating the effort (e.g., number
tions. Some of these reports represent evident charlatanry directed of steps or general motor activity) required in going from one
to the general public. Some others, however, are rigorous and highly point to another. Counting in such a case could be linked to some
controlled scientific studies (e.g., Rugani et al., 2009). proprioceptive and kinesthetic information.
There is, in general, agreement that some rudimentary numeri- In the human brain Kansaku et al. (2006) identified a network
cal concepts are observed in animals. These basic numerical skills of areas involved in enumerating small numbers of auditory, visual,
can be considered as the real origin of the calculation abilities found and somatosensory stimuli, and in enumerating sequential move-
in contemporary man. For instance, pigeons can be trained to peck a ments of hands and feet, in the bilateral premotor cortex, presup-
specific number of times on a board, and rats can be trained to press plementary motor area, posterior temporal cortex, and thalamus.
a lever a certain amount of times to obtain food (Koehler, 1951; The most significant consistent activation across sensory and
Frontiers in Evolutionary Neuroscience www.frontiersin.org June 2010 | Volume 2 | Article 7 | 1

Ardila Evolution of calculation abilities
motor counting conditions was observed in the lateral premotor arithmetic proficiency depends on a more extended activation than
cortex. Lateral premotor activation was not dependent on move- the network found in beginners. In expert individuals with solid,
ment preparation, stimulus presentation timing, or number word extensive mathematical training, specific brain activation changes
verbalization. Furthermore, movement counting, but not sensory are also observed (Zamarian et al., 2009).
counting, activated the anterior parietal cortex. Oneness, twoness, and threeness seemingly are basic perceptual
Not only chimpanzees, but also rats and many other animals, can qualities that our brain can distinguish and process without count-
distinguish numerosity (i.e., global quantification); for instance, ing. It can be conjectured that when pre-historic humans began to
they prefer a bowl containing a larger number of nutritive elements speak, they may have been able to name only the numbers one, two,
(such as chocolates or pellets) when selecting between two bowls and three, corresponding to specific perceptions. To name them was
containing different amounts (Davis and Perusse, 1988). It may be probably no more difficult than naming any other sensory attribute
conjectured that global quantification (numerosity perception) and (Dehaene, 1997). Of note, all world languages can count up to three,
counting (at least the approximate counting of motor responses) even though three may represent many, several, or a lot (Hurford,
represent kinds of basic calculation abilities found at the animal 1987).One is obviously the unit, the individual (the speaker may also
level. Rats prefer a bowl containing 20 pellets to a bowl containing be one). Two conveys the meaning of another (for example, in
only 10 pellets; however, they do not prefer a bowl containing 20 English and also in Spanish, second is related with the verb to sec-
pellets to a bowl containing 21 pellets. Obviously, numerosity per- ond and the adjective secondary).Three may be a residual form of
ception is related to the size and shape of the visual image projected a lot,beyond the others, or many (for example,troppo, which in
to the retina. It can be assumed that 20 pellets in a bowl result in Italian means too much, is seemingly related with the word three
a larger and more complex retinal image than 10 pellets. But the -tre). In the original Indo-European language, spoken perhaps some
visual image corresponding to 20 pellets is difficult to distinguish 1500020000 years ago, apparently the only numbers were one, one
from the visual image corresponding to 21 pellets. and another (two), and a lot, several, or many (three) (Dehaene,
1997). Interestingly, in some contemporary languages, two different
CALCULATION ABILITIES IN PRE-HISTORIC MAN plurals are found: a plural for small quantities (usually two, sometimes
Chimpanzees are capable of various forms of numerical compe- three and four) and a second plural for larger quantities; for instance,
tence, including some correspondence constructions (that is, com- in Russian, one house is odin dom, two, three, or four houses is
paring two collections of elements) for low quantities (Premack, dva, tri, cheterye doma but five houses is pyat domov.
1976; Davis and Perusse, 1988). Most likely, these numerical abilities Of note, in different world languages, the word one does not
also existed in pre-historic man. Homo sapiens antecessors may have any apparent relationship with the word first; and the word
have been capable of using correspondence constructions in some two is also not related with the word second. Three may some-
social activities, such as food sharing. It has been proposed that times, but not always, hold some relationship with third. Beyond
Homo habilis (ancestor of Homo erectus, living about 2.3 million three, ordinals are clearly associated with cardinal numbers. The con-
to 1.4 million years ago) needed to use correspondence construc- clusion is obvious: for small quantities (one, two, three), cardinals and
tions when butchering large animal carcasses (Parker and Gibson, ordinals must have a different origin. For larger quantities, ordinal
1979). Distributing pieces of a divided whole (e.g., prey) into equal numbers are derived from cardinals. As a matter of fact, one/first and
parts required the ability to construct one-to-one correspondences. two/second correspond to different conceptual categories.
Probably, Paleolithic man was able to match the number of objects It may be speculated that for pre-historic man the first per-
in different groups and, eventually, the number of objects in a col- son and the second person in a line (or the first animal and the
lection with the number of items in some external cue system, e.g., second animal during hunting, or other similar concepts) do
fingers or pebbles (incidentally, calculus means pebbles). not seem to be related with the number one and the number
The immediate recognition of certain small quantities is found two. For small children first has the meaning of initial (e.g.,
not only in animals, but also in small children. Animals and children I go first) whereas second is related to later or after (e.g.,
can readily distinguish one, two, or three objects (Fuson, 1988; you go second). These words have a temporal and also spatial
Wynn, 1990, 1992; Cook and Cook, 2009). Antell and Keating meaning, but not an evident numerical meaning. The associa-
(1983) observed that newborn infants were able to discriminate tion between one and first, and two and second seems
among visual stimulus arrays consisting of few a dots. It was found a relatively advanced process in the development of numerical
that infants were able to discriminate between small numbers (2 concepts. That is, the numerical meaning of first and second
versus 3) but not for larger sets. This ability for discriminating and seems to appear after its temporal and spatial meaning. The
also representing and remembering small numbers of items has also association between ordinals and cardinals becomes evident only
been reported by other authors (e.g., Starkey and Cooper, 1980). for larger quantities (more than three) and seems to represent a
Interestingly, evoked potentials at 3 months are already capable later acquisition in human evolution and the complexization of
of marking changes in the nature and number of a set of objects, numerical concepts. Moreover, in many contemporary languages
and these activation changes relate to the parietal lobe (Dehaene (e.g., the Huitoto language, spoken by the Huitoto Indians in the
and Dehaene-Lambertz, 2009). Noteworthy, in normally develop- Amazonian jungle1) there are no ordinal numbers. For first,
ing children and adults, the increase in arithmetic competence is the Huitoto language uses the beginning; to express second
associated with shift of activation from frontal brain areas to pari- the word another is used.
etal areas. A shift of activation is also observed within the parietal
lobe from the intraparietal sulci to the left angular gyrus; experts http://indian-cultures.com/Cultures/huitoto.html
1

Arithmetical abilities are clearly related with counting. 16: aniha-kae-behe kae-taxu-behe kaetaxuwusito (two hands, one
Counting not simply recording the approximate amount of foot, and one toe)
motor responses required for obtaining reinforcement, but actu- 20: aniha-kabe-behe aniha-taxu-behe (two hands and two feet)
ally saying aloud a series of number words that correspond to
a collection of objects is relatively recent in human history. Fingers are named according to their order in counting (as men-
Counting also occurs relatively late in child development. In tioned above, counting begins always with the little finger of the left
human history, as well as in child development, counting using hand). The Sikuani language possesses number words only up to
number words begins with sequencing the fingers (i.e., using a three (kae, aniha-behe, akueyabi). Four (penayanatsi = accompa-
correspondence construction) (Hitch et al., 1987). The name of nied, together) represents a correspondence construction. Strictly
the finger and the corresponding number can be represented using speaking, the Sikuani language counts only up to three. From four to
the very same word (that means the very same word is used for twenty, they use a correspondence construction, not really counting;
naming the thumb and the number one; the very same word is and for higher quantities, they resort to a global quantification.
used to name the index finger and the number two, etc.). The fin- Sometimes not only the fingers (and toes) but also other body
gers [and toes; as a matter of fact, many languages such as Spanish segments may be used in counting: the wrist, the shoulders, the knees,
use a single word (dedo) to name both the fingers and toes] are etc. (Levy-Bruhl, 1910/1947; Cauty, 1984; Dansilio, 2008). However,
usually sequenced in a particular order. This strategy represents a sequencing the fingers (and toes) represents the most universal pro-
basic procedure found in different ancient and contemporary cul- cedure in counting. Some languages (e.g., some Mayan dialects and
tures around the world (LevyBruhl, 1910/1947; Cauty, 1984; Klein Greenland Eskimo) use the same word to denote the number 20
and Starkey, 1987; Dansilio, 2008). Interestingly, it has been dem- (that is, all the fingers and all the toes) and a person.
onstrated that children with low arithmetical skills also present In different Amerindian languages, for higher than 10 or 20
a finger misrepresentation on the Draw-a-Person test (Pontius, figures, most often many is used (global quantification princi-
1989). This observation has been confirmed in different cultural ple) (Cauty, 1984; Ifrah, 2000). Or, they can refer to other peoples
groups. By the same token, difficulty in recognizing and nam- hands (correspondence construction) (e.g., thirty-five might be
ing the fingers represents a reliable predictor of developmental something like my two hands, my two feet, my fathers two hands,
dyscalculia (Kaufmann, 2008). my fathers one foot). As mentioned, twenty sometimes becomes
Taking a typical example as an illustration, the Colombian something like one person, a sort of higher order numeral. It
Sikuani or Guahibo Amazonian jungle Indians2 count in the fol- is interesting to note that in some contemporary languages (like
lowing way: the person (an adult when counting or a child when English and Spanish) one means the unit, but it is also used as
learning to count) places his/her left hand in supination; to point a sort of indefinite personal pronoun. In English and Spanish we
to number one, the right index points to the left little finger, which can also use one as synonymous with myself . Twenty is found
is then bent (Queixalos, 1989). The order followed in counting is to be the base number in the Mayan numerical system (Swadesh,
always from the little finger to the index. To point to number five, 1967; Cauty, 1984). In many contemporary languages, a 10 and/
the hand is turned and the fingers opened; for six, both thumbs or 20 base is evident.
are joined, the left fingers are closed, and the right opened; they are Digit (from Latin digitus) means both number and finger.
opened one after the other for seven, eight, nine and ten. Between 11 The correspondence construction between numbers and fingers is
and 20, the head points to the feet and the sequence is re-initiated. evident. Latin number notation was originally Etruscan (Turner,
The lexicon used is: 1984) and referred (as in other languages) to the fingers. One, two,
and three were written simply by making vertical strokes. In four,
1:
kae (the unit, one) the Latin system resorts to a simplification. Originally, four was
2:
aniha-behe (a pair, both) written IIII, but later on it became IV. Five (V) represented the
3:
akueyabi whole hand with the arm bent (that is, all the fingers of the hand),
4:
penayanatsi (accompanied; that is, the fingers together) and ten (X) the two arms crossed.
5:
kae-kabe (one hand) From a neuropsychological perspective, a strong relation-
Numbers from six to nine, are formed with one hand and (a ship between numerical knowledge, finger gnosis, and even lat-
certain number) of fingers. Ten becomes two hands: eral (rightleft) knowledge is evident (Ardila and Rosselli, 2002;
6: kae-kabe kae-kabesito-nua (one hand and one finger) Kaufmann, 2008). Finger agnosia (and probably rightleft discrimi-
7: kae-kabe anih-akabesito-behe (one hand and a pair of fingers) nation disturbances) could be interpreted as a restricted form of
10: anih-akabe-behe (two hands) autotopagnosia (inability to recognize or localize the various body
Two hands is maintained between 10 and 20. Toes (taxawu- parts) (Ardila et al., 1989, 2000).
sito) are added between 11 and 14; and one foot (kaetaxu) is It is not surprising to find that a decimal (or vigesimal, i.e., with
used in 15. Twenty is two hands together with two feet: a base of 20) system has been most often developed. Simultaneously
11: aniha-kabe-behe kae-taxuwusito (two hands and one toe) or very close in time, decimal systems appeared in different coun-
12: aniha-kabe-behe aniha-tuxuwusito-behe (two hands and two tries (Sumer, Egypt, India, and Crete). Different symbols were used
toes) to represent 1, 10, 100, and 1000 (Childe, 1936; Dansilio, 2008).
15: aniha-kabe-behe kae-taxu-behe (two hands and one foot) There is, however, an interesting and intriguing exception:
Sumerian and later Babylonians (about 2,000 BC) developed not
2
http://indian-cultures.com/Cultures/guahibo.html only a decimal but also a sexagesimal system: a symbol represented

60 or any 60-multiple and another different symbol represented the As pointed out, recognition of individual marks or elements
number 10 and any 10-multiple. Thus, for example, the number up to three is easy: It represents an immediate perception read-
173 was then represented: 2 s 60 (the symbol for 60 repeated ily recognizable. Beyond three, the number of marks (strokes
twice) 5 s 10 (the symbol for 10 repeated five times) 3 (a sym- or dots) usually has to be counted and errors are more likely.
bol for units repeated three times). A base of 60 has remained for Furthermore, it is rather time-consuming and cumbersome to
some contemporary time measures (e.g., minutes and seconds). be constantly counting marks. Interestingly, the different digit
Twelve is also frequently maintained as a second-order unit (e.g., notational systems always represent one, two and three with
a dozen). Evidently, 60 results from five times twelve. Five obvi- strokes (or points, or any specific mark). In other words, the
ously is one hand, and the question becomes, where does 12 come numbers one, two and three are written making one, two or
from? What are the two additional units? It might be speculated three strokes. But beyond these figures, digit writing may give
that 12 means the 10 fingers plus the two feet or even the two way to other strategies. In our Arabic digit notation system,
elbows or the two shoulders or the two knees (individuality of one is a vertical line whereas two and three were originally
components is easier to appreciate in the hands than in the feet). horizontal lines that became tied together by being handwritten
But this is only speculation, although it is feasible according to our (Ifrah, 2000). This observation may be related with the inborn
knowledge about counting procedures used in different cultural ability to perceptually recognize up to three elements. Beyond
groups (Levy-Bruhl, 1910/1947; Ifrah, 2000; Dansilio, 2008). three, errors become progressively more likely. Perceptually dis-
Maya Indians developed a similar system, but had 20 as a base tinguishing eight and nine strokes is not as easy as distinguishing
(Leon-Portilla, 1986). They used different symbols to represent 20, between two and three strokes. The introduction of a different
400 (20 s 20), and 8000 (20 s 20 s 20) (Cauty, 1984). representation for quantities over three was a useful and practi-
Thus, reviewing the history of numerical concepts, it is found cal simplification.
that world languages developed a base 10 (10 fingers) or 20 (10 The numerical system, along with measurement units, were
fingers plus 10 toes) or even five (five fingers) to group quantities. developed departing from the body dimensions (fingers, hands,
In some contemporary languages a residual 20-base can be found arm, steps, etc.). This tendency to use the human body not only to
(e.g., in French 80 can be four twenties). In many contemporary count but also as measure units is still observed in some contem-
languages, different words are used between 1 and 10. Between 10 porary measurement units (e.g., foot).
and 20, the numerical systems usually become irregular, unpre- Adding, subtracting, multiplying and dividing were possible
dictable, and idiosyncratic. From 20 onward, numbers are formed in the Egyptian system, but of course, following procedures quite
simply with the words twenty plus one, twenty plus two, etc. different from those procedures we currently use. Egyptians based
Some contemporary languages still use a five-base in counting. For multiplication and division on the duplication and halving
instance, in the Amerindian language Tanimuca in South America3, method (Childe, 1936). Interestingly, this very same procedure
speakers count up to five. Between five and 10, numbers are five (duplicating and halving quantities) is also observed in illiter-
one, five two, and so on. ate people when performing arithmetical operations. Thus, to
multiply 12 s 18 in the Egyptian system the following procedure
FURTHER DEVELOPMENTS OF ARITHMETICAL ABILITIES was followed:
Writing numbers appeared earlier in history than writing language. 1 18
Some cultures (e.g., Incas) developed a number representing sys- 2 36
tem, but not a language representing system (Swadesh, 1967). As *4 72
mentioned before, calculus means pebble. Pebbles, marks, knots, *8 144
or any other element were used as a correspondence construction to Total 216
record the number of elements (people, cows, fishes, houses, etc). In (The number 18 is duplicated one or several times, and the
Sumer, the first number writing system has been found (about 3,000 amounts corresponding to 12 (4 8 in this example) are selected
BC) (Childe, 1936; Ifrah, 2000): Instead of using pebbles, fingers, and summed up: 72 144 = 216. To divide, the inverse proce-
or knots, it was simpler just to make a mark (a stroke or a point) dure was used. Therefore, the procedure used to divide 19 by 8
on the floor, on a tree branch, or on a board if one wanted to keep would be:
the record. In Egypt, India and later in Crete, a similar system was 1 8
developed: units were represented by a conventional symbol (usu- *2 16
ally a stroke) repeated several times to mean a digit between one 2 4
and nine; a different symbol was used for 10 and 10-multiples. *4 2
Positional digit value is clearly evident in Babylonians, and *8 1
around 1,000 BC the zero was introduced. Positional value and zero That is, 2 4 8 (2 1/4 1/8), which is 2.375
are also evident in Maya Indians (Leon-Portilla, 1986). Egyptians In brief, different steps were followed in the development of
and Babylonians commonly used fractions. Small fractions (1/2, arithmetical abilities: representing quantities, initially as corre-
1/3, and 1/4) are relatively simple numerical concepts, and even spondence constructions (i.e., one mark represents one element),
chimpanzees can be trained to use small fractions (Woodruff and later new marks were added to represent larger quantities (usually
Premack, 1981). 10); providing a positional value to the marks; using fractions; and
computing quantities (adding and subtracting; and later multiply-
3
http://www.ethnologue.com/show_family.asp?subid=931-16 ing and dividing).

THE NEUROSCIENCE OF CALCULATION ABILITIES examined the ability to enumerate sets of stimuli, and another two
Since primary acalculia (a basic defect in computational ability) was experiments required judgments about two concurrently presented
initially described by Henschen (1925) it has been associated with stimuli that were either identically coded (i.e., two Arabic numerals,
left posterior parietal damage (e.g., Mazzoni et al., 1990; Ardila et al., two number words, or two arrays of dots) or differently coded (e.g.,
2000; Mayer et al., 2003). Furthermore, it was suggested that differ- an Arabic numeral and a number word). Both hemispheres were
ent cerebral pathways are responsible for processing rote numerical equally able to enumerate stimuli and make comparisons between
knowledge (e.g., multiplication tables) and semantic knowledge of numerical representations regardless of stimuli coding. However,
numerical quantities. Dehaene and Cohen (1997) have proposed the left hemisphere was more accurate than the right when the task
that a left subcortical network contributes to the storage and retrieval involved number words.
of rote verbal arithmetical facts, whereas an inferior parietal network Cohen et al. (2000) reported a patient with a lesion in the left per-
is dedicated to the mental manipulation of numerical quantities. isylvian area who showed a severe impairment in all tasks involving
Neuroimaging techniques (e.g., fMRI) have been used to analyze numbers in a verbal format, such as reading aloud, writing to dicta-
the pattern of brain activity during diverse calculation tasks. It tion, or responding verbally to questions of numerical knowledge.
has been demonstrated that different brain areas are active dur- In contrast, her ability to manipulate non-verbal representation of
ing arithmetical tasks, but the specific pattern of brain activity numbers, i.e., Arabic numerals, was comparatively well preserved.
depends on the particular type of task that is used. At minimum, This observation supports the proposal that language and calcula-
the following brain areas become activated during calculation: the tion disorders can be dissociated (Basso et al., 2000). Some authors
upper cortical surface and anterior aspect of the left middle frontal have assumed that language and numerical concepts are differently
gyrus (Burbaud et al., 1995); the supramarginal and angular gyrus organized in the brain and follow distinct developmental patterns
(bilaterally) (Rueckert et al., 1996); the left dorsolateral prefron- in children (Gelman and Butterworth, 2005). Other authors have
tal and premotor cortices, Brocas area, and the inferior parietal suggested that calculation and language are mediated by partially
cortex (Burbaud et al., 1999); and the left parietal and inferior different and also partially overlapped brain systems (Baldo and
occipitotemporal regions (lingual and fusiform gyri) (Rickard Dronkers, 2007). Semenza et al. (2006) have emphasized that cal-
et al., 2000). The diversity of brain areas involved in arithmetical culation and language usually share the same hemisphere.
processes supports the assumption that calculation ability repre- In summary, diverse brain areas are activated during the per-
sents a multifactor skill, including verbal, spatial, memory, body formance of different arithmetical tasks, although contemporary
knowledge, and executive function abilities (Ardila and Rosselli, evidence suggests that the intraparietal sulcus seems to the play the
2002). Dehaene et al. (2004), however, proposed that regardless most crucial role. During the learning of new calculation abilities,
of the diversity of areas that become active during arithmetical additional brain areas become involved and the specific pattern of
tasks, the human ability for arithmetic is associated with activation brain activity depends on the particular type of test that is used.
of very specific brain areas, in particular, the intraparietal sulcus. It can be assumed that during human history, the development
Neuroimaging studies with humans have demonstrated that the of new numerical abilities was correlated with the involvement
intraparietal sulcus is systematically activated during a diversity of of new brain areas during the performance of progressively more
number tasks and could be regarded as the most crucial brain region complex numerical tasks.
in the understanding and use of quantities (Ashkenazi et al., 2008).
These observations have been supported using brain electrostimu- CONCLUSION
lation (Roux et al., 2009). Other brain areas, such as the precentral Arithmetical abilities and number representation have existed
area and the inferior prefrontal cortex, are also activated when for only some 50006000 years. Most likely, during the Stone-
subjects engage in mental calculations. Roca (2009) has proposed Age, only simple counting up to three and, of course, bigger
that there exists a frontoparietosubcortical circuit responsible for and smaller (magnitude judgment) concepts were present.
complex arithmetic calculations and that procedural knowledge Global quantification is observed at the pre-human level.
relies on a visuo-spatial sketchpad that contains a representation Correspondence constructions allowed increasing the amount
of each sub-step of the procedure. of numbers to be used. The most immediate correspondence
Traditionally, calculation defects have been associated with pos- construction is performed with the fingers. Finger knowledge
terior left parietal damage (Henschen, 1925; Boller and Grafman, and counting represent, to a certain extent, the same cognitive
1983), and frequently included in the so-called Gerstmanns (or ability, as is still evident in some contemporary languages, such
angular gyrus) syndrome. Symptoms of Gerstmanns syndrome as Sikuani.
(acalculia with agraphia, disorders in right-left orientation, and Counting, finger gnosis, and even lateral spatial knowledge may
finger agnosia) (Gerstmann, 1940) can be found during direct cor- present a common historical origin. Seemingly, calculation abilities
tical stimulation in the angular gyrus region (Roux et al., 2003). were derived from finger sequencing. Number representation and
Using fMRI, it has been observed that the left angular gyrus is not arithmetical operations are observed only since some 50006000
only involved in arithmetic tasks requiring simple fact retrieval, years ago. Currently, calculation abilities are rapidly evolving due
but may show significant activations as a result of relatively short to the introduction of modern technology and resulting cognitive
training of complex calculation (Delazer et al., 2003). demands. Right-left discrimination (as well as the use of other
Colvin et al. (2005) investigated numerical abilities in a split- spatial concepts) most likely was present in pre-historic man,
brain patient using experiments that examined the hemispheres because requirements of spatial abilities may have been very high,
abilities to make magnitude comparisons. One experiment even higher than in contemporary man (Hours, 1982; Ardila and

Ostrosky, 1984; Ardila, 1993). Rightleft discrimination and finger problems. Also, it has been suggested that there exists a fronto
gnosis are strongly interdependent, and even they can be inter- parietosubcortical circuit responsible for complex arithmetic
preted as components of the autotopagnosia syndrome. It seems, in calculations and procedural knowledge.
consequence, that there is a rationale for finding a common brain It can be conjectured that numerical abilities continue evolving
activity for finger gnosis, calculation, and right-left discrimination due to advances in mathematical knowledge and the introduc-
(and in general, spatial knowledge mediated by language) (Ardila tion of new technologies; for instance, instead of writing numbers
and Rosselli, 2002). down on paper and applying certain computational rules, we more
Contemporary neuroimaging techniques, specifically fMRI, have often require the ability to use a pocket calculator or a compu-
demonstrated that the left parietal lobe, particularly the intrapari- ter program. Doubtless, computation ability is evolving in a new
etal sulcus is systematically activated during a diversity of number direction. Brain representation of calculation may be taking a new
tasks; other areas, particularly the frontal lobe, are also involved direction, and even the acalculia syndrome may present different
in processing numerical information and solving arithmetical clinical manifestations.
REFERENCES Burbaud, P., Degreze, P., Lafon, P., Franconi, phylogense et ontogense. Bull. Acad. Klein, A., and Starkey, P. S. (1987). The
Antell, S. E., and Keating, D. (1983). J. M., Bouligand, B., Bioulac, B., Caille, J. Natl. Med. 193, 883889. origins and development of numerical
Perception of numerical invariance by M., and Allard, M. (1995). Lateralization Dehaene, S., Molko, N., Cohen, L., and cognition: A comparative analysis, in
neonates. Child Dev. 54, 695701. of prefrontal activation during inter- Wilson, A. J. (2004). Arithmetic and Cognitive Processes in Mathematics, eds
Ardila, A. (1993). On the origins of cal- nal mental calculation: a functional the brain. Curr. Opin. Neurobiol. 14, J. A. Sloboda and D. Rogers (Oxford:
culation abilities. Behav. Neurol. 6, magnetic resonance imaging study. J. 218224. Clearendon Press), 125.
8998. Neurophysiol. 74, 21942200. Delazer, M., Domahs, F., Bartha, L., Koehler, O. (1951). The ability of birds
Ardila, A., Concha, M., and Rosselli, M. Capaldi, E. J., and Miller, D. J. (1988). Brenneis, C., Lochy, A., Trieb, T., and to count. Bull. Anim. Behav. 9, 4145.
(2000). Angular gyrus syndrome revis- Counting in rats: its functional sig- Benke, T. (2003). Learning complex Leon-Portilla, M. (1986). Tiempo y
ited: acalculia, finger agnosia, right-left nificance and the independent cog- arithmetic-an fMRI study. Brain Res. Realidad en el Pensamiento Maya
disorientation, and semantic aphasia. nitive processes that constitute it. J. Cogn. Brain Res. 18, 7688. [Time and reality in Maya think-
Aphasiology 14, 743754. Exp. Psychol. Anim. Behav. Process 14, Fuson, K. C. (1988). Childrens Counting ing]. Mxico: Universidad Nacional
Ardila, A., Lopez, M. V., and Solano, E. 317. and the Concepts of Number. New York: Autnoma de Mxico.
(1989). Semantic aphasia recon- Cauty, A. (1984). Taxonomie, syntaxe et Springer-Verlag. Levy-Bruhl, L. (1910/1947). Las Funciones
sidered, in Brain Organization of economie des numerations parlees. Gelman, R., and Butterworth, B. (2005). Mentales en las Sociedades Inferiores
Language and Cognitive Processes, eds Amerindia 9, 111146. Number and language: how are they [Mental functions in lower societies].
A. Ardila and F. Ostrosky-Solis (New Childe, V. G. (1936). Man Makes Himself. related? Trends Cogn. Sci. (Regul. Ed.) Buenos Aires: Lautaro.
York: Plenum Publishing Company), London: Pitman Publishing. 9, 610. Mayer, E., Reicherts, M., Deloche, G.,
77193. Cohen, L., Dehaene, S., Chochon, F., Gerstmann, J. (1940). The syndrome of fin- Willadino-Braga, L., Taussik, I.,
Ardila, A., and Ostrosky, F. (1984). Lehericy, S., and Naccache, L. (2000). ger agnosia, disorientation for right and Dordain, M., Van der Linden, M.,
Some final remarks, in The Language and calculation within left, agraphia and acalculia. Arch. Neurol. and Annoni, J. M. (2003). Number
Right Hemisphere: Neurology and the parietal lobe: a combined cog- Neurochir. Psychiatr. 44, 398408. processing after stroke: anatomo-
Neuropsychology, eds A. Ardila and nitive, anatomical and fMRI study. Henschen, S. E. (1925). Clinical and clinical correlations in oral and writ-
F. Ostrosky-Solis (London: Gordon Neuropsychologia 38, 14261440. anatomical contributions on brain ten codes. J. Int. Neuropsychol. Soc. 9,
and Breach Science Publishers), Colvin, M. K., Funnell, M. G., and pathology. Arch. Neurol. Psychiatry 13, 899912.
265273. Gazzaniga, M. S. (2005). Numerical 226249. Mazzoni, M., Pardossi, L., Cantini, R.,
Ardila, A., and Rosselli, M. (2002). Acalculia processing in the two hemispheres: Hitch, G., Cundick, J., Haughey, M., Pugh, Giornetti, V., and Arena, R. (1990).
and dyscalculia. Neuropsychol. Rev. 12, studies of a split-brain patient. Brain R., and Wright, H. (1987). Aspects of Gerstmanns syndrome: a case report.
179231. Cogn. 57, 4352. counting in childrens arithmetics, in Cortex 25, 459468.
Ashkenazi, S., Henik, A., Ifergane, G., Cook, J. L., and Cook, G. (2009). Child Cognitive Processes in Mathematics eds Mechner, F. (1958). Probability relations
and Shelef, I. (2008). Basic numerical development: Principles and perspec- J. A. Sloboda and D. Rogers (Oxford: within response sequences under ratio
processing in left intraparietal sulcus tives. Boston: Pearson Allyn and Clearendon Press), 2641. reinforcement. J. Exp. Anal. Behav. 1,
(IPS) acalculia. Cortex 44, 439448. Bacon. Hours, F. (1982). Les Civilisations 109121.
Baldo, J. V., and Dronkers, N. F. (2007). Dansilio, S. (2008). Los trastornos du Pal Nolithique. Paris: Presses Parker, S. T., and Gibson, K. H. (1979). A
Neural correlates of arithmetic and del clculo y el procesamiento del Universitaires de la France. developmental model for the evolu-
language comprehension: a com- nmero. Montevideo: Prensa Mdica Hurford, R. (1987). Language and Number. tion of language and intelligence in
mon substrate? Neuropsychologia 45, Latinoamericana. Oxford: Basil Blackwell. early hominids. Behav. Brain Sci. 2,
229235. Davis, H., and Perusse, R. (1988). Ifrah, G. (2000). The Universal History of the 367408.
Basso, A., Burgio, F., and Caporali, A. (2000). Numerical competence in animals: Numbers. New York: Wiley. Available at: Pontius, A. A. (1989). Color and spatial
Acalculia, aphasia and spatial disor- Definitional issues, current evidence indian-cultures.com/Cultures/huitoto. error in block design in stone-age
ders in left and right brain-damaged and a new research agenda. Behav. html. Retrieved December 17, 2009; Auca Indians: Ecological underuse of
patients. Cortex 36, 265280. Brain Sci. 11, 561615. indian-cultures.com/Cultures/guahibo. occipital-parietal system in men and
Boller, F., and Grafman, J. (1983). Dehaene, S. (1997). The Number Sense. html. Retrieved December 17, 2009. of frontal lobes in women. Brain Cogn.
Acalculia: Historical development How the Mind Creates Mathematics. Kansaku, K., Johnson, A., Grillon, M. L., 10, 5475.
and current significance. Brain Cogn. New York: Oxford University Press. Garraux, G., Sadato, N., and Hallett, Premack, D. (1976). Intelligence in Ape and
2, 205223. Dehaene., S., and Cohen, L. (1997). M. (2006). Neural correlates of Man. Hillsdale, NJ: Erlbaum.
Burbaud, P., Camus, O., Guehl, D., Bioulac, Cerebral pathways for calculation: counting of sequential sensory and Queixalos, F. (1989). Numeracin tradi-
B., Caille, J. M., and Allard, M. (1999). double dissociation between rote motor events in the human brain. cional Sikuani. [Traditional Sikuani
A functional magnetic resonance verbal and quantitative knowledge of Neuroimage 31, 649660. numbering] Glotta, 3, 2831.
imaging study of mental subtraction arithmetic. Cortex 33, 219250. Kaufmann, L. (2008). Dyscalculia: Rickard, T. C., Romero, S. G., Basso, G.,
in human subjects. Neurosci. Lett. 273, Dehaene, S., and Dehaene-Lambertz, G. Neuroscience and education. Educ. Wharton, C., Flitman, S., and Grafman,
195199. (2009). Neuro-imagerie cognitive: Res. 50, 163175. J. (2000). The calculating brain: An

fMRI Study. Neuropsychologia 38, calculation with functional MRI. Turner, R. (1984). The Great Cultural Conflict of Interest Statement: The
325335. Neuroimage 3, 97103. Tradition. The Foundation of author declares that the research was
Roca, E. C. (2009). Arithmetic procedural Rugani, R., Fontanari, L., Simoni, E., Civilization. New York: McGraw- conducted in the absence of any com-
knowledge: a cortico-subcortical cir- Regolin, L., and Vallortigara, G. (2009). Hill. mercial or financial relationships that
cuit. Brain Res. 1302, 148156. Arithmetic in newborn chicks. Proc. Woodruff, G., and Premack, D. (1981). could be construed as a potential conflict
Roux, F. E., Boetto, S., Sacko, O., Chollet, Biol. Sci. 276, 24512460. Primitive mathematical concepts in of interest.
F., and Tremoulet, M. (2003). Writing, Semenza, C., Delazer, M., Bertella, L., the chimpanzee: Proportionality and
calculating, and finger recognition Gran, A., Mori, I., Conti, F. M., numerosity. Nature 293, 568570. Received: 25 January 2010; paper pend-
in the region of the angular gyrus: Pignatti, R., Bartha, L., Domahs, F., Wynn, K. (1990). Childrens under- ing published: 01 March 2010; accepted:
a cortical stimulation study of Benke, T,, and Mauro, A. (2006). Is standing of counting. Cognition 36, 07 June 2010; published online: 23 June
Gerstmann syndrome. J. Neurosurg. math lateralized on the same side as 155193. 2010.
99, 716727. language? Right hemisphere aphasia Wynn, K. (1992). Childrens acquisition Citation: Ardila A (2010) On the evolution
Roux, F. E., Boukhatem, L., Draper, L., and mathematical abilities. Neurosci. of the number words and counting of calculation abilities. Front. Evol. Neurosci.
Sacko, O., and Dmonet, J. F. (2009). Lett. 406, 285288. system. Cogn. Psychol. 24, 220251. 2:7. doi: 10.3389/fnevo.2010.00007
Cortical calculation localization using Starkey, P., and Cooper, R. (1980). Available at: www.ethnologue.com/ Copyright 2010 Ardila. This is an open-
electrostimulation. J. Neurosurg. 110, Perception of numbers by human show_family.asp?subid=931916. access article subject to an exclusive license
12911299. infants. Science 210, 10331035. Zamarian, L., Ischebeck, A., and Delazer, agreement between the authors and the
Rueckert, L., Lange, N., Partiot, A., Swadesh, M. (1967). El Lenguage y La M. (2009). Neuroscience of learning Frontiers Research Foundation, which
Appollonio, I., Litvan, I., Le Bihan, Vida Humana [Language and human arithmetic evidence from brain permits unrestricted use, distribution, and
D., and Grafman, J. (1996). Visualizing life]. Mxico: Fondo de Cultura imaging studies. Neurosci. Biobehav. reproduction in any medium, provided the
cortical activation during mental Econmica. Rev. 33, 909925. original authors and source are credited.

Hindawi Publishing Corporation
Behavioural Neurology
Volume 2015, Article ID 872487, 16 pages
http://dx.doi.org/10.1155/2015/872487
Review Article
A Proposed Neurological Interpretation of Language Evolution
Alfredo Ardila
Department of Communication Sciences and Disorders, Florida International University, Miami, FL 33199, USA
Correspondence should be addressed to Alfredo Ardila; ardilaalfredo@gmail.com
Received 17 December 2014; Revised 15 March 2015; Accepted 17 May 2015
Academic Editor: Joao Quevedo
Copyright 2015 Alfredo Ardila. This is an open access article distributed under the Creative Commons Attribution License, which
permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Since the very beginning of the aphasia history it has been well established that there are two major aphasic syndromes (Wernickes-
type and Brocas-type aphasia); each one of them is related to the disturbance at a specific linguistic level (lexical/semantic and
grammatical) and associated with a particular brain damage localization (temporal and frontal-subcortical). It is proposed that
three stages in language evolution could be distinguished: (a) primitive communication systems similar to those observed in
other animals, including nonhuman primates; (b) initial communication systems using sound combinations (lexicon) but without
relationships among the elements (grammar); and (c) advanced communication systems including word-combinations (grammar).
It is proposed that grammar probably originated from the internal representation of actions, resulting in the creation of verbs;
this is an ability that depends on the so-called Brocas area and related brain networks. It is suggested that grammar is the basic
ability for the development of so-called metacognitive executive functions. It is concluded that while the lexical/semantic language
system (vocabulary) probably appeared during human evolution long before the contemporary man (Homo sapiens sapiens), the
grammatical language historically represents a recent acquisition and is correlated with the development of complex cognition
(metacognitive executive functions).
1. Introduction In spite of the potentially significant contribution that

aphasia knowledge can make towards understanding the
Diverse disciplines have contributed to advancing our under- origin of human language, limited interest has been observed
standing on the origins and evolution of language: lin- in using the aphasia model to approach language evolution
guistics, neuroanatomy, archeology, comparative psychology, [29, 30]. Code [31] has clearly stated that some aspects of
and genetics [127]. As a matter of fact, the origins and aphasic symptomatology may represent fossilized clues to the
evolution of human language represent particularly complex emergence of human language and hence aphasia analysis
and intriguing questions. According to Christiansen and can positively contribute to comprehending the evolution of
Kirby [28] understanding language evolution represents the language. He further proposes that the evolution of lexical
hardest problem in contemporary science. speech automatisms (such as language cliches, overused
This paper does not attempt to further review and discuss social expressions, automatic speech, and the like) to agram-
the origins and evolution of language, but rather to relate matism (a type of language pattern frequently found in severe
the origins of human language, with contemporary cognitive nonfluent aphasia) might also provide useful insights into
neurosciences, particularly with the aphasia area. Given the the early evolution of language. Code [32] suggests that
complexity of the topic, evidence not only from aphasia, commonly occurring lexical speech automatisms may reflect
but also from brain evolution theory, linguistics, genetics, substages of development from single repeated expletive and
anthropology, and psychology will be examined to further the syntactically primitive pronoun + modal/aux constructions,
central idea taken from aphasia literature; that is, there exist forming a bridge to a protosyntax stage, to agrammatism,
two language systems supported by different brain circuits, thus bridging a gap between protolanguage and full syntax
probably appearing at different historical moments in time. (page 143). As a matter of fact, he considers that lexical speech
2 Behavioural Neurology
automatisms could represent some of the earliest utterances Table 1: Different names used to refer to the two basic aphasic
appearing in human evolution. This type of analysis presented syndromes.
by Code [31, 32] clearly illustrates that aphasia can indeed sig-
Receptive Expressive
nificantly advance our understanding about human language
Sensory Motor
evolution.
Ventral Dorsal
In this paper, initially, some fundamental observations
Fluent Nonfluent
about language disturbances in the case of brain pathology
Wernicke-type Broca-type
(aphasia) are reviewed. It is emphasized that throughout the
history of aphasia it has been accepted that there are two
fundamental types of aphasia syndromes. Although these two and cognitive neurosciences, including Brocas aphasia, Wer-
fundamental types of aphasia syndromes have been named nickes aphasia, conduction aphasia, amnesic aphasia, and
in different ways (e.g., motor/sensory; anterior/posterior; transcortical aphasia [4146]. The exact number of aphasia
nonfluent/fluent; etc.) (see Table 1), each one of them is asso- subtypes depends on the particular classification, but usually
ciated with the disturbance of one of two different language between four and seven different aphasic syndromes are
elements (lexicon and grammar). This is a basic distinction mentioned. Seemingly, this suggested diversity of aphasic
that has to be considered when analyzing language evolution. syndromes has obscured the major and basic distinction
Lexicon (vocabulary) and grammar (morphosyntax) are sup- in aphasia: there are only two major aphasic syndromes
ported by different neural networks and can be independently (see Table 1) [47].
impaired in cases of brain damage; hence, they present quite Assuming that there is a significant number of aphasic
different cerebral organization. Interestingly, vocabulary and disturbances (usually between four and seven; sometimes
morphosyntax acquisition are also based on different learn- even more) may result in the implicit hypothesis that
ing types (declarative and procedural learning; [3436]) human language includes diverse discrete abilities, such
and probably emerged at quite different historical moments. as phoneme recognition, lexical memory, morphosyntax,
Integrating this basic distinction in an interpretation about repetition ability, and naming. Each one of these abilities
historical language evolution can significantly advance our would consequently be associated with the activity of a
insight of language evolution. particular cerebral area. These diverse aphasia syndromes
It should be noted that Bickerton [5] has emphasized that (such as Brocas aphasia, conduction aphasia, Wernickes
there are two most central issues in language evolution: (a) aphasia, anomic aphasia, and transcortical sensory aphasia)
how did symbolic units (words or manual signs) evolve? (b) are further regarded as the disturbance of a specific language
How did syntax evolve? He considers that symbolic units (i.e., ability: phoneme recognition, morphosyntax, repetition, and
lexicon) and syntax (i.e., grammar) are the only real novelties so forth. In consequence, it can be conjectured that human
in human communication systems and are therefore the most language is based on seven (some times more) language
important points to approach in a theory on language evo- abilities.
lution. He further explicitly points out there is no reason to It is important to emphasize that since the very beginning
believe that the emergence of the two was either simultaneous of the aphasia history, it has been clearly pointed out that
or due to similar causes, and some good reasons for supposing there are only two basic aphasic syndromes (see Table 1),
the contrary (page 512). To support this argument, he named in different ways, but roughly corresponding to
refers to Chomskys [37] distinction between the conceptual Wernickes-type aphasia and Brocas-type aphasia [4043, 46
and the computational aspects of language. According to 56]. This has been most basic idea throughout the history
this proposal the conceptual elements (conceptual structure, since the very beginning of aphasia analysis. For instance,
lexical instantiation) must be significantly older than any Hippocrates (400 BC) in his pioneer analysis of language
computational mechanism (grammar). However, symbolic impairments associated with brain damage clearly referred
units can be understood in different ways, and depending to two different types of language disturbances: aphonos,
on how they are defined, it could be argued that even they without voice, and anaudos, without hearing. Antonio
exist in animal communication systems [38, 39]. We should Guaneiro during the XV century reported two aphasic
assume that Bickerton refers specifically to the symbolic units patients: one with a fluent paraphasic speech and the other
of human language. Bickerton [5] points out that simple logic one with a nonfluent speech. Later, in 1825 Bouillaud (French
indicates that symbolic units (lexicon) must exist before any physician) distinguished two different types of language
procedure to link these units (grammar). That is, lexicon pathology: one had an articulatory basis, and the other
phylogenetically should have appeared long before grammar. pathology was amnesic in nature. In 1843, Jacques Lordat
This is exactly the point of view that will be argued in this (a professor of anatomy and physiology at Montpellier in
paper. France) proposed a similar dichotomy; he described the
inability to produce words, referred to as verbal asynergy,
2. There are Two Fundamental and the disturbance in the ability to recall words, referred
Aphasia Syndromes to as verbal amnesia [57, 58]. This distinction between two
major language disturbances represents the most basic infor-
Aphasia is generally defined as the loss or impairment of mation in aphasia: aphasia is not a single unified language
language caused by brain damage [40]. Different subtypes disturbance, but two rather different (even opposite) clinical
of aphasia syndromes are often mentioned in neurology syndromes (page 29) [48].
Behavioural Neurology 3
These two fundamental aphasic syndromes are associated inaccessible; sometimes they are replaced by more general
with a disturbance at the level of the language elements words (for instance, instead of dog the patient says animal);
(lexical/semantic) in Wernickes aphasia or at the level of (b) there are difficulties in selecting between semantically
the association between the language elements (morphosyn- related words (cat, dog, horse, fox, etc.), and semantic sub-
tactic/grammatical) in Brocas aphasia. It has been further stitutions (so-called semantic paraphasias) are observed.
observed that these two basic dimensions of language (lex- (c) Frequently, these patients fill out their discourse with
ical/semantic and grammatical) are related to two basic so-called circumlocutions (to go around in speech); for
linguistic operations: selecting (that means the language as a instance, the clock is referred to as the instrument used to
paradigm) (paradigm in linguistics is usually understood as know the time.
a set of linguistic items that form mutually exclusive choices Luria [61] reanalyzed the proposal presented by Jakobson
in particular syntactic roles) and sequencing (that means [5962] and suggested that the selection (paradigmatic)
language as syntagm) [5961]. disorder could potentially be observed at three different
Jakobson [62] suggested that aphasia can involve one language levels; the disturbance in each one of these levels
of two potential types of languages defects: language can would be associated with a specific aphasic syndrome: (a)
be impaired as a paradigm (in Wernickes aphasia) or as a disturbance in phoneme selection that is observed in the
syntagm (in Brocas aphasia) (Figure 1). In other words, in so-called acoustic agnosic aphasia (a subtype of Wernickes
aphasia, the lexicon (vocabulary) and the grammar (mor- aphasia according to Luria), (b) disturbance in word selection
phosyntax) can be independently impaired [48, 63] and, associated with a different subtype of Wernickes aphasia
hence, lexicon and grammar depend not only on different referred to by Luria as an acoustic amnesic aphasia, and,
brain areas, but also on different cerebral networks. From finally, (c) error in selecting the word association, that is,
a purely linguistic perspective Chomsky [64, 65] has clearly the semantics of the words, correlated with the so-called
illustrated that the lexical/semantic system is independent amnesic aphasia. Similarly, the sequencing (contiguity) disor-
from the grammatical system; that is, a sentence can be der can potentially be found at two different levels: (a) when
grammatically correct but semantically empty. Ardila [33, 48, sequencing words in a sentence, as is observed in Brocas
63, 66] has explicitly proposed that there are two different aphasia (designated by Luria as kinetic motor aphasia), or
language systems in the brain: lexical/semantic and gram- (b) in sequencing sentences in discourse, found in so-called
matical system, supported by different brain areas (temporal transcortical motor aphasia (named by Luria as dynamic
and frontal) in the left hemisphere, developed at different ages aphasia). It is interesting to keep in mind that different
during childs language acquisition, and appearing at different subtypes of Wernickes aphasia are frequently distinguished
historical moments during human evolution. The mecha- (e.g., [40, 71]). For Luria, so-called acoustic agnosic aphasia,
nisms for learning are also different for both language sys- acoustic amnesic aphasia, and amnesic aphasia are simply
tems: the lexical/semantic knowledge is based on a particular subtypes of the aphasia syndrome usually referred to as
type of memory known as declarative memory (facts and Wernickes (or sensory) aphasia.
knowledge we are aware of), whereas grammatical knowledge As a matter of fact, in Wernickes aphasia different
corresponds to a procedural memory (memory about how language deficits can be found: the lexical knowledge (vocab-
to perform a particular action) [3436]. A disturbance in each ulary) may be decreased resulting in difficulties in under-
one of these types of memory is associated with a specific standing spoken language. Sometimes (particularly in cases
subtype of language impairment: lexical/semantic disorder of damage close to the primary auditory area) phoneme
impairment in Wernickes aphasia and grammatical disorder discrimination defects are also found. Furthermore, words
in Brocas aphasia. Mental lexicon depends on temporal lobe can lack a precise meaning, and semantic disturbances are
substrate of declarative memory, whereas mental grammar observed, associated with left temporal-occipital pathology.
(rule-governed combination of lexical items into complex So, it can be conjectured that three different defects account
representations) depends on specific frontal, basal ganglia, for the language impairments found in Wernickes aphasia: (a)
parietal and cerebellar structures [67]. phoneme discrimination impairments; (b) language memory
Jakobson is usually considered as a pioneer of the struc- abnormalities; and (c) association defects between words and
tural analysis of language. Regardless of the fact that most meanings. Figure 2 presents the model proposed by Ardila
ideas that he proposed have been integrated in contemporary [33] in an attempt to integrate the language abnormalities
linguistics, his interpretations have not been free of critics found in Wernickes aphasia. According to this model, there
(e.g., [68]). Currently, there is a poststructuralism movement are three different levels of language recognition that can
that has not only advanced previous theories, but formulated potentially be impaired in Wernickes aphasia; they are the
new proposals in the structural analysis of language (e.g., phonemic, the lexical, and the semantic level. The impair-
[69, 70]). ment in each one will result in a particular subtype of
Wernickes aphasia.
2.1. The Lexical/Semantic Disorder. The selection disorder Neuroimaging studies clearly reinforce the heterogeneous
observed in Wernickes aphasia limits the patients ability to role of the left temporal lobe in processing auditory informa-
select words (impairment of the paradigmatic axis of the tion. For example, Leaver and Rauschecker [72] analyzed how
language), that is, to select the elements of the vocabulary. the brain processes complex sounds, like voices or musical
Word selection and word-use errors are observed; and there instrument sounds. Using functional magnetic resonance
are some different potential errors: (a) nouns simply become imaging these authors were able to identify category-selective
Wernickes area
4
6
8 5
9 7
46 3 1 2
40
10 39
44
45
43 19
41
11 42
47 22 18 17
37
38 21
Brocas area
20
Figure 1: Traditionally it has been accepted that there are two major areas involved in language: frontal Brocas area (BA44 and probably
BA45) and temporal Wernickes area (BA22, 21, and 37, although BA39 is also frequently included.).
D F G
Long-term memory Short-term memory Long-term memory
for features for features for phonemes
A B C E H
Categorical
Primary auditory Frequency intensity perception
Ear Features recognition Phoneme recognition
analysis recognition language recognition
level I
I
Short-term memory
for phonemes
J
Verbal-acoustic long term

memory
K
Categorical
Verbal-acoustic perception
recognition (lexical) language recognition
level II
L
Verbal acoustic short-term
memory
M
Associations
(visual, tactile, etc.)
N
Semantic Categorical perception
recognition language recognition
level III
Figure 2: Diagram model for language recognition proposed by Ardila [33]. Three levels of language recognition potentially impaired in
Wernicke-type aphasia can be distinguished: phonemic (categorical perception level I), lexical (categorical perception level II), and semantic
(categorical perception level III). Three different subsyndromes can be found: phonemic discrimination defects (acoustic-agnosic or Wernicke
aphasia type I), verbal-acoustic memory defects (acoustic amnesic or Wernicke aphasia type II), and semantic association defects (amnesic,
nominal, or Extrasylvian sensory aphasia).
responses in the anterior superior temporal regions, consist- During recent years a significant interest in cognitive
ing of clusters selective for musical instrument sounds and neurosciences for understanding the specific role of Brocas
for human speech. An additional subregion was found that area has been observed. There has been the implicit assump-
was particularly selective for the acoustic-phonetic content tion that understanding Brocas area is fundamental for
of speech. Regions along the superior temporal plane closer understanding human cognition. As mentioned above, it
to primary auditory cortex were not selective for stimulus can be assumed that Brocas area is not really specialized
category, responding instead to specific acoustic features in producing speech, but rather in a fundamental neural
embedded in natural sounds. process responsible not only for speech movements, but also
It has to be emphasized that, in Wernickes aphasia, the for grammar use. Noteworthy, deafmute individuals (conse-
language abnormality is situated at the level of the language quently not using speech) frequently present difficulties in
elements (words, vocabulary). Phoneme and word selection understanding and using language grammar [86].
can be impaired, but language morphosyntax (grammar) is Meta-analyses of functional neuroimaging studies (par-
not impaired. Nonetheless, sometimes patients with Wer- ticularly fMRI and PET) have indicated that grammatical
nickes aphasia tend to overuse the grammatical elements, processing is clearly related to the left inferior frontal gyrus,
resulting in a phenomenon usually referred to as paragram- including Brodmanns areas 44 and 45, corresponding to
matism [73]. Brocas area [87, 88]. In an illustrative experimental study,
It has been observed that nouns are apparently associated Petersson et al. [89] investigated a group of subjects on
with an organized pattern of cerebral activity. According to a grammaticality classification task; the participants had
contemporary clinical and functional studies, the knowledge been previously exposed to well-formed consonant strings
of different semantic categories (e.g., animals, musical instru- generated from an artificial regular grammar. The aim of the
ments, and body parts) may be separately represented in study was to find whether brain regions related to language
the brain [74, 75]. It is well known that anomia (difficulties processing overlap with the brain regions activated by the
for finding names) may differently impair naming body grammaticality classification task used in this research. The
parts, naming external objects, and naming colors [76]. authors observed that artificial grammaticality violations
Moreover, the naming defect can be limited to a particular activated Brocas region in all participants, emphasizing the
semantic category (for instance, naming living things, tools, involvement of this brain region in grammar knowledge and
and geographical places) [7780]; the naming defect can be use.
so specific as to refer just to medical terms [81]. It has Noteworthy, some authors have suggested that left infe-
been consequently suggested that there is a kind of brain rior frontal gyrus is involved only complex syntactic pro-
mapping of the word memories, associated with different cessing (demanding increased cognitive control and working
semantic categories [71]. memory) (e.g., [9092]).
Departing from studies with monkeys it has been pro-
posed that there is a dual-stream (anterior and posterior) in 3. Three Different Stages in
the auditory cortex [82]. However, close homologies between
human and monkey cortex have been found. Consequently,
Human Language Evolution
two different systems can be distinguished in Wernickes area: Departing for the mentioned linguistic and neurological
dorsal and ventral. The left superior temporal gyrus (ventral observations (i.e., there are two different language systems in
stream) supports auditory word-form recognition, whereas the brain and there are two fundamental types of aphasia)
superior temporal/inferior parietal lobules (dorsal stream) and Bickertons [5] suggestion that symbolic units (lexi-
support functions of inner speech [83]. cal/semantic system) and syntax (grammatical system) are
the only real novelties in human communication system,
2.2. The Grammatical Disorder. Disturbances in grammar which probably emerged at different historical moments,
are observed in Brocas aphasia. Jakobson [5961], departing three different stages in language evolution could be pro-
from a purely linguistic perspective, suggested that in Brocas posed:
aphasia there is a basic defect in the sequencing process. It has
been well established that as a matter of fact, Brocas aphasia (a) Primitive communication systems: they use some
includes two different abnormalities: (a) a motor production sounds but may also include other types of informa-
defect characterized by decreased fluency, abnormalities in tion, such as gestures and grunts. These communi-
the speech kinetic melodies, articulation slowness, and so cation systems obviously correspond to the commu-
forth, referred to as apraxia of speech, and (b) a disturbance nication systems found in other animals, including
in the use of grammar usually known as agrammatism [40, nonhuman primates.
46, 51, 84]. It has been conjectured that if both impairments (b) Initial language systems using combined sounds
(apraxia of speech and agrammatism) are simultaneously to form words but without a relationship among
found, they simply represent two apparent manifestations of the words (grammar): that means language as lexi-
a fundamental defect [85]. It has been suggested that the cal/semantic system but not yet as a grammatical sys-
inability to sequence expressive elements observed at the tem. This type of language is similar to the holophras-
phonological/articulatory level (resulting in so-called apraxia tic period observed in children at the beginning of
of speech) or at the purely linguistic level (resulting the so- language development, around the age 1218 months
called agrammatism) could be such a fundamental defect. [86].
(c) Advanced communication systems using word- concluded that these primates have a limited ability for
combinations (grammar): that means language as the creation of new significant elements in communication
grammatical system. At advanced ages in children, (words).
not just an increase in the vocabulary is observed,
but also the beginning of grammar; around the 3.2. Moving to a Human Language. The origin of human lan-
age of 2430 months children begin to combine guage has been for centuries a particularly controversial topic.
words into simple sentences. Initially, utterances During the 19th century, different hypotheses were presented
including two words without connecting elements; to account for the emergence of human language; however,
later, grammatical connectors appear [93]. these proposals did not include the origins of grammar but
were restricted to the origins of the lexical/semantic system
I am suggesting that human language initially emerged (vocabulary) [102, 103]. Some of these hypotheses are the
as a collection of significant combination of sounds (words; following:
the paradigmatic axis of the language, according to Jakobson
(1) Language began as imitations of natural sounds. In
[5760]) and only later evolved toward a system of relations
other words, onomatopoeias (onomatopoeia is a word
between these words (the paradigmatic axis of the language,
that phonetically reproduces the source of the sound
according to Jakobson [5962]). This point of view is congru-
that it refers to) represent a basic mechanism for
ent with Bickertons [5] proposal about language evolution
the creation of new words. Indeed, this is a very
stages: lexicon phylogenetically should appear long before
important mechanism to create new words, and, as
grammar.
a matter of fact, every human language contains
an important amount of words that originally were
3.1. First Stage: Primitive Communication Systems. Animals
onomatopoeias (e.g., hiccup, zoom, bang, beep, moo,
use different communication systems, based in different
and splash). Some words that currently do not look
sensory modalities: visual, auditory, and even olfactory.
like onomatopoeias originally were onomatopoeias;
Without question, initial human language was similar to the
for instance, the word barbarian is derived from
communication systems observed in other hominid primates,
Greek barbaros foreign, strange, ignorant, from the
such as chimpanzees, orangutans, gorillas, and gibbons.
root barbar (onomatopoeia of unintelligible speech of
It is known that chimpanzees use a diversity of ges-
foreigners) [104].
tures (including facial expressions) to communicate [94].
In addition, they have a limited repertoire of vocalizations (2) Gestures are at the origin of language, and body
(they produce about 12 different vocalizations) that can movement preceded language. Oral language repre-
be used for communication purposes with other chimps. sents the use of oral gestures that began in imitation
Observations of nonhuman primates communication strate- of hand gestures that were already in use for commu-
gies have been collected in different conditions, including nication. Recently, different authors [68, 105] have
natural environments and also laboratory groups in human- argued that gestures represent the most important
controlled environments [95]. Interestingly, chimpanzees can element in creating human language.
learn some artificial languages (such as using tokens) and (3) Language began with interjections, emotive cries,
close to about 200 words (symbols). and emotional expressions. In fact, emotional com-
It has been found that a chimps ability to learn complex munication continues playing a significant role in
communication systems has not been particularly successful. contemporary human communication under certain
K. J. Hayes and C. Hayes [96] trained the chimp Vicki in a particular conditions, such as highly emotional situa-
human environment. Her ability to learn a human language tions [103].
was limited to four different words in several years! Similar (4) Language began with the easiest syllables attached to
experiments have been carried out with other chimps and the most significant objects (e.g., /ma/). Because the
gorillas with similarly limited success [9799]. easiest syllables are the same for every child anywhere
Noteworthy, whereas nonhuman primates can learn a worldwide, some early words are quite similar across
relatively high amount of words (e.g., Kanzi learned to different languages (e.g., /mama/) [103].
use some 200 symbols) they have significant difficulties in
learning to combine these words (to use grammar); that (5) Language arose from rhythmic chants and vocalisms
means it is not evident that nonhuman primates can learn the uttered by people engaged in communal labor [103].
language syntax [100, 101]. (6) It has been observed that there is a certain cor-
The crucial question in language evolution is how to respondence between language sounds (phonemes)
move from the language as a collection of words to a and meanings; that is, words maintain some rela-
grammatical language. For humans, creating new words does tionship with the meaning. Small, sharp, high things
not seem specially complicated. As a matter of fact, it has been tend to have words with high front vowels in many
proposed that in human history certain mechanism could languages (e.g., /i/in little), while big, round, low
have been used to create new words (for instance, words things tend to include back vowels (e.g., /a/in large).
can be created departing from onomatopoeias or emotional This relationship is often referred to as phonetic
expressions) [102]. However, considering the limited amount symbolism [103, 106] and has been demonstrated
of vocalizations found in nonhuman primates, it can be in a diversity of languages (for a review, see [107]).
Phonetic symbolism simply refers to the notion that also simple (consonant-vowel); these simple syllables may
phonemes can convey meaning on their own, apart have been produced in a repeated way (e.g., mamama). (c)
from their configuration in words [108]. They obviously were nouns (real objects), something that is
(7) It has been also suggested that language comes out of directly experienced.
play, laughter, cooing, courtship, emotional mutter- To create articulated words requires the progressive
ings, and the like [102]. development of a series of articulatory oppositions [13].
According to Jakobson [110] the most basic one is the
(8) Considering that there is a need for interpersonal opposition between vowels and consonants. The second most
contact, language may have begun as sounds to signal important articulatory opposition is between oral and nasal
both identity (here I am!) and belonging (Im with phonemes. But the production of these oppositions requires
you!); this is known as contact theory [102]. some anatomical adaptations in the phonatory (vocal folds,
These hypotheses are not contradictory, and indeed all larynx) and articulatory (tongue, lips, palate, etc.) systems.
these mechanisms may have contributed to the creation Human articulatory ability is partially due to the specific
of new words. Nonetheless, these hypotheses attempt to position and also configuration of the larynx. Interestingly,
explain how the language vocabulary was created; hence, how the human larynx descends during infancy and the early ado-
language evolved from the first (primitive communication lescent years; it is assumed that this descent significantly con-
systems) to the second stage (language as a lexical/semantic tributes to the anatomical requirements for speech articula-
system). They do not include any explanation for the devel- tion. Although this developmental phenomenon is frequently
opment of what could be considered as more characteristic considered to be unique to humans, Nishimura [111, 112]
of human language: language grammar. Indeed, these mech- demonstrated that indeed chimpanzees larynx is similar and
anisms for creating new words are still used in contemporary also descends during infancy, as observed in humans. Prob-
languages. For instance, onomatopoeias are still a significant ably, the descent is associated with developmental changes
strategy for the creation of new words (the name of the game of the swallowing mechanisms. But most important, it also
ping-pong is clearly departing from onomatopoeia). contributes morphologically to an increased independence
between the processes of phonation and articulation for
3.3. The Role of Vocalizations (Noises, Grunts) in Human speech production.
Communication. Regardless of the significant amount of Interestingly, laryngeal descent in nonhumans is not
vocalizations (noises, grunts) used in everyday human com- accompanied by descent of the hyoid [113]. In humans,
munication, little mention to them is found. As a matter of lowered larynx increases the vocal tract length, increases the
fact, vocalizations represent a basic communication strategy potential vocal sounds repertoire, reduces the frequency of
in different nonhuman primates including chimpanzees, and resonances, and makes sounds louder [114, 115].
without question, they have continued playing a communica- Different researchers have proposed that there are some
tion function throughout human history. People in everyday universal language characteristics, found across world lan-
life frequently use a diversity of noises (vocalizations) to say guages and even in some attempts to reconstruct extinct
yes, no, to express different emotions, to make emphasis, languages that have been presented [116118], for instance, to
and so forth. These vocalizations are close to interjections, reconstruct the Indo-European language (a central language
and sometimes become real interjections (e.g., ooph!). for most languages in spoken in Europe, the Middle East, and
India) [115, 119122] that disappeared over 10,000 years ago.
3.4. Second Stage: Initial Communication Systems. Bickerton Similarly, some proposals about the initial human vocabulary
[4] proposed that a protolanguage must have preceded the have also been presented; so, Swadesh [123] refers to some
full-fledged syntax of todays discourse. Some echoes of universal words existing across different languages (kind
this initial protolanguage can be found in, (a) in pidgin of basic vocabulary). According to the Swadeshs basic
languages (pidgin is a simplified language that develops for vocabulary [26, 124], the following categories are found
communication among people that do not have a common across different languages and may represent the initial word
language), (b) in the initial words that children develop, categories: (a) grammatical words (e.g., I/me), (b) quantifiers
(c) in the symbols used by trained chimpanzees in artificial (e.g., all), (c) adjectives (e.g., big), (d) human distinctions
conditions, and finally (d) in the syntax-free utterances of (e.g., person), (e) animals (e.g., fish), (f) highly frequent
some children who do not learn to speak at the normal age. elements (e.g., tree), (g) body parts (e.g., hair), (h) actions
Bickerton [109] considers that such a protolanguage existed (e.g., drink), (i) natural phenomena (e.g., sun), and (j) colors
already in the earliest Homo (about 2.3 to 2.4 million years (e.g., red).
ago) and was developed due to the pressure of the behavioral
adaptations faced by Homo habilis (2.3 to 1.4 million years 3.5. Third Stage: Advanced Communication Systems. The
ago). evolution of grammar (grammar or morphosyntax refers
What made up these original words? Again, the analogy to the rules governing the use of language and includes
with the childs initial vocabulary can be taken. (a) They were morphology, the study of word formation, and syntax,
simple and easy to produce from the articulatory point of the study of how words are combined into larger units
view; probably, they included those phonemes regarded as such as phrases and sentences) represents the most com-
universal phonemes (i.e., they are found across all the world plex and poorly understood question in language evolution.
languages, such as /m/, /a/). (b) The phoneme sequence was Noteworthy, human languages, regardless of the diversity in
their details, present profound structural similarities in all (corresponding to so-called the verbal phrase). Analyzing
regions of the world (i.e., there is core syntax or universal language development in children, Brown [129] proposed
grammar) [5], suggesting an original grammar, or at least, that most of the utterances when beginning grammar devel-
some universal principles for expressing ideas resulting from opment could be described by a small set of functional
the specific human brain idiosyncratic organization. relationships between words, such as agent + action (baby
Some proposals have been presented to account for the kiss), action + object (pull car), and agent + object (daddy
historical origins of grammar [125, 126]. Klein and Edgar ball).
[127] proposed that a mutation in the human species may The crucial point in emerging grammar is not just
have occurred about 50,000 years ago, accounting for the full the complexity of the lexical/semantic system, that is, the
human language (i.e., grammatical language). The rationale extension of the vocabulary. What is really important is to
behind Kleins claim refers to the fact that human culture have words corresponding to different classes that can be
significantly accelerated shortly after the date, resulting in a combined to form a higher level unit (syntagm, phrase, and
rapid increase in the amount of produced elements, including sentence). One of the words has to refer to an object (noun);
the first symbolic artifacts (statuettes, cave-paintings, etc.). the other is an action (verb). A sentence is usually regarded as
As will be mentioned below, this acceleration in culture a grammatical unit that is syntactically independent and has
development may have been related to the development of a subject that is expressed or understood (as in imperative
so-called metacognitive executive functions (such as plan- sentences) and a predicate that contains at least one finite
ning, abstracting, problem solving ability, and temporality verb [130]; that means a sentence contains a subject (noun)
of behavior) [128]. Metacognitive executive functions are and a verb, indicating that two different word categories are
strongly linked to the internal representation of actions, to required.
the use of verbs, and to the development of a grammatical Naming actions have been related to left frontal opercu-
language [48, 63]. To determine the exact date in which lum activation [131]. In cases of brain pathology, the ability to
this occurred is obviously extremely difficult, but it could be use verbs is impaired simultaneously with the ability to use
around the date proposed by Klein and Edgar [127] (about grammar as observed in cases of damage involving Brocas
50,000 years) or even later. area [132]. For instance, Ardila and Rosselli [133] reported
So what was the crucial leap for the development of the case of a 33-year-old woman who presented a selective
grammar? (i.e., syntagmatic dimension of the language). defect in finding verbs and naming actions after a head injury
Bickerton [5] stated this question in a direct and clear way: associated with a left frontal posterior hematoma. Objects,
the emergence of our own species released a torrent of colors, body parts, and qualities were named in a normal way.
creativity that is still gathering speed. What caused this dif- In this case it was clear that the ability to name objects and
ference? Clearly, it is some startling increment in cognition. name actions was clearly dissociated.
But what caused cognition to change so dramatically? The It is important to underline that some authors have
emergence of modern syntacticized language is the most argued that lexicon and grammar develop simultaneously in
plausible, indeed perhaps the only serious contender (page human history; for instance, Tomasello [134] assumes that the
520). Here, it will be argued that the modern syntacticized origin of language is related to the use of gestures, and indeed
language and the development of metacognitive executive grammar is already in the action. Consequently, lexicon does
functions (the increased cognition Bickerton refers to) are not appear before grammar. This disagreement emphasizes
simply two sides of the same coin. that there are important ongoing debates and competing
Grammar begins with the ability to combine two words explanations with regard to the origins of human language.
to create a new higher level unit (a syntagm, two or more
linguistic elements that occur sequentially in the chain
of speech and have a specific relationship). But how can 4. Brain Evolution and the Origins of
we pinpoint the particular relationship between these two Human Language
words? Obviously, the procedure has to be the simplest one,
probably similar to the procedure observed during childs 4.1. Origins of the Lexical/Semantic System. To understand
language development. the origins of language, it is crucial to consider the evolution
If we have two nouns such as of the brain areas involved in language processing, such as the
temporal lobe (lexical/semantic system). It is known that in
-baby-toy,
monkeys, the temporal lobes participate in recognizing the
we can suppose that different relations can be established sounds and calls of their own species [135138]. Hence, the
between these two words; but in order to create a simple temporal lobe plays a crucial role in auditory communication
sentence, a verb indicating an action is required, for instance, not only in humans but also in nonhuman primates. However,
baby likes toy, baby has toy, and baby wants toy. This means what is the specific adaptation of the temporal lobe that
that in order to create a syntagmatic relationship between resulted in a significant advance and increase in complexity
two or more vocabulary words, different word categories of human auditory communication?
have to be distinguished, specifically, nouns (objects) and Gannon et al. [139] observed that the anatomic pat-
verbs (actions). As a matter of fact, to create a simple tern and left hemisphere size predominance of the planum
phrase, only two types of elements are indeed required: nouns temporale, a language area of the human brain, are also
(corresponding to the so-called nominal phrase) and verbs present in chimpanzees. Consequently, this is not a critical
difference between the human and chimpanzee brains. Sim- and perception. When a particular word is associated with
ilarly, anatomical temporal lobe asymmetries (favoring the own body information (for instance, the word finger),
left hemisphere) are also found in different monkey species brain representation of the lexicon seems associated with a
[140]. Hopkins and Nir [141] used magnetic resonance images parietal extension; when the word has a visual association (for
to analyze whether chimpanzees present asymmetries in the instance, the word book), an occipital extension is found
planum temporale for grey matter volume and surface area. [158], and so forth.
The results indicated that the chimpanzees present leftward
asymmetries for both surface area and grey matter volumes. 4.2. Origins of the Grammatical System. It has been sug-
Consequently, It could be suggested that leftward asymmetry gested that verbs, grammar, and speech praxis (generated
of the left temporal auditory association area (Wernickes spatiotemporal specifications for skilled purposeful articu-
area) developed prior to the appearance of contemporary latory movements) appeared simultaneously in history [33].
human language and probably even before our divergence Interestingly, grammar, speech praxis, and the ability to
between humans and chimpanzees [142]. use verbs are simultaneously impaired in cases of Brocas
It has been further suggested that temporal lobe differ- area damage, suggesting a common neural activity. So, the
ences between humans and nonhuman primates relate to the origin of the grammar is directly linked to the ability to
temporal lobe volume. Rilling and Seligman [143] studied use verbs and the ability to produce certain articulatory
the temporal lobe volume in several primates including movements. What has been the evolution of Brocas area and
humans. It was found that overall volume, surface area, and associated networks (responsible for grammar and speech
white matter volume were significantly larger in humans praxis) obviously represents a fundamental question.
than predicted by the ape regression lines. This increase Genetic observation has shed some light on the question
in several temporal lobe dimensions may be related to the about the origins of neural systems responsible for language
complexity of the human auditory communication system. production. During recent years, the study of a family affected
It is interesting to note that the temporal lobe directly with language production difficulties significantly advanced
participates in the recognition of the own species sounds, our understanding on the origins of language grammar and
and the superior temporal gyrus contains neurons that are speech praxis. This is an English family usually referred
tuned to species-specific calls [144]. On the other hand, it to as KE family. For over three generations about half of
has been proposed that this significant enlargement of the the family members presented important abnormalities in
temporal lobe may have occurred about 200300 thousand language development. Speech difficulties were evident and
years ago [145] suggesting an increase in complexity in the articulation was deficient. This disorder was associated with
human communication system around this time. Conse- a mutation in a single autosomal-dominant gene, FOXP2,
quently, it can be conjectured that hominids existing before located in the chromosome 7 [10, 149, 152]. In addition to the
the contemporary Homo sapiens could have developed a speech difficulties, affected members also presented defects
certain complex lexical/semantic communication system. For in processing words according to grammatical rules, the
instance, it could be speculated that Neanderthal man (Homo understanding of more complex sentence structure such as
neanderthalensis if classified as a different species or Homo sentences with embedded relative clauses, the ability to form
sapiens neanderthalensis if classified as a subspecies of the intelligible speech, the ability to move the mouth and face in
Homo sapiens) could have had a language relatively complex a way not associated with speaking (relative immobility of the
as a lexical/semantic system. It is worth noting that the lower face and mouth, particularly the upper lip), and general
FOXP2 gene sequence was found in two male Neanderthals intellectual limitations.
dated about 40,000 years ago [146148]. This FOXP2 gene It was suggested that the core deficit in this complex
has strongly implicated in speech and language development syndrome was one involving sequential articulation and oro-
[149152] (see below) suggesting that the Homo sapiens nean- facial praxis [159162]. Brain abnormalities involving so the
derthalensis indeed possessed language somehow similar to cerebral cortex and the subcortical areas were documented.
contemporary human language. Interestingly, Homo sapiens An abnormal gene (SPCH1) in the chromosomal band 7q31
neanderthalensis was mostly right handed in a proportion was localized. The genetic mutation or deletion in this region
similar to contemporary man [153] and obviously there is was suggested to be associated with significant impairment of
a significant association between handedness and language speech and expressive language, including grammar [162].
lateralization in the brain [154, 155]. Enard et al. [10] analyzed the evolution of the gene
Falk et al. [156] examined the endocasts of Australopithe- FOXP2. They emphasized the extremely conservative nature
cus africanus and three species of Paranthropus. They found of FOXP2. The authors point out that the mouse FOXP2
that the brain morphology of Australopithecus africanus differs by just one amino acid from chimpanzee, gorilla, and
appears more human like than that of Paranthropus in terms rhesus monkey. However, human FOXP2 differs from gorilla,
of overall frontal and temporal lobe shape. This finding is chimp, and rhesus macaque by two further amino acids (and
consistent with the hypothesis that Australopithecus africanus thus differs from mouse by three amino acids out of 715).
could have been ancestral to Homo and has implications That means that in 75 million years since the divergence of
for assessing the early hominid neurological and cognitive mouse and chimpanzee lineages only one change occurred in
evolution [157]. FOXP2, whilst in the six million years since the divergence of
Brain organization of the lexicon (vocabulary) depends man and chimpanzee lineages two additional changes have
on the specific type of association between vocabulary words occurred in the human lineage. The authors calculated that
the last two mutations might have occurred between 10,000 For Vygotsky [181], the central point is that complex
and 100,000 years ago and speculated that the mutations have cognition depends on certain mediation (especially, but not
been critical for the development of contemporary human only by language). Thinking is regarded as a covert motor
speech. activity (i.e., inner speech). Lieberman [1517] has on the
This genetic approach to the origins of language seems other hand proposed that the frontal lobes are implicated
particularly important in understanding the appearance and in virtually all cognitive activities, whereas posterior cortical
evolution of language in humans [163, 164]. It has been regions represent active elements in vocabulary knowledge.
pointed out that FOXP2 could have contributed to the Many other authors have presented a similar point of view
evolution of human speech and language by adapting specific [183187]. Some recent research seemingly supports this
corticobasal ganglia circuits to communication purposes interpretation [188].
[150]. It has been further suggested that although FOXP2 The discovery of the so-called mirror neurons (a neuron
is expressed in many brain regions and has multiple roles which fires both when an animal performs an action and
during mammalian development, the evolutionary changes also when the animal observes the same action performed
that occurred in the protein in human ancestors specifically by another animal) [189193] can significantly contribute
affect brain regions that are connected via corticobasal to a better understanding of the brain representation of
ganglia circuits [165]. actions (verbs). Mirror neurons were initially described in
So, it seems that the development of a human lan- monkeys [189]; in the human brain, the existence of mirror
guage grammatical system may be related to some specific neurons has been suggested in the premotor cortex and the
genetic mutations occurring relatively recently in history. inferior parietal cortex [194]. It has also been suggested that
The development of a grammatical language, no question, mirrors neurons exist in Brocas area [192]. The discovery
had a significant impact in the evolution of human complex of mirror neurons in Brocas area may have important
cognition. consequences for understanding brain language organization
and language evolution [195198]. Indeed, mirror neurons
could be involved in the internal representation of actions
5. Grammar and the Origin of Complex and, hence, in the origins of grammar. Notably, it has been
Human Cognition suggested that inner speech is related to activity of Brocas
area [199, 200].
In contemporary cognitive neurosciences, complex cognition Finally, as a note of caution, it is important to keep in
is usually related to so-called executive functions [166 mind that these two recent discoveries, FOXP2 gene and
172]. Nonetheless, it is not clear what is the unitary and mirror neurons, do not represent a direct and easy answer to
fundamental factor underlying and unifying executive func- the question of language evolution. Bickerton [5] clearly and
tions [171, 172]. Ardila [128] has suggested that action overtly criticizes the expectation that these recent discoveries
representation (i.e., internally representing movements or (mirror neurons and the FOXP2 gene) will provide easy
actions) and associated with it behavior temporality, may answers about the evolution of human language. He considers
represent the fundamental executive functions factor. He that mirror neurons cannot, even in principle, shed any light
suggested that both might depend upon one single core ability on how symbols originated or how syntax originated, the
(sequencing?) [173]. two most basic questions in language evolution. On the other
Action representation is evidently related to the use hand, he considers that FOXP2 gene may have something to
of verbs, because verbs usually refer to actions (temporal do with human-ape differences, probably including language;
changes); consequently, action representation is required but he emphasizes that until we know exactly which other
for the development and use of verbs and grammar. Some genes FOXP2 turns on or off, it is premature to claim any
authors have interpreted the use of time concepts and behav- specific function and simply incorrect to consider it as a
ioral temporality in general, as the basic factor accounting to major driving force in language evolution.
executive functions [174, 175]. Notably, executive functions
are directly controlled by the prefrontal cortex; and the 6. Conclusions
prefrontal cortex represents an extension and further increase
in complexity of the frontal motor areas involved in action Without question, aphasia analysis can significantly con-
performance [176, 177]. Furthermore, significant motor contribute to the understanding of human language evolution.
trol abnormalities can be observed in cases of prefrontal According to contemporary aphasia knowledge, in cases of
damage, including primitive reflexes and perseveration [178, brain pathology, language can be disturbed in two rather
179]. different ways: as a lexical/semantic system (Wernicke-type
Different authors have emphasized that complex cog- aphasia) and as a grammatical system (Broca-type aphasia).
nition (such as thought, reasoning, and problem solving) Both language systems not only depend upon different
depends on an internalization of actions. Vygotsky [180182] brain areas (temporal and frontal), but are also supported
argued that thought (and so-called complex psychological by different neuroanatomical circuitries. This observation
processes) is associated with some inner speech. Lieber- is concurrent with contemporary theories of language and
man [1517] has suggested that language in particular and language evolution, distinguishing two major elements in
cognition in general arise from complex sequences of motor language (e.g., symbolic units and syntax; elements and
activities. structure; nouns and verbs).
As it was mentioned before the superior temporal gyrus [10] W. Enard, M. Przeworski, S. E. Fisher et al., Molecular
contains neurons that are tuned to species-specific calls. evolution of FOXP2, a gene involved in speech and language,
Human brain and old-world monkeys show a great deal of Nature, vol. 418, no. 6900, pp. 869872, 2002.
anatomical similarity [201]. The auditory cortical system is [11] S. E. Fisher and G. F. Marcus, The eloquent ape: genes, brains
organized into a ventral and a dorsal pathway in both species. and the evolution of language, Nature Reviews Genetics, vol. 7,
The similar role of the ventral auditory pathway in both no. 1, pp. 920, 2006.
humans and monkeys in the decoding of spectrally complex [12] W. T. Fitch, M. D. Hauser, and N. Chomsky, The evolution of
soundsincluding the perception of speech soundshas the language faculty: clarifications and implications, Cognition,
been well established. The dorsal processing stream plays a vol. 97, no. 2, pp. 179210, 2005.
major role in speech production. This idea is quite similar [13] P. F. MacNeilage, The frame/content theory of evolution of
speech production, Behavioral and Brain Sciences, vol. 21, no.
to the current proposal that there are two different language
4, pp. 499546, 1998.
systems in the brain. The main difference is that anterior
[14] P. M. Greenfield, Language, tools and brain: the ontogeny
and posterior superior temporal cortex are distinguished on
and phylogeny of hierarchically organized sequential behavior,
the basis of nonhuman primate neuroanatomy (e.g., [202]), Behavioral and Brain Sciences, vol. 14, no. 4, pp. 531595, 1991.
which leads to the distinction of ventral and dorsal pathways.
[15] P. Lieberman, Human Language and Our Reptilian Brain,
Observations with childrens language development and Harvard University Press, Boston, Mass, USA, 2002.
experiments with nonhuman primates demonstrate that
[16] P. Lieberman, On the nature and evolution of the neural bases
language initially appears as a lexical/semantic system. Gram- of human language, Yearbook of Physical Anthropology, vol. 45,
mar, on the other hand, is correlated with the ability to use pp. 3662, 2002.
verbs and represent actions. This is an ability that depends [17] P. Lieberman, The evolution of human speech: its anatomical
on, the so-called Brocas area and related brain circuits. and neural bases, Current Anthropology, vol. 48, no. 1, pp. 39
But this ability also depends on, is correlated, and likely 66, 2007.
appeared simultaneously in human history with the ability [18] J. P. Mallory, In Search of the Indo-Europeans, Thames and
to rapidly sequence articulatory movements (speech praxis). Hudson, London, UK, 1989.
Furthermore, language grammar probably represents the [19] G. Marcus and H. Rabagliati, What developmental disorders
departing ability for the development of complex human can tell us about the nature and origins of language, Nature
cognition (executive functions). Neuroscience, vol. 9, no. 10, pp. 12261229, 2006.
[20] P. F. MacNeilage, Origins of Speech, Oxford University Press,
Conflict of Interests 2008.
[21] M. A. Nowak and N. L. Komarova, Towards an evolutionary
The author declares that there is no conflict of interests theory of language, Trends in Cognitive Sciences, vol. 5, no. 7,
regarding the publication of this paper. pp. 288295, 2001.
[22] M. A. Nowak and D. C. Krakauer, The evolution of language,
Proceedings of the National Academy of Sciences of the United
References States of America, vol. 96, no. 14, pp. 80288033, 1999.
[1] H. Ackermann, S. R. Hage, and W. Ziegler, Brain mecha- [23] M. Ruhlen, The Origin of Language, Wiley, New York, NY, USA,
nisms of acoustic communication in humans and nonhuman 1994.
primates: an evolutionary perspective, Behavioral and Brain [24] P. T. Schoenemann, Evolution of brain and language, Lan-
Sciences, vol. 37, no. 6, pp. 529546, 2014. guage Learning, vol. 59, supplement 1, pp. 162186, 2009.
[2] A. Ardila, On the Origins of Human Cognition, Florida Interna- [25] T. C. Scott-Phillips, Evolutionary psychology and the origins of
tional University, Miami, Fla, USA, 2012, http://aalfredoardila language, Journal of Evolutionary Psychology, vol. 8, no. 4, pp.
.files.wordpress.com/2013/07/ardila-a-2012-on-the-origins-of- 289307, 2010.
human-cognition.pdf. [26] M. Swadesh, Language and Human Life, Fondo de Cultura
[3] R. C. Berwick, A. D. Friederici, N. Chomsky, and J. J. Bolhuis, Economica, Mexico City, Mexico, 1967.
Evolution, brain, and the nature of language, Trends in [27] M. Tallerman, Language Origins: Perspectives on Evolution,
Cognitive Sciences, vol. 17, no. 2, pp. 8998, 2013. Oxford University Press, 2005.
[4] D. Bickerton, Language and Species, The University of Chicago [28] M. H. Christiansen and S. Kirby, Eds., Language Evolution: The
Press, Chicago, Ill, USA, 1990. States of the Art, Oxford University Press, New York, NY, USA,
[5] D. Bickerton, Language evolution: a brief guide for linguists, 2003.
Lingua, vol. 117, no. 3, pp. 510526, 2007. [29] N. Geschwind, The development of the brain and the evolution
[6] M. C. Corballis, From Hand to Mouth: The Origins of Language, of language, in Monograph Series on Language and Linguistics,
Princeton University Press, Princeton, NJ, USA, 2002. C. I. J. M. Stuart, Ed., pp. 155169, Georgetown University Press,
[7] M. C. Corballis, From mouth to hand: Gesture, speech, and the Washington, DC, USA, 1964.
evolution of right-handedness, Behavioral and Brain Sciences, [30] S. Walker, The evolution and dissolution of language, in
vol. 26, no. 2, pp. 199208, 2003. Progress in the Psychology of Language, A. W. Ellis, Ed., Lawrence
[8] M. C. Corballis, The gestural origins of language, in The Ori- Erlbaum Associates, London, UK, 1987.
gins of Language: Unraveling Evolutionary Forces, N. Masataka, [31] C. Code, First in, last out?: the evolution of aphasic lexical
Ed., pp. 1123, Springer, 2008. speech automatisms to agrammatism and the evolution of
[9] J. L. Desalles, Why We Talk? The Evolutionary Origins of human communication, Interaction Studies, vol. 6, no. 2, pp.
Language, Oxford University Press, 2007. 311334, 2005.
[32] C. Code, Nonfluent aphasia and the evolution of proto- [58] A. Benton, Exploring the History of Neuropsychology. Selected
language, Journal of Neurolinguistics, vol. 24, no. 2, pp. 136144, Papers, Oxford University Press, New York, NY, USA, 2000.
2011. [59] R. Jakobson and M. Halle, Two Aspects of Language and Two
[33] A. Ardila, Origins of the language: correlation between brain Types of Aphasic Disturbances, Mouton, New York, NY, USA,
evolution and language development, in Foundations of Evolu- 1956.
tionary Cognitive Neuroscience, S. M. Platek and T. K. Shack- [60] R. Jakobson, Studies on Child Language and Aphasia, Mouton,
elford, Eds., Cambridge University Press, New York, NY, USA, New York, NY, USA, 1971.
2009. [61] A. R. Luria, Sobre las dos formas basicas del alteraciones
[34] F. Fabbro, The bilingual brain: cerebral representation of afasicas en el lenguaje, in Psicobiologa del Lenguaje, A. Ardila,
languages, Brain and Language, vol. 79, no. 2, pp. 211222, 2001. Ed., Trillas, Mexico City, Mexico, 1983.
[35] F. Fabbro, The Neurolinguistics of Bilingualism: An Introduction, [62] R. Jakobson, Toward a linguistic typology of aphasic impair-
Psychology Press, New York, NY, USA, 1999. ments, in Disorders of Language, A. V. S. DeReuck and M.
[36] M. Paradis, A Neurolinguistic Theory of Bilingualism, John OConnor, Eds., Little and Brown, New York, NY, USA, 1964.
Benjamins, Philadelphia, Pa, USA, 2004. [63] A. Ardila, Interaction between lexical and grammatical lan-
[37] N. Chomsky, Rules and Representations, Blackwell, Oxford, UK, guage systems in the brain, Physics of Life Reviews, vol. 9, no.
1980. 2, pp. 198214, 2012.
[38] D. L. Cheney and R. M. Seyfarth, The representation of social [64] N. Chomsky, Syntactic Structures, Mouton, The Hague, The
relations by monkeys, Cognition, vol. 37, no. 1-2, pp. 167196, Netherlands, 1957.
1990. [65] N. Chomsky, Aspects of the Theory of Syntax, MIT Press, 1965.
[39] M. Hauser, The Evolution of Communication, MIT Press, Cam- [66] A. Ardila, Orgenes del lenguaje: un analisis desde la perspec-
bridge, Mass, USA, 1997. tiva de las afasias, Revista de Neurologia, vol. 43, pp. 690698,
[40] D. F. Benson and A. Ardila, Aphasia: A Clinical Perspective, 2006.
Oxford University Press, New York, NY, USA, 1996. [67] M. T. Ullman, Contributions of memory circuits to language:
[41] M. Albert, H. Goodglass, N. A. Helm, A. B. Rubers, and M. the declarative/procedural model, Cognition, vol. 92, no. 1-2,
P. Alexander, Clinical Aspects of Dysphasia, Springer, Berlin, pp. 231270, 2004.
Germany, 1981. [68] V. Erlich, Russian formalism: in perspective, The Journal of
[42] K. Goldstein, Language and Language Disturbances, Grune & Aesthetics and Art Criticism, vol. 13, no. 2, pp. 215225, 1954.
Stratton, New York, NY, USA, 1948. [69] J. Angermuller, Poststructuralist Discourse Analysis: Subjectivity
[43] H. Hecaen, Introduction a la Neuropsychologie, Larousse, Paris, in Enunciative Pragmatics, Palgrave Macmillan, 2014.
France, 1972. [70] T. Brown, Mathematics Education and Language: Interpreting
[44] A. Kertesz, Aphasia and Associated Disorders, Grune and Hermeneutics and Post-Structuralism, Springer Science & Busi-
Stratton, New York, NY, USA, 1979. ness Media, 2001.
[45] L. Lichtheim, On aphasia, Brain, vol. 7, pp. 433484, 1885. [71] A. Ardila, Las aphasias [The aphasias], 2006, https://aalf-
[46] A. R. Luria, Basic Problems of Neurolinguistics, Mouton, New redoardila.files.wordpress.com/2013/07/ardila-a-2006-las-afas-
York, NY, USA, 1976. ias-primera-parte.pdf.
[47] A. Ardila, A proposed reinterpretation and reclassification of [72] A. M. Leaver and J. P. Rauschecker, Cortical representation
aphasic syndromes, Aphasiology, vol. 24, no. 3, pp. 363394, of natural complex sounds: effects of acoustic features and
2010. auditory object category, The Journal of Neuroscience, vol. 30,
no. 22, pp. 76047612, 2010.
[48] A. Ardila, There are two different language systems in the
brain, Journal of Behavioral and Brain Science, vol. 1, pp. 2336, [73] E. Bates, A. Friederici, and B. Wulfeck, Grammatical morphol-
2011. ogy in aphasia: evidence from three languages, Cortex, vol. 23,
no. 4, pp. 545574, 1987.
[49] D. C. Bastian, Aphasia and Other Speech Defects, H.K. Lewis,
London, UK, 1898. [74] A. Martin and L. L. Chao, Semantic memory and the brain:
structure and processes, Current Opinion in Neurobiology, vol.
[50] H. Head, Aphasia and Kindred Disorders of Speech, Cambridge
11, no. 2, pp. 194201, 2001.
University Press, Cambridge, UK, 1926.
[75] E. K. Warrington, The selective impairment of semantic
[51] A. Kertesz, Aphasia, in Handbook of Clinical Neurology,
memory, The Quarterly Journal of Experimental Psychology, vol.
Clinical Neuropsychology, J. A. M. Frederiks, Ed., pp. 287332,
27, no. 4, pp. 635657, 1975.
Elsevier, Amsterdam, The Netherlands, 1985.
[76] H. Hecaen and M. L. Albert, Human Neuropsychology, John
[52] L. Lichtheim, On aphasia, Brain, vol. 7, no. 4, pp. 433484,
Wiley & Sons, New York, NY, USA, 1978.
1885.
[77] D. M. Harris and J. Kay, Selective impairment of the retrieval
[53] A. Pick, Aphasia, Charles C. Thomas, New York, NY, USA, 1931. of peoples names: a case of category specificity, Cortex, vol. 31,
[54] H. Schuell, J. J. Jenkins, and E. Jimenez-Pabon, Aphasia in no. 3, pp. 575582, 1995.
Adults, Harper and Row, 1964. [78] H. Goodglass, A. Wingfield, M. R. Hyde, and J. C. Theurkauf,
[55] M. Taylor-Sarno, Acquired Aphasia, Academic Press, New York, Category specific dissociations in naming and recognition by
NY, USA, 1998. aphasic patients, Cortex, vol. 22, no. 1, pp. 87102, 1986.
[56] S. A. K. Wilson, Aphasia, Kegal Paul, New York, NY, USA, 1926. [79] F. Lyons, J. R. Hanley, and J. Kay, Anomia for common names
[57] A. L. Benton, Aphasia: historical perspectives, in Acquired and geographical names with preserved retrieval of names of
Aphasia, M. T. Sarno, Ed., Academic Press, New York, NY, USA, people: a semantic memory disorder, Cortex, vol. 38, no. 1, pp.
1981. 2335, 2002.
[80] E. K. Warrington and T. Shallice, Category specific semantic [102] G. Yule, The Study of Language, Cambridge University Press,
impairments, Brain, vol. 107, no. 3, pp. 829854, 1984. Cambridge, UK, 2nd edition, 1996.
[81] B. Crosson, P. J. Moberg, J. R. Boone, L. J. G. Rothi, and A. [103] O. Jespersen, Language: Its Nature, Development and Origin,
Raymer, Category-specific naming deficit for medical terms Allen & Unwin, Sidney, Ohio, USA, 1922.
after dominant thalamic/capsular hemorrhage, Brain and Lan- [104] Online Etimology Dictionary, 2014, http://www.etymonline
guage, vol. 60, no. 3, pp. 407442, 1997. .com/index.php?term=barbarian.
[82] J. R. Binder, J. A. Frost, T. A. Hammeke et al., Human temporal
[105] D. F. Armstrong and S. E. Wilcox, The Gestural Origin of
lobe activation by speech and nonspeech sounds, Cerebral
Language, Oxford University Press, Oxford, UK, 2007.
Cortex, vol. 10, no. 5, pp. 512528, 2000.
[106] E. Sapir, A study in phonetic symbolism, Journal of Experimen-
[83] I. DeWitt and J. P. Rauschecker, Wernickes area revisited:
tal Psychology, vol. 12, no. 3, pp. 225239, 1929.
parallel streams and word processing, Brain and Language, vol.
127, no. 2, pp. 181191, 2013. [107] P. L. French, Toward an explanation of phonetic symbolism,
Word, vol. 28, pp. 305322, 1977.
[84] H. Goodglass, Understanding Aphasia, Academic Press, New
York, NY, USA, 1993. [108] L. J. Shrum and T. M. Lowrey, Sounds convey meaning: the
[85] R. S. Berndt and A. Caramazza, A redefinition of the syndrome implications of phonetic symbolism for brand name construc-
of Brocas aphasia: implications for a neuropsychological model tion, in Psycholinguistic Phenomena I Marketing Communica-
of language, Applied Psycholinguistics, vol. 1, no. 3, pp. 225278, tions, T. M. Lowrey, Ed., pp. 3958, Erlbaum, Mahwah, NJ, USA,
1980. 2007.
[86] H. Poizner, U. Bellugi, and E. S. Klima, Biological foundations [109] D. Bickerton, Adams Tongue: How Humans Made Language,
of language: clues from sign language, Annual Review of How Language Made Humans, Hill and Wang, New Yok, NY,
Neuroscience, vol. 13, pp. 283307, 1990. USA, 2009.
[87] Brodmanns Interactive Atlas, 2013, http://www.fmriconsulting [110] R. Jakobson, Child Language, Aphasia and Phonological Univer-
.com/brodmann/Introduction.html. sals, Mouton, New York, NY, USA, 1968.
[88] Y. Grodzinsky and K. Amunts, Eds., Brocas Region, Oxford [111] T. Nishimura, Developmental changes in the shape of the
University Press, New York, NY, USA, 2006. supralaryngeal vocal tract in Chimpanzees, American Journal
of Physical Anthropology, vol. 126, no. 2, pp. 193204, 2005.
[89] K. M. Petersson, C. Forkstam, and M. Ingvar, Artificial syntac-
tic violations activate Brocas region, Cognitive Science, vol. 28, [112] T. Nishimura, A. Mikami, J. Suzuki, and T. Matsuzawa, Descent
no. 3, pp. 383407, 2004. of the larynx in chimpanzee infants, Proceedings of the National
Academy of Sciences of the United States of America, vol. 100, no.
[90] N. Dronkers and J. Ogar, Brain areas involved in speech
12, pp. 69306933, 2003.
production, Brain, vol. 127, no. 7, pp. 14611462, 2004.
[113] P. H. Lieberman, D. H. Klatt, and W. H. Wilson, Vocal tract
[91] C. Rogalsky and G. Hickok, The role of Brocas area in sentence
limitations on the vowel repertoires of rhesus monkey and other
comprehension, Journal of Cognitive Neuroscience, vol. 23, no.
nonhuman primates, Science, vol. 164, no. 3884, pp. 11851187,
7, pp. 16641680, 2011.
1969.
[92] D. Saura, B. W. Kreher, S. Schnell et al., Ventral and dorsal
[114] W. T. Fitch, Vocal tract length and formant frequency disper-
pathways for language, Proceedings of the National Academy
sion correlate with body size in rhesus macaques, Journal of the
of Sciences of the United States of America, vol. 105, no. 46, pp.
Acoustical Society of America, vol. 102, no. 2, pp. 12131222, 1997.
1803518040, 2008.
[93] E. Hoff, Language Development, Wadsworth, New York, NY, [115] W. T. Fitch and J. Giedd, Morphology and development of the
USA, 4th edition, 2009. human vocal tract: a study using magnetic resonance imaging,
The Journal of the Acoustical Society of America, vol. 106, no. 3 I,
[94] C. Hobaiter and R. W. Byrne, The gestural repertoire of the wild pp. 15111522, 1999.
chimpanzee, Animal Cognition, vol. 14, no. 5, pp. 745767, 2011.
[116] J. H. Greenberg, Universals of Human Language, Stanford
[95] J. McCrone, The Ape That Spoke: Language and the Evolution of University Press, Palo Alto, Calif, USA, 1978.
the Human Mind, Avon Books, New York, NY, USA, 1991.
[117] C. Hagege, La Structure des Langues, Presses Universitaires de
[96] K. J. Hayes and C. Hayes, Imitation in a home-raised chim- France, Paris, France, 1982.
panzee, Journal of Comparative and Physiological Psychology,
vol. 45, no. 5, pp. 450459, 1952. [118] P. L. van den Berghe, Human Family Systems, Elsevier, New
York, NY, USA, 1979.
[97] R. A. Gardner and B. T. Gardner, Teaching sign language to a
chimpanzee, Science, vol. 165, no. 3894, pp. 664672, 1969. [119] J. M. Anderson, Structural Aspects of Language Change, Long-
man, 1973.
[98] J. Limber, What can chimps tell us about the origins of
language? in Language Development, S. Kuczaj, Ed., pp. 429 [120] W. P. Lehmann, Proto Indoeuropean Syntax, University of Texas
446, Erlbaum, Hillsdale, NJ, USA, 1982. Press, Austin, Tex, USA, 1974.
[99] F. Patterson and E. Linden, The Education of Koko, Holt Rinehart [121] A. Martinet, Evolution of langues et Reconstruction, Presses
& Winston, New York, NY, USA, 1981. Universitaires de France, Paris, France, 1975.
[100] J. Mitani, Review of Savage-Rumbaugh and Lewins Kanzi: the [122] V. Shevoroshkin, The mother tongue: how linguists have
ape at the brink of the human mind, Scientific American, vol. reconstructed the ancestor of all living languages, The Sciences,
272, pp. 4354, 1995. vol. 30, no. 3, pp. 2027, 1990.
[101] S. Savage-Rumbaugh and R. Lewin, Kanzi: The Ape at the Brink [123] M. Swadesh, The Origin and Diversification of Language, Aldine,
of the Human Mind, Wiley, New York, NY, USA, 1994. Los Angeles, Calif, USA, 1971.
[124] M. Swadesh, Lexicostatistic dating of prehistoric ethnic con- of the Royal Society B: Biological Sciences, vol. 277, no. 1691, pp.
tacts, Proceedings of the American Philosophical Society, vol. 96, 21652174, 2010.
pp. 152163, 1955. [143] J. K. Rilling and R. A. Seligman, A quantitative morphometric
[125] M. D. Hauser, N. Chomsky, and W. T. Fitch, The faculty of comparative analysis of the primate temporal lobe, Journal of
language: what is it, who has it, and how did it evolve? Science, Human Evolution, vol. 42, no. 5, pp. 505533, 2002.
vol. 298, no. 5598, pp. 15691579, 2002. [144] J. P. Rauschecker and B. Tian, Mechanisms and streams for
[126] S. Pinker and R. Jackendoff, The faculty of language: whats processing of what and where in auditory cortex, Proceedings
special about it? Cognition, vol. 95, no. 2, pp. 201236, 2005. of the National Academy of Sciences of the United States of
[127] R. G. Klein and B. Edgar, The Dawn of Human Culture, John America, vol. 97, no. 22, pp. 1180011806, 2000.
Wiley & Sons, New York, NY, USA, 2002. [145] V. I. Kochetkova, Paleoneurology, Moscow State University
[128] A. Ardila, On the evolutionary origins of executive functions, Press, Moscow, Russia, 1973.
Brain and Cognition, vol. 68, no. 1, pp. 9299, 2008. [146] P. Endicott, S. Y. W. Ho, and C. Stringer, Using genetic
[129] R. Brown, A First Language: The Early Stages, Harvard Univer- evidence to evaluate four palaeoanthropological hypotheses for
sity Press, Boston, Mass, USA, 1973. the timing of Neanderthal and modern human origins, Journal
of Human Evolution, vol. 59, no. 1, pp. 8795, 2010.
[130] The Free Dictionary, August 2014, http://www.thefreedic-
tionary.com/sentence. [147] R. E. Green, J. Krause, A. W. Briggs et al., A draft sequence of
the neandertal genome, Science, vol. 328, no. 5979, pp. 710722,
[131] H. Damasio, T. J. Grabowski, D. Tranel, L. L. B. Ponto, R.
2010.
D. Hichwa, and A. R. Damasio, Neural correlates of naming
actions and of naming spatial relations, NeuroImage, vol. 13, no. [148] J. Krause, C. Lalueza-Fox, L. Orlando et al., The derived FOXP2
6, pp. 10531064, 2001. variant of modern humans was shared with Neandertals,
Current Biology, vol. 17, no. 21, pp. 19081912, 2007.
[132] A. R. Damasio and D. Tranel, Nouns and verbs are retrieved
with differently distributed neural systems, Proceedings of the [149] G. Coop, K. Bullaughey, F. Luca, and M. Przeworski, The
National Academy of Sciences of the United States of America, timing of selection at the human FOXP2 gene, Molecular
vol. 90, no. 11, pp. 49574960, 1993. Biology and Evolution, vol. 25, no. 7, pp. 12571259, 2008.
[133] A. Ardila and M. Rosselli, Averbia as a selective naming dis- [150] W. Enard, FOXP2 and the role of cortico-basal ganglia circuits
order: a single case report, Journal of Psycholinguistic Research, in speech and language evolution, Current Opinion in Neurobi-
vol. 23, no. 2, pp. 139148, 1994. ology, vol. 21, no. 3, pp. 415424, 2011.
[134] M. Tomasello, Origins of Human Communication, MIT Press, [151] T. Maricic, V. Gunther, O. Georgiev et al., A recent evolutionary
2010. change affects a regulatory element in the human FOXP2 gene,
Molecular Biology and Evolution, vol. 30, no. 4, pp. 844852,
[135] J. P. Rauschecker, B. Tian, and M. Hauser, Processing of
2013.
complex sounds in the macaque nonprimary auditory cortex,
Science, vol. 268, no. 5207, pp. 111114, 1995. [152] F. Vargha-Khadem, K. Watkins, K. Alcock, P. Fletcher, and
R. Passingham, Praxic and nonverbal cognitive deficits in a
[136] J. P. Taglialatela, J. L. Russell, J. A. Schaeffer, and W. D. Hopkins,
large family with a genetically transmitted speech and language
Visualizing vocal perception in the chimpanzee brain, Cere-
disorder, Proceedings of the National Academy of Sciences of the
bral Cortex, vol. 19, no. 5, pp. 11511157, 2009.
United States of America, vol. 92, no. 3, pp. 930933, 1995.
[137] X. Wang, M. M. Merzenich, R. Beitel, and C. E. Schreiner,
[153] D. W. Frayer, I. Fiore, C. Lalueza-Fox, J. Radovcic, and L.
Representation of a species-specific vocalization in the pri-
Bondioli, Right handed Neandertals: vindija and beyond,
mary auditory cortex of the common marmoset: temporal and
Journal of Anthropological Sciences, vol. 88, pp. 113127, 2010.
spectral characteristics, Journal of Neurophysiology, vol. 74, no.
6, pp. 26852706, 1995. [154] S. Knecht, B. Drager, M. Deppe et al., Handedness and
hemispheric language dominance in healthy humans, Brain,
[138] Z. Wollberg and J. D. Newman, Auditory cortex of squirrel
vol. 123, no. 12, pp. 25122518, 2000.
monkey: response patterns of single cells to species-specific
vocalizations, Science, vol. 175, no. 4018, pp. 212214, 1972. [155] T. Rasmussen and B. Milner, The role of early brain injury in
the lateralization of cerebral speech functions, Annals of the
[139] P. J. Gannon, R. L. Holloway, D. C. Broadfield, and A. R. Braun,
New York Academy of Sciences, vol. 299, pp. 355369, 1977.
Asymmetry of chimpanzee planum temporale: humanlike
pattern of Wernickes brain language area homolog, Science, [156] D. Falk, J. C. Redmond Jr., J. Guyer et al., Early hominid
vol. 279, no. 5348, pp. 220222, 1998. brain evolution: a new look at old endocasts, Journal of Human
Evolution, vol. 38, no. 5, pp. 695717, 2000.
[140] P. L. Heilbroner and R. L. Holloway, Anatomical brain asymme-
tries in New World and Old World monkeys: stages of temporal [157] A. Barnard, Social Anthropology and Human Origins, Cam-
lobe development in primate evolution, American Journal of bridge University Press, 2011.
Physical Anthropology, vol. 76, no. 1, pp. 3948, 1988. [158] F.-E. Roux, V. Lubrano, V. Lauwers-Cances, C. R. Mascott,
[141] W. D. Hopkins and T. M. Nir, Planum temporale surface area and J.-F. Demonet, Category-specific cortical mapping: color-
and grey matter asymmetries in chimpanzees (Pan troglodytes): naming areas, Journal of Neurosurgery, vol. 104, no. 1, pp. 2737,
the effect of handedness and comparison with findings in 2006.
humans, Behavioural Brain Research, vol. 208, no. 2, pp. 436 [159] H. Takahashi, K. Takahashi, and F.-C. Liu, FOXP genes, neural
443, 2010. development, speech and language disorders, Advances in
[142] M. A. Spocter, W. D. Hopkins, A. R. Garrison et al., Wernickes Experimental Medicine and Biology, vol. 665, pp. 117129, 2009.
area homologue in chimpanzees (Pan troglodytes) and its rela- [160] S. C. Vernes, J. Nicod, F. M. Elahi et al., Functional genetic
tion to the appearance of modern human language, Proceedings analysis of mutations implicated in a human speech and
language disorder, Human Molecular Genetics, vol. 15, no. 21, [179] M. Victor and A. H. Ropper, Adams & Victors Principles of
pp. 31543167, 2006. Neurology, McGraw Hill, New York, NY, USA, 2001.
[161] F. Vargha-Khadem, D. G. Gadian, A. Copp, and M. Mishkin, [180] L. S. Vygotsky, The problem of the cultural development of the
FOXP2 and the neuroanatomy of speech and language, Nature child II, The Journal of Genetic Psychology, vol. 36, pp. 415432,
Reviews Neuroscience, vol. 6, no. 2, pp. 131138, 2005. 1929.
[162] S. E. Fisher, F. Vargha-Khadem, K. E. Watkins, A. P. Monaco, [181] L. S. Vygotsky, Thought and Language, MIT Press, Boston, Mass,
and M. E. Pembrey, Localisation of a gene implicated in a USA, 1934/1962.
severe speech and language disorder, Nature Genetics, vol. 18, [182] L. S. Vygotsky, Mind in Society, Harvard University Press,
no. 2, pp. 168170, 1998. Cambridge, Mass, USA, 1934/1968.
[163] C. Scharff and J. Petri, Evo-devo, deep homology and FoxP2:
[183] B. Hommel, J. Musseler, G. Aschersleben, and W. Prinz, The
implications for the evolution of speech and language, Philo-
Theory of Event Coding (TEC): a framework for perception and
sophical Transactions of the Royal Society B: Biological Sciences,
action planning, Behavioral and Brain Sciences, vol. 24, no. 5,
vol. 366, no. 1574, pp. 21242140, 2011.
pp. 849878, 2001.
[164] Y. Tanabe, E. Fujita, and T. Momoi, FOXP2 promotes the
[184] M. Jeannerod, The Cognitive Neuroscience of Action, Wiley-
nuclear translocation of POT1, but FOXP2(R553H), mutation
Blackwell, New York, NY, USA, 1997.
related to speech-language disorder, partially prevents it, Bio-
chemical and Biophysical Research Communications, vol. 410, no. [185] A. P. Luria, Frontal lobe syndromes, in Handbook of Clinical
3, pp. 593596, 2011. Neurology, J. P. Vinken and G. W. Bruyn, Eds., pp. 725757,
[165] S. Reimers-Kipping, W. Hevers, S. Paabo, and W. Enard, North-Holland, Amsterdam, The Netherlands, 1969.
Humanized Foxp2 specifically affects cortico-basal ganglia [186] E. Morsella, J. A. Bargh, and P. M. Gollwitzer, Oxford Handbook
circuits, Neuroscience, vol. 175, pp. 7584, 2011. of Human Action, Oxford University Press, Oxford, UK, 2009.
[166] E. Koechlin, C. Ody, and F. Kouneiher, The architecture of [187] W. Prinz, Action representation: crosstalk between semantics
cognitive control in the human prefrontal cortex, Science, vol. and pragmatics, Neuropsychologia, vol. 55, pp. 5156, 2014.
302, no. 5648, pp. 11811185, 2003. [188] E. Clerget, W. Poncin, L. Fadiga, and E. Olivier, Role of Brocas
[167] E. K. Miller and J. D. Cohen, An integrative theory of prefrontal area in implicit motor skill learning: evidence from continuous
cortex function, Annual Review of Neuroscience, vol. 24, pp. theta-burst magnetic stimulation, Journal of Cognitive Neuro-
167202, 2001. science, vol. 24, no. 1, pp. 8092, 2012.
[168] D. R. Stuss and R. T. Knight, The Frontal Lobes, Oxford [189] G. di Pellegrino, L. Fadiga, L. Fogassi, V. Gallese, and G.
University Press, New York, NY, USA, 2002. Rizzolatti, Understanding motor events: a neurophysiological
[169] D. T. Stuss and B. Levine, Adult clinical neuropsychology: study, Experimental Brain Research, vol. 91, no. 1, pp. 176180,
lessons from studies of the frontal lobes, Annual Review of 1992.
Psychology, vol. 53, pp. 401433, 2002. [190] D. I. Perrett, E. T. Rolls, and W. Caan, Visual neurons respon-
[170] D. T. Stuss and M. P. Alexander, Is there a dysexecutive sive to faces in the monkey temporal cortex, Experimental
syndrome? Philosophical Transactions of the Royal Society B: Brain Research, vol. 47, no. 3, pp. 329342, 1982.
Biological Sciences, vol. 362, no. 1481, pp. 901915, 2007. [191] D. I. Perrett, P. A. J. Smith, A. J. Mistlin et al., Visual analysis
[171] J. T. Ustarroz, A. Garca-Molina, M. R. Lago, and A. Ardila, of body movements by neurones in the temporal cortex of
Neuropsicologa de la Corteza Prefrontal y de las Funciones the macaque monkey: a preliminary report, Behavioural Brain
Ejecutivas, Viguera, Barcelona, Spain, 2012. Research, vol. 16, no. 2-3, pp. 153170, 1985.
[172] A. Miyake, N. P. Friedman, M. J. Emerson, A. H. Witzki, A. [192] G. Rizzolatti and M. A. Arbib, Language within our grasp,
Howerter, and T. D. Wager, The unity and diversity of executive Trends in Neurosciences, vol. 21, no. 5, pp. 188194, 1998.
functions and their contributions to complex Frontal Lobe
[193] G. Rizzolatti, L. Fadiga, V. Gallese, and L. Fogassi, Premotor
tasks: a latent variable analysis, Cognitive Psychology, vol. 41,
cortex and the recognition of motor actions, Cognitive Brain
no. 1, pp. 49100, 2000.
Research, vol. 3, no. 2, pp. 131141, 1996.
[173] A. Ardila and B. Bernal, What can be localized in the brain?
Toward a factor theory on brain organization of cognition, [194] G. Rizzolatti and L. Craighero, The mirror-neuron system,
International Journal of Neuroscience, vol. 117, no. 7, pp. 935969, Annual Review of Neuroscience, vol. 27, pp. 169192, 2004.
2007. [195] M. A. Arbib, Aphasia, apraxia and the evolution of the
[174] J. M. Fuster, The prefrontal cortexan update: time is of the language-ready brain, Aphasiology, vol. 20, no. 911, pp. 1125
essence, Neuron, vol. 30, no. 2, pp. 319333, 2001. 1155, 2006.
[175] J. M. Fuster, Frontal lobe and cognitive development, Journal [196] M. A. Arbib, From monkey-like action recognition to human
of Neurocytology, vol. 31, no. 3-5, pp. 373385, 2002. language: an evolutionary framework for neurolinguistics,
Behavioral and Brain Sciences, vol. 28, no. 2, pp. 105124, 2005.
[176] B. L. Miller and J. L. Cummings, The Human Frontal Lobes:
Functions and Disorders, Guilford Press, New York, NY, USA, [197] L. Craighero, G. Metta, G. Sandini, and L. Fadiga, The mirror-
2007. neurons system: data and models, Progress in Brain Research,
[177] J. Risberg, Evolutionary aspects of the frontal lobes, in The vol. 164, pp. 3959, 2007.
Frontal Lobes. Development, Function and Pathology, J. Risberg [198] D. R. Lametti and A. A. G. Mattar, Mirror neurons and the
and J. Grafman, Eds., pp. 120, Cambridge University Press, lateralization of human language, Journal of Neuroscience, vol.
2006. 26, no. 25, pp. 66666667, 2006.
[178] A. Ardila and M. Rosselli, Neuropsicologia Clinica, Manual [199] P. K. McGuire, D. A. Silbersweig, R. M. Murray, A. S. David, R.
Moderno, Mexico City, Mexico, 2007. S. J. Frackowiak, and C. D. Frith, Functional anatomy of inner
speech and auditory verbal imagery, Psychological Medicine,

vol. 26, no. 1, pp. 2938, 1996.
[200] G. de Zubicaray, N. Postle, K. McMahon, M. Meredith, and R.
Ashton, Mirror neurons, the representation of word meaning,
and the foot of the third left frontal convolution, Brain and
Language, vol. 112, no. 1, pp. 7784, 2010.
[201] J. P. Rauschecker and S. K. Scott, Maps and streams in the
auditory cortex: nonhuman primates illuminate human speech
processing, Nature Neuroscience, vol. 12, no. 6, pp. 718724,
2009.
[202] L. M. Romanski, B. Tian, J. Fritz, M. Mishkin, P. S. Goldman-
Rakic, and J. P. Rauschecker, Dual streams of auditory afferents
target multiple domains in the primate prefrontal cortex,
Nature Neuroscience, vol. 2, no. 12, pp. 11311136, 1999.
Applied Neuropsychology Copyright 2001 by
2001, Vol. 8, No. 4, 242247 Lawrence Erlbaum Associates, Inc.
Rehabilitation of Balints Syndrome:

A Single Case Report
Mnica Rosselli
Department of Psychology, Florida Atlantic University, Davie, Florida, USA
Alfredo Ardila
Department of Neuropsychology, Memorial Regional Hospital, Hollywood, Florida, USA
Christopher Beltran
Department of Psychology, Miami Institute of Psychology, Miami, Florida, USA
Very little research has been done in the area of rehabilitation of the visual perceptual deficits
observed after Balints syndrome. This syndrome is characterized by difficulties with visual
scanning, dysmetria secondary to visual perceptual deficits, and an inability to perceive more
than 1 object at a time. This article describes the effects of a neuropsychological treatment
protocol on a 23-year-old patient who suffered a fat embolism involving the territory of the
posterior cerebral arteries. A neuropsychological evaluation carried out 12 months after the
brain embolism disclosed Balints syndrome, alexia without agraphia, visual agnosia,
prosopagnosia, and memory impairments. The rehabilitation protocol included both visuo-
perceptual retraining and a functional program designed to provide rehabilitation in contexts
that were meaningful to the patient. After 1 year of treatment, a second neuropsychological
evaluation was carried out. Significant improvement was demonstrated in terms of both
objective testing and the return of an integrated and productive lifestyle.
Key words: Balints syndrome, case report, neurorehabilitation
Balints first description of this visuoperceptual dis- tor of the visual field at any given time. Simultanag-
order dates back to the early 20th century (Balint, nosia has been described as impaired simultaneous
1909). He described it as a psychic paralysis of visual perception (Thaiss & DeBleser, 1992). Luria (1959)
fixation. Three main clinical signs characterize this noted that although it has been described as a visual
condition. The first sign is fixed gaze, or ocular apraxia. disorientation, it has alternately been characterized as a
This is the inability to scan the visual field despite nor- constriction of visual attention by Holmes (1918).
mality of eye movement, intact visual fields, and intact Hcaen and Angelergues (1963) proposed a similar
reflex motility. The second sign is optic ataxia. This is explanation. Patients presenting with simultanagnosia
the inability to do precise hand movements, such as often report that objects in the visual field disappear
grasping or touching an object, while under visual from direct view. These patients describe a fragmented
guidance. Patients with optic ataxia lack the integration perception of the visual environment (Rizzo & Hurtig,
of visual and hand coordination (Ayuso-Peralta et al., 1987). It has been explained using the analogy of a
1994). The third sign, which may be considered the camera lens (Humphreys & Bruce, 1989). A reduction
essence of Balints syndrome, is simultanagnosia. This in the attention-processing capacity of the wide-angle
is the inability to attend to more than a very limited sec- aspect of the lens leaves only a small diameter of reso-
lution, resulting in a patient being unable to use global
form information to derive a gestalt (Thaiss &
DeBleser, 1992).
Requests for reprints should be sent to Mnica Rosselli, Florida
Balints syndrome is associated with bilateral dam-
Atlantic University, Department of Psychology, 2912 College age to the parieto-occipital areas of the cerebral cortex,
Avenue, Davie, FL 33314-7714, USA. E-mail: mrosell@fau.edu with disrupted connections between the visual cortical
242
BALINTS SYNDROME
regions and the prerolandic motor areas (Ayuso- impairment. Their patients were able to learn new
Peralta et al., 1994). The fixed gaze, or ocular apraxia, visualverbal associations, but the generalization of
may be due to the inability to attend to peripheral tar- this learning was scarce.
gets or to the loss of parietal neurons involved in visu- The aim of this article is to analyze the neuro-
ally guided eye movements (Watson & Rapscak, psychological profile and the cognitive rehabilitation
1989). Hof, Bouras, Constantinidis, and Morrison protocol in a 23-year-old man diagnosed with Balints
(1990) added that ocular apraxia may exist because of syndrome.
damage in Brodmanns area 7b in the posterior parietal
lobe. This cortical area appears to be involved in tar-
geting the gaze on a stimulus and controlling smooth- Case Report
pursuit ocular movements.
Disruption of fronto-parieto-occipital pathways Clinical History
subsequent to parieto-occipital lesions has been impli-
cated in the manifestation of Balints syndrome The patient was a 23-year-old male officer in the
(Girotti et al., 1982; Hof et al., 1990). Patients with Colombian army whose medical history was unremark-
this syndrome present with optic ataxia that implies a able prior to a serious car accident. In this accident, the
dysfunctional transferring of information from sen- patient received multiple limb fractures, including that
sory to motor areas. Damage directly to the fronto- of the left femur; at no time did he experience loss of
parieto-occipital pathways, however, is not necessary consciousness. The patient did not show any neuropsy-
to produce the symptoms of Balints syndrome chological deficits at admission to the hospital. How-
(Girotti et al., 1982). ever, 3 days after the accident, the patient experienced
Lesions in the superior occipital cortex have been a fat embolism, probably due to the multiple fractures,
implicated in difficulty with constriction of the visual which produced an occlusion in the posterior cerebral
field (Hof et al., 1990), as in simultanagnosia. Watson arteries (vertebro-basilar system; Adams, Victor, &
and Rapscak (1989) added that the restricted visual Ropper, 1997). This condition initially resulted in the
attention of simultanagnosia may be explained as result- patient experiencing cortical blindness. One week later,
ing from bilateral damage to the ambient visual system. the patient was given the diagnosis of Balints syn-
It functions to signal peripheral visual stimuli as well as drome, as evidenced by ocular apraxia, optic ataxia,
to shift attention from a stimulus being viewed by the and simultanagnosia. A CT scan was performed and
peripheral system, with a sparing of the foveal system. indicated significant bilateral parieto-occipital damage.
In terms of the specific rehabilitation of visuo- The patient remained hospitalized for several months.
perceptual disorders, such as Balints syndrome, the lit- A neuropsychological assessment was given
erature is virtually silent. Based on the main character- 12 months after the vascular accident and before any
istics of the syndrome, the rehabilitation training cognitive retraining. Table 1 shows the tests used and
should focus on the improvement of visual scanning, the scores obtained by the patient in this first neuro-
the development of visually guided manual move- psychological evaluation.
ments, and the improvement of the integration of visual
elements. Lopera (1996) and Solhberg and Mateer
(1989) recommended exercises that allow practice of Neuropsychological Test Results
spatial analysis and visuomotor skills. Perez, Tunkel,
Lachmann, and Nagler (1996) described the individual- The patient was right-handed and was alert and
ized rehabilitation of three cases of Balints syndrome. cooperative. He was oriented to person, place, and
The mean age of the three patients was 61. The etiology time. No gross motor deficits were observed. Speech
of two of them was degenerative (posterior cortical was abundant and contained no articulation, grammat-
atrophy), and one was vascular (bilateral occipito- ical, or prosodic errors. During conversation, the
parietal infarct). Treatment was focused on compensa- patient did not establish visual contact. There were no
tory strategies to visual impairment and memory defects in language comprehension, and the patient
deficits and strategies to improve reading and calcula- had preserved judgment and insight. He was aware and
tions. A significant improvement in visual recognition concerned about his neurological deficits. Depression
was reported 6 months after treatment initiation. was evident in his subjective report of feelings of
Humphreys and Riddoch (1994) described the cogni- worthlessness due to his visual problems; however, he
tive rehabilitation of two cases with visual memory did not meet the Diagnostic and Statistical Manual of
243
ROSSELLI, ARDILA, & BELTRAN
Table 1. Results Obtained in Different Neuropsychological No visual field defects were presented, but horizon-
Tests Before and After the Treatment Protocol tal diplopia was evident. The patient was able to draw
a clock and a map of Colombia adequately but
Test Before After
painstakingly. The spontaneous drawing of geometri-
WAIS: Verbal IQ 115 118 cal figures was found to be relatively well preserved,
Wechsler Memory Scale: MQ 90 92 but his copying of the same figures was deficient and,
ReyOsterrieth Complex Figure: 2/36 15/36
on occasions, impossible. The copying of the
Copy
Letter Reading 16/20 20/20 ReyOsterreith Complex Figure (ROCF) Test (Oster-
Word Reading 0/20 18/20 reith, 1944) was defective. He was unable to locate a
Verbal Fluency point in the center of a circle.
Phonologic 15 In summary, results from the neuropsychological
Semantic 18 testing led to the conclusion that the patient had
Recognition of Real Objects 7/7
Recognition of Schematized Figures 0/15 12/15
Balints syndrome characterized by ocular apraxia,
Recognition of Overlapped Figures 0/15 9/15 optic ataxia, and simultanagnosia. He also had alexia
Recognition of Colors 6/6 without agraphia, prosopagnosia, visual agnosia for
Boston Naming Test 42/60 50/60 schematized objects, and a diminished memory
Trail-Making Test A 15 min 1 min, capacity.
30 secs
Based on the deficits revealed in the neuropsycho-
Famous People Photographs 2/10 9/10
Ideomotor Praxis 10/10 logical tests (see Table 1), a cognitive rehabilitation
protocol was developed, with focus on the improve-
Note: WAIS ! Wechsler Adult Intelligence Scale. ment of the visuoperceptual symptoms. The rehabilita-
tion protocol was based on the process-specific
Mental Disorders (4th ed., [DSMIV]; American Psy- approach (Solhberg & Mateer, 1989) and attempted to
chiatric Association, 1994) diagnosis criteria for a retrain specifically the impaired visual spatial func-
major depressive episode. His Verbal IQ (Wechsler, tions (Prigatano, 1986, 1999).
1968) was within normal limits. Mild memory prob- According to the report of the patients mother, no
lems were observed. His MQ (Wechsler, 1945) was noticeable improvement of the symptoms had
lower than Verbal IQ, suggesting a mild deficit. occurred during the 6 months prior to the initial test-
A complete Balints syndrome was corroborated. The ing. The effectiveness of the treatment was assessed
patient displayed ocular apraxia, optic ataxia, and simul- through a retest neuropsychological evaluation and
tanagnosia. He was unable to scan moving target objects; generalization of treatment to daily life activities.
eye movements under verbal command were very diffi-
cult. His inability to reach target objects (misreaching)
and his dysmetria secondary to visuoperceptual deficits Rehabilitation Procedure
were interpreted as an optic ataxia. He had very severe
difficulty in visual searching and made compensatory The rehabilitation program used visuoperceptual
head movements. In attempting to describe a compli- retraining and a functional adaptation program, as
cated figure, he could only describe individual details, described in the following.
with failure to recognize the integrated figure (simul-
tanagnosia). He presented significant difficulties in the
identification of photographs of famous people (Ardila Visuoperceptual Retraining
& Rosselli, 1990) and in the recognition of schematized
objects. During 12 months, the patient attended an outpa-
The patient recognized handwritten letters and was tient rehabilitation program 2 hr a week. These ses-
able to perform tactile identification of letters cor- sions were devoted to teaching the patient visual exer-
rectly. He also recognized words spelled by the exam- cises to be practiced at home. Some exercises were
iner but was unable to read them. He was capable of specially developed to target the patients impair-
writing spontaneously and to dictation but was unable ments. The participation of the family (mother and sis-
to copy words or phrases. Given his visuoperceptual ter) was essential within the retraining process. They
difficulties, the distribution of his writing was erratic. were periodically interviewed and informed us about
No components of apraxic agraphia were observed in the patients progress. They also reported on the gen-
his handwriting. eralization of the training at home.
244
BALINTS SYNDROME
The following exercises were used within the train- Trail-making exercises. Exercises similar to the
ing protocol. Trail-Making Test A were developed, using an increas-
ing level of difficulty: The distance between numbers
Eye movements. The patient first had to visually and the amount of numbers increased; the size of the
follow objects moved by the examiner. The aim of this numbers became smaller. Initially, colors were used to
exercise was to improve visual scanning. increase salience, but later these were eliminated.
The patient then had to place the index fingers at
a distance of 15 to 20 cm in front of his face and look Writing exercises. The patient had to write letters
alternately toward the right index finger, then the left and words, spontaneously and to dictation, on paper
index finger, during a 5-min period. Three centime- with lines. Large letters were required to facilitate sub-
ters was the initial distance between the index fin- sequent self-reading.
gers. This distance was progressively increased
based on the improvement in visual targeting with-
out head movements. The distance between the fin- Functional Adaptation Program
gers was increased up to 50 cm. The aim of this exer-
cise was to increase the self-control of visual The applicability to everyday life of skills learned or
scanning. relearned during rehabilitation in a laboratory or clini-
cal setting has come into question (Sbordone & Long,
Convergence exercises. From a central point 1996). Thus, it was advisable to devise a rehabilitation
30 cm away from the face, the patient should bring the program that would, at least in a partial fashion, facili-
right or left index finger gradually closer to the nose, tate improvement in ways that are useful in the patients
maintaining permanent eye contact on the index finger. world. Bearing this in mind, a functional program,
This exercise should be done for approximately 30 min complementary to the visuoperceptual retraining, was
each day. The aim of this exercise was to acquire more developed. The focus of this complementary program
control of eye movements. was to assign tasks that, although part of everyday life
(e.g., reading a newspaper), would address the retrain-
Word reading. Single two-syllable words were ing of the deficits found in Balints syndrome. In
presented on white index cards. The patient was addition, it would increase independence and self-
requested to use the index finger as an aid to find the confidence. With this in mind, the intended goal was
letters of the words and to spell out loud all the words for the patient to return to work at his highest capacity.
he was reading. This task was alternated with the read- The following activities were recommended:
ing of two-syllable words that were written in large
size (2 cm in height) and in two colors. The vowels 1. The patient was encouraged to go out with his
were written in red, and the consonants were written in mother or his sister as frequently as possible and to
green. The patient was asked to read only the vowels, participate in daily life activities, such as shopping,
using the finger as a guide. The letters used in this going to parks, and so on. In addition, social activities
exercise were progressively smaller in size, whereas were reinforced in an attempt to reduce isolation and to
words had more syllables. The aim of this exercise was increase self-confidence. This was an initial step, con-
to teach the patient to scan and read words by spelling sidering that the patient had become socially isolated
them out. after the vascular accident.
2. The use of public transportation to come to the
Visuokinetic functioning. Letters and words rehabilitation sessions was suggested. This required
were placed in front of the patient. He had to read them that the patient use visuoperceptual skills and fostered
aloud and simultaneously reproduce the letter in the air a sense of independence. Initially, the patient was
using the index finger. The aim of this exercise was to escorted to the sessions, but later he independently
develop a kinesthetic learning of words. used public transportation.
3. It was suggested to the patient to remain active
Exercises of visual search. The patient had to at home: do homework, help with housekeeping,
look for letters and words in a crossword puzzle. The watch television, and try to read the newspaper. Dur-
level of difficulty was progressively increased. The ing each session, he had to report at least three topics
aim of this exercise was to improve visual scanning from the previous days newspaper that were interest-
and recognition of words. ing to him.
245
ROSSELLI, ARDILA, & BELTRAN
4. When the rehabilitation program was advanced Table 2. Changes in Community Independence Before and After
enough, it was suggested that the patient look in the the Treatment
newspaper for educational or vocational courses he
Before After
may be capable of attending. It was further suggested
that he go to the corresponding office, initially with a Socially Isolated Frequent Group Activities: Family,
companion and in the following months by himself, to Friends, etc.
Inability to Use Public Independent Use of Public
gather information, brochures, and so on. This would
Transportation Transportation
maximize the use of visuoperceptual skills and also Escorted to Appointments Attending Appointments by Himself
helped the patient regain confidence. Inactive at Home Reading, Writing, Watching Television
5. It was recommended to the patient that he attempt at Home
to rejoin the military. This was the patients desire, and No Attempt to Work Actively Searching for a Job
although he may rejoin at a less demanding position, this Inability to Work Returned to Work
would represent the optimal recovery scenario.
recovery is expected. Despite the difficulty of measur-
Table 2 summarizes the changes in community ing the exact amount of recovery in the months fol-
independence before and after the treatment. lowing a brain injury, it is usually accepted that a good
After 1 year on this treatment protocol, a follow-up amount of spontaneous recovery occurs during the first
neuropsychological evaluation was performed (see year. Converging evidence, however, supports that
Table 1). A significant improvement was observed, increased functioning can be observed with treatment
particularly in those tests that are sensitive to scanning begun later than 1 year (Prigatano, 1999). Recovery in
deficits, such as word reading and the Trail-Making stroke patients is less pronounced than recovery in
Test, as well as measures sensitive to simultanagnosia, posttraumatic brain injury (Kolb & Whishaw, 1996). It
such as the ROCF: copy, recognition of schematized is suitable to propose that, in the case described here,
figures, recognition of overlapping figures, and most of the spontaneous recovery had occurred at the
famous people photographs. No significant improve- beginning of the rehabilitation treatment. Most new
ment was seen in the Wechsler Memory Scale or in visual spatial improvements may be attributed to the
Verbal IQ. At this point, the patient reported feeling introduction of the rehabilitation protocol.
more confident. He returned to the army and was Age may affect the recovery from brain damage.
appointed to an activity not requiring special visuo- Better recovery is expected from younger patients. Our
perceptual abilities (a communication section). patient was a young adult, which may have contributed
to his good recovery. The recovery after brain damage
of soldiers in the 21-to-25 age group is greater than in
Discussion the 26-and-over age group (Kolb & Whishaw, 1996).
The patient described here had a neuropsychologi-
This article describes a case report of a 23-year-old cal evaluation before the beginning of the treatment
man presenting with Balints syndrome and analyzes protocol. These scores were used as baseline for treat-
the rehabilitation program that was effective for him. ment effectiveness. One year after the introduction of
The patient developed the visuoperceptual disorder of treatment, the same neuropsychological assessment
Balints syndrome, characterized by ocular apraxia, was done. Improvement was observed in all visual
optic ataxia, and simultanagnosia, secondary to a cere- tests but was absent in memory and intelligence tests.
brovascular accident resulting from a fat embolism. The rehabilitation program consisted of a visuopercep-
The patient also presented with pure alexia (alexia tual training, however, and no memory retraining was
without agraphia), visual agnosia to superimposed and included. Thus, it is reasonable to propose that the
schematized figures, prosopagnosia, and a diminished rehabilitation protocol used was effective. This case
memory capacity. study suggests that a successful rehabilitation program
The patient began a rehabilitation program 1 year for visuoperceptual disorders should also include a
after the embolism. Based on a family report, no spon- functional component.
taneous recovering had occurred in the previous 6 There are some limitations to this single case report.
months. Most spontaneous recovery in cases of brain There was no control over confounding variables, such
damage occurs during the first 3 months after the as the testretest effects, that may have affected the
injury, with some recovery occurring in the following neuropsychological test scores. The patient attended
6 months (Kertesz, 1979). Thereafter, very little or no the rehabilitation program a limited amount of time
246
BALINTS SYNDROME
(2 hr weekly), so there was no control over activities Holmes, G. (1918). Disturbances of visual orientation. British
performed by the patient in his own time. There was a Journal of Ophthalmology, 2, 449486.
Humphreys, G. W., & Bruce, V. (1989). Visual cognition. London:
significant similarity between some of the testing tasks
Lawrence Erlbaum Associates, Ltd.
and some of the rehabilitation exercises. For example, Humphreys, G. W., & Riddoch, M. J. (1994). Visual object pro-
the patient was practicing plenty of times (even hun- cessing in normality and pathology: Implications for rehabili-
dreds of times) in exercises similar to the Trail-Making tation. In M. J. Riddoch & G. W. Humphreys (Eds.), Cognitive
Test. At the beginning of the program, the patient was neuropsychology and cognitive rehabilitation (pp. 3976).
socially isolated and overtly depressed, whereas at the Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Kertesz, A. (1979). Aphasia and related disorders. New York:
end, depression had cleared, and he was significantly Grune & Stratton.
positive and active in his extratreatment environments. Kolb, B., & Whishaw, I. Q. (1996). Fundamentals of human neuro-
All these variables may have affected the final testing psychology. New York: Freeman.
results. Nonetheless, using a functional criterion, dur- Lopera, F. (1996). Estrategias de tratamiento en las agnosias
ing the 12 months of the rehabilitation program, a sig- [Strategies in agnosia treatment]. In F. Ostrosky, A. Ardila, &
R. Chayo (Eds.), Rehabilitacion neuropsicologica
nificant improvement was observed. The patient was
(pp. 180202). Mexico, DF: Planeta.
able to return to the army, securing a limited position Luria, A. R. (1959). Disorders of simultaneous perception in a
that did not require special visuoperceptual abilities. case of bilateral occipito-parietal brain injury. Brain, 82,
In summary, significant improvement in objective 437449.
testing and the return to a productive life can be Osterreith, P. A. (1944). Le test de copie dune figure complexe.
attained via cognitive rehabilitation, even after the Archives de Psychologie, 30, 206356.
Perez, F. M., Tunkel, R. S., Lachmann, E. A., & Nagler, W. (1996).
period in which the majority of spontaneous recovery Balints syndrome arising from bilateral posterior cortical
is likely to occur. atrophy or infarction: Rehabilitation strategies and their limi-
tation. Disability and Rehabilitation, 18, 300304.
Prigatano, P. G. (1986). Neuropsychological rehabilitation after
References brain injury. Baltimore: The Johns Hopkins University Press.
Prigatano, P. G. (1999). Principles of neuropsychological rehabili-
tation. New York: Oxford University Press.
Adams, R. A., Victor, M., & Ropper, A. H. (1997). Principles of Rizzo, M., & Hurtig, R. (1987). Looking but not seeing: Attention,
neurology (6th ed.). New York: McGraw-Hill. perception, and eye movements in simultagnosia. Neurology,
American Psychiatric Association. (1994). Diagnostic and statisti- 37, 16421648.
cal manual of mental disorders (4th ed.). Washington, DC: Sbordone, R. J., & Long, C. J. (Eds.). (1996). Ecological validity
Author. of neuropsychological testing. Delray Beach, FL: St. Lucie.
Ardila, A., & Rosselli, M. (1990). Prueba de reconocimiento de Solhberg, M. M., & Mateer, C. A. (1989). Introduction to cognitive
caras [Test of face recognition]. Unpublished manuscript. rehabilitation. New York: Guilford.
Ayuso-Peralta, L., Jiminez-Jiminez, F. J., Tejeiro, J., Vaquero, A., Thaiss, L., & DeBleser, R. (1992). Visual agnosia: A case of
Cabrera-Valdivia, F., Madero, S., Cabello, A., & Garcia- reduced attentional spotlight? Cortex, 28, 601621.
Albea, E. (1994). Progressive multifocal leukoencephalopa- Watson, R. T., & Rapscak, S. Z. (1989). Loss of spontaneous blink-
thy in HIV infection presenting as Balints syndrome. Neurol- ing in a patient with Balints syndrome. Archives of Neurol-
ogy, 44, 13391340. ogy, 46, 567570.
Balint, R. (1909). Seelenlahmung des Schauens, optische Wechsler, D. (1945). A standardized memory test for clinical use.
Ataxie, raumliche Storung der Aufmerksamkeit. Mo- Journal of Psychology, 19, 8795.
natsschrift fr Psychiatrie und Neurologie, 25, 5181. Wechsler, D. (1968). Escala de Inteligencia Wechsler para Adul-
Girotti, F., Milanese, C., Casazza, M., Allegranza, A., Corridori, F., tos. New York: Plenum.
& Avanzini, G. (1982). Occulomotor disturbances in Balints
syndrome: Anatomoclinical findings and electrooculographic
analysis in a case. Cortex, 18, 603614.
Hcaen, H., & Angelergues, T. (1963). La Ccit psychique. Paris:
Masson et Cie.
Hof, P. R., Bouras, C., Constantinidis, J., & Morrison, J. (1990).
Selective disconnection of specific visual association path-
ways in cases of Alzheimers disease presenting with Balints
syndrome. Journal of Neuropathology and Experimental Neu- Original submission December 20, 2000
rology, 49, 168184. Accepted June 14, 2001
247
Applied Neuropsychology Copyright 2000 by
2000, Vol. 7, No. 1, 316 Lawrence Erlbaum Associates, Inc.
ARTICLES
Syntactic Comprehension, Verbal Memory, and Calculation Abilities

in SpanishEnglish Bilinguals SPANISHENGLISH
ARDILA
BILINGUALS
ET AL.
Alfredo Ardila
Miami Institute of Psychology, Miami, Florida, USA
Mnica Rosselli
College of Liberal Arts, Florida Atlantic University, Davie, Florida, USA
Feggy Ostrosky-Sols
Department of Psychophysiology, National Autonomous University of Mexico,
Mexico City, Mexico
Jos Marcos
El Colegio de Mxico, Mexico City, Mexico
Instituto Mexicano de Psiquiatria, Mexico City, Mexico
Gladys Granda and Miguel Soto
Miami Institute of Psychology, Miami, Florida, USA
This article analyzes the interfering effect of the second language (L2) on the first language (L1) in
native Spanish speakers living in the United States. We examined 3 linguistic aspects: (a) syntactic
comprehension, (b) verbal memory, and (c) calculation abilities. We carried out 2 different studies.
In the 1st study, we studied syntactic understanding in 50 SpanishEnglish bilinguals. For all par-
ticipants, L1 was Spanish and L2 was English, and all learned English early in life and had attended
English schools. Results for the Spanish Syntactic Comprehension Test (Marcos & Ostrosky, 1995)
were compared with the normative results obtained with 40 Spanish monolingual participants. We
observed that the closer to the English syntax the sentences were, the easier it was for the partici-
pants to understand them. Participants who had been exposed to English between the ages of 5 and
12 outperformed participants exposed to English before 5 years of age. Language preference corre-
lated with syntactic comprehension. Women outperformed men. In the 2nd study, verbal memory
and calculation abilities were examined in L1 and L2 in a group of 85 SpanishEnglish bilinguals.
Parallel versions of the different tests were administered in Spanish and English. The results indi-
cated some significant differences between the 2 languages in several verbal learning and calcula-
tion ability subtests. Most of the verbal memory subtests were better performed in L1. Scores on
tasks measuring speed and calculation accuracy were higher in the participants native language.
Best spoken language proved to be a significant variable in some verbal memory subtests per-
formed in English but not in Spanish. We analyze implications of bilingualism in
neuropsychological testing. We also present some suggestions to minimize the bilingualism effect.
Requests for reprints should be sent to Alfredo Ardila, 12230 NW 8th Street, Miami, FL 33182, USA.
3
ARDILA ET AL.
Key words: bilinguals, Spanish, English, verbal memory, syntactic comprehension
Bilingualism is not an unusual phenomenon. More neme productions (most frequently, /r/, /p/, /l/, and
than half of the worlds population is bilingual or multi- /t/), and Spanish verbal fluency decreases (Rosselli
lingual at some level of proficiency. Bilinguals and et al., 1998). Even morphological and semantic ab-
multilinguals, however, represent an extremely hetero- normalities can be observed. This deleterious effect
geneous population. They can vary along a significant from L2 to L1 has been usually overlooked and evi-
number of dimensions, including (a) sociolinguistic dently deserves much more attention. This article an-
background, (b) type of bilingualism, (c) degree of pro- alyzes the interference from L2 to L1 in native
ficiency, (d) age and sequence of acquisition of the lan- Spanish speakers living in the United States. Three
guages, (e) method of acquisition of the second linguistic aspects are approached: (a) syntactic com-
language (L2), (f) language-specific factors, and (g) an- prehension, (b) verbal memory, and (c) calculation
atomical dimensions (Dupont, Ardila, Rosselli, & abilities. Understanding linguistic characteristics of
Puente, 1992). Although there is extensive research on Hispanic populations may be extremely important in
bilingualism, bilingualism in many regards remains a psychological and neuropsychological evaluation of
poorly understood phenomenon. U.S. Spanish speakers.
SpanishEnglish bilingualism in the United States Each language has its own syntactic idiosyncrasies.
has been the center of interest from different perspec- In contrast to English, which is a strong word-order lan-
tives, ranging from the linguistic to the political sphere. guage with relatively weak morphology, Spanish, like
The Hispanic population represents about 10% of the Italian and other Latin languages, is a weak word-order
general population. When one takes into account not language with a rich morphological system. Spanish
only the significant number of Spanish speakers but syntactic rules allow for various displacements of the
also the fact that there is one Spanish-speaking U.S. as- constituents of a sentence without modifying its essen-
sociate state (Puerto Rico), the United States can be re- tial meaning (Seco, 1980). For example, the sentence
garded, to a certain extent, as a Latin American John loves Mary can be expressed in six different
country. However, Spanish is not socially, academi- ways: Juan ama a Mara, Juan a Mara ama, A
cally, economically, and politically equivalent to Eng- Mara ama Juan, A Mara Juan ama, Ama Juan a
lish, and frequently it is maintained as a marginal Mara, and Ama a Mara Juan. In English (as in
language. Native Spanish speakers are usually required many other languages), if the sentence does not have a
to speak English at work and in general everyday activ- passive form or subordinate clauses, the canonical or-
ities. Children attend English-speaking schools. Span- der is the only one permitted.
ish books in the United States are very limited, and in It has been observed that, in English, Italian, and
general, cultural activities in Spanish are scarce. Some German, the subjectverbobject (SVO) word order
interference from English to Spanish may be conjec- is the canonical one (Bates, Friederici, & Wulfeck,
tured. Furthermore, in this particular context, Span- 1987). However, in Japanese, the canonical order is
ishEnglish bilingualism represents an active SOV (Hagiwara & Caplan, 1990). In Spanish, al-
bilingualsm (i.e., both languages are used in everyday though the canonical order is also SVO, canonicity
life), and in consequence, not only interference but also seems to be related to the distance between the nouns
mixture can be anticipated. (Ns) and the V (Marcos & Ostrosky, 1995). The N that
Interference from the first language (L1) to L2 is a appears after the V (NVN) is considered as the one
well-established observation in bilinguals (e.g., that receives the action, thus becoming the O of the sen-
Bahrick, Hall, Goggin, & Bahrick, 1994; tence, whereas the N that is left is given the role of
Dongryong, 1990; Dupont et al., 1992). Nonetheless, agent (A). Thus, in Spanish, due to its flexibility in the
interference from L2 to L1 is rarely mentioned in bi- position of the elements, the canonical order seems to
lingualism literature. It is an intuitive observation be related to the VO structure: The N that follows the
that L1 tends to deteriorate when the speaker is ex- V is the O; this is the most reliable cue to grammatical
posed to an alien linguistic environment. Everyday roles. Thus, once the O is identified, the N that is left is
observation illustrates that Spanish speakers living given the role of A.
for some time in the United States frequently present Marcos and Ostrosky (1995) observed that, in the
difficulties in using Spanish orthography, minor pho- interpretation of sentences, normal participants fol-
netic abnormalities may be observed in some pho- lowed different strategies. The first depends on a
4
SPANISHENGLISH BILINGUALS
grammatical knowledge of the function of the prepo- Theoretical discussions have debated the existence of
sition a (to) as a sign of the direct O. Further, the au- two separate storing places versus one integrated storing
thors observed a tendency to interpret the nominal place in the bilingual persons memory (Hummel, 1986;
phrase that precedes the transitive V as the S (A) and McCormack, 1976; Paradis, 1993). Differences may in-
the one that appears immediately after the V as the O. volve the encoding and retrieval systems used in tasks
For example, in sentences such as Un len golpe un performed in both languages (Durgunoglu & Roediger,
tigre (A lion hit a tiger), the tiger is given the role 1987). Because of the heterogeneity of bilingual popula-
of the O because it is placed immediately after the V, tions, results may vary between these two systems based
whereas the lion is assigned the role of A. However, on the bilingual sample that is selected.
when the syntactic ambiguity of this sentence be- Paivio and Lambert (1981) predicted and found that
comes apparent and the elements that compose it are words coded bilingually were retained better than words
moved (e.g., Golpe un len un tigre), the role of O coded unilingually. They suggested that the bilinguals
is usually assigned to the lion because of its position processing is composed of translation, which is a deeper
following the V. This differs from the descriptions and more intense encoding and decoding than that in-
made in other languages such as English (Bates, volved in encoding one language. McCormack (1976)
Friederici, Wulfeck, & Juarez, 1988; Caplan & Futter, argued that verbal memory, as other aspects of memory,
1986; Wulfeck, Bates, & Capasso, 1991). When the may include a significantly high number of attributes,
preposition is lacking or when the order does not cor- such as speakers voice, spatial position, and orthogra-
respond to canonical order in Spanish (SVO), other phy. Kolers (1978, 1979) hypothesized that knowledge
cues are observed for understanding. Among these is organized according to the means and procedures by
other strategies is the use of articles to mark the A or which it was acquired and forms part of our memory. He
patient. When presenting an N in a sentence preceded concluded that the organization of knowledge and mem-
by the definite article el (the) or the indefinite article ory are language dependent.
un (a), different information about the N is being pro- Few studies have approached the question of verbal
vided for assigning thematic roles. The different strat- memory in SpanishEnglish bilinguals. Harris,
egies interact very peculiarly and consistently. Cullum, and Puente (1995) developed equivalent
Bilinguals, and particularly balanced bilinguals, are list-learning tests in Spanish and English and adminis-
required to simultaneously rely on two different tered them to three different groups of participants:
morphosyntactic systems. A certain degree of interfer- monolingual English, balanced SpanishEnglish
ence may be anticipated. Furthermore, it might be as- bilinguals, and nonbalanced SpanishEnglish
sumed that difficulties in complex syntactic bilinguals. They observed that nonbalanced bilinguals
comprehension should be evident in those syntactic assessed in English learned fewer words overall and
principles separating the two languages. It can be pre- demonstrated lower retention scores compared to
dicted that SpanishEnglish bilinguals may prefer the monolinguals. Comparisons of groups assessed in their
canonical order when speaking Spanish. Moreover, dominant languages (Spanish or English) revealed no
they may face some difficulties in understanding Span- significant differences on verbal memory performance.
ish noncanonical word order as a result of the English The second study presented in this article continued
influence. We selected a bilingual SpanishEnglish the analysis of verbal memory in SpanishEnglish
sample and analyzed syntactic understanding in Span- bilinguals. The ability to perform calculation tasks in
ish in the first study reported in this article. Spanish and English was further analyzed. It was hy-
Verbal memory in bilinguals has been the focus of pothesized that (a) verbal memory and calculation tasks
interest of several research studies (Peynircioglu & would be better performed in L1 than in L2, and (b)
Durgunoglu, 1993); however, the results have been when similar syntactic structures between L1 and L2
mixed. Sometimes performance in memory tasks is were used, syntactic understanding would be higher.
similar in both languages, sometimes bilinguals seem
to be at a disadvantage, and sometimes bilinguals ap- Study 1
pear to be at an advantage (Magiste, 1980; Ransdell &
Fischer, 1987; Snodgrass, 1984). Peynircioglu and Method
Durgunoglu suggested that, depending on the specific
context of the memory task performed, bilingualism Participants. Fifty middle socioeconomic status
may increase or decrease performance and that a simple SpanishEnglish bilinguals (18 men and 32 women;
and unidirectional effect may not be expected. age M = 28.7 years, SD = 7.1) with a minimum of 12
5
ARDILA ET AL.
years of schooling were selected. In addition to gather- tion of the matching picture is randomized within
ing the general demographic data, we gave each partic- sentence-distracter types. In the Spanish Syntactic Com-
ipant a special questionnaire to assess the characteris- prehension Test, the effects of different factors are stud-
tics of their bilingualism. For all the participants, L1 ied: sentence type, presence and absence of the preposi-
was Spanish and L2 was English, and all learned Eng- tion, and the use of the definite or indefinite article.
lish early in life and had attended English schools. Fre- The Spanish Syntactic Comprehension Test has the
quently, they mastered English better than Spanish. following characteristics:
Twenty-eight participants had been exposed to Eng-
lish before the age of 5, and 22 had been exposed be- 1. Ninety active, 65 passive, and 35 pseudocleft re-
tween ages 5 and 12. All participants were living in a versible sentences are included, with a total of
bilingual community (Miami, FL) and using both lan- 190 sentences.
guages in everyday life activities. Some of the partici- 2. According to the position of the nominal phrases
pants were second-generation Latin American immi- representing the A and the O of the transitive V,
grants born in the United States, but other participants there are three different orders in the corpus for
were born abroad and had arrived in the United States the active, six for the passive, and four for the
before the age of 12. All the participants parents were pseudocleft sentences.
native Spanish speakers born in Spanish-speaking 3. All the orders of constituents allowed in Spanish
countries, and Spanish was always the language spo- for each type of sentence are used, not including
ken at home during childhood. All participants were agrammatical sentences.
nonpaid volunteers. 4. In half of the sentences, the first N is the A, and
Results of the Spanish Syntactic Comprehension in the other half, the second N is the A.
Test (Marcos & Ostrosky, 1995) were compared to the 5. The function of the preposition a (to) is ana-
normative results obtained from the Spanish monolin- lyzed. Active sentences with and without a prep-
gual participant sample (N = 40) from Mexico City with osition are included.
a similar educational, age, and sex distribution. Norma- 6. The definite and indefinite articles are distrib-
tive results of this Mexican sample have been presented uted in the same proportion in all the sentences.
elsewhere (Marcos & Ostrosky, 1995). In Miami, it is
virtually impossible to find a normal Spanish monolin- Thirty-eight different types of sentences and five ex-
gual population sample with similar education, age, amples of each type of sentence are included in the test.
and sex characteristics. All native Spanish speakers in Examples are presented in Table 1.
Miami, except a few elderly people and some very re-
cently arrived immigrants, are bilingual to some de-
gree. This was the only reason for using the normative Procedure. Before participants started the test,
results obtained from Mexico City. the ability to name the animals and actions was deter-
mined by means of picture identification. The partici-
Instrument. The Spanish Syntactic Comprehen- pant was then told that a pair of animals would be in-
sion Test developed by Marcos and Ostrosky (1995) was cluded in each sentence, one of which would perform an
used. This is a normalized instrument used to assess the action and the other that would be affected by it. The
syntactic comprehension of Spanish. A forced election participant was further instructed to point at the picture
task is used in which the participant listens to 190 differ- corresponding to the sentences read by the examiner. A
ent reversible sentences and is asked to select, by point- set of 20 different plates, each one containing four pic-
ing, one of four options presented on a plate. Each option tures, was used during sentence presentation. Sentences
contains a pair of animals performing a specific action, were read with a normal intonational contour by the ex-
and only one option is correct. The four options corre- aminer. Participants were not limited in their time to re-
spond to (a) correct action and correct animal, (b) same spond. Each sentence was repeated once to the partici-
action but performed by the wrong animal, (c) same ani- pant on request. No feedback with regard to the
mals but performing a different action, and (d) different accuracy of the answer was provided. The 190 random-
pair of animals and the same action. Four pairs of animals ized sentences (38 different types of sentences, each one
(liontiger, camelhorse, duckrooster, rabbitskunk) appearing five times but using different Ns and Vs)
and five different transitive Vs (hit, kick, bite, pull, and were presented in a single session lasting about 30 to 40
pick) are used based on pragmatic knowledge that both min. A native Spanish-speaking graduate psychology
animals are equally able to perform the action. The posi- student performed the testing.
6
Results active reversible sentences without a preposition.

Differences in passive sentences were virtually non-
Table 1 compares the number of correct responses existent. The order of difficulty in the bilingual sam-
between the normative sample and the SpanishEng- ple was passive sentences (easiest), active reversible
lish bilingual group. We observed that, in general, the sentences with preposition, active reversible sen-
performance on the Spanish Syntactic Comprehen- tences without preposition, and sentences with a
sion Test was higher in the monolingual sample than pseudocleft A (hardest).
in the bilingual sample. Differences were most evi- Differences in the understanding of the following
dent in sentences with a pseudocleft A, followed by five types of sentences were statistically significant:
Table 1. General Results: Percentage of Correct Responses
Order Article Preposition Example Monolingualsa Bilingualsb 2 p
I. Active Reversible Sentences Without Preposition

1. AVO D/I No El len golpe un tigre. 100.0 98.0 0.10 ns
2. OVA I/D No Un tigre golpe el len. 66.5 17.2 92.07 .001
3. AOV D/I No El len un tigre golpe. 94.5 65.2 23.10 ns
4. OAV I/D No Un tigre el len golpe. 83.5 75.6 5.25 ns
5. VAO D/I No Golpe el len un tigre. 94.5 74.0 11.40 ns
6. VOA I/D No Golpe un tigre el len. 65.5 55.6 3.53 ns
II. Active Reversible Sentences With Preposition
7. AVO D/I Yes El len golpe a un tigre. 100.0 99.6 0.00 ns
8. AVO D/I Yes Un len golpe al tigre. 98.5 98.8 0.00 ns
9. AOV D/I Yes El len a un tigre golpe. 100.0 87.2 4.20 ns
10. AOV I/D Yes Un len al tigre golpe. 98.5 85.2 4.68 ns
11. VAO D/I Yes Golpe el len a un tigre. 98.5 92.8 0.91 ns
12. VAO I/D Yes Golpe un len al tigre. 98.5 91.6 1.31 ns
13. OVA D/I Yes Al tigre lo golpe un len. 100.0 98.8 0.03 ns
14. OVA I/D Yes A un tigre lo golpe el len. 100.0 96.0 0.40 ns
15. OAV D/I Yes A el tigre un len lo golpe. 97.5 95.6 0.06 ns
16. OAV I/D Yes A un tigre el len lo golpe. 100.0 92.4 1.44 ns
17. VOA D/I Yes Golpe al tigre un len. 96.0 86.0 2.60 ns
18. VOA I/D Yes Golpe a un tigre el len. 97.5 83.6 4.75 ns
III. Passive Reversible Sentences
19. OVA D/I Yes El tigre fue golpeado por un len. 100.0 99.6 0.02 ns
20. OVA I/D Yes Un tigre fue golpeado por el len. 98.5 96.0 0.19 ns
21. AVO D/I Yes Por el len fue golpeado un tigre. 96.0 91.2 0.60 ns
22. AVO I/D Yes Por un len fue golpeado el tigre. 86.5 92.8 1.03 ns
23. OAV D/I Yes El tigre por un len fue golpeado. 98.5 95.6 0.25 ns
24. OAV I/D Yes Un tigre por el len fue golpeado. 98.5 96.8 0.10 ns
25. VOA D/I Yes Fue golpeado el tigre por un len. 97.0 96.4 0.01 ns
26. VOA I/D Yes Fue golpeado un tigre por el len. 100.0 94.8 0.67 ns
27. VOA D/I Yes Fue golpeado el tigre por un len. 93.5 83.2 2.68 ns
28. VAO I/D Yes Fue golpeado por un len el tigre. 98.5 89.2 2.33 ns
29. AOV I/D Yes Por el len un tigre fue golpeado. 94.5 96.4 0.06 ns
30. AOV D/I Yes Por el len un tigre fue golpeado. 98.5 96.8 0.10 ns
IV. Pseudocleft Agent With Preposition
31. VAVO D/I Yes Fue el len lo que golpe a un tigre. 98.5 90.4 1.78 ns
32. VAVO I/D Yes Fue un len lo que golpe al tigre. 100.0 95.6 0.48 ns
33. VOVA D/I Yes Lo que golpe al tigre fue un len. 97.0 94.0 0.26 ns
34. VOVA I/D Yes Lo que golpe a un tigre fue el len. 98.5 88.8 2.53 ns
35. VAOV D/I Yes Fue el len lo que a un tigre golpe. 93.5 34.8 91.48 .001
36. VOAV I/D Yes Fue a un tigre lo que el len golpe. 97.0 46.0 67.42 .001
37. OVVA D/I Yes Lo que al tigre golpe fue un len. 98.5 34.4 104.94 .001
38. OVVA I/D Yes Lo que a un tigre golpe fue el len. 98.5 31.6 114.26 .001
Note: All sentences mean The lion hit the tiger. A = agent; V = verb; O = object; D = definite; I = indefinite.
a
n = 40. bn = 50.
7
ARDILA ET AL.
1. Active reversible sentences without a preposi- cally significant for all sentences with pseudocleft A
tion (OVA; e.g., Un tigre golpe el len [A and a preposition.
tiger hit the lion]). Table 3 compares the effect of the preferred language
2. Pseudocleft A with a preposition (VApa- variable. Participants preferring Spanish outperformed
tientV; e.g., Fue el tigre lo que a un len those participants preferring English. In two sentences
golpe [It was the tiger that a lion hit]). (pseudocleft A with preposition with the order VAOV,
3. Pseudocleft A with a preposition (Vpa- and pseudocleft A with a preposition with the order
tientAV; e.g., Fue a un tigre lo que el len OVVA), the differences were statistically significant.
golpe [It was a tiger that the lion hit]). Table 4 compares the effect of the sex variable. We
4. Pseudocleft A with a preposition (pa- noted that the number of correct responses was about
tientVVA; e.g., Lo que al tigre golpe fue twice as high in women than in men. Except for the first
un len [What the tiger hit was a lion]). sentences (active reversible sentences without preposi-
5. Pseudocleft A with a preposition (pa- tion), these differences were statistically significant.
tientVVA; e.g., Lo que a un tigre golpe fue
el len [What {to} the tiger hit was the lion]). Study 2
Three variables were further analyzed: age when ex- Method

posed to the L2, preferred language, and sex. Only the
five sentences understood significantly differently be- Participants. Participants were 69 middle socio-
tween monolinguals and bilinguals were included in this economic status SpanishEnglish bilinguals. They were
analysis. Table 2 compares the effect of age when ex- nonpaid volunteer graduate students recruited at an ad-
posed to the L2. We observed that those participants ex- vanced studies institution in Miami, Florida. Their ages
posed to the L2 (English) between ages 5 and 12 ranged from 18 to 49 (M = 30.28 years, SD = 7.97). There
performed better than those individuals who were ex- were 34 men and 35 women, of whom 27 were born in the
posed before 5 years of age. Differences were statisti- United States and 42 were born in Latin America (Cuba =
Table 2. Age Exposed to Second Language Effect: Percentage of Correct Responses
Type of Sentence and Order Article Preposition < 5a 512b 2 p

2. OVA I/D No 11.4 24.5 0.84 ns
35. VAOV D/I Yes 16.4 58.2 10.51 .01
36. VOAV I/D Yes 22.8 75.4 12.71 .01
37. OVVA D/I Yes 12.1 61.8 19.93 .01
38. OVVA I/D Yes 13.5 54.5 13.13 .01
Note: O = object; V = verb; A = agent; I = indefinite; D = definite.

a
n = 22. bn = 28.
Table 3. Preferred Language Effect: Percentage of Correct Responses
Type of Sentence and Order Article Preposition Spanisha Englishb 2 p

2. OVA I/D No 23.3 13.7 3.68 ns
35. VAOV D/I Yes 46.6 28.1 6.06 .05
36. VOAV I/D Yes 55.5 40.6 3.06 ns
37. OVVA D/I Yes 44.4 28.7 4.08 .05
38. OVVA I/D Yes 38.8 27.5 2.71 ns

a
n = 18. bn = 32.
8
Table 4. Sex Effect: Percentage of Correct Responses
Type of Sentence and Order Article Preposition Mena Womenb 2 p

2. OVA I/D No 12.2 20.0 1.63 ns
35. VAOV D/I Yes 16.6 43.7 11.75 .01
36. VOAV I/D Yes 27.7 56.2 9.70 .01
37. OVVA D/I Yes 20.0 42.5 8.52 .01
38. OVVA I/D Yes 17.7 39.3 7.96 .01

a
n = 18. bn = 32.
19; Puerto Rico = 6; Colombia = 6; Venezuela = 3; Chile = Materials and procedure. A demographic ques-
3; Panama, Ecuador, Dominican Republic, Peru, and Ar- tionnaire that we designed was administered to evaluate
gentina = 1 in each country). In everyday life, they used
English, Spanish, and Spanglish (i.e., mixture of Span-
ish and English with frequent code switching, and also
1. In Spanish, there is not an exact word corresponding to the
Spanish spoken including a significant amount of English borrowed word (e.g., driveway).
borrowings) for communicating. All participants claimed 2. The borrowed word presents some cultural salience in the
fluency in both languages, although the ability to write and new U.S. cultural context but not in the native one (e.g., suit as an ap-
read Spanish was limited in most participants born in the peal or legal demand).
United States or arriving early in life. On a self-evaluation 3. The borrowed word has a very exact referent, or it is a sort of
proper name (e.g., income tax)
scale ranging from 1 (virtually nothing) to 5 (excellent), 4. In Spanish, there are several potentially correct words, but no
they rated themselves on spoken Spanish and English at one has the exact meaning (e.g., the word ratio corresponds in Span-
the level 3 (well) or above. All the participants had Spanish ish to razn or relacin, but for both, the meaning corresponding to
as their L1 but had used English for more than 3 years. ratio is the secondary meaning, not the primary one), or there are
Some participants had been exposed to both languages significant regional variations with regard to that particular word.
That is, although several words exist, no one word seems to be the
since birth, and both languages could be considered as
superordinate word (e.g., tag is placa, chapa, or tablilla; balloon is
their L1. Twenty-seven participants stated that they pre- bomba or globo; carpet corresponds in Spanish to both tapete and
ferred Spanish and spoke better Spanish than English (9 alfombra). In different Spanish-speaking areas, one of the two is
had learned English before age 12, and 18 had learned after preferred. Some mild variations with regard to the exact meaning
age 12); 39 participants stated that they preferred English can exist. So, tapete may denote it is smaller, alfombra may imply it
and spoke better English than Spanish (38 had learned is woven, and so on. To avoid having to make a selection from sev-
eral alternatives, the superordinate English word is preferred.
English before age 12, and 1 had learned after age 12). 5. From the phonological point of view, the borrowed word is
Three pointed out that they could speak either language at accessible with the Spanish phonology (i.e., it is quite unlikely that a
the same level, and they did not have any significant differ- word with an extremely difficult phonology for Spanish speakers,
ence in preference. These 3 participants were deleted from such as girl, will be borrowed).
the sample in some analyses. 6. When the English word is phonologically simpler than the
corresponding Spanish word, the English word can be borrowed
The age of acquisition of English ranged from 0
(e.g., the English word pin is a notoriously simpler word than the
(Spanish and English were simultaneously learned) to Spanish word alfiler).
35 years (M = 10.5 years, SD = 8.88). As a result of liv- 7. Very high frequency words and, in consequence, overlearned
ing in a bilingual social environment (Miami, FL), bor- words are not usually borrowed from the L2 (e.g., words such as
rowed words from English were frequently observed casa [house] and mano [hand] are never borrowed).
when participants spoke Spanish.1 8. Use of technical words usually are also borrowed in other lan-
guages, becoming international words (e.g., software).
9. When a word is frequently used in everyday life in English,
there is a tendency to continue using it when speaking native Span-
1
Rules governing borrowing are not well defined yet. Why is a ish language (e.g., mailbox).
particular word (and not another) borrowed? Everyday observation 10. Finally, when the word has been learned in English, the
makes it apparent that in the majority of cases, the borrowed word is meaning is more directly accessed in English than in Spanish. This
a N, and only sporadically an adjective or V. Grammatical connectors may be particularly true with some professional and academic
are not borrowed. One may hypothesize that there are several simul- words (e.g., standard deviation). The very same principles may be
taneously acting principles: assumed to apply in other similar linguistic contexts.
9
ARDILA ET AL.
the characteristics of participants bilingualism. Five Time and errors were scored. However, errors were
subtests taken from the Wechsler Memory Scale so low that no statistical analysis of errors was possible.
(WMS) English version (Wechsler, 1945, 1987) and its Results, in consequence, refer only to time (sec).
equivalent translations to the Spanish language were After participants performed the calculation tasks,
used to compose an instrument that could evaluate we assessed delayed recall with the following tests:
memory in both languages. In addition, the Serial Ver-
bal Learning Test (SVL; Ardila, Rosselli, & Puente, 1. Delayed Associative Learning (Spanish).
1994) was administered in both languages. The result- 2. Delayed Logical Memory (English).
ing verbal memory section consisted of eight verbal 3. Delayed SVL (Spanish).
memory subtests administered in the following order 4. Delayed Associative Learning (English).
and language: 5. Delayed Logical Memory (Spanish).
6. Delayed SVL (English).
1. WMS Logical Memory (Spanish).
2. SVL (English). WMS administration and scoring followed the guide-
3. Digits ForwardBackward (Spanish). lines provided by the WMS manual (Wechsler, 1945,
4. WMS Associate Learning (English). 1987). In the SVL Test, three different scores were used:
5. SVL (Spanish). (a) number of words recalled in the first trial, (b) number
6. WMS Logical Memory (English). of trials required to recall the 10-word list, and (c) de-
7. WMS Associate Learning (Spanish). layed recall of the words. For the calculation portion,
8. Digits ForwardBackward (English). time was recorded for each subtest. Testing (including
the demographic and history of bilingualism sections)
lasted approximately 45 min. Performance in Spanish
Instructions were presented in the corresponding
and English was compared using t tests. The age of ac-
language (Spanish or English), and participants were
quisition effect and preferred language effect were ana-
requested to use only Spanish or English according to
lyzed using 2 2 factorial analyses of variance.
the specific subtest.
To assess the participants calculation ability in
Spanish and English, three basic arithmetical opera-
tions and one numerical problem were performed aloud Results
in both languages. This section was presented in the
following order and languages: Table 5 summarizes the results on the verbal and ar-
ithmetical tests used in this study. In general, perfor-
1. Successive subtraction from 100 (to subtract 7 in mance was better in Spanish than in English. In the
a successive way from 100; Spanish). verbal memory subtests, out of 10 memory subtest
2. One multiplication (12 23; Spanish). scores, 4 (Delayed Logical Memory, Digits Forward,
3. One division (140 15; Spanish). Total Digits, and Total Words) reached a statistical
4. Successive subtractions from 100 (to subtract 9 level of significance. Performance in English was
in a successive way from 100; English). better in Delayed Logical Memory and the 2 Digits
5. One multiplication (13 26; English). scores. Performance was higher in Spanish in the Total
6. One division (170 13; English). Words of the SVL. On the calculation abilities test, per-
7. One numerical problem (Si tengo 18 libros formance was significantly faster in Spanish in 3 out of
distribuidos en dos paquetes, y en uno de los the 4 scores.
paquetes hay el doble de libros que en el otro, The effects of age of acquisition of the L2 was ana-
cuntos libros hay en cada paquete? [If I lyzed (Table 6). A language effect favoring Spanish
have 18 books distributed into two groups, and was observed in the WMS Delayed Logical Memory
in one of the groups there are twice as many and WMS Delayed Associative Learning, SVL De-
books as in the other, how many books are there layed Recall, and SVL Total Words. An age effect was
in each group?]; Spanish). found only in the WMS Digits subtest. Few interactions
8. One numerical problem (The sum of the ages of were significant: WMS Logical Memory (both condi-
a father and a son is 48. If the father is triple the tions) and SVL Total Words and number of trials. A
age of the son, how old are each one of them?; language effect was observed in three out of the four
English). calculation abilities scores. Age effect was found in the
10
Table 5. General Results in the Verbal Memory and Calculation ory (both conditions), Total Digits, and SVL Total
Tests Words. In the calculation abilities subtests, a language
effect was observed in all scores, whereas the preferred
Test English Spanish t p
language affected the numerical problem. Two interac-
Wechsler Memory Scale tions were significant: multiplication and division tests.
Logical Memory
M 10.28 10.36 0.19 .847
SD 3.46 2.93
Delayed Logical Memory Discussion
M 8.53 6.91 3.94 .001
SD 3.22 3.21
Associate Learning Errors in the Spanish Syntactic Comprehension Test
M 15.44 15.12 0.83 .409 refer to identification of who the A is and, in conse-
SD 3.06 3.53 quence, who the patient (O) is. Four alternative re-
Delayed Associate Learning sponses were available, but the overwhelming majority
M 5.03 5.73 0.52 .598 of the errors did not refer to the misinterpretation of the
SD 1.53 1.35
Digits Forward
action or the names of the animals included in the fig-
M 6.72 6.21 3.48 .001 ures. Because of this, although four answers were avail-
SD 1.19 0.97 able, only two answers were likely: The lion [tiger,
Digits Backwards camel, horse, duck, rooster, rabbit, or skunk] hits
M 5.72 4.65 1.43 .157 [kicks, bites, pulls, or picks] the tiger [lion, camel,
SD 1.25 1.10
horse, duck, rooster, rabbit or skunk] or the inverse.
Total Digits
M 11.72 10.86 3.89 .001 The task for the participant was to identify the A and the
SD 2.03 1.75 O. Consequently, the chance level may be supposed to
Serial Verbal Learning be 50%.
First Trial Passive sentences are more frequently used in Eng-
M 5.60 5.42 0.83 .408
lish than in Spanish, so it is not surprising that perfor-
SD 1.58 1.63
Total Words mance in understanding Spanish passive sentences was
M 39.36 49.09 2.29 .025 virtually perfect in our bilingual sample. However,
SD 5.77 5.34 when Spanish syntax moved apart from English syntax,
Number of Trials difficulties emerged. Active reversible sentences with
M 4.15 3.98 1.05 .296 prepositions were almost normally understood. Differ-
SD 1.02 1.16
Calculation Abilities
ences between monolingual and bilingual participants
Successive Subtractions were slight and nonsignificant due to an appropriate un-
M 35.50 42.39 2.06 .043 derstanding of the use of the preposition a. The critical
SD 25.42 30.86 point is not the word order per se but understanding the
Multiplication markers distinguishing the A and the O. When the ac-
M 16.38 12.86 3.25 .002
tive sentences did not include the preposition a, how-
SD 8.10 7.28
Division ever, understanding difficulties became evident. In this
M 40.61 22.29 5.76 .001 type of sentence without a preposition, the definite arti-
SD 25.11 22.21 cle marks the A, whereas the undetermined article
Numerical Problem points to the patient. This is a type of strategy not ob-
M 51.89 37.82 3.53 .001
served in English.
SD 30.53 28.44
Sentences with a pseudocleft A were extremely dif-
Note: n = 69. ficult for bilingual participants. Significant difficulties
were observed in the last four sentences. These sen-
multiplication, and interaction was significant in the tences included a segment with an N plus a V. When, in
multiplication and division. this segment, the N was the A (Sentence 36), perfor-
The effect of the preferred language was analyzed mance was about 50% (chance level). When the N was
(Table 7). A language effect was found in 6 of the 11 the O, the sentences were more frequently understood
verbal memory scores, whereas the preference effect in the opposite sense, that is, as if the N were the A. This
was evident only in the WMS Digits (all conditions). type of sentence, with a pseudocleft A, is not usually
Only four interactions were significant: Logical Mem- found in English.
11
ARDILA ET AL.
Table 6. Effect of the Age of Acquisition of Second Language on Verbal Memory and Calculation Tests
Age of Acquisition
< 12a > 12b Language Effect Age Effect Interaction
Test English Spanish English Spanish F p F p F p
Wechsler Memory Scale

Logical Memory
M 10.7 10.2 9.26 10.6 0.98 .325 0.62 .430 4.27 .042
SD 2.7 3.1 4.6 2.5
Delayed Logical Memory
M 8.8 6.5 7.7 7.6 7.84 .007 0.00 .984 6.12 .016
SD 2.8 3.2 3.9 3.1
Associate Learning
M 5.6 5.2 5.6 5.8 0.11 .740 0.63 .427 1.22 .272
SD 1.6 1.7 1.3 1.7
Delayed Associate Learning
M 4.9 6.9 4.8 6.9 28.31 .001 0.01 .916 0.00 .943
SD 2.6 1.9 2.4 2.7
Digits Forward
M 7.0 6.4 6.1 5.7 9.30 .003 12.38 .001 0.12 .723
SD 1.1 0.9 1.2 1.1
Digits Backward
M 5.2 4.8 4.5 4.4 2.78 .100 4.21 .040 1.19 .278
SD 1.2 1.1 1.1 1.0
Total Digits
M 12.2 11.2 10.7 10.1 10.18 .002 9.75 .003 0.99 .322
SD 1.9 1.7 1.9 1.8
Serial Verbal Learning
First Trial
M 5.6 5.3 5.6 5.8 0.11 .740 0.63 .427 1.22 .272
SD 1.7 1.7 1.4 1.7
Total Words
M 39.4 39.7 39.1 44.2 12.14 .001 3.15 .080 9.97 .002
SD 5.5 5.1 6.4 4.6
Number of Trials
M 4.2 4.3 4.1 3.4 3.80 .054 4.13 .046 5.95 .017
SD 1.0 1.0 1.1 1.3
Delayed Recall
M 4.9 6.9 4.8 6.9 28.31 .001 0.01 .916 0.00 .941
SD 2.6 2.0 2.4 2.7
Successive Subtractions
M 32.8 41.1 40.9 45.2 2.84 .096 0.91 .342 0.26 .606
SD 25.5 19.1 24.9 35.1
Multiplication
M 14.3 12.6 21.1 13.4 17.32 .001 5.78 .019 7.00 .010
SD 7.9 6.1 7.8 9.9
Division
M 34.8 23.1 53.8 20.4 49.22 .001 2.39 .126 11.45 .001
SD 23.5 25.4 27.4 12.48
Numerical Problem
M 50.4 33.8 55.1 46.9 5.12 .027 1.24 .270 0.58 .447
SD 39.9 26.2 42.9 37.3
Note: N = 69.
a
n = 48. bn = 21.
12
Table 7. Preferred Language Effect on Memory and Calculation Test Performance
Prefered Language
Spanish English Language Effect Preference Effect Interaction
Test English Spanish English Spanish F p F p F p
Wechsler Memory Scale

Logical Memory
M 8.8 10.9 11.4 10.0 1.20 .276 1.62 .207 21.19 .001
SD 3.8 2.8 2.8 3.0
Delayed Logical Memory
M 7.2 7.7 9.6 6.4 13.98 .001 0.72 .398 25.62 .001
SD 3.4 3.0 2.7 3.3
Associate Learning
M 14.4 14.7 16.2 15.4 0.33 .563 2.92 .092 1.71 .195
SD 3.0 3.5 3.0 3.7
Delayed Associate Learning
M 4.8 5.6 5.1 5.8 26.8 .001 0.71 .401 0.12 .728
SD 1.5 1.6 1.5 1.2
Digits Forward
M 6.0 5.8 7.2 6.4 8.20 .006 16.56 .001 3.55 .064
SD 1.2 1.0 1.0 0.9
Digits Backward
M 4.3 4.4 5.5 4.8 2.87 .095 13.14 .001 5.79 .019
SD 1.0 1.0 1.1 1.2
Total Digits
M 10.3 10.3 12.6 11.3 10.15 .002 19.78 .001 9.10 .004
SD 1.7 1.6 1.7 1.8
Serial Verbal Learning
First Trial
M 5.4 5.8 5.7 5.2 0.06 .798 0.20 .653 4.71 .034
SD 1.5 1.5 1.6 1.9
Total Words
M 38.0 42.4 40.2 29.9 8.67 .005 0.01 .913 11.30 .001
SD 6.1 5.2 5.1 5.4
Number of Trials
M 4.4 3.9 4.1 4.1 1.39 .241 0.13 .714 2.66 .108
SD 1.0 1.2 1.0 1.1
Delayed Recall
M 4.5 6.9 5.2 7.1 31.31 .001 0.06 .418 0.39 .533
SD 2.4 2.5 2.7 2.0
Successive Subtractions
M 40.9 52.0 30.1 35.1 4.90 .030 5.61 .021 0.71 .040
SD 28.3 38.4 22.3 21.7
Multiplication
M 19.7 12.1 14.1 13.41 16.1 .001 1.94 .168 11.34 .001
SD 7.8 8.8 7.6 6.3
Division
M 50.9 22.7 33.1 22.7 37.3 .001 3.07 .089 7.90 .007
SD 31.0 16.6 20.0 26.1
Numerical Problem
M 58.3 47.3 44.0 32.6 7.63 .008 2.20 .142 0.02 .966
SD 40.5 42.3 32.8 25.1
Note: N = 66.
a
n = 27. bn = 39.
13
ARDILA ET AL.
In brief, the closer to English syntax the Spanish could be considered balanced bilinguals. This similar
sentences were, the easier it was for participants to un- level of proficiency as found in the majority of the par-
derstand them. The farther from English syntax the sen- ticipants may represent an additional factor responsible
tences were, the harder it became for the participants to for the relatively modest differences found between
interpret them correctly. both languages.
The significant effect of age exposed to the L2, pre- Although Spanish can be considered the L1 for this
ferred language, and sex variables deserve some com- group, a significant percentage of the participants
ment. We found a very strong effect of age exposed to learned Spanish early in life, later attended school in
the L2. Participants exposed to English between ages 5 English, and were currently living in a mostly Eng-
and 12 outperformed the participants exposed to Eng- lish-speaking environment. They frequently reported
lish before the age of 5. This means that the interfering English to be their best spoken and preferred language.
effect of English was stronger in the participants ex- Hence, although they were native Spanish speakers,
posed to English before age 5. The preferred language they had a better knowledge of English. Preferred spo-
effect is quite understandable. Participants preferring ken language turned out to be a more significant vari-
Spanish (or either) had a better syntactic understanding able than age of acquisition of the L2. Preferred
than those participants preferring English. Preference language was significantly associated with the school
means that the participant feels more comfortable language. Preferred language, however, was not a sig-
speaking that language. Preference, however, not only nificant variable in the calculation subtests.
may relate to knowledge of the language but can also Age of acquisition turned out to be a significant vari-
include affective relation with the language. Both are able only with regards to the WMS Digits and SVL
correlated. number of trials. These results should be interpreted as
The sex effect on language has been frequently re- a difference in the speed with which participants
ported (Kimura, 1992; MacCoby & Jacklin, 1974). The learned tasks presented in their L1 or L2. Although ver-
sex effect has been pointed out with regard to language bal learning is possible using either language, the speed
acquisition (McGuiness, 1976; Smolak, 1986), sponta- of learning may be faster in one language.
neous language during aging (Ardila & Rosselli, 1996), Overall, differences in the calculation subtests were
verbal fluency (Kolb & Whishaw, 1996), and other ver- larger than differences in verbal memory tasks. An ex-
bal abilities. This sex effect on linguistic abilities, how- planation could be that the language used in calculation
ever, remains controversial (Halpern, 1992; Hyde & tasks is the thinking language, which is minimally in-
Linn, 1988). Our results were not completely unex- fluenced by social linguistic context, and that bilinguals
pected: Women outperformed men. To the best of our continue using their native language in counting and
knowledge, however, sex differences with regard to other numerical issues. Consequently, better perfor-
syntactic comprehension had not been mentioned in the mance in Spanish was not unexpected.
available literature. Digit span was larger in English than in Spanish. It
In our second study, several significant differences has been found that English speakers accomplish a
were found in verbal memory and calculation subtests digit span close to 7 digits, whereas digit span scores
performed in Spanish and English. However, it must in Spanish speakers are usually about 5.8 (Ardila et
be emphasized that our sample was quite heteroge- al., 1994). Phonological length (number of phonemes)
neous from a linguistic point of view, although spo- in Spanish and English digits represents a significant
ken and written mastery of both languages was factor partially accounting for this difference
notoriously variable. This heterogeneity may have ob- (Olazaran, Jacobs, & Stern, 1996). The average num-
scured the results. ber of phonemes in English digits is about three,
Some significant differences in memory and calcu- whereas in Spanish it is about four (33% higher). In-
lation abilities were nevertheless found in Span- terestingly, digit span in Chinese (two phonemes per
ishEnglish bilinguals when tested in Spanish and digit) is about 9 to 10 (Dehaene, 1997). Additionally,
English. Performance was overall better in the partici- the 7-digit span is coincidental with the amount of
pants L1 (Spanish) than in their L2 (English). Our re- digits in phone numbers, a highly overtrained every-
sults support Harris et al.s (1995) findings. It is day activity in English speakers in the United States.
important to note that our bilingual participants were English speakers are in the habit of repeating phone
proficient in both languages and over half of them re- numbers digit by digit (5517975), perhaps as-
ported a better knowledge of English than of Spanish. sociated with the tendency to spell words, whereas
In fact, a significant percentage of these participants Spanish speakers cluster telephone numbers by 2 and
14
even 3 digits (5517975) and do not routinely prac- trained in the scientific and clinical understanding of
tice repeating 7-digit series. It is probable that, if a se- bilingualism.
ries of 2-digit numbers (i.e., 348756, etc.) were 3. The scores should be adjusted to neutralize the
used instead of single-digit-span numbers (i.e., 573, penalizing effect of bilingualism. The problem is that
etc.), Spanish speakers would do as well or maybe we do not know how much the test scores should be ad-
even outperform English speakers. It is noteworthy justed to be fair at best. Of course, it depends on the test,
that immediate verbal memory, measured according the idiosyncrasies of the clients bilingualism, and the
to the number of words repeated after single presenta- testing situation, including the examiners language.
tion (SVL, first trial score), was virtually identical in
Spanish and English. None of these three solutions alone or combined
In the numerical section of the test, the differences seem easy to put in action. However, when U.S. His-
between solving numerical problems in Spanish and panic bilinguals are tested, it should be emphasized that
English should be emphasized. When a numerical the results do not necessarily reflect the participants
problem is being solved, not only is the application of real abilities, and his or her real performance may be
an arithmetical procedure required but the correct un- higher than observed. Clearly, in any bilingual
derstanding of the problem conditions is also neces- neuropsychological report, the participants degree of
sary. A numerical problem represents a reasoning bilingualism (at least age of acquisition of the L2,
task. This significant difference between both lan- schooling language, and use of both languages in ev-
guages on this task suggests a better ability for rea- eryday life), language used in testing, norms used, and
soning when using a native language or L1 than when caution regarding the potentially penalizing effect of
an L2 is used. It should be emphasized, however, that bilingualism should be mentioned.
we were exploring the ability to reason with a particu- It should be noted that the penalizing bilingualism
lar language, not reasoning per se. It is noteworthy effect is expected in neuropsychological testing when
that, in our experimental conditions, it took longer to verbal materials are included. Bilinguals, however, are
say the Spanish problem (51 syllables) than the Eng- not expected to be at a disadvantage in other circum-
lish one (36 syllables). stances, for example, when attending rehabilitation
These results have an extremely important conse- programs. In such a cases, bilingualism may even pres-
quence for the neuropsychological testing of Hispanics ent some advantage.
in the United States: SpanishEnglish bilinguals may
be at a disadvantage when using either language. For
bilingual participants, both languages were active lan- References
guages, and their functional language was not either
Spanish or English but a mixture of both. Interference Ardila, A., & Rosselli, M. (1996). Spontaneous language produc-
is expected to be high. To use either Spanish or English tion: Age and educational effects. International Journal of
testing materials and norms can penalize U.S. Span- Neuroscience, 87, 7178.
Ardila, A., Rosselli, M., & Puente, A. E. (1994). Neuropsychological
ishEnglish bilinguals. No clear solution to this diffi-
evaluation of the Spanish speaker. New York: Plenum.
culty is available. Bahrick, H. P., Hall, L. K., Goggin, J. P., & Bahrick, E. (1994). Fifty
Three procedures, however, could at least reduce the years of language maintenance and language dominance in bi-
bilingualism effect in psychological and lingual Hispanic immigrants. Journal of Experimental Psy-
neuropsychological testing: chology: General, 123, 264283.
Bates, E., Friederici, A., & Wulfeck, B. (1987). Comprehension in
1. Special norms should be established for Span- aphasia: A cross-linguistic study. Brain and Language, 32,
ishEnglish bilinguals. This solution does not seem 1967.
Bates, E., Friederici, A., Wulfeck, B., & Juarez, L. A. (1988). On the
easy when the tremendous heterogeneity of U.S. bilin- preservation of word-order in aphasia: Cross-linguistic evi-
gualism is taken into consideration. dence. Brain and Language, 33, 232365.
2. The examiner could be a bilingual mastering a Caplan, D., & Futter, C. (1986). Assignment of thematic roles to
similar type of biligualism. Testing could be performed nouns in sentence comprehension by agrammatic patients.
in Spanish, English, or both languages. Instructions and Brain and Language, 27, 117134.
Dehaene, S. (1997). The number sense. New York: Oxford Univer-
answers in either language or both languages could be
sity Press.
acceptable; both English and Spanish norms could be Dongryong, K. (1990). A study of linguistic interference in bilingual
used, with the one favoring the participant being pre- childrenA case of students in a Korean school. Japanese
ferred. Nonetheless, the examiner should also be Journal of Educational Psychology, 38, 205212.
15
ARDILA ET AL.
Dupont, S., Ardila, A., Rosselli, M., & Puente, A. (1992). McGuiness, D. (1976). Sex differences in the organization of per-
Neuropsychological assessment in bilinguals. In A. E. Puente ception and cognition. In B. Lloyd & J. Archer (Eds.), Ex-
& R. J. McCaffrey (Eds.), Handbook of neuropsychological as- ploring sex differences (pp. 123159). New York: Academic.
sessment: A biopsychosocial perspective (pp. 193210). New Olazaran, J., Jacobs, D. M., & Stern, Y. (1996, February). Verbal and
York: Plenum. visual span in English and Spanish speakers: Testing a linguistic
Durgunoglu, A., & Roediger, H. L. (1987). Test differences in as- hypothesis. Paper presented at the Twenty-Fourth Meeting of the
sessing bilingual memory. Journal of Memory & Language, International Neuropsychological Society, Chicago.
26, 377391. Paivio, A., & Lambert, W. E. (1981). Dual coding and bilingual mem-
Hagiwara, H., & Caplan, D. (1990). Syntactic comprehension in ory. Journal of Verbal Learning & Verbal Behavior, 20, 532.
Japanese aphasics: Effects of category and thematic role order. Paradis, M. (1993). Multilingualism and aphasia. In J. C. Marshall
Brain and Language, 38, 159170. (Ed.), Handbook of linguistics and communication sciences
Halpern, D. F. (1992). Sex differences in cognitive abilities (2nd (Vol. 9, pp. 278288). New York: de Gruyter.
ed.). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Peynircioglu, Z. F., & Durgunoglu, A. Y. (1993). Effects of bilin-
Harris, J. G., Cullum, C. M., & Puente, A. E. (1995). Effects of bilingual context on memory performance. In J. Altarriba (Ed),
gualism in learning and memory in Hispanic adults. Journal of Cognition and culture: A cross-cultural approach to cognitive
the International Neuropsychological Society, 1, 1016. psychology (pp. 5775). Amsterdam: North-Holland.
Hummel, K. (1986). Memory for bilingual prose. In J. Vaid (Ed.), Ransdell, S. E., & Fischer, I. (1987). Memory in a monolingual
Language processing in bilinguals (pp. 4764). Hillsdale, NJ: mode: When are bilinguals at a disadvantage? Journal of Mem-
Lawrence Erlbaum Associates, Inc. ory and Language, 26, 392405.
Hyde, J. S., & Linn, M. C. (1988). Gender differences in verbal abili- Rosselli, M., Ardila, A., Araujo, K., Weekes, V., Volk, L., &
ties: A meta-analysis. Psychological Bulletin, 106, 5359. Caracciolo, V. (1998, November). The aging of language in
Kimura, D. (1992, September). Sex differences in the brain. Scien- SpanishEnglish bilinguals. Paper presented at the 18th An-
tific American, pp. 8187. nual Conference of the National Academy of
Kolb, B., & Whishaw, I. Q. (1996). Fundamentals of human Neuropsychology, Washington, DC.
neuropsychology (4th ed.). New York: Freeman. Seco, R. (1980). Manual de gramtica espaola [Handbook of Spanish
Kolers, P. A. (1978). On the representations of experience: Lan- grammar]. Madrid, Spain: Espasa-Calpe.
guage interpretation and communication. New York: Plenum. Smolak, L. (1986). Infancy. Englewood Cliffs, NJ: Prentice Hall.
Kolers, P. A. (1979). Reading and knowing. Canadian Journal of Snodgrass, J. G. (1984). Concepts and their surface representations.
Psychology, 33, 106117. Journal of Verbal Learning, 23, 322.
Lenneberg, E. H. (1967). Biological foundations of language. New Wechsler, D. (1945). A standardized memory test for clinical use.
York: Wiley. Journal of Psychology, 19, 8795.
MacCoby, E., & Jacklin, C. (1974). The psychology of sex differ- Wechsler, D. (1987). Wechsler Memory ScaleRevised. San Anto-
ences. Stanford, CA: Stanford University Press. nio, TX: Psychological Corporation.
Magiste, E. (1980). Memory for numbers in monolinguals and Wulfeck, E., Bates, D., & Capasso, R. (1991). A cross-linguistic
bilinguals. Acta Psychologica, 46, 6368. study of grammaticality judgements in Brocas aphasia. Brain
Marcos, J., & Ostrosky, F. (1995). Estrategias para la asignacin de and Language, 45, 448464.
papeles temticos en la interpretacin de enunciados en
espaol [Strategies for the assignment of thematic roles in the
interpretation of Spanish statements]. In D. Pool (Ed.),
Estudios en linguistica formal (pp. 4762). Mexico City,
Mexico: El Colegio de Mxico.
McCormack, P. D. (1976). Language as an attribute of memory. Ca- Original submission May 28, 1999
nadian Journal of Psychology, 30, 238248. Accepted September 10, 1999
16
Intern. J. Neuroscience, 117:935969, 2007
Copyright C 2007 Informa Healthcare
ISSN: 0020-7454 / 1543-5245 online

DOI: 10.1080/00207450600912222
WHAT CAN BE LOCALIZED IN THE BRAIN?

TOWARD A FACTOR THEORY ON BRAIN
ORGANIZATION OF COGNITION
ALFREDO ARDILA
Department of Communication Sciences and Disorders
Miami, Florida, USA
BYRON BERNAL
Department of Radiology
Miami Childrens Hospital
Miami, Florida, USA
A theoretical integration attempt among the lesional (neuropsychological), phys-

iological (functional), and psychometric models of cognition is presented in this
article. Recent neuroimaging techniques particularly fMRI have shown that there
are some brain functions (i.e., simple) that can be localized into single brain areas
whereas there are others (i.e., complex) that cannot. Clinical neuropsychology has
been able to propose some cognitive factors based on empirical observations
in patients with brain lesions. Factor analysis in psychometry may provide an
additional tool to extract some constitutive elements of psychological functions
(factors). Factors in factor analysis, however, may have different levels of
specificity. Some times they refer to functional systems (complex cognition); in
other occasions to elements of cognition (cognitive factors). It is emphasized
that the very same brain areas (and cognitive factors) may be potentially involved
in different types of cognition. It is proposed that complex cognition depends
on specific patterns of activation of different brain areas and specific circuitries
(modules), each one making its own contribution to the whole system (functional
Received 25 March 2006.

Address correspondence to Alfredo Ardila, Ph.D., ABPN, Florida International University,
Department of Communication Sciences and Disorders, HLS139 Miami, FL 33199 USA.
E-mail: ardilaa@fiu.edu
935
936 A. ARDILA AND B. BERNAL
system). Impairment in a specific cognitive factor, on the other hand, may result
in diverse types of impairments. At the moment, it seems feasible to suppose
some cognitive factors responsible for normal neuropsychological performance.
Theoretically, the impairment in any of these factors could be responsible for some
specific neuropsychological syndromes.
Keyword brain imaging, brain localization, cognition, cognitive testing, factor

analysis, factor theory, fMRI, neuropsychology
INTRODUCTION
The localization of the brain cortical functions has been a major endeavor since
the first work of Paul Broca (1861), who associated a left inferior frontal gyrus
lesion to an expressive language deficit. Based on this correlation between
neurological deficits and cerebral post-mortem findings, the nineteen-century
neuroanatomists developed the brain function localization model. This model
basically stated that a brain function is performed by a given region of the
brain. Korvinian Brodmann gave a strong support to this model in 1909
by describing 50 pair areas of different laminar organization of the cerebral

cortex. These areas would provide the bases for discrete localization of several
brain functions. It was not clear enough, however, what exactly the term
brain function meant. Is language comprehensionas an example, a brain
function? Certainly it is, but such a complex function seems to demand several
localized areas at least to account for different input forms. Primary sensory and
motor functions have been found to be localizable in rather specific brain regions
across different subjects. Nevertheless, complex functions, like naming, have
been found disperse in different brain areas in studies of electrocortical stimu-
lation in patients worked up for epilepsy surgery (Ojemman, 1979; Ojemann &
Mateer, 1979; Ojemann et al., 2002). From these findings, it appears that there
are some brain functions (i.e., simple) that can be localized into single brain
areas and others (i.e., complex) that cannot. Complex functions are the result
of a multistage integration of different components contributing to the whole
function. If one accepts a wide meaning for brain function, then what elements
of the complex function can be localized? In other words, which elements of
the complex functions could be assigned to a specific region of the brain?
This article presents a theoretical integration attempt among the lesional
(neuropsychological), physiological (functional), and psychometric approaches
to cognition. An effort is made to discern those localizable elements
contributing to more complex systems. Lesional model refers to the information
derived from the neuropsychological clinical observation of individuals
FACTOR THEORY OF COGNITION 937
presenting a brain pathology; that is, the cognitive sequelae resulting from focal
brain damage. The neurophysiological model refers to the contribution given
by modern neuroradiological techniquesparticularly functional Magnetic
Resonance Imaging (fMRI) and Positron Emission Tomography (PET), when
performing different intellectual tasks. Finally, the psychometric studies on
cognition refer to the knowledge yielded by standardized measures of cognition,

that is, psychological and neuropsychological tests.
During recent decades, modular theories of cognition have been particu-
larly influential in neuropsychology (e.g., Caplan, 1981; Caramazza, 1992;
Fodor, 1983; Humphreys & Riddoch, 1987; Moscovitch & Ulmita, 1990;
Shallice, 1988). These theories have attempted to find out the architectural
organization of cognition, and the elements, steps, or modules required in
specific forms of complex cognition (e.g., reading). The proposal that we are
presenting here representsat least in some pointsan extension of these
modular theories of cognition.
This article shall use the term factor to mean an element of cognition.
It can be related with a specific pattern of brain activity. The same word will
be used with the statistical meaning (factor analysis); that is, communality
(shared variance) among different tests. Although they can overlap at some
extent, it does not happen in all cases. For example Verbal comprehension
may appear as a factor when analyzing the psychometric tests, but it refers
to something ampler than the specific pattern of brain activity in the sense
aforementioned.
Three departing points will be taken:
1. The results of neuropsychological assessments, seeking in particular
for association (communality) existing among different measures of
psychological and neuropsychological tests. Factor analysis, a statistical
technique aimed to discover patterns among variations in the values of
multiple variables, has been applied to neuropsychological studies. These
studies will be reviewed, pointing out that factor analysis may contribute to
find out basic elements in cognition.
2. The neuropsychological theories about brain organization of cognition, in
particular, Lurias (1966, 1970, 1973, 1976) interpretation of psychological
processes as functional systems. The idea of functional system has
been directly or indirectly taken by different contemporary authors to
explain cognition, even without a direct reference to its origin. For instance,
Lieberman (2000) refers to a functional language system to explain how
the language is organized in the brain and how a diversity of cortical and
subcortical structures contributes to this functional language system.
3. Contemporary data provided by modern neuroimaging techniques about

brain activation. Neuroimaging includes different techniques to see the
working regions of the brain over an anatomical framework. Two main
types are considered: PET and fMRI. The findings of brain activation during
simple and complex tasks will be reviewed focusing on descriptions of
core and ancillary centers to specific cognitive tasks.
MEASURING COGNITION
There are two different sets of instruments directed to the appraisal of cognitive
abilities: psychometric intelligence tests (e.g., Wechsler Adult Intelligence
Scale in its different versions), and neuropsychological assessment tests
(e.g., Wisconsin Card Sorting Test, Boston Diagnostic Aphasia Examination,
Halstead-Reitan Battery). It is assumed that psychometric intelligence tests
are directed to normal populations, whereas neuropsychological instruments
are directed to braindamaged populations (e.g., Lezak, 1995). There is,
however, a significant overlap between both. In general, neuropsychological
tests have a lower ceiling and are more appropriate to detect abnormalities in
cognition (e.g., naming defects, perceptual impairments). Neuropsychological,
but not intelligence, tests were developed specifically to detect brain integrity,
departing from the observation of the deficits found in diverse types of brain
abnormalities. During recent decades, a significant effort has been directed in
neuropsychology to establish clinical/anatomical correlations (i.e., correlating
impaired performance in a particular test with anatomical abnormalities) (e.g.,
Damasio & Damasio, 1989; Kertesz, 1994). Neuropsychology evaluation, for
instance, analyzes how different abilities contributing to the arithmetical test
performance can be impaired with left angular, left prefrontal, right parietal,
pathology, and so on (Ardila & Rosselli, 2002).
Localization by means of a non-anatomical-imaging strategy (i.e., neu-
ropsychological tests) is a fuzzy concept. How precise can a localization
diagnosis be? Reading can illustrate the complexity of brain localization of
any psychological process (functional system) (Price et al. 1994). Reading is
based in certain fundamental abilities (or elements of cognition or simply
factors) (e.g., complex shape perception, crossmodal learns) already
existing 8,000 years ago, and of course, existing in illiterate individuals. What
might be localizable in the brain is not reading per se, but certain basic
abilities or elements of cognition or factors (information processing levels)
required to readalbeit not only to read (Ardila, 2004). Learning to read
may be supposed to involve the activation and longterm potentiation (Levy
& Steward, 1983; Perkins & Teyler, 1988) of several high order modules,
using some particular types of information (visual information related with
the perception of letter shapes, crossmodal association between letters and
language sounds, etc.).
Furthermore, the same information processing level may be useful for
apparently different types of cognition. As example, it has been pointed out

that mathematics, painting, playing chess, reading and writing, mechanics,
and music abilities may be impaired in cases of right hemisphere damage of
those very same areas that in an Eskimo or Amazonian Indian could simply
imply an impossibility to move around the snow or the jungle (Ardila, 1993
a). Brain assemblies (basic circuitry) or modules able to perform a specific
information processing level can be potentially useful for many different types
of cognition. Any type of complex cognition will require the participation of
multiple cognitive elements.
Individual differences in cognition have to be recognized. Visual areas in
blind people are activated with auditory stimulation, haptic stimulation (Rosler
et al., 1993) or Braille reading (Uhl et al., 1993). The pattern of activation
(and the recruited brain area) observed when moving the little finger of the left
hand is more complex and extensive in professional musicians playing a string
instrument (Elbert et al., 1995). Learning to read supposes significant changes
in the language processing activity in the brain (Castro-Caldas et al., 1998;
Petterson et al., 2000). Practicing a specific ability, in consequence, results
in changing the pattern of brain activity and even increasing the size of that
cortical brain area (Levin & Grafman, 2000). Every individual has a unique set
of experiences. Individual experiences are different not only between overtly
different people living in overtly different conditions (e.g., the Amazonian
Indians and New York dwellers), but also between any two individuals. Even
the performance of a relatively simple task (e.g., telling animal names) may be
carried out using quite different individual strategies (e.g., imagining moving
around the zoo; selecting specific categories such as mammals; finding the
animal names beginning with a specific letter) and depends on the personal
experiences with animals. Personal experience with animals may be sensory
based (i.e., obtaining through the senses in a directvisual, auditory, olfactory,
tactile or any combination, or indirectpictures, way) or conceptually based
(i.e., reading about a particular animal). It is not surprising that the pattern
of brain of different people when telling animal names will not be totally
coincidental. There is something in common, but there is also something
variable depending on the individual experiences.
A neuropsychological test result may point to a hemisphere or gyrus

activity, although usually the results of neuropsychological tests point
to systems (functional systems), subsystems, or functions, that may
involve one or several lobes. They usually yield performance results as
scalar magnitudes (usually compared to standards) but without a precise
anatomical direction. Frequently, the abnormal test result may be due to

multiple causes and even multiple regions of lesion. Neuropsychological test
localization was an enormous contribution in the past. Nowadays with the
advent of neuroimaging, the neuropsychology-test localization is challenged.
Neuroimaging techniques, nonetheless, have generally supported the findings
of the clinical neuropsychology; they have shown that many intervening
cognitive factors, as units of localizable brain functions, can account for
the same or similar deficits. These factors should be understood as units
of functional brain localization related to specific tasks. They should be able
to be demonstrated as isolated elements involved in systems but independent
to be recruited in different tasks.
Although neither the results, nor the usual methodology of the neuropsy-
chological or psychometric tests yield information about the discrete factors

involved, there are methods to extract them, and ways to validate the resulting
conclusions.
Before tackling the analysis of the psychometric assessments it is important
to define what this article is referring to with functional system, brain
function, module, and factor. The authors are naming the pattern of
cognitive abilities participating in complex psychological processes (complex
cognition; e.g., reading) as functional system, following the Lurian tradition.
This article shall refer to the brain activity supporting the functional system
as brain system. The authors are referring to the elements of the functional
systems or complex forms of cognition as cognitive factors. The cognitive
factors would be related with a specific way of processing information,
associated with an integrate activity of a specific brain module or supramodule.
Thence, factor will be understood as an element of cognition. Module will
be used to refer to a brain assembly performing a particular operation (specific
way of processing information). The article shall be arguing that the elements
of cognition (cognitive factors) may be potentially analyzed and pinpointed
using three different procedures: neuropsychologic, neuroradiologic, and
psychometric. Theoretically, a coincidence among the three approaches should
exist.
The term factor has been extensively used in psychometry (factor
analysis), but psychometric procedures can just suggest the existence of some
Figure 1. Cognitive factors may include subfactors.
general cognitive factors (elements of cognition). The factors found in factor

analytic studies are frequently too broad (frequently they indeed represent
functional systems), and obviously depend on the tests included in that specific
factor analytic study. Comparing diverse factor analytic studies, however,

converging results may shed light on basic elements of cognition (cognitive
factors). The use of the term cognitive factor (i.e., element of cognition) is
only partially equated with the meaning usually given in psychometry. Finally,
the existence of factors of different levels (subfactors, suprafactors) will be
assumed. An example taken from clinical neuropsychology can be illustrative.
Finger naming may be a particular case of body-parts naming defects. Finger
naming, however, may be impaired without generalized defects in body-parts
naming; but, when body-parts naming impairments are found, finger naming
is always defective (Hecaen & Albert, 1978). In consequence, there is a single
dissociation, not a double dissociation. A double dissociation would mean two
different syndromes depending on the activity of two different brain areas.
Therefore, a kind of hierarchy can be assumed. Certain specific brain modules
can participate in naming fingers (factor). Naming body parts requires the
recruitment of more extended brain areas (supramodules) representing a higher
level ability (suprafactor) (Figure 1)
ANATOMICAL CORRELATES OF LOCALIZATION

Brain localization is a term intimately related to anatomical landmarks.
Following a macro-to-micro order the first localization category is given by
the lateralization of functions. For instance, motor function is lateralized to
the opposite hemisphere. In a more precise categorization, the brain is divided

into lobes and further into gyri yielding a smaller regionalization of the cortex.
Heschls gyrus, for example, is a well defined anatomical structure of the
temporal lobe; it is the seat for the primary auditory cortex. In a smaller
level of division, the brain cortex is regionalized according to the laminar
organization of its six constitutive layers of neurons. Differences in these

histological properties convey, as expected, differences in function. The visual
area located in both lips of the calcarine fissure has particularities not shared by
any other region of the cortex. This area is, therefore, tuned up to serve as the
cortical first processor for visual inputs. Beyond this anatomical subdivision it is
found a columnar organization of the function. This is, at least represented in the
visual cortex, where Hubel and Wiesel (1961, 1962, 1963, 1965) demonstrated
a columnar organization responsible only for lines specifically oriented. These
columns are the unit of function response and are also called modules. Thus, a
decrescendo chain will have:
Hemisphere Lobe Gyrus Brodmann area Module
It is expected that a similar progression of complexity can be described with
respect to different brain functions. However, there is a lack of such functional

stepwise categorization. One is constrained to a simple chain between a global
level of Functional System to a rather basic level of minimal discrimination
or minimal unit of output probably yielded by modules. Intermediate levels
of processing should exist. For instance, in anomia it has been traditionally
recognized that naming body-parts, external objects and colors depend (and
are altered) upon the activity of different brain areas (Hecaen & Albert 1978).
It has also been found that finer distinctions can be made with regard to naming
defects, which can be limited to a rather specific semantic category (e.g.,
peoples names, living things, geographical names) (e.g., Harris & Kay, 1995;
Goodglass et al., 1986; Lyons et al., 2002; Warrington & Shallice, 1984) and
even as specific as medical terms (Crosson et al., 1997). The factor theory
presented here represents a development of the classical modular explanations
creating a transitional link between functional systems (that supposedly
encompass larger structural areas), and the minute modular level.
FROM COGNITIVE FACTORS TO BRAIN FUNCTIONS

The problems of finding the constituent parts of a given brain function are
exemplified by this question: Is naming ability due to a non-reducible factor
(element) of brain activity or is it the result of a subsystem consisting of several
functional modules susceptible to be affected at different levels with similar
deficit in the output? If the answer points to a non-reducible element, the brain
localization would be circumscribed to a specific region, whereas if the answer
is inclined to a subsystem, the brain localization would be circumscribed to
a sort of circuitry showing different participating elements. These are two
different approaches of brain localization. Furthermore, what exactly does the
term brain function mean? Is language comprehension a brain function?

Certainly it is, but such complex function seems to demand the participation
of several localized areas. Cortical electrical stimulation in patients worked
up for epilepsy surgery have shown that each subject has several areas that
respond with anomia when stimulated (Ojemman, 1979; Ojemann & Mateer,
1979; Ojemann et al., 2002). A diversity of category specific anomia have
been observed in clinical practice (e.g., Harris & Kay, 1995; Goodglass et al.,
1986; Lyons et al., 2002; Warrington & Shallice, 1984). Consequently, it may
be inferred that namingas brain functionis the resultant sum of several
intervening cortical areas, and cannot be explained as the output of a single
naming module. A more rational way to look at the localization concept may
be taking it in a brain system level, involving a diversity of supramodules, active
according to the task demand, and following a specific activation pattern.
FACTOR ANALYSIS
Neuropsychological inventories and psychometric tests may be refined to
provide clues of brain function localization if their results are factorized to
extract their constitutive elements. The results will be named also as factors
but as pointed out earlier, they do not necessarily refer to non-reducible units of
brain function as cognitive factor does. Often, they are closer to a functional
system (e.g., verbal comprehension factor).
Factor analysis represents a strong and relatively sophisticated statistical
tool in measurement research. Factor analysis allows deducing underlying
factors accounting for variance in individual tests. Communality, and in
consequence, relative distance among different subtests can be deduced.
Departing from correlation matrixes among different measures, factors
responsible for observed dispersions can be deduced.
Factor analysis has been extensively used in psychological measurement
sphere, and different factorial studies of general intelligence have been
presented in psychological literature (e.g., Carroll, 1993; Cattell, 1971;
Guilford, 1967; Matarazzo, 1972, 1992; Wechsler, 1997).
Using factor analysis, it has been attempted to determine which could
be considered as the basic, primary, or fundamental cognitive abilities.
For instance, Thurstone (1947) proposed that seven primary mental abilities
existed: verbal, reasoning, numeric, spatial, perceptual speed, memory, and
verbal fluency. Guilford (Guilford, 1967, 1968; Guilford & Hoepfner, 1971)
proposed a threedimensional classification of intelligence, including: contents
(letters, numbers, words, and behavioral descriptions); operations (memory,
evaluation, convergent thinking, and divergent thinking); and products (units,

classes, relations, systems, transformations, and implications). In consequence,
according to Guildford 120 different intellectual abilities could be distin-
guished.
The Wechsler Intelligence ScaleRevised (Wechsler, 1987) represents a
good example of an assessment instrument used not only in psychological but
also in neuropsychological evaluation. Most often, a threefactor structure has
been found in the WAIS (Matarazzo, 1972). Factor I (Verbal Comprehension)
is measured by Information, Vocabulary, Similarities, and Comprehension sub-
tests. Factor II (Perceptual Organization) is measured with Object Assembly,
Block Design, Picture Completion, and Picture Arrangement subtests. Factor
III (Freedom of Distractibility) is specially measured with DigitSymbol
subtest. A fourth (albeit weaker) factor has been sometimes reported, measured
with Arithmetic subtest (Cohen, 1957). From a neuropsychological perspective,
the distinction between verbal (Verbal Comprehension) and visuoperceptual
(Perceptual Organization) abilities simply represents the most evident
dichotomy for left and right hemisphere activity. Freedom of Distractibility
is just an attentional factor, often conforming an independent factor in factor
analytic studies (Bornstein & Chelune, 1988; Roid et al., 1988; Wechsler, 1987;
Ardila & Rosselli, 1994). From a neuropsychological perspective, attention
may be altered in cases of frontal pathology (Luria, 1966; Stuss & Benson,
1986).
The factor analysis of the WAIS-III (Wechsler, 1997) resulted in a more
complex factor solution. New tests were included in the WAIS-III, not existing
in the previous versions (Matrix Reasoning, Letter-Number Sequencing,
Symbol Search). According to the Manual of the WAIS-IIII, and using a
sample of 2,450 1689-year-old adult participants, the following factors were
disclosed: Verbal Comprehension, Perceptual Organization, Working Memory
and Processing Speed (Table 1). A fifth factor (Arithmetic) defined by only
one subtest (Arithmetic) was also suggested. This factor solution has been in
general confirmed by other authors (e.g., Burton et al., 2001; Ryan & Paulo,
2002).
Tulsky and Price (2003) further presented a series of confirmatory factor
analyses to determine the joint WAIS-III and WMS-III factor structure. Using
Table 1. WAIS-III factor analysis (Wechsler, 1997)
Factor Subtests
Verbal comprehension Vocabulary

Similarities
Information
Comprehension
Perceptual organization Block design
Matrix reasoning
Picture completion
Picture arrangement
Working memory Digit span
Letter-Number sequencing
Arithmetic
Processing speed Digit symbol
Symbol search
a structural equation modeling approach, a 6-factor model that included verbal,

perceptual, processing speed, working memory, auditory memory, and visual

memory constructs provided the best model fit to the data. Allowing select
subtests to load simultaneously on 2 factors improved model fit and indicated
that some subtests are multifaceted. The results were then replicated in a large
cross-validation sample (n = 858).
Carroll (1993) analyzed 461 factoranalytic studies presented in the
literature up to date. He observed that some factors tend to appear with a
significant frequency across different factorial studies. This is observed in
different cognitive domains: reasoning, language, memory, visual perception
abilities, and so on. Table 2 presents a summary of these relatively constant
factors found across different factor analytic studies. That is, regardless of the
diversity of tests included in the different 461 factor analyses, an importance
convergence in the factor structures is observed. These common factors found
across diverse studies may represent general and fundamental elements of
cognition. From a neuropsychological perspective, these factors (fundamental
elements of cognition) are potentially impaired in cases of brain pathology.
Some neuropsychological syndromes may be conjectured to be found in cases
of disruption of these basic cognitive factors.
In neuropsychology, factor analysis has been specially applied to some
specific tests and scales directed to measure single cognitive abilities. The
Wechsler Memory Scale (WMS) represents a good example. To date, several
factoranalytic studies with the WMS have been published, usually reporting
Table 2. Some relatively constant factors found across different factoranalytic studies
(Carroll, 1993), and the neuropsychological syndromes with which they could be associated
Factors Neuropsychological syndrome
Language
Lexical knowledge Wernicke aphasia

Grammatical sensitivity Broca aphasia
Communication ability Prefrontal syndrome
Oral production Verbal apraxia?
Speech sound discrimination Worddeafness
Naming facility Anomia
Expressional and word fluency Extransylvian (transcortical) motor aphasia
Reasoning
Sequential reasoning Prefrontal syndrome
Inductive reasoning Prefrontal syndrome
Quantitative reasoning Frontal acalculia
Visual perception
Spatial relations Spatial agnosia
Serial perceptual integration Topographic agnosia?
Perceptual speed Visual agnosia
Numerical
Number facility Acalculia
Attention and concentration abilities
Attention and Concentration Prefrontal syndrome
a twofactor structure (General Memory and Attention factors) (Bornstein &

Chelune, 1988; Roid et al., 1988; Wechsler, 1987; Ardila & Rosselli, 1994).
However, Elwood (1991) analyzed the factor structure of the WMSRevised
in a clinical sample and observed that only one general memory component
accounted for 54% of the variance; a second factor that contributed 9.4% of
the total variance was found only after IQ scores were included. Ardila and
Rosselli (1994) used the original WMS version, but included delayed recall for
the Logical Memory, Visual Reproduction, and Associative Learning subtests.
One general memory component accounted for 51% of the variance. The same
second factor reported by Wechsler (1987) as an Attentional Factor was also
found. Finally, Ardila and Rosselli (1994) observed a third, and indeed very
weak factor, related with verbal memory, specially measured by means of the
Associative Learning subtests (Verbal Memory factor). These results with
the WMS were very similar to the results obtained by Wechsler; Bornstein
and Chelune (1988); and Elwood (1991) using the WMSR. In consequence,
seemingly here is not too much difference between the WMS and the WMSR
factor structure.
Ardila and Pineda (2000) analyzed the factor structure of nonverbal
cognition. A neuropsychological test battery for assessing different nonverbal
cognitive domains (attention, memory, visuoperceptual and visuoconstructive
abilities, executive functions, praxis abilities, and written calculation abilities)

was administered. Initially, independent factor analyses were carried out for
each domain. Three attention factors (Sustained Attention, Divided Attention,
and Processing Speed, 73.1% of the variance); two memory factors (Categorical
and Non-Categorical Memory, 59.7% of the variance); two visuoperceptual
and visuoconstructive factors (Sequential and Simultaneous, 54.0% of the
variance); and two executive function factors (Categorization and Trial/Error,
82.0% of the variance) were found. Subsequently, several sequential factor
analyses using Varimax orthogonal rotations for noncorrelated variables
were performed. The 32 test variables were included, but progressively
some variables were removed. This procedure finally selected 13 variables
corresponding to five factors accounting for 72.6% of variance. Factor 1 was
an Executive Function factor (30% of variance). Factor 2 corresponded to a

Sequential-Constructional factor (14.7%). Factor 3 represented a Processing
Speed factor and accounted for 10.6% of the variance. Factor 4 was a
Visuoperceptual factor (9.5% of the variance). Finally, Factor 5 (7.8% of
the variance) was a Nonverbal Memory factor. It was concluded that several
different cognitive dimensions are included in nonverbal cognition.
Only a few factorial analyses of extensive neuropsychological test batteries
have been occasionally presented in literature (e.g., Ardila et al., 1994, 1998;
Haut et al., 1992; Wilhem & Franzen, 1992; Loewenstein et al., 2001; Ostrosky
et al., 1985, 1986; Livingston et al., 2000; Ponton et al., 1994, 2000). Thus,
Ponton et al. (1994) gave a test battery to 300 normal subjects; which included
10 different neuropsychological tests. A factor analysis yielded five factors:
a Verbal Factor (measured with Verbal fluency, Naming, and Digit span); a
Learning Factor (measured specially with the Auditory Verbal Learning test); a
Speeded Processing Factor (including the DigitSymbol, Color Trails, and the
Block Design), a Visual Processing Factor (measured with the ReyOsterrieth
Complex Figure and the Raven test), and a Psychomotor Speed Factor
(measured with the Pin Test). Ardila et al. (1994) used a neuropsychological test
battery, including 10 basic and common neuropsychological tests. Twentyfive
different scores were calculated. A factor analysis with varimax rotation
disclosed nine different factors, accounting for about 70% of the variance
(Table 3).
Table 3. Factors observed in Ardila et al. (1994) neuropsychological test battery, tests
most saturated by these factors, and probable neuropsychological syndromes theoretically
associated with impairments in those factors
Probable
neuropsychological
Factor Tests syndromes
I Verbal Production Verbal fluency Convexital left prefrontal

syndrome
II Constructional ReyOsterrieth Figure Constructional apraxia
Visuospatial Visual memory WMS Spatial agnosia
III Verbal Memory Logical memory Verbal amnesia
Wernicke aphasia
IV Fine Movements Tapping test Premotor syndrome
Kinetic apraxia
V Verbal Knowledge Information Wernicke aphasia
Boston Naming Test anomia
VI Praxis Ability Ideomotor apraxia test Ideomotor apraxia
VII Delayed Associative Delayed Associative Hippocampus amnesia
Learning Learning WMS
VIII Arithmetic Digits Acalculia

IX Attentional Mental control WMS Orbital prefrontal syndrome
In another study, an extensive neuropsychological test battery was

assembled and individually administered to a relatively homogenous
300subject sample (Ardila et al., 1998). The participants were 1725-year-old
righthanded middle socioeconomic status male university students. The
battery included some basic neuropsychological tests directed to assess
language, calculation abilities, spatial cognition, praxis abilities, memory,
perceptual abilities, and executive functions. In addition, the Wechsler Adult
Intelligence Scale was also administered. Fortyone different scores were
calculated. Correlations among the different test scores were analyzed. It was
found that some of the tests presented a quite complex intecorrelation system,
whereas other tests did present few or no significant correlations. Mathematical
ability tests and orthography knowledge represented the best predictors of Full
Scale IQ. A factor analysis with varimax rotation of the neuropsychological
battery tests was performed. Five different factors with an eigenvalue higher
than 1.00 were disclosed. These five factors accounted for 63.6% of the total
variance. The first factor accounted for about onefourth of the total variance.
Table 4 presents the general results of the factor analysis.
Table 4. Factor analysis of an extensive neuropsychological test battery (Ardila et al.,

1998)
Factor Pct. of variance Most saturated tests
Verbal 26.7% WAIS Similarities

Visuoperceptual 12.5% Hidden Figures

Executive function 9.8% WCST Categories
Fine movements (fluency) 7.9% Finger Tapping Test
Memory 6.9% Serial Verbal Learning
It could be supposed that some shared brain systems may underlie

to the performance of those tests measuring common factors. For example,
in Ardila et al. (1994) factor analysis, Factor II (Nonverbal Memory and
Constructional factor) represented a quite independent and isolated factor.
It could be proposed that the two tests saturated by this factor (Visual
Reproduction from the Wechsler Memory Scale, and ReyOsterrieth Complex
Figure) were the only two tests associated with the activity of certain specific
brain systems (according to current neuropsychological knowledge, most
likely related with the activity of some right hemisphere areas). It could be
further expected their performance were impaired in cases of pathology of
this brain system (i.e., some righthemisphere pathology). Correlations
between these two tests and other tests were low and non-significant. The
brain activity supporting the performance in these two tests should be quite
independent of the brain activity supporting the performance in the rest of the
battery. But both tests share the same brain system. Functional distance
between them is very low. Conversely, functional distance between these
two tests and the rest of the test battery is high. Functional distance should
be understood as the commonality in the brain organization of the cognitive
processes supporting their performance. Excepting Visual Reproduction and
ReyOsterrieth Complex Figure scores, for the rest of the tests a quite complex
matrix of intercorrelations was obtained. Nonetheless, albeit correlated,
different factors were extracted. These tests were not homogenous in their
scores; and likely, they depended on somehow different (but relatively
close and partially overlapped) brain systems. For example, it could be
hypothesized that performance in the Boston Naming Test and the Information
subtest of the Wechsler Intelligence Scale (highly correlated, and grouped
into a single factor: Verbal Knowledge factor) depend on closer brain
systems, than the Boston Naming Test and, say, the Orientation subtest of
Figure 2. High correlations (high communality) and low or none communality can be observed
among different cognitive abilities.
the Wechsler Memory Scale (with a correlation close to zero). It could be

supposed that the stronger the correlation between two tests, the higher the
commonality of the brain activity supporting their performance. The lower
the correlation, the higher the functional distance in the brain organization
of those cognitive abilities required for the correct performance. Figure 2

illustrates how arithmetic can be significantly associated with vocabulary and
perceptual speed, but vocabulary and perceptual speed represent independent
factors, associated with different brain systems.
A factor might be somehow interpreted as a kind of information
processing type or information processing level. A specific information
processing level can be incorporated and required for the performance of
different tests. In a psychometric language, a factor can be saturating different
tests. In consequence, it would seem reasonable to attempt establishing
correlations between brain pathology and disturbances at specific information
processing levels (correlations between brain pathology and factors), rather
than attempting to correlate brain pathology with performance in specific tests.
However, when different factor analyses are applied, it does not seem
necessary to expect a fixed and limited number of factors responsible for
psychological activity. Depending on the included tests, factor structure
can eventually vary (Carroll, 1993). Besides, not all factors have the same
weight (eigenvalue); they account for different percentages of the total
variance. Furthermore, subfactors can exist, as usually it is recognized in
the psychometric tradition (Anastasi, 1982; Cronbach, 1990). So, if a verbal
factor were found (e.g., Factor V: Verbal knowledge; Ardila et al., 1994),
it seems reasonable to expect that, when carrying out a sufficiently extensive
language evaluation, some subfactors included in this general verbal factor
would eventually emerge. These verbal subfactors eventually found, would

correspond to different language processing levels, or language elements
(factors of cognition).
Moreover, if the conditions of the tests were changed, the level of
performance can also change. For example, performance in verbal fluency tasks
(to search for words according to certain characteristics) varies depending on

the category that is used (words beginning with a particular letter, animals,
flowers, chemical elements, abstract words, etc.). With regard to naming, as it
was mentioned earlier, brain organization of different naming categories can
be partially different (Benson & Ardila, 1996).
Comparing different factor analytic studies in neuropsychology, some
factors can appear to be relatively constant and stable throughout various
analyses. Verbal knowledge (Naming, Information, Vocabulary) usually
appears as a single factor across various factoranalytic studies. Constructional
and perceptual abilities usually appear conforming a single factor, quite
independent from othersverbally based, factors. Attention (measured in
diverse ways) also appears to represent an independent factor. Verbal fluency
could be also considered as another relatively constant and independent factor.

If each one of these factors were correlated with a focal neuropsychological
syndrome, a notorious approximation between psychometric results and clinical
data would be achieved. Conversely, if the specific factor responsible for every
neuropsychological syndrome could be pinpointed and further, a statistical
demonstration of its existence in normal populations were obtained, no doubt,
a solid mathematical ground for clinical observations would be provided.
As it was stated before, the factors yielded by the factorial analysis of the
psychometric tests frequently correspond to functional systems rather than
elements of cognition. So, they can be subdivided into more simple elements
approaching the level of units of brain function localization (modules). In
order to dissect the functional systems given by the factorial analysis into
cognitive factors one needs the theoretical input provided by the lesional
model and the empirical validation given by the neuroimages.
THE NEUROPSYCHOLOGICAL APPROACH (LESIONAL MODEL.

A. R. LURIAS THEORETICAL APPROACH)
For Luria, psychological processes represent functional systems. The concept
of functional system was taken from Anokhin (1974), and is understood
as a group of interconnected biological operations that produce a particular
biological effect. The functional system is based on a complex dynamic
constellation of stages, situated at different levels of the nervous system, which

in performing an adaptive task, may be changed without the task itself being
changed. Writing, for instance, represents a complex psychological process
(functional system) that requires the participation of multiple areas of the brain;
each of these areas making its particular contribution to the whole system. A
focal lesion of the brain will disrupt the ability to write at a particular level
(e.g., the ability to perform the skilled movements required for writing, the
spatial organization of writing, the selection of words, the ability of sequencing
graphemes). However, such particular brain focal damage will also disrupt
all the functional systems for which that particular operation is required.
For instance, the patient will not only have spatial organization of writing
difficulties, but also, difficulties in spatial organization of numbers, figures,
drawings, and so on. In all the functional systems in which such ability is
included, the defect will be apparent.
The brain damage does not produce the loss of a specific cognitive
process (functional system), which is poorly localized, but its disturbance
at a specific level. This implies that neuropsychological assessment will be
aimed at disclosing the fundamental defects underlying the apparent deficits

(syndromatic analysis). For this purpose and according to Luria, it will be
necessary to administer to the patients different types of tasks and to analyze
how the particular difficulties in performing each one of them are manifested.
For Luria, the most important observation when testing a patient refers to
the nature of deviations or errors, and how such mistakes could be explained.
This implies that the pass/fail criterion is not enough, or at best, it is simply an
initial gross approximation to the characteristics of the deficit. The qualitative
analysis of errors will be particularly informative as concerns the underlying
deficit of the patient. It is not enough to know that a patient cannot understand
language, or cannot write. For Luria, the most important information is the
precise nature of the patients inability to understand language or to write, and
the specification of the level in the functional system that is disrupted. What
are the actual mistakes the patient presents when trying to understand language
or trying to write? The errors produced by patients with frontal, parietal, or
temporal lesions when performing calculation tasks are quite different, although
all of them may present a certain degree of acalculia (Ardila & Rosselli, 1990,
2002; Rosselli & Ardila, 1989). All these patients can fail in exactly the same
tasks, but because of totally different reasons. Their errors will be the key clues
for understanding the underlying deficit.
Clinical/anatomical correlations were widely developed by Luria. As a
matter of fact, he pioneered the method of the superimposition of lesions to
disclose critical areas in a particular type of disorder. His study of 800 patients
to determine the critical brain area for phonemic discrimination deficits, has
become classic (Luria, 1947/1970). This procedure of superimposing lesions
to highlight critical areas responsible for clinical syndromes is extensively used
in the present day neuropsychological research (e.g., Ardila & Ostrosky, 1989;
Damasio & Damasio, 1989; Kertesz 1983, 1994).

Luria strived to establish correlations between brain pathology and
disturbances at specific levels of information processing (e.g., phonemic
discrimination)not to correlate brain pathology with performance in specific
tests. Tests may be changed, but while some specific level of information
processing will still be required, impairment will be manifested. Besides, the
performance on even apparently very simple tests can require the participation
of different brain areas. Hence, even performance on simple tests can be altered
as a consequence of very different brain pathology, although the specific error
patters will be different. Many different types of brain pathology can alter for
instance calculation abilities (Ardila & Rosselli, 2002); however, in each case
the difficulty (and the errors) will be the result of a disturbance at a different
level in the calculation process. Patients with frontal lobe damage and patients
with angular gyrus damage can both present serious difficulties in simple
calculation tests. However, the underlying impaired mechanism and the type of
errors manifested are quite different (Rosselli & Ardila, 1989). Consequently,
the validity derived from correlating the site of the brain pathology with
performance on a particular test appears, in Lurias interpretation, as a very
preliminary and crude approximation.
Lurias interpretation of semantic aphasia and acalculia represents a
good illustration of what a cognitive factor is, according to Lurias
approach. For him, both syndromes are just two clinical manifestations of
the very same disrupted component (i.e., a defect in the ability to understand
logicalgrammatical relationships holding a quasispatial content). It would
be reasonable according to this interpretation, to expect a virtually perfect
correlation between both syndromes. Furthermore, if one assumed that acalculia
and semantic aphasia are also associated with finger agnosia and right left
disorientation (Ardila et al., 2000), one may conjecture that they represent
specific manifestations of an impaired cognitive factor (Figure 3).
Cognitive factors would, in consequence, represent the basic elements
of cognition, or the basic cognitive abilities, the functional counterpart of
the anatomical modules. Luria discussed this question with some detail only
with regard to language. In his last book, Basic Problems of Neurolinguistics
(1976), Luria analyzed the cognitive factors that can underlie to the different
Figure 3. The impairment in a common cognitive factor (verbal mediated mental rotations?) may
account for the different manifestations observed in case of left angular damage.
aphasic syndromes (Table 5). However, it is not easy to deduce the impaired
cognitive factors in other neuropsychological syndromes (e.g., agnostic or
apraxic disorders). This cognitive factor theory represents one of the most
interesting and outstanding ideas in Lurias neuropsychological perspective.
Unfortunately, Luria left his cognitive factor theory of psychological activity

without a complete development.
Evidently, the core problem is how to determine those fundamental
components underlying normal cognition. Correlation procedure among
performance in different cognitive tasks seems to represent a provocative
possibility.
THE NEUROIMAGING APPROACH (FUNCTIONAL MODEL)

In contrast to the lesional model that derives assumptions from neuropsycholog-
ical deficits in patients with brain lesions, the neuroimaging approach explores
Table 5. Factors underlying different aphasia syndromes, according to Luria (1976)
Aphasia type Impaired factor
Acoustic-Agnosic Phoneme discrimination

Acoustic-Amnesic Verbal memory
Amnesic Semantic structure of words
Semantic Understanding logical-grammatical (quasi-spatial) structures
Afferent Motor Articuleme discrimination
Efferent Motor Disturbances in speech kinetic structure
Dynamic Verbal initiative
normal or abnormal brain activity associated with a specific task. A reasonable

expectancy for the neuroscientist is that, both the lesional and the functional
models were totally convergent. However, they are not. Such divergence is
seen in other neurological fields contrasting lesional and functional models.
For example, in the creation of maps of dermatomes there are two models that
are not entirely coincidental. One model is derived from the area of anesthesia
after a lesion or anesthetic blockage of one spinal root (Keegan & Garret,
1948). The other is based in the area that remains with sensitivity after the
rest of roots are blocked, or the area responding with electrical stimulation
of the dorsal root (Foerster, 1933). The explanation for this phenomenon is
functional overlapping. A simpler example is demonstrated with the visual
field. What does the right eye see? If only the right eye is closed (lesional
model), a small portion of the far lateral vision is lost. If only the right eye is
opened (functional model) the entire visual field of the right eye is revealed. The
lesional model points more to proprietary functions but could underestimate
the extent of function of a given structure. On the other hand, functional models
may fail to isolate critical nodes of a distributed circuitry or overlapped function.
Functional neuroimaging procedures are in essence functional models.

The understanding of these differences is important when deriving
assumptions from exploratory models that neither rely on lesional experiences
nor on functional examinations, but on normalization data, as it happens with
the psychometric tests.
The two most important functional neuroimaging procedures are positron
emission tomography (PET), and functional magnetic resonance imaging
(fMRI). These techniques have provided the most valuable information about
brain activity during performance of different cognitive tasks. PET images are
based on metabolic and regional perfusion changes and require radiotracers.
fMRI is based in minute regional changes of oxyhemoglobin/deoxyhemoglobin
blood levels that influence the magnetic signal. Both techniques allow
visualizing levels of brain activity and focal involvement during different
conditions. The procedures entail typically a probe task and a control condition
that are repeated several times accounting for their small magnitude and noise.
A common characteristic of PET and fMRI is the possibility of cognitive
subtraction. This is a modeling technique of the experiments that allow the
researcher to isolate a target function. For example, the experimenter can
target the areas involved in verbal comprehension, when the subject is reading.
To accomplish it, the session is divided into two blocks. During one block
(condition A), the subject reads a normal text from a screen. In a second block
(condition B), the subject is presented with a meaningless arrangement of
letters. The subject will search for a specific letter during this block to control
visual, attention and working memory loads. The rest of the conditions are kept
the same. Then, a subtraction is made between the constitutive elements.
Condition A = Attention + visual processing + ocular movement + working
memory + decoding of letters + grammar comprehension. Condition B =
Attention + visual processing + ocular movement + working memory + de-

coding of letters. Grammar comprehension is derived by subtracting B from A.
PET results are based in series and do not account for intra-subject
variability. Its spatial resolution is poorer than fMRIs, and assumptions derived
from the technique are applicable to populations, not to individuals. This
advantage in terms of normalization turns out to be a disadvantage in clinical
practice.
fMRI does not require pooling data from subject series. Its results are
applicable to individuals and the exam may be repeated as many times as
deemed necessary since it is not based on ionizing radiation. PET and fMRI
show great reproducibility (Casey et al., 1998; Carey et al. 2000) both among
subjects and groups. Reproducibility, resolution, and localization are better
in fMRI because there is no risk for re-exposing the subject to the MR

environment and because its matrix of reconstruction is more precise. However,
the reproducibility of findings differs according with the complexity of the tasks.
Neuroimaging, particularly fMRI, has shown, for instance, a total
reproducibility of activation in groups and subjects when mapping primary
visual areas (Engel et al., 1994; Thulborn et al., 1997; Moradi et al.,
2003). Several tasks have been used. Passive viewing of a black and white
checkerboard, for example, actives regions along the boundary of the calcarine
fissure (primary visual area) (Hirsch, 1994).
Primary auditory area has also been consistently mapped using tones
(Wessinger et al., 1997), chords (Bernal et al., 2004), music (Li et al.,
2000; Bernal & Altman, 2001), noise (Colder & Tannenbaum, 1999), etc.
(for a review see Bernal et al., 2004). The same reproducibility shows the
primary motor-sensory cortex (Freund, 2002; Lotze, 2000; Fera et al., 2004).
In all of these cases the activation takes place in restricted areas common
to different subjects with minimal variations. Recognition of syllables, for
example, activates very restricted areas of the left temporal lobe corresponding
to the primary auditory cortex (Black & Behrmann, 1994). Viewing nouns,
hearing nouns, and generating verbs activate the left occipital, temporal, and
frontal areas, respectively (Raichle, 1994). In all cases, activation is limited to
some rather specific brain areas.
In contrast to the restricted and limited activation of the primary areas

elicited by simple tasks, complex tasks produce activation of a circuitry in which
different brain areas participate holding specific contributions to the whole
performance. Figure 4 shows the activation elicited by a silent word generation
task. In addition to the Brocas area (main target of this paradigm), involved in
speech production, activation is also seen in the left premotor area accounting
for motor planning; supplementary motor area (SMA) accounting for initiation,
and probably effort; left auditory association areas most like related to some
semantic analysis; visual areas related to picturing; and left thalamus. The
involvement of thalamus in this task is not well understood, but it is congruent
with cases reported of aphasia resulting from thalamic lesions (Crosson et al.,
1986; Ozeren et al., 1994). In the experience of one of the authors of this paper
(BB), Brocas and lateral frontal areas are consistently activated across subjects
(core areas), whereas SMA, thalamus, caudate nucleus, and temporal areas may
vary. This has also been described by Friedman et al. (1998). Although some
of the ancillary activation findings may be explained because of the different
individual strategies chosen on the basis of personal skills, word generation in
groups demonstrated that core areas are not restricted to speech areas (Holland
et al., 2001). Similar circuitry distribution is also seen in many other cognitive
tasks. Calculation tasks, for example, may activate a constellation of different
cortical places (Figure 5) depending on their complexity (Bernal et al., 2003).
Core areas of cognitive tasks are revealed in averaging of groups. Figure 6,
for example, is the averaging of activation of a group of 16 right handed
volunteers realizing a Color Stroop task. In total, 16 subjects activated 13
different areas, but after group analysis only 67 areas of the left hemisphere
proven communality in this task supporting a core circuitry.
In an effort to isolate these core centers (modules?) in high complex
tasks Cabeza and Nyberg (2000) reviewed 275 PET and fMRI studies on 5
different categories of cognitive tasks: language, memory, attention, imagery,
and perception. Each category was divided in subgroups. For language, it
was considered written/spoken word recognition, with and without spoken
responses. For memory, they included studies on working memory, semantic
memory, episodic memory encoding, episodic memory retrieval (rather fluency
tasks), and procedural memory.
Attention was subdivided into sustained, selective, Stroop, orientation,
and divided. Imagery encompasses the subgroups of object and navigation.
For perception, subgroups were objects, faces, and space/motion. In all these
studies there was an extended participation of different brain regions. Pooled
data of receptive language tasks showed activation in 25 different Brodmanns
Figure 4. Demonstrative example of a functional MRI on a right-handed subject performing a

semantic fluency task. The axial images appear in radiological conventional orientation with left
hemisphere on the right. The ON condition consisted of silently retrieval of words belonging to a
given category auditory presented. The OFF (control) condition consisted of thinking of a blue
sky. The subject kept his eyes closed and remained still, while hearing the magnets gradient noise
throughout the task subtracting the primary auditory area activation. The task elicits a complex
activation pattern. The circuitry includes: left Brodmanns 44 area (long arrows), left Brodmanns
6 and 8 areas (short arrows), left Brodmanns area 22 (arrow head); and supplementary motor area
(encircled). In addition there are some activation in the inferior aspect of the right occipital lobe
(small cluster in the lower left image), left thalamus (thin arrow head), and left middle frontal
gyrus (round bubble in lower right image). (Images from the Department of Radiology, Miami
Childrens Hospital.)
areas and 3 subcortical structures. Some tasks activated up to 17 different

areas (naming), whereas 11 of 36 tasks activated 3 or less areas. Areas more
frequently activated were left Brodmanns 22 and 21 (in 24 and 10, respectively
of 36 experiments), more frequently with left dominance. Brodmanns 47 and 45
areas were activated in 11 and 10 studies also with left dominance. Obviously,
the tasks pooled had different designs, and thresholding, and some of them
had more auditory load whereas others were more visually clued. However,
evident differences appear when comparing a number of areas involved in
tasks like hearing words and reading. The former activated an average of 3.4
areas in 9 studies, whereas the latter activated 8 areas in average, having more
frontal areas involved. The authors pooled expressive language tasks within
the category of semantic memory retrieval. Twenty studies of word generation
were analyzed. Twenty different areas activated at least in 3 studies. In average
Figure 5. Activation circuitry elicited by a calculation task. The subject subtracts 7 from one
hundred, and over again from each subsequent obtained result. Several areas activated. Brodmanns
area 44 appears activated bilaterally, more on the left (arrows, left image), extensive activation is
seen in the left parietal lobe encompassing both banks of the intraparietal sulcus, and the posterior
inferior parietal lobule (encircled). In addition there are activation in right Brodmanns 22 area,
SMA, and prefrontal areas. Interestingly there are two symmetrical activations occurring in the
posterior third of the superior frontal sulcus (short arrows). (Images from the Department of
Radiology, Miami Childrens Hospital.)
Figure 6. Activation circuitry obtained with a Stroop color task. The image is an average of
activation of a group of 16 right-handed subjects. The task consisted of names of colors printed
in congruent color ink (for example, Blue in blue ink), for the ON condition, and incongruently
(Blue in red ink) for the OFF condition. The subject was asked to silently name the ink color. Thus,
visual stimuli, reading, inner speech, attention, ocular motion, and working memory are controlled.
Activation is seen in different areas of the left hemisphere: Brodmanns area 37 (encircled),
Brodmanns area 44 (long arrow), and Brodmanns areas 22, 39, and 40 involving the supramarginal
gyrus (short arrows). There are also activation in left thalamus, left striattum, left cerebellum, and
anterior cingulate gyrus with left predominance. (Images from the Department of Radiology, Miami
Childrens Hospital.)
5.7 areas were involved. Left Brodmanns 45 and 44 (except one study that
reports bilateral), are reported in 13 and 10 studies, respectively. Brodmanns
area 21, 6 and SMA (3 cingulate gyrus areas) are reported in 13 studies.
Verbal working memory activated between 4 and 11 different areas with a
mean of 7.6. Brodmanns area 44 was reported activated in 14 of 18 studies, 11
of them in the left side, and the rest bilaterally. Brodmanns area 6 is reported
in 15 studies, 11 of them bilaterally. Area 40, in 15 studies, 8 bilaterally. Area 7
appears 11 times. Interestingly, the cerebellum appears activating in 10 studies.
No study found less than four different areas involved.
Results are similar for the rest of the pooled study categories: several
areas are simultaneously involved in complex tasks with some core areas
that appear in more than 80% of the studies. An interesting corollary of
this study is finding several of Brodmanns areas participating in different
types of tasks. For example Brodmanns area 9 appears activated in sustained
attention (right), verbal and spatial working memory tasks (bilaterally),
episodic memory encoding (left), and episodic memory retrieval. Area 46 is
found activated in all types of working memory, episodic memory encoding
(left) and episodic memory retrieval (right). Left Brodmanns area 45 is

activated in tasks of verbal comprehension, word generation, episodic memory
encoding and conceptual priming. Brodmanns area 6 activates in orientation,
divided attention, space/motion perception, reading, verbal and spatial working
memory (bilaterally), problem solving, and learning of motor schemes. Similar
involvement is seen for Brodmanns area 7 without lateralization. Brodmanns
area 40 is activated in orientation, space/motion perception, space/motion
imaginery (left), verbal working memory (left), spatial working memory
(bilateral), spatial episodic memory encoding (left), verbal retrieval, and motor
skill learning. Brodmanns area 19 and cerebellum, bilaterally, are also involved
in several different types of tasks.
In conclusion, it could be stated that: (1) during relatively simple tasks
(e.g., listening syllables, watching a checkerboard), brain activity changes are
restricted to somehow specific brain areas; and (2) during complex cognitive
tasks (reading, speaking, memory encoding), a participation of multiple brain
areas is observed. Each of these areas makes its particular contribution to the
whole system.
TOWARD A SYNTHESIS
It has been assumed that the cerebral cortex is organized in modules, which
probably range in size from a hundred thousand to a few million neurons
(Carlson, 1994). Each module receives information from other modules,

performs certain processing and then passes the results to other or others
modules (De Valois & De Valois, 1988; Livingstone & Hubel, 1988). In
the primary visual cortex it has been proposed some 2,500 modules, each
approximately 0.5 0.7 mm. The neurons in each module are devoted the
analysis of various features in one specific portion of the visual field. Basic
similarities in the brain organization of different perceptual systems have been
proposed, and neural organization for different sensory systems is assumed to
be alike (Konorski, 1967). In consequence, the very same principles (neural
operations) required for visual pattern recognition (Hubel & Wiesel, 1961,
1962, 1963, 1965) can be also expected for auditory pattern recognition, for
example, phoneme perception (Ardila, 1993b). Although the comprehension
of how these modules work and are organized is of the utmost importance
for neurocognitive sciences, it is of poor aid for clinical research, and has no
medical decision impact. Indeed, there is no neurological condition affecting
single modules. Tumors, infarcts, trauma, degenerative diseases, and infections
injure the brain, affecting areas larger than modules. Even when they are small
as a module can get, it is very unlikely that the location matches exactly that one
of a single module. For this reason, in clinical practice, the observed deficit is
due to impairments of small units of processing, usually greater than a module.
Clinical observation has shown that apparently rather different tasks (e.g.,
solving numerical problems and recognizing fingers) can be altered as a
consequence of the very same brain pathology. Certain common cognitive
factors should underlie in those apparently different, but simultaneously
altered tasks (e.g., to solve numerical problems and to recognize fingers).
Evidently, the higher the correlation between performance in two different
tasks, the higher the probability should be that the abilities tapped by these two
tasks will be simultaneously impaired in case of focal brain pathology. Broca
aphasia can represent an illustrative example. It is usually recognized that Broca
aphasia has two different distinguishing features: (a) a motor component (lack
of fluency, disintegration of the speech kinetic melodies, verbalarticulatory
defects, etc., that is, apraxia of speech); and (b) agrammatism (e.g., Benson
& Ardila, 1996; Luria, 1976; Goodglass, 1993; Kertesz, 1985). If both defects
are simultaneously observed (i.e., they are very highly correlated), it simply
means they both are just two different manifestations of a single underlying
defect. It is not easy to understand which one could be the single factor
responsible for these two clinical manifestations; it may be kind of an inability
to sequence expressive elements (Figure 7). But, anyhow, a single common
factor underlying both defects should be assumed. Brocas area, most likely, is
Figure 7. The impairment in a common cognitive factor (ability to sequence expressive elements?)
may account for the different manifestations observed in case of Broca aphasia.
not specialized in producing language, but in certain activity that can fundament
not only skilled movements required for speech, but also morphosyntax. It is
interesting to note that deafmute subjects (who, in consequence have not ever
produced verbal articulatory movements) present a virtually total impossibility
to learn, understand, and use language grammar (Poizner et., 1987). Probably,
the lack of verbal articulatory normal development is necessarily associated

with a lack of normal grammatical development.
Tentatively, it could be concluded that,
1. Recent neuroimaging techniques show that cognitive functions are yielded

by a distributed brain activity involving several discrete areas (modules)
arranged into circuitries. These areas are presenting a specific contribution
to the whole system.
2. Clinical neuropsychology has been able to propose some cognitive factors
based on empirical observations in patients with brain lesions.
3. Neuropsychology inventories can provide rough approximation about the
localization of different functional systems.
4. Factor analysis may provide an additional tool to extract some constitutive
elements of functions (factors). Factor in factor analysis may have different
levels of specificity. Some times it refers to a functional system; in other
occasions to elements of cognition.
5. The very same brain areas (and cognitive factors) may be potentially
involved in different types of cognition (functional systems). Thus, phoneme
discrimination is required for language understanding, speaking, writing,
and so on.
6. Impairment in a specific cognitive factor may result in diverse types of
impairments. Thus, the ability to sequence expressive movements, for
example, may be associated with agrammatism and also with apraxia of

speech.
7. Currently, it seems feasible to suppose that some cognitive factors are
responsible for normal neuropsychological performance. Theoretically, the
impairment in any of these factors could be responsible for some specific
neuropsychological syndromes.
REFERENCES
Anastasi, A. (1982). Psychological testing, 5th edition. New York: Macmillan Publishing
Co.
Anokhin, P. K. (1974). Biology and neurophysiology of the conditioned reflex and its
role in adaptive behavior. New York: Pergamon Press.
Ardila, A. (1993a). Historical evolution of spatial abilities. Behavioral Neurology, 6,
8388.
Ardila, A. (1993b). Toward a model of phoneme perception. International Journal of
Neuroscience, 70, 112.
Ardila, A. (2004). There is not any specific brain area for writing: From cave painting
to computers. International Journal of Psychology, 39, 6167.
Ardila, A., Concha, M., & Rosselli, M. (2000). Angular gyrus syndrome revisited: Acal-
culia, finger agnosia, right-left disorientation, and semantic aphasia. Aphasiology,
14, 743754.
Ardila. A., Galeano, L. M., & Rosselli M. (1998). Toward a model of neuropsychological
activity. Neuropsychology Review, 8, 171190.
Ardila, A., & Ostrosky, F. (eds.) (1989). Brain organization of language and cognitive
processes. New York: Plenum Press.
Ardila, A., & Pineda, D. A. (2000). Factor structure of nonverbal cognition. International
Journal of Neurosciences, 104, 125144.
Ardila, A., & Rosselli, M. (1990). Acalculias. Behavioural Neurology, 3, 39
48.
Ardila, A., & Rosselli, M. (1994). Development of language, memory and visuospatial
abilities in five- to 12-year-old children using a neuropsychological battery test.
Developmental Neuropsychology, 10, 97120.
Ardila, A., & Rosselli, M. (2002). Acalculia and dyscalculia. Neuropsychology Review,
12, 179232.
Ardila, A., Rosselli, M., & Bateman, J. R. (1994). Factorial structure of cognitive
activity. Behavioural Neurology, 7, 4958.
Benson, D. F. & Ardila, A. (1996). Aphasia: A clinical perspective. New York: Oxford
University Press.
Bernal, B., & Altman, N. R. (2001). Auditory functional MR imaging. American Journal
of Roentgenology, 176, 10091015.
Bernal, B., Altman, N. R., & Medina, L. S. (2004). Dissecting non verbal auditory
cortex. asymmetry: An fMRI study. International Journal of Neurosciences, 114,
661680.
Bernal, B., Ardila, A., & Altman, N. R. (2003). Acalculia: An fMRI Study
with implications with respect to brain plasticity. International Journal of

Black, S. E., & Behrmann, M. (1994). Localization in alexia. In Kertesz, A.
(ed.), Localization and neuroimaging in neuropsychology. New York: Academic
Press.
Bornstein, R. A., & Chelune, G. J. (1988). Factor structure of the Wechsler Memory
Scale-Revised. The Clinical Neuropsychologist, 2, 107115.
Broca, M. P. (1861). Remarques sur le siege de la facute du langage articule,
suiveies dune observation daphemie (perte de la parole). Bulletin de la Societe
Anatomique, 6, 330357.
Brodmann, K. (1909). Vergleichende lokalisationslehre der grosshirnrinde in ihren
prinzi pen dargestellt auf grund des zellenbaus. Leipzig: Barth.
Burton., D. B., Ryan, J. J., Axelrod, B. N., & Schellenberger, T. (2002). A confirmatory
factor analysis of the WAIS-III in a clinical sample with cross validation in the
standardization sample. Archives of Clinical Neuropsychology, 17, 371387.

Cabeza, R., & Nyberg, L. (2000). Imaging cognition II: An empirical review of 275
PET and fMRI studies. Journal of Cognitive Neurosciences, 12, 147.
Caplan, D. (1981). On the cerebral localization of linguistic functions: Logical and
empirical issues surrounding deficit analysis and functional localization. Brain
and Language, 14, 120137.
Caramazza, A. (1992). Is cognitive neuropsychology possible? Journal of Cognitive
Carey, L. M., Abbott, D. F., Egan, G. F., Tochon-Danguy, H. J., & Donnan, G. A. (2000).
The functional neuroanatomy and long-term reproducibility of brain activation
associated with a simple finger tapping task in older healthy volunteers: A serial
PET study. Neuroimage, 11, 124144.
Carlson, N. R. (1994). Physiology of behavior. Boston: Allyn and Bacon, 5th edition.
Carroll, J. B. (1993). Human cognitive abilities: A survey of factor analytic studies.
Cambridge: Cambridge University Press.
Casey, B. J., Cohen, J. D., OCraven, K., Davidson, R. J., Irwin, W., Nelson, C. A., Noll,
D. C., Hu. X., Lowe, M. J., Rosen, B. R., Truwitt, C. L., & Turski, P. A. (1998).
Reproducibility of fMRI results across four institutions using a spatial working
memory task. Neuroimage, 8, 249261.
Castro-Caldas, A., Petersson, K. M, Reis, A., Stone-Elander, S., & Ingvar, M. (1988).
The illiterate brain. Learning to read and write during childhood influences the
functional organization of the adult brain. Brain, 121, 10531063.
Cattell, R. B. (1971). Abilities: Their structure, growth and action. San Francisco:
Houghton-Mifflin.
Cohen, J. A. (1957). A factor analyticallybased rationale for the Wechsler Adult

Intelligence Scale. Journal of Consulting Psychology, 21, 451457.
Colder, B. W., & Tanenbaum, L. (1999). Dissociation of fMRI activation and awareness
in auditory perception task. Cognitive Brain Research, 8, 177184.
Cronbach, L. J. (1990). Essentials of psychological testing, 5th edition. New York:
Harper & Row.

Crosson, B., Moberg, P. J., Boone, J. R., Rothi, L. J., & Raymer, A. (1997). Category-
specific naming deficit for medical terms after dominant thalamic/capsular
hemorrhage. Brain and Language, 60, 407442.
Crosson, B., Parker J. C., Kim, A. K., Warren, R. L., Kepes, J. J., & Tully R. (1986). A
case of thalamic aphasia with postmortem verification. Brain and Language, 29,
301314.
Damasio, H., & Damasio, A. R. (1989). Lesion analysis in neuropsychology. New York:
Oxford University Press.
De Valois, R. I., & De Valois, K. K. (1988). Spatial vision. New York: Oxford University
Press.
Elbert, T., Pantev, C., Wienbruch, C., Rockstroh, B., & Taub, E. (1995). Increased
cortical representation of the fingers of the left hand in string players. Science,
270, 305307.
Elwood, R. W. (1991). Factor structure of the Wechsler Memory Scale-Revised
(WMSR) in a clinical sample: A methodological reappraisal. The Clinical
Neuropsychologist, 5, 329337.
Engel, S. A., Rumelhart, D. E., Wandell, B. A., Lee, A. T., Glover, G. H., Chichilnisky,
E. J., & Shadlen, M. N. (1994). fMRI of human visual cortex. Nature, 369, 525.
Fera, F., Yongbi, M. N., van Gelderen, P., Frank, J. A., Mattay, V. S., & Duyn, J. H. (2004).
EPI-BOLD fMRI of human motor cortex at 1.5 T and 3.0 T: sensitivity dependence
on echo time and acquisition bandwidth. Journal of Magnetic Resonance Imaging,
19, 1926.
Fodor, J. A. (1983). The modularity of mind. Boston: Bradford Books/MIT Press.
Foerster, O. (1933). The dermatomes in man. Brain, 56, 139.
Freund, H. J. (2002). fMRI studies of the sensory and motor areas involved in movement.
Advances in Experimental Medical Biology, 508, 389395.
Friedman, L., Kenny, J. T., Wise, A. L., Wu, D., Stuve, T. A., Miller, D. A., Jesberger, J.
A., & Lewin, J. S. (1998). Brain activation during silent word generation evaluated
with functional MRI. Brain and Language, 64, 231256.
Goodglass, H. (1993). Understanding aphasia. New York: Academic Press.
Goodglass, H., Wingfield, A., Hyde, M. R., & Theurkauf, J. C. (1986). Category specific
dissociations in naming and recognition by aphasic patients. Cortex, 22, 87102.
Guilford, J. P. (1967). The nature of human intelligence. New York: McGraw-Hill.
Guilford, J. P. (1968). Intelligence has three facets. Science, 160, 615620.
Guilford, J.P. & Hoepfner, R. (1971). The analysis of intelligence. New Yrok: McGraw-
Hill.
Harris, D. M., & Kay, J. (1995). Selective impairment of the retrieval of peoples
names:A case of category specificity. Cortex, 31, 575582.
Haut, J. S., Haut, M. W., Callahan, T. S., & Franzen, M. D. (1992). Factor analysis of
wide range of assessment of memory and learning (WRAML) scores of a clinical
sample. The 12th Annual Meeting of the National Academy of Neuropsychology,
Hirsch, J. (1994). Functional imaging of the human visual system using magnetic
resonance. Aging and Vision News, 6, 416.
Holland, S. K., Plante, E., Weber, B. A., Strawsburg, R. H., Schmithorst, V. J., & Ball,
W. S. Jr. (2001). Normal fMRI brain activation patterns in children performing a
verb generation task. Neuroimage, 14, 837843.
Hubel, D. H., & Wiesel, T. N. (1961). Receptive field of single neurons in the cats
striate cortex. Journal of Physiology, 148, 574591.
Hubel, D. H., & Wiesel, T. N. (1962). Receptive field, binocular interactions and
functional architecture in the cats visual center. Journal of Physiology, 160, 106
154.
Hubel, D. H., & Wiesel, T. N. (1963). Receptive field of cells in striate cortex of
very young, visually inexperienced kitten. Journal of Neurophysiology, 26, 994
1002.
Hubel, D. H., & Wiesel, T. N. (1965). Receptive fields and functional architecture in
two nonstriate visual areas (18 and 19) of the cat. Journal of Neurophysiology, 28,
229289.
Humphreys, G. W. & Riddoch, M. J. (1987). Visual object processing: A cognitive
neuropsychological approach. New York: Lawrence Erlbaum Associates.
Keegan, J. J., & Garret, F. D. (1948). Anatomical Record, 102, 409437.
Kertesz, A. (ed.). (1983). Localization in neuropsychology. New York: Academic Press.
Kertesz, A. (1985). Aphasia. In Frederiks, J. A. M. (ed.), Handbook of clinical neurology,
vol 45: Clinical neuropsychology. Amsterdam: Elsevier.
Kertesz, A. (ed.). (1994). Localization and neuroimagening in neuropsychology. New
York: Academic Press.
Konorski, J. (1969). The integrative activity of the brain. Chicago: The Chicago
University Press.
Levin, H. S., & Grafman, J. (2000). Cerebral reorganization of function after brain
damage. New York: Oxford University Press.
Levy, W. B., & Steward, O. (1983). Temporal contiguity requirements for long-term
associative potentiation/depression in the hippocampus. Neuroscience, 8, 791797.
Lezak, M. D. (1995). Neuropsychological assessment, 3rd edition. New York: Oxford
University Press.
Li, E., Weng, X., Han, Y., Wu, S., Zhuang, J., Chen, C., Feng, L., & Zhang, K. (2000).
Asymmetry of brain functional activation: fMRI study under language and music
stimulation. Chinese Medical Journal, 113, 154158.
Lieberman, P. (2000). Human language and our reptilian brain. Boston: Harvard
University Press.
Livingstone, M. S., & Hubel, D. H. (1988). Psychophysical evidence for separate
channels for the perception of form, color, movement, and depth. Journal of
Neuroscience, 7, 34163468.
Livingston, R. B., Gray, R. M., Haak, R. A., & Jennings, E. (2000). Factor structure of the
Reitan-Indiana Neuropsychological Battery for Children. Assessment, 7, 189199.
Loewenstein, D. A., Ownby, R., Schram, L., Acevedo, A., Rubert, M., & Arguelles, T.
(2001). An evaluation of the NINCDA-ADRDA neuropsychological criteria for
the assessment of Alzheimers disease: A confirmatory factor analysis of single
versus multi-factor models. Journal of Clinical & Experimental Neuropsychology,
23, 274284.
Lotze, M., Erb, M., Flor, H., Huelsmann, E., Godde, B., & Grodd, W. (2000).
fMRI evaluation of somatotopic representation in human primary motor cortex.
Neuroimage, 11, 473481.
Luria, A. R. (1966). Higher cortical functions in man. New York: Basic.
Luria, A. R. (1970). Traumatic aphasia. New York: Mouton. (Original work published
1947)
Luria, A. R. (1973). The working brain. New York: Basic.

Luria, A. R. (1976). Basic problems of neurolinguistics. New York: Mouton.
Lyons, F., Hanley, J. R., & Kay, J. (2002). Anomia for common names and geographical
names with preserved retrieval of names of people: a semantic memory disorder.
Cortex, 38, 2335.
Matarazzo, J. D. (1972). Wechslers measurement and appraisal of adult intelligence.
New York: Oxford University Press.
Matarazzo, J. D. (1992). Psychological testing and assessment in the 21st century.
American Psychologist, 47, 10071018.
Moradi, F., Liu, L. C., Cheng, K., Waggoner, R. A., Tanaka, K., & Ioannides, A. A.
(2003). Consistent and precise localization of brain activity in human primary
visual cortex by MEG and fMRI. Neuroimage, 18, 595609.
Moscovitch, M. & Ulmita, C. (1990). Modularity and neuropsychology: Modules
and central processing in attention and memory. In Lister, R. G. & Weingartner,
H. J. (eds.), Modular deficits in Alzheimers type dementia. New York: Oxford
University Press.
Ojemann, G., & Mateer, C. (1979). Human language cortex: localization of memory,
syntax, and sequential motor-phoneme identification systems. Science, 205,
14011403.
Ojemann, G. A. (1979). Individual variability in cortical localization of language.
Journal of Neurosurgery, 50, 164169.
Ojemann, J. G., Ojemann, G. A., & Lettich, E. (2002). Cortical stimulation mapping of
language cortex by using a verb generation task: Effects of learning and comparison
to mapping based on object naming. Journal of Neurosurgery, 97, 3338.
Ostrosky, F., Canseco, E., Quintanar, L., Navarro, E., & Ardila, A. (1985). Sociocultural
effects in neuropsychological assessment. International Journal of Neuroscience,
27, 5366.
Ostrosky, F., Quintanar, L., Canseco, E., Meneses, S., Navarro, E., & Ardila, A. (1986).
Habilidades cognoscitivas y nivel sociocultural. Revista de Investigacion Clnica,
38, 3742.
Ozeren, A., Sarica, Y., & Efe, R. (1994). Thalamic aphasia syndrome. Acta Neurological
Belgian, 94, 205208.
Perkins, A. T., & Teyler, T. J. (1988). A critical period for long-term potentiation in the
developing rat visual cortex. Brain Research, 439, 2547.
Petersson, K. M., Reis, A., Askelof, S., Castro-Caldas, A., & Ingvar, M. (2000). Lan-
guage processing modulated by literacy: A network analysis of verbal repetition
in literate and illiterate subjects. Journal of Cognitive Neurosciences, 12, 364
382.
Poizner, H., Klima, E. S., & Bellugi, U. (1987). What the hands reveal about the brain.
Boston: MIT Press.
Ponton, M. O., Gonzalez, J. J., Hernandez, I., Herrera, L., & Higareda, I. (2000). Factor
analysis of the Neuropsychological Screening Battery for Hispanics (NeSBHIS).
Applied Neuropsychology, 7, 3239.

Ponton, M. O., Satz, P., & Herrera, L. (1994). Factor analysis of a neuropsychological
screening battery for Hispanics (NeSBHIS). Angers (France). XVII European
Meeting International Neuropsychological Society,
Price, C. J., Wise, R. J., Watson, J. D., Patterson, K., Howard, D., & Frackowiak, R. S.
(1994). Brain activation during reading: The effects of exposure duration and
task. Brain, 117, 12551269.
Raichle, M. E. (1994). Visualizing the mind. Scientific American, 270, 5865.
Roid, G. H., Prifitera, A., & Ledbetter, M. (1988). Confirmatory analysis of the factor
structure of the Wechsler Memory ScaleRevised. The Clinical Neuropsychology,
2, 116120.
Rosler, F., Roder, B., Heil, M., & Hennighausen, E. (1993). Topographic differences
of slow event-related brain potentials in blind and sighted adult human subjects
during haptic mental rotation. Cognitive Brain Research, 1, 145159.
Rosselli, M., & Ardila, A. (1989). Calculation deficits in patients with right and left
hemisphere damage. Neuropsychologia, 27, 607617.
Ryan, J. J., & Paolo, A. M. (2002). Exploratory factor analysis of the WAIS-III in a
mixed patient sample. Archives of Clinical Neuropsychology, 16, 151156.
Shallice, T. (1988). From neuropsychology to mental structure. Cambridge: Cambridge
University Press.
Stuss, D. T., & Benson, D. F. (1986). The frontal lobes. New York: Raven Press.
Thulborn, K. R., Chang, S. Y., Shen, G. X., & Voyvodic, J. T. (1997). High-resolution
echo-planar fMRI of human visual cortex at 3.0 tesla. NMR Biomedic, 10, 183190.
Thurstone, L. L. (1947). Multiple factor analysis. Chicago: University of Chicago Press.
Tulsky, D. S., & Price, L. R. (2003). The joint WAIS-III and WMS-III factor structure:
Development and cross-validation of a six-factor model of cognitive functioning.
Psychology Assessment, 15, 149162.
Uhl, F., Franzen, P., Podreka, I., Steiner, M., & Deecke, L. (1993). Increased regional
cerebral blood flow in inferior occipital cortex and cerebellum of early blind
humans. Neuroscience Letter, 150, 162164.

Warrington, E. K., & Shallice, T. (1984). Category specific semantic impairments.
Brain, 107, 829854.
Wechsler, D. (1987). Manual: Wechsler Adult Intelligence Scal Revised. San Antonio:
Psychological Corporation.
Wechsler, D. (1997). WAISIII: Administration and scoring manual. San Antonio:
The Psychological Corporation.
Wessinger, C. M., Buonocore, M. H., Jussmaul, C. L., & Mangun, G. R. (1997).
Tonotopy in human auditory cortex examined with functional magnetic resonance
imaging. Human Brain Mapping, 4, 1825.
Wilhem, K. L. & Franzen, M. D. (1992). A factor analysis of several brief
neuropsychological instruments. Pittsburgh, PA: The12th Annual Meeting of the
National Academy of Neuropsychology.
How Localized are Language Brain Areas? A Review of Brodmann Areas

Involvement in Oral Language
Alfredo Ardila1,*, Byron Bernal 2, Monica Rosselli3
1
2
Radiology Department/Research Institute, Miami Childrens Hospital, Miami, FL, USA
Downloaded from http://acn.oxfordjournals.org/ at Florida International University on January 19, 2016

3
Department of Psychology, Florida Atlantic University, Davie, FL, USA
*Corresponding author at: Department of Communication Sciences and Disorders, 11200 SW 8th Street, AHC3-431B, Florida International University, Miami,
FL 33199, USA. Tel.: +1-305-348-2750; fax: +1-305-348-2710.
E-mail address: ardilaa@fiu.edu (A. Ardila).
Accepted 27 October 2015
Abstract
The interest in understanding how language is localized in the brain has existed for centuries. Departing from seven meta-analytic studies of
functional magnetic resonance imaging activity during the performance of different language activities, it is proposed here that there are two
different language networks in the brain: first, a language reception/understanding system, including a core Wernickes area involved in
word recognition (BA21, BA22, BA41, and BA42), and a fringe or peripheral area (extended Wernickes area: BA20, BA37, BA38,
BA39, and BA40) involved in language associations (associating words with other information); second, a language production system
(Brocas complex: BA44, BA45, and also BA46, BA47, partially BA6-mainly its mesial supplementary motor area-and extending toward
the basal ganglia and the thalamus). This paper additionally proposes that the insula (BA13) plays a certain coordinating role in interconnecting
these two brain language systems.
Keywords: Language; Brodmann areas; Meta-analysis; Wernickes area; Brocas area
Introduction
The interest in understanding the localization of language in the brain has existed for centuries (Eling, 1994; Tesak & Code,
2008). This interest notoriously increased during the late 18th and early 19th centuries with the development and spread of the
so-called phrenology movement (Simpson, 2005) and received a scientific anatomical impulse with Brocas (1861, 1863) and
Wernickes (1874/1970) clinical studies. During that historical moment (the second half of the 19th century), it was supposed
that language was associated with the activity of three areas in the left hemisphere: the posterior frontal lobe, the upper
segment of the temporal lobe, and the insula. Wernicke stated, The whole area of the convolution encircling the Sylvian
fissure, in association with the cortex of the insula, serves as speech center. The first frontal gyrus, being motor, is the center
for representation of movements, and the first temporal gyrus, being sensory, is the center for word-images (Wernicke, 1874/
1970, p. 280). Dejerine (1914) integrated the available clinical/anatomical observations and proposed there was a language
area (or language zone) in the brain. This area included Brocas area, located in the left inferior frontal gyrus corresponding
to Brodmann Area (BA) 44; Wernickes area (posterior segment of the first temporal gyrus; BA22); and a written language
center (angular gyrus; BA39). Although he suggested that the insula (BA13) may also participate in language, he did not expli-
citly include it within the brain language area. As a result, the interest in the potential participation of the insula in language vir-
tually disappeared until the end of the 20th century. Dronkers (1996) demonstrated that the insula is a critical brain region for the
coordination of complex articulatory movements (speech praxis). Further articles were published emphasizing the participation of
the insula in language during the following years (e.g., Ardila, 1999; Ardila, Benson & Flynn, 1997).
# The Author 2015. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
doi:10.1093/arclin/acv081 Advance Access publication on 12 December 2015
A. Ardila et al. / Archives of Clinical Neuropsychology 31 (2016); 112122 113
During the last decades of the 20th century, the aphasia model was generally used to illustrate the brain organization of language
(e.g., Brown, 1976; Damasio, 1992; Obler & Gjerlow, 1999). That is, brain organization of language was explained departing from
language pathology. Different types of language disorders were reported in cases with brain damage, depending upon the specific
location of the damage. Consequently, it was proposed that specific brain areas are involved in processing particular aspects of
language. Two major aphasia models were generally used during the middle and late 20th century: Geschwind and associates
model and Lurias model.
(1) Geschwind and associates model (so-called Wernicke Geschwind model of language) (Benson, 1979; Geschwind,
1970, 1972; Kertesz, 1985) represents an extension of the Wernicke Lichtheim model (Lichtheim, 1885). In this
model, seven different types of aphasia, each one related with the pathology of a specific brain area, are distinguished:
motor, sensory, conduction, transcortical motor, subcortical motor, transcortical sensory, and subcortical sensory.
Minor variations to the basic classification, proposed by Wernicke, were sometimes used (e.g., Hecaen & Albert,
1978; Lecours, Lhermitte, & Bryans, 1983). According to the Wernicke Geschwind model, three major dimensions

of language can be impaired in cases of aphasia: fluency (in cases of pre-rolandic aphasias), comprehension (in temporal
lobe aphasias), and repetition (in perisylvian aphasias).
(2) Lurias model (1964, 1970, 1976). According to Lurias aphasia interpretation, seven different forms of aphasia can be
recognized: efferent motor or kinetic, dynamic, afferent motor or kinesthetic, semantic, acoustic-agnosic,
acoustic-amnesic, and amnesic. Each one suggests a specific level (factor) of language processing impairment.
Following Jakobsons linguistic analysis of aphasia (Jakobson, 1964; Jakobson & Halle, 1956), Luria (1964)
assumed that these diverse aphasic syndromes could be grouped into two major language axes or dimensions: paradig-
matic and syntagmatic. Some aphasias can be due to a paradigmatic disorder (i.e., inability to select language elements),
whereas others can be the result of a syntagmatic impairment (i.e., inability to sequence language elements). In pre-
rolandic aphasias (motor kinetic and dynamic), the syntagmatic dimension of language is impaired (sequencing
words in a sentence in the first one; sequencing sentences in language discourse in the second one); whereas the retro-
rolandic aphasias are due to a defect in the selection process (selecting phonemes in acoustic-agnosic; selecting words in
acoustic-amnesic; selecting the word meanings in amnesic; and selecting the meaning of word spatial relations in se-
mantic aphasia) (paradigmatic axes of language).
During the early 21st century, Ardila (2010) further developed the idea concerning the two fundamental dimensions in language
impairments. He proposedfollowing Jakobson linguistic analysisthat there are only two fundamental forms of aphasia which
are linked to impairments in the lexical-semantic (paradigmatic axes) and grammatical (syntagmatic axes) systems of language,
namely Wernicke-type aphasia and Broca-type aphasia, respectively (Ardila, 2011, 2012). He further suggested that grammar cor-
relates with the ability to internally represent actions (verbs), depending on the functioning of what is known as Brocas area (BA44
and BA45) and its related brain circuits; it is also associated with the ability to quickly carry out the sequencing of articulatory
movements required for speaking (speech praxis). He argued that lexical-semantic and grammatical systems not only depend
on different brain circuitries but also on different types of memory and learning (declarative and procedural). These systems
tend to appear at different moments during language development in the child and language evolution in humankind. Ardila
emphasized that other aphasic syndromes do not impair language knowledge per se, but rather peripheral mechanisms required
to produce language (in conduction aphasia and the aphasia of the supplementary motor area), or the executive control of language
(in extra-Sylvian, transcortical motor, or dynamic aphasia).
The introduction of contemporary neuroimaging techniques, especially positron emission tomography and functional magnet-
ic resonance imaging (fMRI), significantly advanced the understanding of the brain organization of language. Regardless of the
diversity of methodological limitations that can be pointed out through the use of these neuroimaging techniques (Logothetis,
2008), it has become clear that language is associated with brain networks or circuits rather than with specific brain areas
(Ferstl, Neumann, Bogler, & Von Cramon, 2008; Gitelman, Nobre, Sonty, Parrish, & Mesulam, 2005; Papathanassiou et al.,
2000). These contemporary techniques have allowed the localization of language processing areas in the brain to be re-analyzed.
Attempts have been made to pinpoint the language-comprehensive area in the temporal lobe (BA22, BA21) (e.g., DeWitt &
Rauschecker, 2013; Dronkers, Redfern, & Knight, 2000; Poeppel & Hickok, 2004), as well the language production area
in the frontal lobe (BA44, BA45) (Foundas, Daniels, & Vasterling, 1998; Foundas, Eure, Luevano, & Weinberger, 1998;
Grodzinsky & Amunts, 2006; Grodzinsky & Santi, 2008; Ullman, 2006). For instance, DeWitt and Rauschecker (2013) suggested
two different systems can be distinguished in Wernickes area: dorsal and ventral. The left superior temporal gyrus (ventral stream)
(roughly BA22) supports auditory word-form recognition, whereas the superior temporal/inferior parietal lobule (dorsal stream)
(roughly BA22, BA39) supports functions of inner speech.
114 A. Ardila et al. / Archives of Clinical Neuropsychology 31 (2016); 112122
There is a significantly large body of research today based on functional neuroimaging techniques. Now is the time to put these
significant amounts of information collected during these last decades together, develop more meta-analytic studies to integrate
and reinterpret the available fragmented pieces of information, and ultimately reach a clearer understanding of the brain organ-
ization of language.
Contribution of Different BAs to the Language System
To further advance understanding about the brain organization of language and departing from contemporary neuroimaging
techniques, a series of meta-analytic studies were recently executed (Ardila, Bernal, & Rosselli, 2014a, 2014b, 2015, in press;
Bernal, Ardila, & Rosselli, 2015; Rosselli, Ardila, & Bernal, 2015). These studies aimed to analyze the specific contribution of
different BAs to the language system. A meta-analytic approach integrating a significant amount of recent neuroimaging
studies potentially allows an overall perspective of the different circuits and areas involved in language processing. Some
areas potentially involved in language reception and understanding (lexical-semantic system) as well as areas involved in language
production (grammatical system) were analyzed. Two frontal areas (BA44 and BA46) were selected. BA44 is regarded as the core

Brocas area and BA46 is considered a major frontal lobe executive functioning area. In the temporal lobe, three areas surrounding
the classical Wernickes area (BA22 and BA21) were analyzed: BA20, BA37, and BA38. Because the angular gyrus (BA39) is at
times included within the Wernickes area (e.g., Mesulam, 1998), it was also studied. Finally, the insula (BA13) was notably ana-
lyzed, considering its unclear, but perhaps crucial involvement in language processes.
Brain areas co-activated when performing a particular task suggest that they belong to a common specific network or brain
circuit related to the function selected as filter criterion (e.g., language). Thus, it is assumed that if two or more areas are activated
within the same task they are functionally connected and consequently participate in a single network.
Currently, there are several techniques that potentially demonstrate brain circuitries or networks. These techniques are grouped
under the term brain connectivity. Recently, a new alternative approach to study brain connectivity has been proposed by Robinson,
Laird, Glahn, Lovallo, and Fox (2010) known as meta-analytic connectivity modeling or MACM. MACM is based on automatic
meta-analysis done by pooling co-activation patterns. The technique takes advantage of Brainmap.orgs repository of functional
MRI studies (e.g., Fox & Lancaster, 2002; Laird, Lancaster, & Fox, 2005; Van Essen, 2002) and of Sleuth, a special software provided
by the same group to find, filter, organize, plot, and export the peaks coordinates for further statistical analysis of its results. Sleuth
provides a list of foci, in Talairach or MNI coordinates, each one representing the center of mass for a cluster of activation. The method
takes the region of interest (for instance, BA44), makes it the independent variable, and interrogates the database for studies showing
activation of the chosen target. The query is easily filtered by different conditions (such as age, normal vs. patients, type of paradigm,
domain of cognition, etc.). By pooling the data with these conditions, this tool provides a universe of co-activations that can be stat-
istically analyzed for significant commonality. As a final step, activation likelihood estimation (ALE) (Laird, Fox, et al., 2005;
Turkeltaub, Eden, Jones, & Zeffiro, 2002) is performed utilizing GingerALE, another software provided by Brainmap, which
assesses the probability of an event to occur at voxel level across studies. As mentioned, this methodology assumes coactivation
of brain areas within the same task to indicate interconnectivity between activated areas and their participation in a common network.
In the papers included in the current review, meta-analyses of seven discrete BAs were performed. The following common criteria
were used: first, studies reporting activation of a specific BA (i.e., BA20, BA37, BA38, BA39, BA44, BA46, and BA13); second,
studies using fMRI; third, context: normal subjects; fourth, activations: activation only; fifth, handedness: right-handed subjects;
sixth, age 1860 years; seventh, domain: cognition; subtype: language; eighth, Language: English. The final condition was used
because today a significant amount of fMRI research utilizes Chinese language; including a distant language from the English phon-
ology, lexicon, and grammar could introduce a potential uncontrolled confounding variable. The other seven conditions were used in
an attempt to have a relatively homogenous normal population, studied using a single technical procedure (i.e., fMRI).
Meta-analyses of three types of brain areas were developed: first, areas involved in language reception and understanding
(lexical-semantic system) (BA20, BA37, BA38, and BA39); second, areas involved in language production (grammatical
system) (BA44 and BA46); and third, the insula (BA13), a brain area potentially involved in language control and coordination.
Activation foci associated to each specific area (search criteria) were obtained automatically from the Sleuth software. This
automatic report lists a number of clusters defined by the center of mass (in MNI coordinates), volume in cubic millimeter,
maxima intensity (peak), and neighboring BAs-peaks within 5 mm of the maxima plus and minus with respect the orthogonal
coordinates. Clusters are labeled accordingly with location of maxima. These coordinates, per subject/task/paper/BA, were
exported as text files (pooled results) for analysis on the following step.
Statistical significance of clusters found on the pooled-data was then analyzed utilizing the ALE method (Eickhoff et al., 2009).
This step was performed with the open source software GingerALE (http://brainmap.org). ALE treats reported peaks of activation
as spatial probability distributions centered at the given coordinates. ALE computes the union of activation probabilities for each
voxel, allowing differentiation between true convergence of activation foci from random clustering (noise). ALE scores obtained
from thousands of random iterations are used to assign p-values to the observed clusters of activation. Our ALE maps were thresh-
old at p , .01 corrected for multiple comparisons utilizing the false discovery rate method. Only clusters of 200 or more cubic
millimeter were accepted as valid clusters. ALE results were overlaid onto an anatomical template suitable for MNI coordinates,
also provided by BrainMap.org. For this purpose, we utilized the Multi-Image Analysis GUI (Mango) (http://ric.uthscsa.edu/
mango/). A mosaic of transversal-cut insets of fusioned images was obtained utilizing the same tool.
Language Reception and Understanding: Lexical-Semantic System
In four different meta-analytic studies, the role of BA20, BA37, BA38, and BA39 in language was analyzed (Table 1).
BA20 (inferior temporal gyrus and anterior part of the fusiform gyrus) (Ardila, Bernal, & Rosselli, unpublished). Eleven
papers, corresponding to 12 experimental conditions, and 207 participants were used in this analysis. Our results demonstrated
seven clusters of activation: The first cluster (Cluster #1) included the left temporal lobe, BA20 and BA21, whereas Cluster #2
was located at the left insula (BA13) and left prefrontal BA46. Cluster #3 involved the left inferior frontal lobe (BA47) and

Cluster #4 was situated in the left inferior temporal lobe (BA37). Cluster #5 was again situated in the left prefrontal cortex
(BA9). The last two clusters involved the cingulate gyrus (BA30) (Cluster #6) and the left occipital gyrus (BA19) (Cluster #7).
There were few activated areas outside the temporal lobe: insula (BA13), prefrontal cortex (BA46 & BA9), cingulate gyrus
(BA30), and the occipital lobe (BA19). All seven clusters were located in the left hemisphere.
BA37 (inferior temporal gyrus, fusiform gyrus) (Ardila et al., 2015). Twenty papers, with 28 suitable experiments, and 403
participants were obtained. Twelve different clusters were found, six related to the left and six to the right hemisphere.
Significantly higher connectivity values, as represented by higher ALE scores, were located in the left hemisphere. BA37 pre-
sented significant connection with left inferior temporal lobe (BA20), left prefrontal cortex (BA9, BA46, BA45 and BA47),
left insula (BA13), bilateral precuneus (BA7, BA19), cerebellum, and occipital areas (BA18). It was concluded that left BA37
is a common node of two distinct networks: visual recognition (perception) and semantic language functions. It is well known
that BA37 is involved in lexico-semantic associations (i.e., associating words with visual percepts) (see Brodmanns
Interactive Atlas). Clinical observations demonstrate that damage in the left BA37 is associated with significant word-finding dif-
ficulties (anomia) (e.g., Antonucci, Beeson, Labiner, & Rapcsak, 2008; Kraft et al., 2014; Luria, 1976; Raymer et al., 1997),
impaired naming of pictures, significant amounts of semantic paraphasias, and relatively preserved word comprehension
(Foundas, Daniels, & Vasterling, 1998; Foundas, Eure, Luevano, & Weinberger, 1998; Raymer et al., 1997).
BA38 (temporal pole). Eleven papers, corresponding to 12 experimental conditions, and 201 participants were selected (Ardila
et al., 2014b). Four different clusters of activity were found: two in the left (BA38; BA13, BA22) and two in the right hemisphere
(BA7, BA21). The first cluster included BA38, while the second contained the left insula (BA13) and superior temporal lobe
(BA22). The last two clusters involved were the right parietal (BA7) and temporal (BA21) lobes. Seemingly, this area has two
major connection pathways: one within the left hemisphere (language) and the other involving the right hemisphere plausibly par-
ticipating in visuospatial and integrative audiovisual functions.
BA39 (angular gyrus). Eight papers, corresponding to 13 experimental conditions, and 155 participants were selected (Rosselli
et al., 2015). Sixteen activation clusters made a network that included the activation of the two angular gyri (BA39), the superior
right parietal lobe (BA7) and right supramarginal gyri (BA40), the left middle temporal lobe (BA21), and the frontal lobe (bilateral
premotorBA6and left prefrontalBA46). BA39 functional connectivity in language tasks were observed with other
regions of the left hemisphere: frontal lobe including BA6, BA9, BA44, BA46, and BA8, parietal lobe BA40, and temporal
lobe BA37. Some connectivity with the limbic cortex (cingulate gyrus) was also observed. In addition, clusters of activations
were observed with right superior (BA7) and inferior parietal lobe (BA40) and the right insula (BA13). One of the clusters indi-
cated an activation of the right precuneus (BA31).
These results are in agreement with previous findings using structural connectivity techniques and support the integrative role
of the angular gyrus in language functions (Binder, Desai, Graves, & Conant, 2009; Margulies & Petrides, 2013; Seghier, Fagan, &
Price, 2010; Sroka et al., 2014).
Considering these studies, it can be concluded that BA20, BA37, BA38, and BA39 have a partial participation in language,
specifically associating language with other types of information. They can consequently be regarded as language associations
areas. Our results suggest that regardless of BA20, BA37, BA38, and BA39 having participation in language processes, they
cannot be considered core receptive language processing areas (Wernickes area), as the contrast with clinical model findings
does not support that claim. Nonetheless, these parts could be regarded as language processing marginal areas participating in
an extended Wernickes area or simply Wernickes system.
Table 1. Lexical-semantic system: BA20, BA37, BA38, and BA39
Region (BA) x y z Volume (mm3)
BA20
Cluster #1 L Inferior temporal gyrus (20) 258 252 214 1880
L Middle temporal gyrus (21)
Cluster #2 L Insula (13) 244 24 2 968
L Frontal lobe (46)
Cluster #3 L Interior frontal lobe (47) 234 30 210 584
Cluster #4 L Inferior temporal lobe (37) 232 238 214 528
Cluster #5 L Frontal lobe (9) 244 8 28 432
Cluster #6 L Cingulate gyrus (30) 212 256 16 248
Cluster #7 L Superior occipital gyrus (19) 238 280 36 208
BA37
Cluster #1 L Fusiform gyrus (37) 246 258 214 9568

L Sub-gyralgrey matter (37)
L Occipital-temporal gyrus (37)
L Inferior temporal lobe (20)
Cluster #2 L Inferiorfrontal gyrus (9) 242 8 28 7552
L Middle frontal gyrus (46)
L Insula (13)
L Inferior frontal gyrus (45)
Cluster #3 L Middle frontal gyrus (32) 24 14 48 2184
Cluster #4 L Precuneus (19) 230 264 48 1872
Cluster #5 L Inferior frontal gyrus (47) 234 28 26 1336
Cluster #6 R Precuneus (7) 30 266 44 1304
Cluster #7 R Fusiform gyrus (37) 46 260 218 1168
Cluster #8 R Occipital-temporal gyrus (37) 52 268 2 824
Cluster #9 R Posterior lobe of the cerebellum & pyramid of vermix 30 268 232 424
Cluster #10 R Middle occipital gyrus (18) 34 286 8 400
Cluster #11 R Anterior lobe 40 258 230 288
Cluster #12 L Inferior frontal gyrus 252 20 22 240
BA38
Cluster #1 L Superior temporal (38) 252 8 218 536
Cluster #2 L Insula (13) 240 8 214 393
Lsup temporal (22)
Cluster #3 R sup parietal (7) 22 268 52 368
Cluster #4 R mid temporal (21) 250 22 222 288
BA39
Cluster #1 R angular gyrus (39) 36 262 42 2704
R sup parietal (7)
R inf parietal (40)
Cluster #2 R Frontal superior (6) 4 16 48 1872
Cluster #3 R Insula (13) 36 24 24 1792
Cluster #4 L Frontal inferior (9) 250 8 24 1616
L Precentral gyrus (6)
Cluster #5 L Temporal middle (21) 250 258 30 1496
Cluster #6 L Fusiform gyrus (37) 250 054 218 1360
Cluster #7 L Angular gyrus (39) 232 256 44 1312
Cluster #8 L Frontal inferior (44, 45) 244 20 8 1232
L Frontal middle (46)
Cluster #9 R Head caudate nucleus 10 8 0 1040
Cluster #10 R Parietal, precuneus 6 250 36 952
Cluster #11 L Frontal middle (8) 226 32 42 872
Cluster #12 L Inferior parietal (40) 244 240 46 696
Cluster #13 L Parahyppocamal gyrus (36) 228 238 212 680
Cluster #14 L Posterior cingulate gyrus (30) 212 256 18 648
Cluster #15 R Precuneus (31) 6 236 38 472
Cluster #16 L Frontal left (9) 238 22 40 424
Language Production: Grammatical System
In two different meta-analytic studies, the roles of BA44 and BA46 were analyzed (Table 2).
BA44 (Brocas area; inferior frontal gyruspars opercularis). Fifty-seven papers, 84 experiments, and 883 participants were
selected (Bernal et al., 2015). A network consisting of 16 clusters of activation were obtained. Main clusters were located in
the left frontal operculum (BA44 and further, BA47), supplementary motor area (BA6), left parietal lobe (BA7, BA39), left tem-
poral lobe (BA22), and left secondary visual areas (BA18, BA19). There were also clusters in subcortical structures including
the left thalamus, left putamen, and right cerebellum (see Fig. 1). The main cluster encompasses left BA44 and its abutting
areas. These are the anterior insula (BA13), the middle frontal (BA46) gyrus, and the precentral gyrus (BA6). The second
cluster is located in the left pre-SMA and anterior cingulate gyrus involving BA6 and BA32. The involvement of the right anterior
cingulate gyrus could be real, or most likely, an effect of the post-processing utilized to remove pixelation and aliasing of the cluster
borders. This procedure, also known as smoothing, grows the cluster, resulting in fabricated adjacent activation areas. This effect

could explain the aspect of activation of the neighbor contralateral homologue anterior cingulate gyrus. The third cluster is located
in the left superior and inferior parietal lobule, an area shared by BA7, BA39, and BA40. The fourth cluster involves some mirror
areas of the left Broca (right BA44, right anterior insula, and right BA9), plus one subcortical structure, namely the putamen. The
fifth cluster refers to the left fusiform gyrus. The sixth cluster represents the core of the receptive language area or Wernickes area
(left BA22). The next cluster in importance is located in the left thalamus. The other clusters listed in the automated report by
GingerALE are the left putamen, the right parietal lobe (BA7), the occipital lobes (BA18, BA19), the cerebellum, and the right
precentral gyrus (BA4). BA44 (Brocas area) has been related to language production, grammar, language fluency, and sequencing
(Ardila, 2012; Grodzinsky & Amunts, 2006).
BA46 (convexity prefrontal cortex: anterior middle frontal gyrus). Nineteen papers, corresponding to 60 experimental condi-
tions, and 245 participants were selected (Ardila et al., in press). Eleven different clusters of activation were found, mostly
related to the left hemisphere, including BA6, BA44, BA45, BA13, BA37, BA18, BA32, BA21, BA22, and BA19. The first
cluster includes the left frontal areas BA6, BA44, BA45, BA46, and BA47. That is, the whole frontal system involved in language
production. Noteworthy, this as an extensive cluster with a volume approximately eight times larger than Cluster #2 and nearly 11
times larger than Cluster #3. The rest of the activation clusters are relatively small. The second cluster includes the right insula
(BA13). Cluster #3 includes the left fusiform gyrus (BA37) (most likely, the activation of the culmen of the cerebellum is explained
by the smoothing effect of the adjacent activation of the left fusiform gyrus). Cluster #4 includes the left fusiform gyrus (BA37), as
well as its anterior extension (BA20). Cluster #5 and Cluster #6 refer to the left occipital lobe (BA18) and medial frontal lobe (32),
whereas Cluster #7 corresponds to the right BA46. Additionally, Cluster #8 is located in the superior parietal lobe (BA7) and
Cluster #9 corresponds to Wernickes area (BA22, BA21). Finally, Cluster #11 refers to the left lenticular nucleus. In conclusion,
BA46 plays a central role in the language production system, most likely as related to its executive control.
It is essential to observe the pattern of activation in BA44 almost encompasses entirely BA46 activation (Fig. 1). Nonetheless,
first, significant connectivity of BA44 with basal ganglia (lenticular nucleus) and the thalamus was observed while second, BA46
was mostly related with other frontal areas; its connections with posterior and subcortical areas were very limited, suggesting a
fundamental role in the executive control of language production.
The Insula as a Language Coordinating System
BA13 (insula) (Ardila et al., 2014a). Twenty-six papers, corresponding to 39 paradigms, and 522 participants were selected.
Thirteen different activation clusters were found. Insula connections included not only areas involved in language production
(such as the Brocas area) and language understanding (such as the Wernickes area) but also areas involved in language repetition
(such as the supramarginal gyrus) and other linguistic functions such as BA9 in the left prefrontal lobe (part of the prefrontal cortex
probably involved in complex language processes) and BA37 (involved in lexico-semantic associations) (Table 3).
The first cluster includes the left claustrum (the insular subcortical gray matter). This focus extends not only subcortically but
also anteriorly toward the BA9 (middle frontal gyrus in the prefrontal cortex involved in complex language processes including the
use of verbal strategies in executive functions; see: Brodmanns Interactive Atlas). The second cluster includes the anterior cin-
gulate gyrus (BA24) (involved in motor organizationmotor preparation/planning, cognitive/motor inhibitionand language
initiative) and left BA6 (medial frontal gyrus). BA6 includes the supplementary motor area (SMA), clearly included in language
initiation and maintenance of voluntary speech production (Ardila, 2010, 2014). Thus, this second cluster suggests a participation
of the insula in a brain circuit controlling verbal initiative and maintenance of speech production.
Table 2. Language production/grammatical system: BA44, BA46
BA44
Cluster# 1 L Inferior Frontal gyrus (44) 247 15 15 39,992
L Anterior Insula (13)
L Precentral gyrus (6)
L Inferior frontal (47)
L Middle frontal (46)
Cluster #2 L Superior frontal (6) 21 15 47 12,904
L Medial frontal (32)
R Anterior cingulate (32)
Cluster #3 L Superior parietal (7) 234 256 46 12,184
L Inferior parietal (39)
L Supramarginal gyrus (40)

Cluster #4 R Inferior frontal (44) 42 19 1 10,408
R Anterior Insula (13)
R Inferior frontal (9)
R Putamen
Cluster #5 L Fusiform gyrus (37) 243 255 216 5,600
L Cerebellum. Culmen.
Cluster #6 L Middle temporal (22) 256 243 7 4,568
L Inferior Parietal Lobule (39)
Cluster #7 L Thalamus. Medial Dorsal Nucleus 28 213 9 1,968
Cluster #8 L Lentiform Nucleus. Putamen 221 2 3 1,728
Cluster #9 R Superior parietal (7) 30 257 46 1,288
Cluster #10 L Inferior occipital (18, 19) 228 289 22 1,072
Cluster #11 R Middle occipital (18) 33 284 4 728
Cluster #12 R Fusiform gyrus (19) 41 272 216 536
Cluster #13 L Cerebellum. Culmen. 227 261 224 248
Cluster #14 R Precentral (4) 52 29 39 224
Cluster #15 R Cerebellum. Declive. 25 267 224 200
Cluster #16 L Superior Temporal Gyrus (22) 261 215 6 200
BA46
Cluster # 1 L Middle frontal gyrus (46) 246 34 8 18904
L Precental frontal gyrus (6)
Cluster # 2 R Insula (13) 48 16 24 2,424
R Insula (13)
Cluster # 3 L Fusiformgyrus (37) 242 250 220 1,728
L Cerebellum culmen
Cluster # 4 L Fusiformgyrus (37) 252 248 22 1,288
L Inferior temporal gyrus (20)
Cluster # 5 L Occipital (18) 224 294 24 568
Cluster # 6 L Medial frontal lobe (32) 28 18 44 512
Cluster # 7 R Middle frontal gyrus (46) 50 30 18 488
Cluster # 8 L Superior parietal (7) 226 266 50 432
Cluster # 9 L Superior temporal lobe (22) 246 224 0 296
L Middle temporal lobe (21)
Cluster # 10 L Middle occipital (19) 232 276 26 288
Cluster # 11 L Lenticular 226 14 2 256
Cluster #3 includes the right insula and the insular subcortical gray matter (claustrum) and indicates an integrated activity of the
left and right insula. Cluster #4 refers to left BA7 (superior parietal lobe). This area of the brain participates in ideomotor praxis
(Tonkonogy & Puente, 2009), motor imagery (Solodkin, Hlustik, Chen, & Small, 2004), motor learning, (Tonkonogy & Puente,
2009), language processing (Seghier et al., 2004), and temporal context recognition (Zorrilla, Aguirre, Zarahn, Cannon, &
DEsposito, 1996). Therefore, the insula is also part of the language system related to some contextual and motor learning
aspects of speech.

Fig. 1. Brain connectivity of BA44 and BA46. Left panel, left lateral view; right panel, mesial view. The networks are color coded. BA44 is represented by black
and gray (blue and purple in the online version); BA46 is represented by darker gray and gray (red and purple in the online version). Hence, gray (purple in the online
version) represents the shared network of BA44 and BA46. As observed, the activated areas are quite coincidental. BA46 is almost completely included within the
BA44 local network which also encompasses BA45 and part of BA47. The mesial part of BA6 (pre-SMA) is largely represented within the BA44 network, sharing a
small central part with BA46 connectivity. The left thalamus appears activated as part of the BA44 network. BA44 connectivity with the lenticular nucleus is not
seen.
Table 3. The insula: language coordinating system

Cluster #1 L claustrum 234 16 2 15,504

L insula (13)
L inferior frontal gyrus (9)
L inferior frontal gyrus (44)
Cluster #2 L cingulate gyrus (24) 22 10 46 6,672
L medial frontal gyrus (6)
R cingulated gyrus (32)
Cluster #3 R claustrum 34 20 0 3,672
R insula (13)
Cluster #4 L parietal lobe precuneus (7) 224 266 42 2,720
L superior parietal (7)
Cluster #5 L anterior culmen 238 244 222 2,024
L fusiform gyrus (37)
Cluster #6 L middle temporal gyrus (22) 254 248 4 1,872
L superior temporal gyrus (22)
Cluster #7 L supramarginal gyrus (40) 244 240 42 1,336
L inferior parietal lobe (40)
Cluster #8 L frontal precentral (4) 250 210 44 800
Cluster #9 R thalamus medial dorsal n. 10 216 6 488
Cluster #10 L inferior parietal lobe (40) 254 224 36 400
Cluster #11 R superior temporal gyrus (41) 46 232 6 320
Cluster #12 L fusiform occipital (19) 224 288 28 304
Cluster #13 R cerebellum posterior lobe 28 264 224 248
The following cluster (#5) involves BA37 (posterior inferior temporal gyrus, middle temporal gyrus, and fusiform gyrus) and
the cerebellar culmen. However, considering BA37 is exactly above the culmen, this activation can most likely be referred to
BA37. Consequently, Cluster #5 is suggested to only include the left BA37. The following two clusters (#6 and #7) refer to
two areas traditionally involved in language: the left BA22 (superior temporal gyruspart of Wernickes area), and the left
BA40 (supramarginal gyrus). The left BA22 is considered to be a crucial area in language understanding, whereas the left
BA40 has been related to language repetition (Tonkonogy & Puente, 2009) and semantic processing (Chou et al., 2006).
The next activation clusters (#8 and #9) include the left BA4 (primary motor cortexprecentral gyrus) and the medial dorsal
nucleus of the thalamus. They receive input from the hypothalamus and project to the prefrontal cortex. These clusters are known to
been related to attention and memory. Although a direct relation with language is not evident, Clusters #8 and #9 may contribute to

Fig. 2. Brain language areas. The frontal language area (Brocas complex: language production and grammar: BA44, BA45, BA46, BA47) also partially includes
BA6 and extends subcortically to the basal ganglia. The posterior language area (language reception and understanding: lexical-semantic system) includes a core
Wernickes area (BA21, BA22, BA41, and BA42) and an extended Wernickes area also including BA20, BA37, BA38, BA39, and BA40.
the motor aspects of speech and the attention control of language. Cluster #10 on the other hand is similar to Cluster #7 and includes
the left BA40 (inferior parietal lobe). The last three clusters (notoriously smaller, with 300 mm3 or less) include the BA41 (primary
auditory cortexHeschls gyrus), the BA19 (secondary visual cortexinferior occipital or fusiform gyrus), and the posterior lobe
of the cerebellum (that could be an extension of the fusiform gyrus activation). The activation of these clusters may suggest some
participation of the insula in language recognition and visual associations.
In conclusion, the insula represents a core area in language processing and it is related, not just with language production func-
tions but also with language understanding processing. It can be conjectured that the insula is a core hub for language. Its strategic
location between the anterior and posterior language areas would be crucial to play a language coordinating function.
A note of caution should be introduced. Meta-analyses give population data about brain structures involved in language but we
have to keep in mind there is important inter-individual variability of cortical language representation (Dehaene et al., 1997;
Ojemann & Whitaker, 1978; Tzourio-Mazoyer, Josse, Crivello, & Mazoyer, 2004). Current findings consequently refer only to
populationsnot individualresults.
A Proposed Model of the Brains Organization of Language
The above information proposes:
There is a core Wernickes area including not only BA22 and BA21 (as usually suggested) but also BA41 and BA42.
There is a fringe, or peripheral zone, around this core Wernickes area involved in language associations. It corresponds to
BA20, BA37, BA38, BA39, and BA40 (Fig. 2). An extended Wernickes area could be assumed.
In addition to the well-recognized Brocas area (BA44 and BA45), there is a complex frontal-subcortical circuit involved in
language production and grammar (Brocas complex). It includes not only BA44 and BA45 but also BA46, BA47, partially
BA6 (mainly its mesial supplementary motor area), and extends subcortically toward the basal ganglia.
The insula (BA13), probably plays a coordinating role in interconnecting these two brain language systems (lexical-
semantic and grammatical).
Acknowledgements
Our sincere gratitude goes to Adriana Ardila for her editorial support and to Deven Christopher for her editorial and graphic
support.
References
Antonucci, S. M., Beeson, P. M., Labiner, D. M., & Rapcsak, S. Z. (2008). Lexical retrieval and semantic knowledge in patients with left inferior temporal lobe
lesions. Aphasiology, 22 (3), 281 304.
Ardila, A. (1999). The role of insula in language: An unsettled question. Aphasiology, 13 (1), 7987.
Ardila, A. (2010). A proposed reinterpretation and reclassification of aphasic syndromes. Aphasiology, 24 (3), 363 394.
Ardila, A. (2011). There are two different language systems in the brain. The Journal of Behavioral and Brain Science, 1, 2336.
Ardila, A. (2012). Interaction between lexical and grammatical language systems in the brain. Physics of Life Reviews, 9, 198 214.
Ardila, A. (2014). Aphasia handbook. Miami, FL: Florida International University.
Ardila, A., Benson, D. F., & Flynn, F. G. (1997). Participation of the insula in language. Aphasiology, 11 (12), 11591169.
Ardila, A., Bernal, B., & Rosselli, M. (2014a). Participation of the insula in language revisited: A meta-analytic connectivity study. Journal of Neurolinguistics, 29,
31 41.
Ardila, A., Bernal, B., & Rosselli, M. (2014b). The elusive role of left temporal pole (BA38) in language: A meta-analytic connectivity study. International Journal
of Brain Sciences, 2014, Article ID 946039. http://dx.doi.org/10.1155/2014/946039
Ardila, A., Bernal, B., & Rosselli, M. (2015). Language and visual perception associations: Meta-analytic connectivity modeling of Brodmann area 37. Behavioral
Neurology. 2015 (2015), Article ID 565871. http://dx.doi.org/10.1155/2015/565871

Ardila, A., Bernal, B., & Rosselli, M. (in press). Executive control of language: Meta-analytic connectivity modeling of Brodmann area, 46. Psicothema.
Ardila, A., Bernal, B., & Rosselli, M. How extended is Wernickes area? Meta-analytic connectivity study of BA20 and integrative proposal. Unpublished.
Benson, D. F. (1979). Aphasia, alexia and agraphia. New York: Churchill Livingstone.
Bernal, B., Ardila, A., & Rosselli, M. Brocas area network in language functions. (2015). A Meta-analytic connectivity map based on Analysis Likelihood
Estimates. Frontiers in Psychology, 6, 687.
Binder, J. R., Desai, R. H., Graves, W. W., & Conant, L. L. (2009). Where is the semantic system? A critical review and meta-analysis of 120 functional neuroima-
ging studies. Cerebral Cortex, 19 (12), 27672796. Retrieved January 1, 2015, from brainmap.org.
Broca, P. (1861). Remarks on the Seat of the Faculty of Articulated Language, Following an Observation of Aphemia (Loss of Speech). Bulletin de la Societe
Anatomique, 6, 330 357.
Broca, P. (1863). Localisations des fonctions cerebrales. Siege de la faculte du langage articule. Bulletin de la Societe dAnthropologie, 4, 200208. Brodmanns
Interactive Atlas. Retrieved May 9, 2015, from http://www.fmriconsulting.com/brodmann/Introduction.html.
Brown, J. W. (1976). The neural organization of language: Aphasia and lateralization. Brain and Language, 3 (4), 482 494.
Chou, T. L., Booth, J. R., Bitan, T., Burman, D. D., Bigio, J. D., Cone, N. E., et al. (2006). Developmental and skill effects on the neural correlates of semantic
processing to visually presented words. Human brain mapping, 27 (11), 915924.
Damasio, A. R. (1992). Aphasia. New England Journal of Medicine, 326 (8), 531 539.
Dehaene, S., Dupoux, E., Mehler, J., Cohen, L., Paulesu, E., Perani, D., et al. (1997). Anatomical variability in the cortical representation of first and second lan-
guage. Neuroreport, 8 (17), 38093815.
DeWitt, I., & Rauschecker, J. P. (2013). Wernickes area revisited: Parallel streams and word processing. Brain and Language, 127 (2), 181191.
Dronkers, N. F. (1996). A new brain region for coordinating A. Nature, 384, 14.
Dronkers, N. F., Redfern, B. B., & Knight, R. T. (2000). The neural architecture of language disorders. The New Cognitive Neurosciences, 2, 949960.
Eickhoff, S. B., Laird, A. R., Grefkes, C., Wang, L. E., Zilles, K., & Fox, P. T. (2009). Coordinate based activation likelihood estimation meta-analysis of neuroima-
ging data: A random-effects approach based on empirical estimates of spatial uncertainty. Hum Brain Mapp, 30 (9), 29072926.
Eling, P. (Ed.). (1994). Reader in the history of aphasia: From Franz Gall to Norman Geschwind (Vol. 4). Amsterdam: John Benjamins Publishing.
Ferstl, E. C., Neumann, J., Bogler, C., & Von Cramon, D. Y. (2008). The extended language network: A meta analysis of neuroimaging studies on text compre-
hension. Human brain mapping, 29 (5), 581593.
Foundas, A., Daniels, S. K., & Vasterling, J. J. (1998). Anomia: Case studies with lesion localization. Neurocase, 4, 35 43.
Foundas, A. L., Eure, K. F., Luevano, L. F., & Weinberger, D. R. (1998). MRI asymmetries of Brocas area: The pars triangularis and pars opercularis. Brain and
language, 64 (3), 282296.
Fox, P. T., & Lancaster, J. L. (2002). Mapping context and content: The BrainMap model. Nature Reviews Neuroscience, 3 (4), 319 321.
Geschwind, N. (1970). The organization of language and the brain. Science, 170 (961), 940 944.
Geschwind, N. (1972). Language and the brain. New York: WH Freeman.
Gitelman, D. R., Nobre, A. C., Sonty, S., Parrish, T. B., & Mesulam, M. M. (2005). Language network specializations: An analysis with parallel task designs and
functional magnetic resonance imaging. Neuroimage, 26 (4), 975 985.
Goodglass, H. (1992). Diagnosis of conduction aphasia. In S. E. Kohn (Ed.), Conduction aphasia (pp. 3949). Hillsdale, NJ: Lawrence Erlbaum Associates.
Grodzinsky, Y., & Amunts, K. (Eds.). (2006). Brocas region. New York: Oxford University Press.
Grodzinsky, Y., & Santi, A. (2008). The battle for Brocas region. Trends in Cognitive Sciences, 12 (12), 474480.
Hecaen, H., & Albert, M. L. (1978). Human neuropsychology. New York: John Wiley & Sons.
Jakobson, R. (1964). Toward a linguistic typology of aphasic impairments. In A. V. S. DeReuck, & M. OConnor (Eds.), Disorders of language. Boston: Little and
Brown.
Jakobson, R., & Halle, M. (1956). Fundamental of language. The Hague: Moutin.
Kertesz, A. (1985). Aphasia. In J. A. M. Frederiks (Eds.), Handbook of clinical neurology, vol. 45. Clinical neuropsychology. Amsterdam: Elsevier.
Kraft, A., Grimsen, C., Kehrer, S., Bahnemann, M., Spang, K., Prass, M., et al. (2014). Neurological and neuropsychological characteristics of occipital, occipito-
temporal and occipito-parietal infarction. Cortex, 56, 3850.
Laird, A. R., Fox, P. M., Price, C. J., Glahn, D. C., Uecker, A. M., Lancaster, J. L., et al. (2005). ALE meta-analysis: Controlling the false discovery rate and perform-
ing statistical contrasts. Human Brain Mapping, 25, 155 164.
Laird, A. R., Lancaster, J. J., & Fox, P. T. (2005). Brainmap. Neuroinformatics, 3 (1), 65 77.
Lecours, A. R., Lhermitte, F., & Bryans, B. (1983). Aphasiology. London: Baillere-Tindall.
Lichtheim, L. (1885). On aphasia. Brain, 7, 433 484.

Logothetis, N. K. (2008). What we can do and what we cannot do with fMRI. Nature, 453 (7197), 869 878.
Luria, A. R. (1964). Factors and forms of aphasia. In A. V. S. DeReuck, & M. OConnor (Eds.), Disorders of language. London: Churchill Livingstone.
Luria, A. R. (1970). Traumatic aphasia: Its syndromes, psychology, and treatment. Boston: Mouton de Gruyter.
Luria, A. R. (1976). Basic problems of neurolinguistics. The Hague: Mouton.
Margulies, D. S., & Petrides, M. (2013). Distinct parietal and temporal connectivity profiles of ventrolateral frontal areas involved in language production. The
Journal of Neuroscience, 33 (42), 16846 16852.
Mesulam, M. M. (1998). From sensation to cognition. Brain, 121 (6), 10131052.
Obler, L. K., & Gjerlow, K. (1999). Language and the brain. Cambridge: Cambridge University Press.
Ojemann, G. A., & Whitaker, H. A. (1978). Language localization and variability. Brain and Language, 6 (2), 239 260.
Papathanassiou, D., Etard, O., Mellet, E., Zago, L., Mazoyer, B., & Tzourio-Mazoyer, N. (2000). A common language network for comprehension and production:
A contribution to the definition of language epicenters with PET. Neuroimage, 11 (4), 347357.
Poeppel, D., & Hickok, G. (2004). Towards a new functional anatomy of language. Cognition, 92 (1), 1 12.
Raymer, A., Foundas, A. L., Maher, L. M., Greenwald, M. L., Morris, M., Rothi, L. G., et al. (1997). Cognitive neuropsychological analysis and neuroanatomical
correlates in a case of acute anomia. Brain and Language, 58, 137156.
Robinson, J. L., Laird, A. R., Glahn, D. C., Lovallo, W. R., & Fox, P. T. (2010). Meta-analytic connectivity modeling: Delineating the functional connectivity of the

human amygdala. Human Brain Mapping, 31 (2), 173184.
Rosselli, M., Ardila, A., & Bernal, B. (2015). Modelo de conectividad de la circunvolucion angular en el lenguaje: Meta-analisis de neuroimagenes funcionales.
Revista de Neurologia, 60, 495 503.
Seghier, M. L., Fagan, E., & Price, C. J. (2010). Functional subdivisions in the left angular gyrus where the semantic system meets and diverges from the default
network. The Journal of Neuroscience, 30 (50), 16809 16817.
Seghier, M. L., Lazeyras, F., Pegna, A. J., Annoni, J. M., Zimine, I., Mayer, E., et al. (2004). Variability of fMRI activation during a phonological and semantic
language task in healthy subjects. Human Brain Mapping, 23 (3), 140 155.
Simpson, D. (2005). Phrenology and the neurosciences: Contributions of FJ Gall and JG Spurzheim. ANZ Journal of Surgery, 75 (6), 475 482.
Solodkin, A., Hlustik, P., Chen, E. E., & Small, S. L. (2004). Fine modulation in network activation during motor execution and motor imagery. Cerebral Cortex, 14
(11), 1246 1255.
Sroka, M. C., Vannest, J., Maloney, T. C., Horowitz-Kraus, T., Byars, A. W., & Holland, S. K., & CMIND Authorship Consortium. (2014). Relationship between
receptive vocabulary and the neural substrates for story processing in preschoolers. Brain Imaging and Behavior, 9 (1), 4355.
Tesak, J., & Code, C. (2008). Milestones in the history of aphasia: Theories and protagonists. New York: Psychology Press.
Tonkonogy, J., & Puente, A. (2009). Localization of clinical syndromes in neuropsychology and neuroscience. New York: Springer Publishing Company.
Turkeltaub, P. E., Eden, G. F., Jones, K. M., & Zeffiro, T. A. (2002). Meta-analysis of the functional neuroanatomy of single-word reading: Method and validation.
NeuroImage, 16 (3, Part 1), 765.
Tzourio-Mazoyer, N., Josse, G., Crivello, F., & Mazoyer, B. (2004). Interindividual variability in the hemispheric organization for speech. Neuroimage, 21 (1),
422435.
Ullman, M. T. (2006). Is Brocas area part of a basal ganglia thalamocortical circuit? Cortex, 42 (4), 480485.
Van Essen, D. C. (2002). Windows on the brain: The emerging role of atlases and databases in neuroscience. Current Opinion in Neurobiology, 12 (5), 574 579.
Wernicke, C. (1874/1970). The aphasic symptom-complex: A psychological study on an anatomical basis. Archives of Neurology, 22 (3), 280.
Zorrilla, L. T., Aguirre, G. K., Zarahn, E., Cannon, T. D., & DEsposito, M. (1996). Activation of the prefrontal cortex during judgments of recency: A functional
MRI study. Neuroreport, 7 (15 17), 2803 2806.

Ardila A 2016 La Neuropsicologia Como Una Forma de Vida Neuropsychology As A Way of Living

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ardila A 2016 La Neuropsicologia Como Una Forma de Vida Neuropsychology As A Way of Living

Uploaded by

Copyright:

Available Formats

!

La neuropsicologa como una forma de vida

Neuropsychology as a way of life

Septiembre 4, 2016 September 4th, 2016

0. Un viaje hacia lo desconocido 0. A trip to the unknown

1. Asimetra cerebral 1. Brain asymmetry

Ardila, A. (2010). A proposed re- Ardila, A. (2010). A proposed re-

3. Alexias y agrafias 3. Alexias and agraphias

4. Modelos cognitivos de las alexias 4. Cognitive models of alexias

6. Desarrollo de pruebas de diagnstico 6. Development of neuropsychological

7. Envejecimiento y demencia 7. Aging and dementia

8. Neuropsicologa infantil 8. Child neuropsychology

9. Problemas especficos de aprendizaje 9. Specific learning disabilities

11. Intoxicacin por cocana 11. Cocaine intoxication

12. Estandarizacin de pruebas 12. Standardization of assessment tests

13. Variables educacionales 13. Educational variables

14. Neuropsicologa transcultural 14. Cross-cultural neuropsychology

15. Sndromes neuropsicolgicos 15. Paroxysmal neuropsychological

16. Intoxicacin por escopolamina 16. Scopolamine intoxication

17. Inteligencia 17. Intelligence

Ardila, A. (1999). A neuropsychological Ardila, A. (1999). A neuropsychological

18. Sndrome disejecutivo 18. Dysexecutive syndrome

19. Neuropsicologa de la criminalidad 19. Neuropsychology of criminality

20. Origen de la cognicin humana 20. Origins of human cognition

Ardila, A. (2010). On the evolution of Ardila, A. (2010). On the evolution of

21. Evolucin del lenguaje 21. Language evolution

Ardila, A. (2015). A Proposed Neurological Ardila, A. (2015). A Proposed Neurological

22. Rehabilitacin neuropsicolgica 22. Neuropsychological rehabilitation

23. Bilingismo 23. Bilingualism

0. A trip to the unknown

0. Un viaje hacia lo desconocido When I was approximately 15 or 16 years

2. 1976-1980: Universidad Nacional de 2. 1976-1980: National University of

En Colombia me vincul a la Universidad Returning to Colombia I contacted the

3. 1981-1992: Neurological Institute of

4. 1992-1993: Universidad de San 4. 1992-1993: University of San

5. 1994-1998: Miami Institute of 5. 1994-1998: Miami Institute of

6. 1999-2007: Memorial Regional Hospital 6. 1999-2007: Memorial Regional Hospital

Mi vinculacin al Memoria Regional Hospital My time during the Memorial Regional

7. 2002-hasta el da de hoy: Florida 7. 2002- up-to-date: Florida International

En forma inesperada surgi la posibilidad de I was unexpectedly given the opportunity to

Durante mi vida profesional he trabajado en Throughout my career I have been involved

A continuacin, cinco fotografas corres- The following includes five pictures

International Conference of Neuropsychology, Bogota, 1981

Fundacin de la Sociedad Latinoamericana de Neuropsicologa (SLAN). Foundation of the

2nd International Conference Brain-Mind (Latin American Association of

Journal of Neurolinguistics 28 (2014) 6380

Contents lists available at ScienceDirect

Bilateral representation of language: A critical

Article history: It is well known that for right-handed individuals, language is

64 B. Bernal, A. Ardila / Journal of Neurolinguistics 28 (2014) 6380

2. Normal and anomalous right-sided lateralization of language

B. Bernal, A. Ardila / Journal of Neurolinguistics 28 (2014) 6380 65

3. Right hemisphere involvement in language across life span

66 B. Bernal, A. Ardila / Journal of Neurolinguistics 28 (2014) 6380

4. Bilateral brain representation of language

4.1. Wada test

B. Bernal, A. Ardila / Journal of Neurolinguistics 28 (2014) 6380

68 B. Bernal, A. Ardila / Journal of Neurolinguistics 28 (2014) 6380

group proposes four bilaterality sub-patterns: interhemispheric dissociation, double representation,

4.2. Modern neuroimaging studies

B. Bernal, A. Ardila / Journal of Neurolinguistics 28 (2014) 6380 69

4.3. Insights from tractography

5. Toward a synthesis of bilateral representation of language