Anne Shumway-Cook, Motor Control Theory and Practical Applications Williams & Wilkins (LK)

Motor Control
Theory AND Practical Applications
^s »» •
*
Anne Shumway-Cook
Marjorie WoollacoU
Digitized by the Internet Archive
in 2010
http://www.archive.org/details/motorcontroltheoOOshum
]
Motor Control
Theory and Practical Applications
Motor Control
Theory and Practical Applications
Anne Shumway-Cookj Ph.D.

Research Coordinator
Department of Physical Therapy
Northwest Hospital
Seattle, Washington
Marjorie H. Woollacott^ Ph.D.

Professor
Department of Exercise and Movement Science
Institute of Neuroscience
University' of Oregon
Eugene, Oregon
Williams & Wilkins

3AITIM0RE • PHIIADEIPHIA • HONG KONG
LONDON • MUNICH • SYDNEY • TOKYO
A WAVERLY COMPANY
Editor:John P. Butler
Development Editor: Nana- H. Evans
Copy Editor: Judith F. Minkovc
Desijiner: Wilmi E. Rosenberger
Illustration Planner: Ray Lowman
Production Coordinator: Charles E. Zeller
Photojjrapher: Daxid Trees
Cop\Tight © 1995
Wlliams & Wilkins
428 East Preston Street
Baltimore, Manland 21202, USA
book is protected by copyright. No part of this book may be

All rights reserved. This
reproduced anv form or by any means, including photocop\-ing, or utilized by any
in
information storage and retrieval system without written permission from the cop\nght
Printed in the United States of America
Library of Congress Cataloging in Publication Data
Shumway-Cook, Anne, 1947-

Motor control theory and practical
: applications / Anne Shumway-Cook, Marjoric H.
WooUacott. — 1st ed.
p. cm.
Includes index.
ISBN 0-683-07757-0
1. Physical therapy. 2. Motor learning. I. Woollacon, Marjoric H., 1946- II.
Title.
RM70I.S55 1995
612.7—dc20 94-26889
CIP
97 98 99
23456789 10
It is ivithjireat love and jfratittide that we dedicate this book to the many
people, incliidinff professional colleagues, reviewers, and patients, who have
contributed to the development of the ideas presented here. We gratefully
acknowledjje the divine source of our enthusiasm, wisdom, and joy. We
dedicate this book, as we do all our actions, to the One who set it before us
to do and provided us steadfast wisdom and support throufjhout its
creation.
Photographs by Da\id Trees, Education and Training Department, Northwest Hospital.

PREFACE
In recent years there has been a tremen- This first section leads into the major
dous interest among clinicians regarding new thrust of the book which addresses motor
theories of motor control and the role of these control issues as they relate to the control of
theories in guiding clinical practice. The ex- posture and balance (Section II), mobility'
plosion of new research in the field of neu- (Section III), and upper extremit)' manipula-
roscience and motor control has widened the tor}' functions (Section IV). The chapters in-
gap between research/theon' and clinical cluded in each of these sections follow a stan-
practices related to helping patients regain dard format. The first chapter discusses issues
motor control. This book is an attempt to related to normal control processes. The sec-
bridge the gap between theor\' and practice. ond (and in some cases third) chapter de-
The book stresses the scientific and experi- scribes age-related issues. The third chapter
mental basis of new motor control theories, presents research on abnormal fianction, while
and explains how principles fi-om this science the final chapter discusses the applications of
can be applied to clinical practice. While many current research to the assessment and treat-
theories ofmotor control are discussed, the ment of motor dyscontrol in each of the three
major thrust of the book is to present a sys- functional areas.
tems theory of motor control and a clinical We envision that this text will be of use
approach to assessment and treatment of mo- in both undergraduate and graduate courses
tor control problems based on a systems on normal motor control, motor develop-
model. We refer to this clinical approach as a ment across the life span, and rehabilitation in
"task oriented approach." The book is di- the areas of physical and occupational therapy
\ided into four sections. Section I, entided as well as kinesiolog}'.
"Theoretical Framework," reviews current Motor Control: Tljcory and Practical Ap-
theories of motor control, motor learning, plications seeks to provide a framework that
and recoven,' of function following neurolog- will enable the clinician to incorporate theory
ical insult. The clinical implications of various into practice. More importantiy it is our hope
theories of motor control are discussed. In ad- that this book will serve as a springboard for
dition, this section reviews the physiological developing new, more effective, approaches
basis of motor control and motor learning. to assessing and treating patients with motor
Finally, this section includes a chapter that dyscontrol.
presents a suggested conceptual framework
for clinical practice.
CONTENTS
Preface
Section I THEORETICAL FRAMEWORK

1. Theories of Motor Control 3
2. Motor Learning and Recovery of Function 23
3. Physiology of Motor Control 45
4. Physiological Basis of Motor Le.arning and
Reco\'ery of Function 85
5. A Conceptual Framework for Clinical
Practice 99
Section II POSTURE/BALANCE
6. Control of Posture and Balance 119
7. Development of Postural Control 143
8. Aging AND Postural Control 169
9. Abnoral\l Postur.\l Control 185
10. Assessment and Treatment of Patients wtth
Postural Disorders 207
Section III MOBILITY FUNCTIONS

11. Control of Normal Mobilit\' 239
12. A Life Span Perspectfv^ of Mobility' 269
13. Abnormal Mobility" 295
14. Assessment and Treatment of the Patient
WITH MoBiLiT\^ Disorders 315
Section IV UPPER EXTREMITY CONTROL

15. Upper Extremit\' M.\nipul.-\tion Skills 357
16. Upper Extremit\' Manipul.\tion Skills:
CH.ANGES Across the Life Sp.\n 377
17. Abnormal Upper Extremity' Mânipulation
Control 399
X Contents
18. Assessment and Treatment of the Patient

WITH Upper Extremity Manipulatory
Dyscontrol 417
Appendix A 447
Glossary 457
Index 463
Section I
THEORETICAL FRAMEWORK
Chapter 1
Theories of Motor Control

Introduction Hierarchical Theory
What Is Motor Control? Limitations
Study of Action Clinical Implications
- Study of Perception Motor Programming Theories
Study of Cognition Limitations
J Interaction of Individual, Task, and Clinical Implications
Environment Systems Theory
Why Should Clinicians Study Motor Limitations
Control? Clinical Implications
What Is a Theory of Motor Control? Dynamical Action Theory
< What Is the Relationship Between Theory Limitations
1 and Practice? Clinical Implications
1 Framework for Interpreting Behavior Parallel Distributed Processing Theory
t Guide for Clinical Action Limitations
I
New Ideas: Theories Are Dynamic and Clinical Implications
- Evolving Task-Oriented Theories
Working Hypotheses for Assessment and Limitations
Treatment Clinical Implications
Theories of Motor Control Ecological Theory
Reflex Theory Limitations
Limitations Clinical Implications
Clinical Implications Which Theory of Motor Control Is Best?
Summary
Introduction relation to specific actions or acti\ities. For ex-

ample, motor control physiologists might ask:
What Is Motor Control? how do people walk, run, talk, smile, reach,
or stand still.- Researchers often study move-

In this textbook we define motor con- ment control within the context of a specific
trol as the study of the nature and cause of
acti\iri', like walking, hoping that understand-
movement. \\Tien we talk about motor con-
ing control processes related to this acti\it\'
trol, we are actually talking about two issues.
will pro\ide insight into principles for how all
The first body
issue deals with stabilizing the
of movement is controlled. So the study of
in space, that motor control as it applies to
is,
motor control includes the study of action.
postural and balance control. The second is-
sue deals with mo\ing the body in space, that STUDY OF PERCEPTION
is, motor control as it applies to movement.
Thus, the term motor control is defined Unfortunately, the term motor control
broadl\- here to encompass the control of in itself is somewhat misleading, since move-
both movement and posture. ment arises from the interaction of multiple
processes, including perceptual, cognitive,
STUDY OF ACTION and motor processes. Perception is essential

to action, just as action is essential to percep-
Movement is often described within the tion. Actions are performed within the con-
context of accomplishing a particular action. text of an enNironment. Sensor\'-perceptual
As a result, motor control is usually studied in systems provide information about the body
4 Section I THEORETICAL FIL\MEWORK
and the enxironment, and are clearly integral

to the abilit)' to act effectively within an en-
Nironment (1). Thus, understanding motor
control requires the study oi perception.
STUDY OF COGNITION
In addition, since movement is not usu-
ally performed in the absence of intent, cog-
niti\'e processes are essential to motor control.
In this book we define cognitive processes
broadly to include attention, motivation, and
emotional aspects of motor control that un-
Figure 1.1. Motor control emerges from an interaction
derlie the establishment of intent or goals.
between the and the environment.
individual, the task,
Motor control includes perceptual and action
systems, which are organized to achieve spe-
cific goals or intents. Thus, the study of motor "applied motor control physiologists" (2).
control must include the study of cognitive Their actions are based on the belief that
processes as they relate to the control of per- movement control is important, even essen-
ception and action. tial, to the achievement of functional com-
petence. Sincemotor control is the study of
INTERACTION OF INDIVIDUAL, the nature and cause of movement, under-
TASK,AND ENT\TRONMENT standing motor control is essential to clinical
practice.
While each of these aspects of motor Understanding motor control is easier
control —
perception, action, and cognition said than done. This is not a
because there is
can be studied in isolation, we believe a true universal agreement among scientists or cli-
picture of the nature of motor control cannot nicians about the nature and cause of move-
be achieved without a synthesis of informa- ment. There is no single theon- of motor con-
tion fi-om all three. trol that evePi'one accepts. Among the many
However, motor control research that theories that will be discussed in this chapter,
focuses only on processes within individuals each has made specific contributions to the
without taking into account the environments field of motor control and each has implica-
in which they mo\e, or the tasks they are per- tions for the clinician retraining motor dys-
forming, produce an incomplete picture.
will control.
Accordingly, in this book our discussion of
motor control will focus on the interaction of What Is a Theory of
the individual, the task, and the environment. Motor Control?
Figure I.I illustrates this concept that move-
ment emerges from an interaction between A theor-y of motor control isa group of
these three factors. abstract ideas about the nature and cause of
movement. Theories are oft:en, but not al-
Why Should Clinicians Study ways, based on models of brain function.

Motor Control? What is a model? A model is a represen-
tation of something, usually a simplified ver-

Why should clinicians be concerned sion of the real thing. The better the model,
with the study of motor control? They spend the better it will predict how the real thing
a considerable amount of time retraining mo- will behave in a real situation. Wliy is a model
tor dyscontrol in patients who have fiinctional of brain fijnction needed? Because the brain
limitations. Clinicians have been referred to as is so complex, a model can represent and to
Chapter One THEORIES of Motor Control 5
some extent simplify complex concepts. A • new ideas; and

model of brain function, related to motor • working hypotheses for assessment and
control, is a simplified representation of the treatment.
structure and function of the brain as it relates
to the coordination of movement. Theories of FRAMEWORK FOR INTERPRETING

motor control and models of brain function BEHAVIOR
then go hand in hand.
The idea that there may be more than Theory can help therapists to interpret
one dieor\' of motor control may be a new the behavior or actions of patients they treat.
concept to many therapists. Different theories Theory allows the go beyond the
therapist to
of motor control reflect philosophically dif- behavior of one and broaden the
patient,
ferent views about how the brain controls application to a much larger number of
movement. These theories often reflect dif- cases (3).
ferences in opinion about the relative impor- Theories can be more or less helpfiil de-
components of move-
tance of various neural pending on their ability to predict or explain
ment. For example, some theories stress the behavior of an individual patient. WTien a
peripheral influences, others may stress central theor\' and its associated assumptions does
influences, while still others may stress the not provide an accurate interpretation of a pa-
role of information fi-om the environment in tient's behavior, it loses its usefulness to the
controlling behavior. Thus, motor control therapist. Thus, theories can potentially limit
theories are more than just an approach to ex- a therapist's abilit}' to observe and interpret
plaining action. Often they stress different as- movement problems in patients.
pects of the organization of the underlying For example, look at the patient pic-
neurophysiology and neuroanatomy of that tured in Figure 1.2. Mrs. Johnson is a 67-
action. Some motor control look
theories of year-old woman referred for rehabilitation
at the brain as a black box and simply study following a cerebral vascular accident, which
the rules by which this black box interacts has produced motor dyscontrol in her left
with changing environments. side. The patient habitually sits with her left:
arm and drawn close to her body.

flexed
What Is the Relationship Between When asked to extend her left: arm, she cannot
Theory and Practice? actively extend at the elbow. If you try to ex-
tend her arm, there is resistance. In addition,
Do theories really influence what ther-
when she walks, her knee is stiflF and hyper-
apists do with their patients.^ YES! Rehabili-
extended, and she uses a toe-heel pattern.
tation practices reflect the theories, or basic
Prior to deciding how to retrain arm
ideas, we have about the cause and nature of
function and gait, as her therapist, you must
function and dysfiinction (3). In general,
decide what the underlying problems are.
then, the actions of therapists are based on
What is preventing her from actively extend-
assumptions that are derived from theories.
ing her arm.' Why is she unable to walk with
The specific practices used to assess and treat
a heel-toe gait? You might assume the pa-
the patient with motor dyscontrol are deter-
tient's inability to extend her arm is the result
mined by underlying assumptions about the
of spasticity in the elbow flexors. In addition,
nature and cause of movement. Thus, motor
her inability to walk heel-toe is die result of
control theory is part of the theoretical basis
spasticity in the gastrocnemius muscle. This
for clinical practice.
assumption might be based on a theory of
What are the advantages and disadvan-
motor control that suggests that reflexes are
tages of using theories in clinical practice?
an important part of movement control, and
Theories provide:
that abnormal reflexes are a major reason pa-
• a tramework for interpreting behavior; tients cannot move normally. Based on this
• a guide for clinical action; theory, you might attribute the loss of arm
6 Section I THEORETICAL FPtÂIEWORK
is assumed to be a major determinant of ab-

normal function. As a result, numerous ap-
proaches have been developed to assess and
treat spasticit\' in the course of retraining
function. However, because there are many
different theories about the nature and cause
of movement, there are potentially many
other therapeutic approaches for retraining
motor dyscontrol.
NEW IDEAS: DYNAMIC AND

EVOLVING
Theories are dynamic, and change to re-
flect greater knowledge relating to the theor\'.

How does this affect clinical practices related
to retraining motor dyscontrol.' Changing
and expanding theories of motor control need
Figure 1.2. Mrs. Johnson is a 67-year-old woman re-
ferred for therapy because of a right cerebral vascular not be a source of fioistration to clinicians. Ex-
accident resulting in a left hemiparesis. Pictured is her panding theories can broaden and enrich the
habitual sitting posture. possibilities for clinical practice. New ideas for
the assessment and treatment of motor dys-
function, specifically the inability to acti\'el\-

control will e\'ol\'e to reflect new ideas about
extend the elbow, to be primarily the result of the nature and cause of movement.
spasticity, defined as a release of the stretch
reflex, in the elbow flexors. -WORKING HYPOTHESES FOR
Has your theoretical framework helped ASSESSMENT AND TREATMENT
you to correctiy interpret this patient's behav-
ior.' Only if this patient's problems are in fact
A theoPi' is not cfirecdy testable, since it
is abstract. Rather, theories generate hypoth-

solely the result of spasticity-. The theon- has
eses, which are testable. Information gained
not helped you as a cUnician if it has limited
through hvpotltesis testing is used to validate
your abilit\' to explore other possible expla-
or invalidate a theon,*. This same approach is
nations for your patient's behavior. What are
usefiil in clinical practice. So-called bypothesis-
some of the other factors that potentially im-
driveti clinical practice (4) transforms the
pair arm fianction in your stroke patient? Later
therapist into an active problem solver. Using
in this chapter we will discuss other theories
this approach to retrain motor dyscontrol calls
of motor control that will provide alternative
for the therapist to generate multiple hypoth-
explanations for loss of function.
eses (explanations) for why patients move (or
don't move) in ways to achieve fimctional in-
GUIDE FOR CLINICAL ACTION
dependence. During the course of therapy the
Theories pro\ide therapists with a pos- therapist will test various hypotheses, discard-
sible guide for action. Clinical practices de- ing some, and generating new explanations
signed to treat the patient with motor dys- that are more consistent with their results.
control are based on an understanding of the

nature and cause of normal movement, as well Theories of Motor
as an understanding of the basis for abnormal Control
movement. Therapeutic strategies aimed at
retraining motor control reflect this basic un- There is tremendous enthusiasm among
derstanding. In the above example, spasticit)' therapists for critically examining the models
Chapter One Theories of Motor Control 7
upon which much of clinical practice is based. Response Response

Stimulus '
(stimulus) (stimulus)
Therapists are recognizing the limitations of
past theories and the expanding possibilities Figure 1.4. Reflex chaining as a basis for action. A stim-
of new solutions based on new models of mo- which becomes the stimulus for
ulus leads to a response,
tor control and recoven,' of function. the next response, which becomes the stimulus for the
next response.
In this section we will re\iew theories of
motor control and explore some of their hm-
itations and possible clinical implications. It is
important to understand that all models are insequence, to achieve a common purpose
unified bv the desire to understand the nature (5).
and cause of movement. The difference is in Sherrington performed elegant experi-
the approach. It is not unlike the stor\- of the ments with cats, dogs, and monkeys to show
fi\e men trying to understand the nature and the existence of the reflex, and to carefiilly de-
hinction of an elephant. One carefiilly and scribe and define reflexes. The conception of
systematically studies the trunk, and learns ev- a reflex requires three different structures, as
er\thing there is to know about the nature shown in Figure 1 .3: a receptor, a conducting
and flinction of the trunk. Another studies the ner\'ous pathway, and an effector. The con-
nature and function of the feet; another, the ductor consists of at least X\\o nerve cells, one
tail. Each in his own way has provided essen- connected to the eflfector, the other con-
tial information about the elephant. How- nected to the receptor. The reflex arc then
ever, a true understanding about the nature consists of the receptor, the conductor, and
and hinction of an elephant is only possible the effector (6).
by combining information from all sources. In Sherrington went on to describe com-
this spirit, we approach the following section plex behavior in terms of compound reflexes,
on theories of motor control, their limita- and their successive combination or chaining
tions, and possible clinical applications. together. Sherrington gave the following ex-
ample of a frog capturing and eating a fly. Pic-
Reflex Theory ture Mr. Toad sitting in the sun on his lily
pad. i\long comes the fly; seeing the fly (stim-
Sir Charles Sherrington, a neurophysi- ulus) results in the reflex activation of the
ologist in the late 1800s and early 1900s, tongue darting out to capture the fly (re-
wrote the book Tlje Integrative Action of the sponse). If he is successfial, the contact of the
Nervous System in 1906. His research formed flv on the tongue causes reflex closure of the
the experimental foundation for a classic re- mouth, and closure of the mouth results in
flex theor\' of motor control. For Sherrington, reflex swallowing.
reflexes were the building blocks of com- Sherrington concluded that with the
plex beha\ior. Reflexes worked together, or whole nervous system intact, the reaction of
the various parts of that system, the simple
reflexes, are combined into greater actions,
which constitute the behavior of the indi\id-
ual as a whole. Figure 1.4 represents this con-
Receptor I
-<o- cept of reflex chaining. Sherrington's \iew of
Muscle/ a reflexive basis for movement persisted un-
effector
challenged for 50 years, and continues to in-
fluence thinking about motor control today.
Stimulus
Response
LIMITATIONS
Figure 1 .3. The basic structure of a reflex consists of a Because Sherrington looked primarily at
receptor, a conductor, and an effector. reflexes, and asked questions about the central
8 Section I THEORETICAL FRAMEWORK
nervous system (CNS) related to reflexes, he CLINICAL IMPLICATIONS

drew a picture of the CNS and motor control
that was skewed towards reflex control. There How might a reflex theory of motor
are a number of limitations of a reflex theor>' control be used to interpret a patient's behav-
of motor control (1). ior, and serve as a guide for the therapist's
The reflex cannot be considered the ba- actions?
sic unit of behavior if both spontaneous and

If chained or compounded reflexes are
voluntar}' movements are recognized as ac- the basis for ftincdonal movement, clinical
ceptable classes of behavior, since the reflex strategies designed to test reflexes should al-
must be activated by an outside agent. low therapists to predict function. In addi-
Another limitation of the reflex theor\' tion, a patient's movement behaviors would
of motor control is that it does not adequately be interpreted in terms of the presence or ab-
explain and predict movement that occurs in sence of controlling reflexes. Finalh', retrain-
the absence of a sensory stimulus. More re- ing motor control for functional skills would
cendy, it has been shown that animals can focus on enhancing or reducing the effect of
move in a relatively coordinated fashion in the various reflexes during motor tasks. Appl^dng
absence of sensory input (7). a reflex theor)' to interpreting motor dyscon-
Yer another limitation is that the theory trol was shown in our previous example of
does not explain fast movements, that is, se- Mrs. Johnson. Clinical strategies for improv-
quences of movements that occur too rapidly ing motor control using a reflex model would
to allow for sensory feedback from the pre- focus on methods to reduce fle.xor spasticity,
ceding movement to trigger the next. For ex- which should enhance normal movement ca-
ample, an experienced and proficient t>'pist pacity.
moves from one key to the next so rapidly that Despite the limitations in Sherrington's
there isn't time for sensory information from conclusions, many of his assumptions about
one keystroke to activate the next. how the CNS controls movement have been
An additional limitation is that the reflex reinforced and have influenced current clinical
chaining model fails to explain the fact that a practices.
single stimulus can result in varying responses

depending on context and descending com- Hierarchical Theory
mands. For example, there are times when we
need to override reflexes to achieve a goal. For Many researchers contributed to the
example, normally touching something hot \'iew that the ner\'ous sx'stem is organized as
results in the reflexive withdrawal of the hand. a hierarchy. Among them, Hughlings Jack-
However, if our child is in a fire, we may over- son, an English physician, argued that the
ride the reflexive withdrawal to pull the child brain has higher, middle, and lower levels of
out. control, equated with higher association ar-
does not explain
Finally, reflex chaining eas, the motor cortex and spinal levels of mo-
the ability produce novel movements.
to tor function (8).
Novel movements put together unique com- Hierarchical control in general has been
binations of stimuh and responses according defined as an organizational structure that is
to rules previously learned. A violinist, who top down. That is, each successively higher
has learned a piece on the violin, and also level exerts control over the level below it, as
knows the technique of playing the cello, can shown in Figure L5. In a strict vertical hier-
play that piece perfecdy on the cello without archy, lines of control do not cross and there
necessarily having practiced the piece on the is never bottom up control.
cello. The violinist has learned the rules for In the 1920s, Rudolf Magnus began to
playing the piece and has applied them to a explore the frinction of different reflexes
novel or new situation. within different parts of the nerx'ous system.

Chapter One THEORIES OF MOTOR CONTROL 9
A
A A
Top
within hierarchicalh' organized levels of the
CNS.
and Myrde
In the 1940s, Arnold GeseU (13, 14)
McGraw (15), two well-knowTi
developmental researchers, offered detailed
descriptions of the maturation of infants.
These researchers applied the current scien-
A AAA Down
tific thinking about reflex hierarchies of motor
control to explain the behaviors they saw in

infants. Normal motor development was at-
Figure 1.5. The hierarchical control structure is char-
tributed to increasing corticalization of the
acterized by a top-down structure, where higher centers
are always in charge of lower centers.
CNS resulting in the emergence of higher lev-
els of control o\er lower level reflexes. This
has been referred to as a nciiromaturatioiial

He found that reflexes controlled by lower thcoi-y of development. An example of this
levels of the neural hierarchy are only present model is illustrated in Figure 1.6. This theory-
when cortical centers are damaged. These re- assumes that CNS maturation is the primar%'
sults were later interpreted to imply that re- agent for change in development. It mini-
flexes are partof a hierarchy of motor control, mizes the importance of other factors such as
in which higher centers normally inhibit these musculoskeletal changes during develop-
lower reflex centers (9-10). ment.
Later, Georg Schaltenbrand (11) used Since Hughlings Jackson's original
the concepts developed by Magnus to explain work, a new concept of hierarchical control
the development of mobiliDi' in children and has evolved. Modem neuroscientists ha\e
adults. He described the development of hu- confirmed the importance of elements of hi-
man mobilit)' in terms of the appearance and erarchical organization in motor control. The
disappearance of a progression of reflexes. He concept of a strict hierarchy,where higher
went on hirther to say that patholog\" of the centers are always in control, has been modi-
brain may result in the persistence of primitive fied. Current concepts describing hierarchical
reflexes. He suggested that a complete un- control within the nervous system recognize
derstanding of all the reflexes would allow the the fact that each level of the ner\ous system
determination of the neural age of a child or can act upon other levels (higher and lower)
a patient. depending on the task. In addition, the role
In the late 1930s, Stephan Weisz (12) of reflexes in movement has been modified.
reported on reflex reactions that he felt were Reflexes are not considered the sole deter-
the basis for equilibrium in humans. He de- minant of motor control, but only one of
scribed the ontogeny of equilibrium reflexes many processes important to the generation
in the normally developing child and pro- and control of movement.
posed a relationship bersveen the maturation
of these reflexes and the child's capacity- to sit, LIMITATIONS
stand, and walk.
The results of these experiments and One of the limitations of a reflex/hier-
observations were dra\\n together and are of- archical theon,- of motor control is that it can-
ten referred to in the clinical literature as a not explain the dominance of reflex beha\ior
reflex/hierarchical theor\- of motor control. in certain situations in normal adults. For ex-
This reflex/hierarchical theon- of motor con- ample, stepping on a pin results in an imme-
trol combines reflex and hierarchical theories diate withdrawal of the leg. This is an example
into one. This theon,- suggests that motor of a of the hier-
reflex within the lowest lexel
control emerges from reflexes that are nested archv dominatine; motor fianction. It is an ex-
Neuroanatomical Postural reflex Motor

structures development development
Equilibrium Bipedal
reactions function
Righting Quadrupedal
reactions function
Brainstem
spinal cord Primitive Apedal
reflex
function
Figure 1.6. Neuromaturational theory of motor control attributes motor development to the maturation of neural
processes, including the progressive appearance and disappearance of reflexes.
ample of bottom-up control. Thus, one must to be major deterrents to normal motor
be cautious about assumptions that all low- control.
level behaviors are primiti\e, immature, and A number of treatment approaches have
nonadaptive, while all higher level (cortical) been developed which focus on enhancing or
behaviors are mature, adaptive, and appropri- reducing the efficacy of reflexes as an impor-
ate. tant step in retraining motor control. The
goal of treatment is to achieve greater fiinc-
CLINICAL UVIPLICATIONS tion through the modification of reflex action.
One of the difficulties in using a reflex ap-
Abnormalities of reflex organization proach to retraining motor control is that suc-
have been used by many clinicians to explain cessfiil modification of reflex activirs' is not al-
disordered motor control in the neurological ways mirrored in impro\'ements in functional
patient. Berta Bobath, an English physical skills. Part of the difficult)' may lie in the issue
therapist, in her discussions of abnormal pos- of focusing treatment on reactions instead of
tural reflex activity in children with cerebral preparing patients for action.
palsy, states that"the release of motor re-
sponses integrated at lower levels from re- Motor Programming Theories
straining influences of higher centers, espe-
cially that of the cortex, leads to abnormal More
current theories of motor control
postural reflex activity'" (16). have expanded our understanding of the
Based on a reflex/hierarchical theory of CNS. They have mo\'ed away from \iews of
motor control and development, a number of the CNS as a mosdy reactive system and have
reflex tests have been developed as part of the begun to explore the physiology of actions
clinical assessment of patients with neurolog- rather than the physiology,' of reactions.
ical impairments (17). These reflex assessment Reflex theories have been useful in ex-
profiles are used to estimate the level of neural plaining certain stereot\ped patterns of move-
maturation and predict flinctional abilit\'. In ment. However, an interesting way of viewing
addition, reflex profiles are used to document reflexes is to consider that one can remove the
the presence of persisting and dominating stimulus, or the atierent input, and still have
primitive and pathological reflexes believed a patterned motor response (IS). If we re-
move the motor response from its stimulus, A significant amount of research in the field
we are left with the concept of a central motor of psychology has supported the existence of
pattern. This concept of a motor pattern is hierarchically organized motor programs that
more flexible than the concept of a reflex be- store the rules for generating movements so
cause it can either be activated by sensor^' that we can perform the tasks with a varien,'
stimuli or by central processes. of effector systems.

A motor program theory' of motor con-
trol has considerable experimental support.
For example, experiments in the early 1960s
ACTIVE LEARNING MODULE
studied the grasshopper or locust and showed
wing beat in
that the timing of the animal's You can see this for yourself. Try
flight depended on a rh\thmic pattern gen- writing your signature as you normally
erator. Even when the sensor}' nerves were

would on a small piece of paper. Now
write it on a blackboard. Now try it with
larger,
cut, the ner\'Ous system by itself could gen-
your other hand. While you may be much more
erate the output with no sensory input; how-
proficient with one hand versus the other, you will
ever, the wing beat was slowed (20). This sug-
see elements of your signature that are common
gested that movement is possible in the
to all situations. As shown in Figure 1 .7, the rules
absence of reflexive action. Sensory input, for writing your name are stored as a motor pro-
while not essential in driving mo\'ement, has gram at higher levels within the CNS. As a result,
an important flmction modulating action.

in neural commands from these higher centers to
These conclusions were ftirther supported by write your name can be sent to various parts of the
work examining locomotion in cats (21 ). The body. Yet, elements of the written signature re-
results of these experiments showed that in main constant regardless of the part of the body
the cat, spinal neural networks could produce used to carry out the task (23).
a locomotor rhxthm without either sensor\'

inputs or descending patterns from the brain.
By changing the intensity' of stimulation to LIMITATIONS
the spinal cord, the animal could be made to
walk, trot, or gallop. Thus, it was again shown The concept of central pattern genera-
that reflexes do not drive action, but that cen- tors expanded our understanding of the role
tral pattern generators by themselves can gen- of the nerx'ous system in the control of move-
erate such complex movements as the walk, ment. However, we must be careful to realize
trot,and gallop. Further experiments showed that the central pattern generator concept has
the important modulator)' etfects of incoming never been intended to replace the concept of
sensory inputs on the central pattern genera- the importance of sensory input in controlling
tor (22). movement. It simply expanded our under-
These experiments led to the motor standing of the flexibilits' of the nerx'ous sys-
program theory of motor control. This term tem in creating movements, to include its
has been used in a number of ways by different abilit}' to create movements in isolation from
researchers, so care should be taken in deter- feedback.
mining how the term is being used. The term An important limitation of the motor
motor program may be used to identify' a program concept is that a central motor pro-
central pattern generator (CPG), that is, a gram cannot be considered to be the sole de-
specific neural circuit like that for generating terminant of action (23). Two identical com-
walking in the cat. In this case the term rep- mands to the elbow flexors, for example, will
resents neural connections that are stereo- produce very different movements depending
typed and hardwired. on whether your arm is resting at your side,
But the term motor program is also used or if you are holding your arm out in front of
to describe the higher level motor programs you. The forces of gravity will act differendy
that represent actions in more abstract terms. on the limb in the two conditions, and thus
12 Section I THEORETICAL FR.\.\iEWORK
Abstract
motor program
Synergy Sy-nergy Synergy
Right nana Right arm Left hand

muscles muscles rûSCleS
Figure 1.7. Levels of control for motor programs and their output systems. Rules for action are represented at the
highest level, in abstract motor pro-ams. Lower lev'els of the hierarchy contain information essential for effecting
action.
modift- the movement. In addition, if your In patients whose higher levels of motor
muscles axe fatigued, similar nenous s>-stem programming are affected, motor program
commands will give ver\" different results. theor\' suggests the importance of helping pa-
Thus, the motor program concept does not tients releam the correa rules for action. In
take into account the fact that the nervous s\'5- addition, treatment should focus on retrain-
tem must take into account both musculo- ing movements imp>ortant to a fiinctional
skeletal and emironmental variables in achiev- task, not just on reeducating specific muscles
ing movement control. in isolation.
CLINICAL LMPLICATIONS Systems Theory
Motor program theories of motor con- Even before motor program concepts
trol have allowed clinicians to move beyond a were developed, another researcher, Nicolai
reflex explanation for disordered motor con- Bernstein 1 1896-1966 1, a Russian scientist,
trol. Explanations for abnormal movement was looking at the ner\-ous system and body
have been expanded to include problems re- in a whole new way. Pre\ious neurophysiol-
sulting from abnormalities in central pattern ogists had focused primarily on neural control
generators, or in higher le\el motor pro- aspects of movement. Bernstein recognized
grams. that you cannot understand the neural control
Mrs. Johnson, our stroke patient, may of movement without an understanding of
indeed have flexor spasricin," in her arms wiiich the characteristics of the svstem you are mov-
may afJcct her abilin." to mo\'e. However, it ing, and the external and internal forces acting
will be important to determine wliat levels of on the body.
motor programming are involved. If her In describing the characteristics of the
higher levels of motor programming are not s^-sTc'tti being moved, he looked at the whole
affected, she will h>e able to continue to use body as amechanical s\'Stem, with mass, and
such programs as handwriting, but will find subject to both external forces, like gra\it\
alternate effectors, for example, her unaf- and internal forces, including both inertial
fected hand, to cany out the tasks. Of course, and movement-depvendent forces. During the
these less used lower level s\"nerg\' and mus- course of any movement the amounts of force
cular s\"Steras will have to be trained to carry acting on the body will change as potential
out these higher level programs. and kinetic energ) change. He thus showed
Chapter One THEORIES OF Motor Control 13
that the same central command could result Thus, Bernstein believed that SNTiergies
in quite different movements because of the play an important role in sohing the degrees
interplay between external forces and varia- of freedom problem. This is achieved by con-
tions in the initial conditions (23). For the straining certain muscles to work together as
same reasons, different commands could re- a unit. He hypothesized that though there are
sult in the same movement. tew s\Tiergies, make possible almost the
they
Bernstein also suggested that control of whole \"ariet\' of movements N\"e know. For ex-
integrated movement was probably distrib- ample, he considered some simple synergies
uted throughout many interacting s\"stems to be the locomotor, postural, and respiratory
working cooperatively to achieve movement. synergies.
This gave rise to the concept of a distributed
model of motor control. LIMITATIONS
Ho\%' does Bernstein's approach to mo-
tor control differ from the reflex, hierarchical, What are the limitations of Bernstein's
or motor program approaches presented pre- systems approach.- .\s you can see, it is the
\iously? Bernstein asked questions about the broadest of the approaches we have discussed
organism in a continuously changing situa- thus far, and since it takes into account not
tion. He found answers that were different only the contributions of the nervous system
from pre\ious researchers about the nature to action, but also the contributions of the
and cause of movement, since he asked dif- muscle and skeletal systems, as well as the
ferent questions, such as: How does the body forces of gra\it>' and inertia, it predicts actual
as a mechanical s\stem influence the control beha\ior much better than pre\ious theories.
process? How do the initial conditions affect However, as it is presented today, it does not
the properties of the movement? focus as hea\ily on the interaction of the or-
In describing the body as a mechanical ganism with the en\ironment, as do some
system, Bernstein noted that we have many other theories of motor control.
degrees of freedom that need to be con-
trolled. For example, we have many joints, all CLINICAL IMPLICATIONS
of which flex or extend and many of which
can be rotated as weU. This complicates move- The svstems theon,' has a number of im-
ment control incredibly. He said, "Coordi- plications for therapists. First, it stresses the
nation of movement is the process of master- importance of understanding the body as a
ing the redundant degrees of freedom of the mechanical system. Movement is not solely
mo\'ing organism" (23). In other words, it determined by the output of the nenous s\'s-
in\ olves converting the body into a control- tem, but is the output of the ner\ous system
lable system. as filtered through a mechanical system, the
As a solution to the decrees offreedom bodv. WTien working with the patient who
problem, Bernstein h\pothesized that hierar- has a central ner\"ous s\stem deficit, the ther-
chical control exists to sLmplif\' the control of apist must be carefiil to assess the contribu-
the body's multiple degrees of freedom. In tion of impairments in the musculoskeletal
this \\3\\ the higher levels of the ner\'Ous sys- s\stem, as well as the neural system, to overall
tem activate lower levels. The lower levels ac- lossof motor control.
tivate synergies, or groups of muscles that are In our example of Mrs. Johnson, the
constrained to act together as a unit. We long-term loss of mobilin.- in her arm and leg
can think of our movement repertoire like will potentiaUy affect the musculoskeletal sys-
sentences made up of many words. The tem. She may show shortening of the elbow
letters N\ithin the words are the muscles; flexors and loss of range of motion at the an-
the words themselves are the s\Tiergies, kle joint. These musculoskeletal limitations
and the sentences are the actions them- will have a significant effect on her abilin.' to
selves. recoser motor control.
The systems theon- suggests that assess- What is nonlinear beha\iorr It is a situation in
ment and treatment must focus not only on which, as one parameter is altered and reaches
the impairments within indi\idual systems a critical value, the system goes into a whole
contributing to motor control, but the effect new behavior pattern. For example, as an an-
of interacting impairments among multiple imal walks faster and faster, there is a point at
systems. A good example of this in Mrs. John- which, suddenly, it shifts into a trot. As the
son is the interacting impairments in the mus- animal continues to move faster, there is a sec-
culoskeletal and neuromuscular systems that ond point at which it shifts into a gallop. This
constrain her abilit}' to move her arm. is shown in Figure 1.8.
The dynamical action approach does

Dynamical Action Theory not seek to explain these shifts in terms of the
nervous system circuitry', but instead simplv
The dynamical action theor\' approach attempts to describe mathematically the ftinc-
to motor control has begun to look at the tion of these systems. This allows the predic-
moving person from a new perspective (24- tion of the ways that a given system will act in
26). The perspective comes from the broader different situations. One of the points that
study of dynamics or s\'nergetics within the proponents of this perspective put forth is that
physical world, and asks the questions: How- many body movement transitions may be ex-
do the patterns and organization we see in the plainable without invoking a specific neural
world come into being from their orderless pattern generator to cause the transition. The
constituent parts? And, how do these systems transitions instead may be due to the oscilla-
change o\er time.' For example, we ha\'e tor\' or pendulum-like properties of the limbs
thousands of muscle cells in the heart that themselves. Thus, the dynamical action per-
work together to make the heart beat. How- spective has deemphasized the notion of com-
ls this system of thousands of degrees of free- mands from the central ner\'ous system in
dom (each cell we add contributes a new de- controlling mo\-ement and has sought physi-
gree of freedom to the system reduced to one )
cal explanations that may contribute to move-
of few- degrees of freedom, so that all the cells ment characteristics as well (28).
ftinction as a unitr The dynamical action theon,' has re-
This phenomenon, which we see not centiy been modified to incorporate many of
only in heart muscle, but in the patterns of Bernstein's concepts. This has resulted in the
cloud formations and the patterns of move- blending of these two theories of motor con-
ment of water as it goes from ice to liquid to trol into a dynamical systems model (24).
boiling to e\-aporation, illustrates the principle This model suggests that movement under-
of self-organization, which is a ftindamental lying action results from the interaction of
dynamical systems principle. It says that when both physical and neural components (29).
a system of indi\idual parts comes together,
its elements behave collectively in an ordered LIMITATIONS
way. There is no need for a "higher" center
issuing instructions or commands to achieve This approach has added to our under-
coordinated action. This principle applied to standing of the elements contributing to
motor control suggests that movement movement itself, and serves as a reminder that
emerges as a result of interacting elements, understanding the ner\'Ous system in isolation
without the need for specific commands, or will not allow the prediction of mo\-ement.
motor programs within the nervous s\'Stem. However, a limitation of this model can be
The dynamical action or synergetics per- the presumption that the ner\'ous system has
spective also tries to find mathematical descrip- a fairly unimportant role, and that the rela-
tions of these self-organizing systems. Critical tionship berw een the physical system of the
features that are examined are what are called animal and the environment in which it op-
the nonlinear properties of the system (27). erates primarily determines the animal's be-
Chapter One THEORIES of Motor Control 1
Velocity
Gallop
Behavioral
state
Figure 1.8. A dynamical action model predicts discrete changes in behavior resulting from changes in the linear
dynamics of a moving system. For example, as velocity increases linearly, a threshold is reached that results in a change
in behavioral state of the moving animal from a walk, to a trot, and a gallop.
havior. The focus of the dynamical action the- therefore can help a weak patient move with
on' is usualh' at the level of this interface, not greater ease.
at understanding the neural contributions to In our example of Mrs. Johnson, mov-
the system. ing slowly may not be the best strategy tor
getting from sit to stand, if weakness is a pri-
t^CLINICAL IMPLICATIONS mar\' impairment. Instead, teaching her to in-
crease the speed of tnmk motion may allow
One of the major implications of the dy- her to generate sufficient momentum to suc-
namical action theon,' is the view that move- ceed in standing.
ment is an emergent propertv. That is, it
emerges from the interaction of multiple el- Parallel Distributed Processing

ements that self-organize based on certain dy- Theory
namical properties of the elements them-
selves. This means that shifts or alterations in The parallel distributeci processing
movement behavior can often be explained in ( PDF) theoPi' of motor control cHescribes ho\\-
terms of physical principles rather than nec- the nervous system processes information for
essarily in terms of neural structures. action. This theon,' hasbeen used to explain
What are the implications of this for how we acquire newsince it makes pre-
skills,
treating motor dyscontrol in patients? If as cli- dictions about the processes used by the ner-
nicians we understood more about the phys- vous system during tiie development or ac-
ical or dynamical properties of the human quisition of new skills (30).
body, we could make use of these properties The PDP theor\' is consistent with cur-
in helping patients to regain motor control. rent knowledge in neurophysiolog\' that the
For example, velocit\- can be an important nerx'ous system operates both through serial
contributor to the tiynamics of movement. processing, that is, processing information
Often, patients are askeci to move slowh' in an through a single pathway, and through par-
effort to move safely. Yet, this approach to allel processing, that is, processing informa-
retraining fails to take into account the inter- tion through multiple pathways that process
action between speed and phvsical properties the same information simultaneously in dif-
of the body, which produce momentum, and ferent ways (31).
Figure 1.9. Parallel distributed processing

model showing three layers, the input, hidden,
Input
and output layers, hypothetically equivalent to
units
sensory, interneuron, and motor units.
Hidden units
Output units
Scientists have begun to model neural ference between actual and desired activit)' is
processing using computer programs. These called the error. The error is used to modify
programs have been de\'eloped with sophis- the connections among those elements that
ticated circuitn,' similar to brain networks. have produced the error.
This is how the modeling isdone: Models The process is run over and o\er, sim-
consist of elements that are hooked together ulating the repetition of a task performed
in circuits. Like neuronal synapses, each ele- again and again. With this activity', the system
ment can be affected in a positive or negative self-corrects until it solves the output prob-
way by the other elements. Also, like neuronal lem.
synaptic transmission, each element can have The model has correctiy predicted pro-
different magnitudes of either positive or neg- cesses in both perceptual and action systems.
ative effect on the next element. Each element For example, a PDF has been used to simulate
then summates all the incoming positive and the processing of \isual stimuli underlying the
negative inputs. These models have been abilit)' to recognize and identify letters. In ad-
made into layered networks containing input dition, the models have been used to predict
elements, intermediate processing layers how we calculate the correct joint angles as-
called hidden layers, and output elements. sociated with mo\ing a limb to a particular
This is shown in Figure L9. These layers are position in space (31 ).
equivalent to sensor\' neurons, interneurons, PDF is somewhat unique in its emphasis

and motor neurons (30). on explaining neural mechanisms associated
Just as in the nerxous system, the effi- with motor control. This theon,' and its re-
ciency of performance in this system depends lated models are of great interest right now
on two factors. The first is the pattern of con- because, though they are not exact replicas of
nections betxveen the layers, and the second the ner\ous system, they have many of the
is the strength of individual connections. The properties that are also seen in the nenous
beaut\' of this model is that the researcher can system. Thus, they may help us understand
determine the most efficient connections to how the ner\'ous system solves particular
perform a particular ffinction through a tech- movement problems.
nique called back propagation. Through the
process of back propagation, the most effi- LIMITATIONS
cient output from the "motor neuron" layer
is determined. It starts with a random set of This theor\' is not intended to be an ex-
inputs to the system. The system then calcu- act replica of the nervous system, and there-
lates the difference bet^veen the desired and fore many of its fiinctions, such as back prop-
the actual activity- of the output unit. The dif- agation, do not mimic ner\ous system
Chapter One THEORIES OF MOTOR CONTROL 17
processing of information during perfor- theory' of tasks. By tasks, Greene was referring
mance and learning. to the fundamental problems that the CNS
was required to sohe in order to accomplish
CLINICAL IMPLICATIONS motor tasks. According to Greene, an exam-
ple of a fundamental task inherent in motor
The PDP theor\' is relatively new, and control is the degrees of freedom problem de-
thus its clinical applications are relatively un- scribed by Bernstein.
known. There are several ways that PDP mod- According to Greene, a theor\' of tasks
els could be integrated into clinical practice. would help neuroscientists find obser%'able
A PDP model could be used to predict behaviors to measure that are relevant to the
how injur)' within the ner\ous system affects tasks the brain is called upon to perform.
flinction. The theon,' predicts that because of Thus, an understanding of motor control re-
the availability' of parallel redundant path- quires more than an understanding of circuits.
ways, the loss of just a few elements uill not It requires an understanding of the underly-
necessarily affect function. However, the the- ing problems the CNS is required to solve in
ory' might predict that once a certain level or orcier to accomplish motor tasks. A task-ori-
threshold is attained, the loss of additional el- ented approach to the study of motor control
ements will affect the capacity' of the system would provide the basis for a more coherent
to function. This concept of a threshold for picture of the motor system. Greene suggests
dysfunction can be seen in many cases of pa- that once the essentials of a task have been
thology'. For example, in Parkinson's disease organized into a becomes
coherent picture, it
there is a gradual loss of cells in the basal gan- possible to know and understand more.
less
glia. Clinical symptoms may not be apparent An adaptation of Greene's theor\' of

initially, until the number of neurons lost tasks has been elaborated by Gordon (33) and
reaches a critical threshold. Horak (34). The task-oriented approach pre-
Redundant pathways suggest the possi- sented by Gordon and Horak, however, de-
bility' of multiple roads to recover)'; thus, the fines task from a more fiinctional perspective.
theor\' could be used to suggest approaches That is, what control issues are inherent in the
to retraining motor dyscontrol. It suggests accomplishment of functional tasks in mean-
that recover}' might be best when rehabilita- ingful environments.' The task-oriented ap-
tion training is applied to multiple pathways. proach is based on the recognition that the
For example, Mrs. Johnson's rehabilitation goal of motor control is the control of move-
program might include both voluntan,' acti- ment to accomplish a particular task, not the
vation of the gastrocnemius muscle to help elaboration of movement for the sake of mov-
improve muscle strength, but also practice, ing alone (except in unusual cases such as
using that muscle in postural and locomotor dance). The task-oriented approach assumes
tasks. that of movement is organized
control
around goal-directed fiinctional behaviors
Task- Oriented Theories such as walking or talking.
In the last 50 years, a tremendous LIMITATIONS

amount of information on the basic structure
of the CNS has emerged from neuroscience A limitation of a task-oriented theor\' of
research. But there is still the recognition that motor control is a lack of consistent agree-
we know a lot but understand ver\' littie. That ment about what the fundamental tasks of the
is, we know much about neural circuitn', but CNS are. In addition, scientists don't always
little about how the neurons operate together agree on what the essential elements being
to achieve flinction. Peter Greene (32), a the- controlled within a task are. For example,
oretical biologist, suggested that what was some scientists studying postural control be-
needed in the field of motor control was a lieve that controlling head position is the es-
sential goal of the postural system. However,

other scientists studying postural control be-
lieve that controlling center of mass position
to achie\e bod\' stability is the essential goal
of postural control.
CLINICAL IMPLICATIONS
The most significant implication of a

task-oriented theor>' of motor control is the
concept that motor retraining needs to focus
on essential functional tasks. It suggests the
importance of understanding the role of per- ^^WronM^^'
and action systems in ac-
ceptual, cognitive,
Figure 1.10. Ecological approach stresses the interac-
complishing these tasks. One of the chal- tion between the individual and the environment. The
lenges for clinicians is to analyze essential individual actively explores the environment, which in
components of ever^'day tasks we are called turn supports the individual's actions.
upon to retrain. This requires more than an

understanding of the biomechanical features
of the task, that is, the motor strategies used ecological approach to motor control. It sug-
to accomplish the task. It also requires un- gests that motor control e\olved so that ani-
derstanding the perceptual basis for action, mals could cope with the en\ironment around
and the cognitive contributions to action. them, moving in it effectively in order to find
In our example of Mrs. Johnson, what food, run away from predators, build shelter,
are the essential tasks that will be retrained and even play (28). What is new about this
during the course of her reco\'enir How will approach? It was really the first time that re-
these tasks be retrained? How much time will searchers began focusing on how actions are
the clinician spend on retraining fiinction, as geared to the environment. Actions require
opposed to working on some of the essential perceptual information that is specific to a de-
elements contributing to hinction, such as sired goal-directed action performed within a
strength and range of motion.' How can the specific emironment. The organization of ac-
clinician ensure that tasks learned in a clinical tion is specific to the task and the environ-
setting will be retained when Mrs. Johnson ment in which the task is being performed.
finally returns to her own home.' WTiereas many pre\ious researchers had
seen the organism as a senson'-motor system,
Ecological Theory Gibson stressed that it was not sensation per
se thatwas important to the animal, but per-
In the 1960s, independent of the reception. Specifically, what is needed is the per-
search in physiolog)', a ps\-chologist named ception of environmental factors important to
James Gibson was beginning to explore the the task. He stated that perception focuses on
way in which our motor systems allow us to detecting information in the environment
interact most elfecti\ely with the en\'ironment that will support the actions necessar\' to
in order to perform goal-oriented behavior achieve the goal. From an ecological perspec-
(35). His research focused on how we detect tive, it is important to determine how an or-
information in our en\'ironment that is rele- ganism detects information in the environ-
vant to our actions, and how we use this in- ment that is rele\'ant to action, what form this
formation to control our movements (see Fig. information takes, and how this information
1.10). is used to modifx- and control movement (28 ).
This view was expanded b\- the students In summan,-, the ecological perspective
of Gibson (36, 37) and became known as the has broadened our understanding of ner\ous
system function from that of a sensor\'-motor plete theori' of motor control, the one that
system, reacting to emironmental \ariables, really predicts the nature and cause of moye-
to that of a perception-action system which ment and is consistent with our current
acti\eh' explores the enyironment to satisfy its knowledge of brain anatomy and ph\'siolog\'.'
own goals. As you no doubt can already see, there
is no one theor\' that has it all. We belieye the
LIMITATIONS best theon,' of motor control is one that com-

bines elements from all of the presented the-
^•Mthough the ecological approach has ories. A comprehensiye, or integrated, theor\-
expanded our knowledge significandy con- recognizes the elements of motor control we
cerning the interaction of the organism and do know about and leayes room for the iJiings
the enyironment, it has tended to giye less we don't. Any current theor\' of motor con-
emphasis to the organization and function of trol is in a sense unfinished, since there must
the ner\'ous system, which led to this inter- alwa\'s be room to re\"ise and incorporate new
action. Thus, the research emphasis has information.
shifted from the nenous system to the organ- Many people ha\e been working to de-
ism-emironment interface. velop an integrated theor\' of motor control
(24, 34, 38-43). In some cases, as theories
CLINIC\L IMPLICATIONS are modified, new names As a re-
are applied.
sult, becomes difficult to distinguish among
it
A major contribution of this yiew is in eyohing theories. For example, systems, dy-
describing the indiyidual as an actiye explorer namical, dynamical action, and dynamical ac-
of the enyironment. The actiye exploration of tion systems are all terms that are often used
the task and the enyironment in which the interchangeably.
task is performed allows the indiyidual to de- In pre\ious articles we (40, 42 haye )
yelop multiple ways to accomplish a task. called the theon,' of motor control on which
AdaptabilitN'is important not only in the way
we base our research and clinical practice a
we organize moxements to accomplish a task, s^'stcms approach. We ha\e continued to use
but also in the way we use our senses during this name, though our concept of Ji'^rfw.? the-
action. or\' differs from Bernstein's systems dieor\-
An important part of treatment is help- and has eyoKed to incorporate many of the
ing Mrs. Johnson explore the possibilities for concepts proposed by other theories of motor
achieying a functional task in multiple ways. control. In tliis book we will continue to refer
The ability to deyelop multiple adaptiye so- to our theon,' of motor control as a systems
lutions to accomplishing a task requires that approach. This approach argues that it is crit-
the patient explore a range of possible ways to ical to recognize that movement emerges
accomplish a task, and discoyer the best so- from an interaction betvyeen the indiyidual,
lution for them, giyen the patient's set of lim- the task, and the enyironment in which the
itations. In Mrs. Johnson's case, this abilin,' to task is being carried out. Thus, movement is
actiyely discoyer a range of solutions is ham- not solely the result of muscle-specific motor
pered by a reduced abilip,- to moye, inaccurate programs, or stereoti,ped reflexes, but results
perceptions, and possible cognitiye limita- from dynamic interplay bet\\'een perceptual,
a
tions. cognitive, and action systems.
Action systems are defined here to in-
Which Theory of Motor clude both the neuromuscular aspects and the
Control Is Best? physical or dynamic properties of the muscu-
loskeletal system itself The organizational
So \\hich motor control theor\- best properties of the system emerge as a fiinction
suits the current theoretical and practice of the task and the environment in which the
needs of therapists? Which is themost com- task is performed.
This theoretical framework will be used 2. Brooks VB. The neural basis of motor con-
throughout this text, and is the basis for clin- trol. New York: Oxford University Press,
ical methods for assessing and treating motor 1990.

3. Shepard K. Theory: criteria, importance and
dyscontrol in the patient with neurological
impact. In: Contemporary management of
problems. We have found the theory usefril in
motor control problems: proceedings of the
helping us to generate research questions and
II Step Conference. Alexandria, VA: APTA,
hypotheses about the nature and cause of
1991:5-10.
movement. 4. Rothstein JM, Echternach JL. Hypothesis-
oriented algorithm for clinicians: a method
and treatment planning. Phys
for evaluation
Summary Ther 1986;66:1388-1394.
5. Sherrington, C. The integrative action of the
1 The study of motor control is the study of the
nervous system. 2nd ed. New Haven: Yale
nature and cause of movement. It deals with
University' Press, 1947.
both stabilizing the body in space, that is, pos-
6. Gallistel, CR. The organization of action: a
tural and balance control, and with moving
new synthesis. Hillsdale, NJ: Lawrence Erl-
the body in space.
baum, 1980.
2. The specific practices used to assess and treat
7. Taub E, Herman AJ. Movement and learning
the patient with motor dyscontrol are deter-
in the absence of sensor)' feedback. In: Freed-
mined by underlying assumptions about the
man SJ, ed. The neurophysiology of spatially
nature and cause of movement that come
oriented behavior. Homewood: Dorsey
from specific theories of motor control.
Press, 1968:173-192.
3. Atheory of motor control is a group of abstract
8. Foerster O. The motor cortex in man in the
ideas about the nature and cause of move-
light of Hughlings Jackson's doctrines. In:
ment. Theories provide: (a) a framework for
interpreting behavior; (b) a guide for clinical
Payton OD, Hirt S, Newman, R, eds. Scien-
tific bases for neurophysiologic approaches to
action; (c) new ideas; and {d) working hy-
therapeutic exercise. Philadelphia: FA Davis,
potheses for assessment and treatment.
1977:13-18.
4. Rehabilitation practices reflect the theories, or
9. Magnus R. Animal posture (Croonian lec-
basic ideas, we have
about the cause and na-
ture). Proc Roy Soc London 1925;98:339.
ture of function and dysfunction.
5. In this chapter we have reviewed many motor
10. Magnus R. Some results of studies in the
physiology of posture. Lancet 1926;2:531-
control theories that have an impact on our
585.
perspective regarding assessment and treat-
11. Schaltenbrand G. The development of hu-
ment, including the reflex theory, hierarchical
theory, motor programming theories, systems
man motility and motor disturbances. Arch
Neurol Pyschiatr 1928;20:720-730.
theory, dynamical action theory, parallel dis-
tributed processing theory, task-oriented the-
12. Weisz S. Studies in equilibrium reaction. J
ories, and ecological theory. Nen' Ment Dis 1938;88:150-162.
13. Gesell A, Amatruda CS. Developmental di-
6. In this text we use our systems theory ap-
agnosis. 2nd ed. New York: Paul B. Hoeber,
proach as the foundation for many clinical ap-
plications. According to this theory, move- 1947.
ment arises from the interaction of multiple 14. Gesell A. Behaviorpatternsof fetal-infant and
processes, including (a) perceptual, cognitive, child. Genetics. Proceedings of the Associa-
and motor processes within the individual, tion for Research in Nervous and Mental Dis-
and (h) interactions between the individual, ease 1954;33:114.
the task, and the environment. 15. McGraw M. Neuromuscular maturation of

the human infant. New York: Hafner Press,
1945.
16. Bobath B. Abnormal postural reflex activity
References
caused by brain lesions. London: Heine-
1. Rosenbaum D. Human motor control. New mann, 1965:8.
York: Academic Press, 1991. 17. Fiorentino M. Reflex testing methods for
evaluating CNS Development. Springfield, tor systems. In: Rosen R, Snell FM, eds.
IL: Charles C Thomas, 1963. Progress in theoretical biology. San Diego:
19. van Sant AP. Concepts of neural organization Academic Press, 1972:304-338.
and movement. In: Connolly BH, Mont- 33. Gordon J. Assumptions underlying physical
gomery PC, eds. Therapeutic exercise in de- therapy inter\'ention: theoretical and histori-
velopmental disabilities. Chattanooga, TN: cal perspectives. In: Carr JH, Shepherd RB,
Chattanooga Corp, 1987T-8. Gordon J, et al., eds. Movement sciences:
20. Wilson DM. The central ner\'Ous control of foundations for physical therapy in rehabili-
flight in a locust. J Exp Biol 1961;38:471- tation. Rockville, Md: Aspen Publishers,
490. 1987:1-30.
21. Grillner S. Control of locomotion in bipeds, 34. Horak F. Assumptions underlying motor
tetrapods and fish. In: Geiger SR, ed. Hand- control for neurologic rehabilitation. In:
book of physiology, vol 2. Bethesda, MD: Contemporary management of motor con-

American Physiological Society, 1981:1179- trol problems. Proceedings of the II Step
1236. Conference. Alexandria, VA: APTA, 1992:
22. Forssberg H, Grillner S, Rossignol S. Phase 11-28.
dependent reflex reversal during walking in 35. Gibson, JJ. The senses considered as percep-
chronic spinal cats. Brain Res 1975;85:103- tual systems. Boston: Houghton Mifflin,
107. 1966.
23. Bernstein, N. The coordination and regula- 36. Reed ES. An outline of a theory of action sys-
tion of movement. London: Pergamon Press, tems. Journal of Motor Behavior. 1982;14:
1967. 98-134.
24. Thelen E, Kelso JAS, Fogel A. Self-organiz- 37. Lee DN. The functions of vision. In: Pick H,
ing systems and infant motor development. Saltzman E, eds. Modes of perceiving and
Developmental Review 1987;7:39-65. processing information. Hillsdale, NJ: Erl-
25. Kamm K, Thelen E, Jensen J. A dynamical sys- baum, 1978.
tems approach to motor development: In: 38. Mulder T, Geurts A. Recover)' of motor skill
Rothstein JM,ed. Movement science. Alexan- following nervous system disorders: a be-
dria, VA: APTA Association, 1991: 1-23.1 havioral emphasis. Clinical Neurology. In
26. Kelso JAS, Tuller B. A dynamical basis for ac- press.
tion systems. In: Gazanniga MS, ed. Hand- 39. Patla A. The neural control of locomotion.
book of cognitive neuroscience. NY: Plenum In: Spivack BS, ed. Mobility and gait. In
Press, 1984:321-356. press.
27. Kugler PN, Turvey MT. Information, natural 40. Woollacott M, Shumway-Cook A. Changes
law and self assembly of rhythmic movement. in posture control across the life span — a sys-
Hillsdale, NJ: Erlbaum, 1987. tems approach. Phys Ther 1990;70:799-
28. Schmidt R. Motor and action perspectives on 807.
motor behaviour. In: Meijer OG, Roth K, 41. Shumway-Cook A. Equilibrium deficits in
eds. Complex movement behavior: the mo- children. In: Woollacott M, Shumway-Cook
tor-action controversy. Amsterdam: Elsevier, A, eds. Development of posture and gait
1988:3^4. across the life span. Columbia, SC: Univ of
29. CrutchfieldCA, Hcriza CB, Herdman S. SC Press, 1989: 229-252,
Motor control. Morgantown, WV: Stokes- 42. Woollacott M, Shumway-Cook A, Williams
ville Publishers, in press. H. The development of posture and balance
30. Rumclhart DE, McCelland JL, eds. Parallel control. In: Woollacott MH, Shumway-Cook
distributed processing, explorations in the A, eds. Development of posture and gait
microstructure of cognition, vol1: Founda- across the life span. Columbia, SC: Univ of
Cambridge, Mass: MIT Press, 1986.

tions. SC Press, 1989:77-96.
31. Kandel E, Schwartz JH, JessellTM, eds. Prin- 43. Horak F, Shumway-Cook A. Clinical impli-
ciples of neuroscience. 3rd ed. New York: El- cations of postural control research. In: Dun-
1991:420^39,
sevier, can P, ed. Balance. Alexandria, VA: APTA,
32. Green PH. Problems of organization of mo- 1990.
Chapter 2
Motor Learning and Recovery

OF Function
J Introduction to Motor Learning and Therapy

What Is Motor Learning?
Feedback
Extrinsic
Knowledge of Results
Early Definitions of Motor Learning Practice Conditions
Broadening the Definition of Motor Massed and Distributed Practice
Learning Variable Practice
Relating Performance and Learning Contextual Interference
Forms of Learning Whole Vs. Part Training
Nonassociative Forms of Learning Transfer
Associative Forms of Learning Mental Practice
Classical Conditioning Guidance
Operant Conditioning Recovery of Function
Procedural and Declarative Concepts Related to Recovery of Function
Learning Function
Theories Related to Skilled Learning Recovery
Adam's Closed Loop Theory' Recovery Vs. Compensation
Limitations Sparing of Function
Schmidt's Schema Theory Stages of Recovery
Limitations Factors Contributing to Recovery of
Fitts and Posner: Stages of Motor Learning Function
Limitations Effect of Age on Recovery
m Newell's Theory of Learning as Exploration Quality of the Lesion and Recovery
Limitations Effect of Experience on Recovery
Factors Contributing to Motor Learning Effect of Training on Recovery
Feedback Clinical Overvievi'
Intrinsic Feedback Summary
,. - ••v:..'-'iMV,.V'
Introduction to Motor What Is Motor Learning?

Learning
In Chapter 1, we deftned the field of
Mr. Smith has been receiving iJierapy motor control as the stud>'of the nature and
for 5 weeks now, following his stroke. He has cause of movement. We define the field of
gradually regained the ability to stand, walk, motor learning as the study of the acquisi-
and feed himself again. What is the cause of tion and or modification of movement. While
Mr. Smith's recover^' of motor fianction.' How motor control focuses on understanding the
much due to "spontaneous recover)'"?
is control of movement already acquired, motor
How much of his recover,' may be attributed learning focuses on understanding the acqui-
to therapeutic inter\'entions.* How many of sition and or modification of movement.
his re-acquired motor skills will he be able to The field of motor learning has tradi-
retain and use when he leaves the rehabilita- tionally referred to the study of the acquisi-
tion facilit}' and returns home.^ These question or modification of movement in normal
tions and issues reflect the importance of mo- subjects. In contrast, recovery of function
tor learning to clinicians involved in retraining has referred to the re-acquisition of move-
the patient with motor control problems. ment skills lost through injur\'.
23
While there is nothing inherent in the strategies for sensing as well as moving. Thus,
term motor learning to distinguish it from motor learning, like motor control, emerges
processes involved in the recovery of move- from a complex of perception-cognition-ac-
ment function, the two
thought of
are often tion processes.
as separate. This separation between recovery Previous views of motor learning have
of function and motor learning may be mis- focused primarily on changes in the individ-
leading. Issues facing clinicians concerned ual. But the process of motor learning can be
with helping patients reacquire skills lost as described as the search for a task solution that
the result of injury are similar to those faced emerges from an interaction of the individual
by people in the field of motor learning. with the task and the environment. Task so-
Questions common to both include: how can lutions are new strategies for perceiving and
I best structure practice (therapy) to ensure acting (2).
learning.^ How can I ensure that skills learned Similarly, the recovery of fiinction in-
in one context transfer to others? Will simpli- volves the reorganization of both perceptual
fying a task, that is making it easier to per- and action systems in relationship to specific
form, result in more efficient learning? tasks and environments. Thus, one cannot
In this chapter we use the term motor study motor learning or recovery of function
learning to encompass both the acquisition outside of the context of how individuals are
and reacquisition of movement. We will de- solving flanctional tasks in specific environ-
scribe various theories of motor learning and ments.
recover)' of fimction.
Relating Performance and
EARLY DEFINITIONS OF MOTOR Learning
LEARNING
Traditionally, the study of motor learn-
Learning has been described as the pro- ing has focused solely on motor outcomes.
cess of acquiring knowledge about the world; Earlier views of motor learning did not always
motor learning has been described as a set of distinguish it from performance (3). Changes
processes associated with practice or experi- in performance that resulted from practice
ence leading to relatively permanent changes were usually thought to reflect changes in
in the capability for producing skilled action learning. However, this view failed to con-
(1). This definition of motor learning reflects sider that certain practice effects improved
four concepts: (I) learning is a process of ac- performance initiallv but were not necessarily
quiring the capabilit}' for skilled action; (2) retained, a condition of learning. This led to
learning results from experience or practice; the notion that learning could not be evalu-
(3) learning cannot be measured direcdy in- — ated during practice, but rather during spe-
stead, it is inferred based on behavior; and (4) cific retention or transfer tests. Thus, learning,
learning produces relatively permanent defined as a relatively permanent change, has
changes in behavior, thus short-term altera- been distinguished from performance, de-
tions are not thought of as learning (I). fined as a temporar)' change in motor behav-
ior seen during practice sessions.
BROADENING THE DEFINITION OF We \'iew the term performance from a
MOTOR LEARNING slighdy different perspective. Performance is
behavior observed at any specific moment in
In this chapter the definition of motor time, and not limited to describing behaviors
learning has been expanded to encompass obscA'cd during practice sessions. Perfor-
many aspects not traditionally consicHereci part mance, whether observed during practice ses-
of motor learning. sions or during retention and transfer tasks, is
Motor learning involves more than mo- the result of a complex interaction among
tor processes. Rather, it involves learning new many variables, one of which is the level of
Chapter Two Motor Learning and Recovery of Function 25
learning. Some other variables that may affect in patients with certain tj'pes of vestibular dys-
performance include fatigue,and anxiet)', fiinction. Patients are asked to repeatedly
motivation. Thus, performance, regardless of move in ways that provoke their dizziness.
when it is measured, is not necessarily a mea- This repetition results in habituation of the
sure of absolute learning. This is because dizziness response. Habituation forms the ba-
changes in performance can reflect not only sis of therapy for children who are termed
changes in learning, but changes in other vari- "tactile defensive," that is, who show exces-
ables as well. sive responsiveness to cutaneous stimulation.
Children are repeatedly exposed to gradually
Forms of Learning increasing levels of cutaneous inputs in an ef-
fort to decrease their sensitivit^' to this stim-

The recovery of function following in- ulus.
jury involves the reacquisition of complex Sensitization is an increased respon-
tasks. However, it is ciifficult to understand siveness following a threatening or noxious
the processes involved in learning using the stimulus (4). For example, if I receive a pain-
study of complex tasks. Therefore, many re- fiil stimulus on the skin, and then a light
searchers have begun by exploring simple touch, 1 v\'ill react more strongly than normal
forms of learning, with the understanding to the light touch. After a person has habit-
that these simple forms of learning are the ba- uated to one stimulus, another painful stim-
sis for the acquisition of skilled behavior. ulus can dishabituate the first. That is, sensi-
However, there is very litde information tization counteracts the effects of habitua-
about how these simple forms of learning tion.
contribute to the acquisition of more complex There are times when increasing a pa-
skills. tient's sensitivit)' to a threatening stimulus is
We begin by reviewing these simple important. For example, increasing a patient's

forms of learning and discussing some of their awareness of stimuli indicating a likelihood
clinical applications. We then consider theo- for impending falls might be an important as-
ries of motor learning that have been devel- pect of balance retraining.
oped to describe the acquisition of skilled be- Not all nonassociati\'e forms of learning
havior and suggest how each might be used are simple. SensoPi' learning, where you form
to explain the acquisition of the skill such as a sensory experience, is an example of non-
reaching for a glass of water. At the outset, we associative learning. It is learning that relates
provide a review of simple nonassociative to understanding about a stimulus, in this
forms of learning such as habituation and sen- case, the sensor}' inputs. Helping patients to
sitizadon. explore their perceptual space, as it relates to
learning a particular skill — like reaching or
Nonassociative Forms of Learning transferring —would be an example of non-
associative learning.
Nonassociative Icarninjj occurs when
animals are given a single stimulus repeatedlv. Associative Forms of Learning
As a result, the ner\'ous system learns about
die characteristics of that stimulus. Habitua- What is associative learninjj} One pos-
tion and sensitization are two very simple sible answer is that it involves the association
forms of nonassociative learning (4). Habit- of ideas. For example, if you tell your patients
uation is a decrease in responsiveness that oc- who are having problems walking to try to
curs as a result of repeated exposure to a non- of gravit}' with
associate shifting their center
painful stimulus. you are helping them com-
lifting their leg,
Habituation is useci in many different bine two aspects of a movement into one in-
ways in the clinical setting. For example, ha- tegrated whole. It is through associative
bituation exercises arc used to treat dizziness learning that a person learns to predict rela-
tionships, either relationships of one stimulus Before learning
to another (classical conditioning) or the re- CS No response

lationship of one's behavior to a consequence
(operant conditioning).
A UCS UCR
It has been suggested that associative
learning has evolved to help animals learn to After learning
detect causal relationships in the environment
B CS CR (formerly called UCR)
(4). Establishing lawhiland therefore predic-
tive relationships among
events is part of the Figure 2.1. The process of classical conditioning is di-
process of making sense and order of our agrammed showing the relationship between the con-
ditioned stimulus (CS), unconditioned stimulus (UCS),
world. Recognizing key relationships between
conditioned response (CR), and unconditioned response
events is an essential part of the ability to
(UCR) before learning (A) and during the course of learn-
adapt behavior to novel situations (4). ing (B).
Patients who have suffered an injury

that has drastically altered their abilit>' to sense
and move about their world have the task of tually begin to make the movement with only
reexploring their body in relationship to their the verbal cue.
world, in order to determine what new rela- Thus, we see them
as patients gain skills
tionships exist between the two. move along continuum of assistance from
the
Pavlov studied how humans and animals hands-on assistance from the therapist, to per-
learn the association of tvvo stimuli, through forming the task with verbal cues, and even-
the simple form of learning that is now called tually to performing the action unassisted.
classical conditioning. It shown that we gen-
has recently been
erally learn relationships that are relevant to
CLASSICAL CONDITIONING our survival; it is more difficult to associate
biologically meaningless events. These find-

Classical conditioning consists of learn- ings underscore an important learning prin-
ing to pair two stimuli. During classical con- ciple: the brain is most likely to perceive and
ditioning an weak stimulus (the con-
initially integrate aspects of the environment that are
ditioned stimulus) becomes highly effective in most pertinent. With regard to therapy, learn-
producing a response when it becomes asso- ing in patientsis most likely to occur in tasks
ciated with another stronger stimulus (the un- and environments that are relevant and mean-
conditioned sdmulus). The conditioned stim- ingftil to them.
ulus is usually something that initially
produces no response (like a bell). In contrast, OPERANT CONDITIONING

the unconditioned stimulus (UCS), which
could be food, always produces a response. Operant or instrumental conditioning is
After repeated pairing of the conditioned and a second t\'pe of associational learning (4). It
the unconditioned stimulus, one begins to see is basically trial and error learning. During op-
a conditioned response (CR) to the condi- erant conditioning we learn to associate a

tioned stimulus (CS). Remember — it origi- certain response, from among many that we
nally produced no response (4). This relation- have made, with a consequence. The classic
ship isshown in Figure 2.1. experiments in this area were done with ani-
What the subject is doing in this type of mals who were given food rewards whenever
learning is to predict relationships bet\veen they randomly pressed a lever inside their
two stimuli or events that have occurred and cages. They soon learned to associate the lever
to respond accordingly. For example, in a press with the presentation of food, and lever
therapy setting, if we repeatedly give patients pressing frequency became ver)' high.
a verbal cue in conjunction with giving them The principle of operant conditioning
aid in making a movement, they may even- could be stated as follows: beha\'iors that are
rewarded tend to be repeated at the cost of mance of the task that was practiced. Proce-
other behaviors. And likewise, behaviors fol- dural learning does not depend on awareness,
lowed bv aversive stimuli are not usually re- attention, or other higher cognitive processes.
peated. This has been called the law of effect During motor skill acquisition, repeating a
(4). movement continuously under \'ar\ing cir-
Operant conditioning plays a major role cumstances would topically lead to procedural
in determining the behaviors shown by pa- learning. That is, one automatically learns the
tients referred for therapy. For example, the movement itself, or the rules for mo\ing,
frail elderlv person who leaves her home to go called a movement schema.
shopping and experiences a fall is less likely to For example, when teaching a patient to
repeat that activit}- again. A decrease in activ- transfer ft-om chair to bed, we ofi:en have the
it\' results in declining physical functions, patient practice an optimal movement strat-
which in turn increases the likelihood she will eg>' to move fi-om one to the other. To better
tall. This increased likelihood for falls will re- prepare patients to transfer effecti\ely in a
inforce her desire to be inactive, and on it wide variety' of situations and contexts, pa-
goes, showing the law of effect in action. tients learn to move fi-om chairs of differing
Therapists may make use of a variet\' of tech- heights, and at different positions relative to
niques to assist this patient in regaining her the bed. They thus begin to form the rules
activir\' level and in reducing her likelihood of The de\'el-
associated with the task of transfer.
falling. One technique is to use the process of opment of rules for transferring will allow
desensitization to decrease her anxien,- and them to safely transfer in unfamiliar circum-
fear of falling. stances. This constant practice and repetition
Operant conditioning can be an effec- results in efficient procedural learning and ef-
tive tool during clinical intervention. Verbal fective and safe transfers.
praise by a therapist for a job well done serves On the other hand, declarative learn-
as a reinforcer for some (though not all!) pa- ing results in knowledge that can be con-
tients. Setting up a therapy session so that a sciously recalled and thus requires processes
particular mo\'ement is rewarded by the suc- such as awareness, attention, and reflection
cessfiilaccomplishment of a task desired by (4). Declarative learning can be expressed in
the patient is a powerftil example of operant declarative sentences, like: first 1 button the
conditioning. Using biofeedback to help a pa- top button, then the next one. Constant rep-
tient learn to control the foot during the etition can transform declarative into proce-
swing portion of gait is also an example of dural knowledge. For example, when patients
operant conditioning. are first relearning a skill, they may \erbally
describe each movement as they do it. How-
PROCEDURAL AND DECLARATIVE ever, with repetition, the movement becomes
LEARNING an automatic motor activit}', that is, one that
does not require conscious attention and
Some researchers have begun to classih,' monitoring.
associative learning based on the t\'pe of The advantage of declarative learning is
knowledge that is acquired by the learner. Us- that itcan be shown in other forms than it
ing this t\pe of classification, two t\'pes of was learned. So, for example, expert ski racers,
learning have been identified based on the when preparing to race down a slalom hill at
D,pe and recall of information learned (4). 120 miles an hour, rehearse in their minds the
Procedural learning refers to learning race and how they will run it. So, too, figure
tasks that can be performed automatically skaters preparing to perform will often men-
without attention or conscious thought, like tally practice the sequences to be skated prior
a habit. Procedural learning develops slowly to getting on the ice.
through repetition of an act over many trials, In therapv, when helping patients reac-
and is expressed through improved perfor- quire skills lost through injur\', the emphasis
is often on practices leading to procedural tended movement (6). This internal reference
learning, learning a movement, rather than on of correctness, which Adams calleti a percep-
declarative learning. Declarative learning re- tual trace, is built up over a period of practice.
quires the abilit>' to verbally express the pro- Adams predicted that the perceptual
cess to be performed and is often not possible traceby itself could not lead to the accurate
with patients who have both cognidve and production of skilled movement. He pro-
language deficits that impair their ability' to posed that a second trace, the memory trace,
recall and express knowledge. Teaching is used to select and initiate the movement
movement skills declaratively would, how- ( 1 ). After movement is initiated by the mem-
ever, allow patients to rehearse their move- ory' trace, he proposed that the perceptual
ments mentally, increasing the amount of trace would take over to carry out the move-
practice available to them when physical con- ment and detect error.
ditions such as fatigue would normally limit According to Adams' theory, when
it. learning to pick up a glass, you would grad-
ually develop a perceptual trace for the move-
Theories Related to ment, which would serve as a guide toward
your endpoint. The more you practiced the
Skilled Learning
specific movement, the stronger the percep-
Just as there are theories of motor con- tual trace would become. The accuracy of the
trol, there are theories of motor learning, that movement would be directiy proportional to
is,a group of abstract ideas about the nature the strength of the perceptual trace.
and cause of the acquisition or modification

of movement. Theories of motor learning,
LIMITATIONS
like theories of motor control, must be based Adams' closed-loop theory of motor
on current knowledge regarding the structure learning has been criticized for several rea-
and ftmction of the ner\'ous system. The fol- sons. It has been shown that animals and hu-
lowing section reviews current theories of mans can make movements even when they
motor learning. Included in this section is a
have no sensor\' feedback (7-9). In addition,
brief discussion of several theories related to animals are capable of conditioned learning in
recovery of function, the reacquisition of skills order to avoid shocks, even after deafteren-
lost through injur\'. tation (7). Thus, Adams' closed-loop theor\'
could not explain these open-loop move-

Adams' Closed-Loop Theory ments, that is, movements made in the ab-
sence of sensoPi' feedback.

Adams (5), a researcher in physical ed-
ucation, was the first person to attempt to cre- Schmidt's Schema Theory
ate acomprehensive theory of motor learning.
This theor\' generated a lot of interest during In the 1970s, in response to many of the
the 1970s as researchers attempted to deter- limitations of the closed-loop theor\' of motor
mine its applicability to motor skill acquisi- learning, Richard Schmidt, another researcher
tion. from the field of physical education, proposed
The most important aspect of the theory a new motor learning theor*', which he called
is it is founded on the concept of closed-

that the schema theory. It emphasized open-loop
loop processes in motor control. In a closed- control processes and the motor program
loop process sensory feedback is used for the concept (10). Though the concept of motor
ongoing production of skilled movement. programs was considered essential to under-
This theory hypothesizes that, in motor learn- standing motor control, no one had yet ad-
ing, sensor\' feedback from the ongoing dressed the question of how motor programs
movement is compared within the nervous can be learned. As other researchers before
system with the stored memory of the in- him, Schmidt proposed that motor programs
do not contain the specifics of movements, recognition schema is primarily used for
but instead contain generalized rules for a learning rather than on-line control.
specific class of movements. He predicted that Wlien the movement is over, the error
when learning a new motor program, the in- signal is fed back into the schema and the
dividual learns a generalized set of rules that schema is modified as a result of the sensory
can be applied to a variet}' of contexts. feedback and KR. Thus, according to this the-
At the heart of this motor learning theory', learning consists of the ongoing process
ory is the concept of schema, which has been of updating the recognition and recall sche-
important in psychology' for many years. The mas with each movement that is made.
term schema originallv referred to an abstract According to schema theory, when
representation stored in memory' following learning to reach for a glass, you would op-
multiple presentations of a class of objects. timally practice many variations on the task
For e.xample, after seeing many different t\'pes itself This would allow you to develop a set
of ciogs, it is proposed that we begin to store of rules for reaching, which you would then
an abstract set of rules for general dog quali- applv when reaching for the glass in front of
ties in our brain, so that whenever we see a you. The better your rules for reaching, the
new dog, no matter what size, color, or shape, more optimal would be your strategy for pick-
we can identify' it as a dog. ing up an unfamiliar glass, and the less likely
Schmidt expanded the concept of that you would drop the glass or spill the milk.
schema and applied it to the area of motor
control. He proposed that, after an individual LIMITATIONS
makes a movement, four things are stored in
memory: (a) the initial movement conditions, Is schema theory supported by research?
such as the position of the body and the Yes and no. One of the predictions of schema
weight of the object manipulated; (b) the pa- theor)' is that when practicing a skill, variable
rameters used in the generalized motor pro- forms of practice produce the most eflfec-
will
gram; (
c) the outcome of the movement, in tive schema or motor program. Research to
terms of knowledge of results (KR); and (d) test this prediction has used the following par-
the sensor}' consequences of the movement, adigms. Two groups of subjects are trained in
that is, how it felt, looked, and sounded. This a new task, one given constant practice con-
information is abstracted and stored in the ditions and the other given variable practice
form of a recall (motor) schema and a rec- conditions. Both groups are then tested on a
ognition (sensor)') schema, as you see in Fig- new, but similar movement. According to
ure 2.2. schema theor}', the second group should
The recall schema is used for the selec- show higher level performance than the first,
tion of a specific response. When making a because they have developed a broad set of
given movement, the initial conditions and rules about the task, which should allow them
desired goal of the movement are inputs to to apply the rules to a new situation. On the
the recall schema. Other inputs are the ab- other hand, the group should have de-
first
stract memor\' of previous response specifi- veloped a very narrow schema with limited
cations in similar tasks. rules that would not be easily applicable to
The recognition schema is used for the new situations.
evaluation of the response. In this case the on normal adults, the support
In studies
sensory consequences and outcomes of pre- is mixed. Many
studies show large effects of
vious movements are coupled with the cur- variable practice, while some studies show
rent initial conditions to create a representa- very small effects or no effect at all. However,
tion of the expected sensory consequences. with regard to studies in children, there has
This is then compared to the senson,' infor- been strong support. For example, 7- and 9-
mation from the ongoing movement in order \'ear-old children were trained to toss bean
to evaluate the efficiency of the response. The bags over a variable distance or a fixed dis-
Figure 2.2. Diagram ot Schmidt's schema theory, illustrating the critical elements in the acquisition of movement.
EXP PFB = expected proprioceptive feedback; EXP EFB = expected exteroceptive feedback. (Adapted from Schmidt
RA. A schema theory of discrete motor skill learning. Psychol Rev 1975;82:225-260.)
tance. When asked to throw at a new distance, Another limitation of the theor\' is that
the variable practice group produced signifi- it lacks specificit\'. Because of its generalized
candy better scores than the fixed practice nature, there are few recognizable mecha-
group (11). VVhy might there be differences nisms that can be tested. Thus, it is not clear
between children and adults in these experi- how schema processing itself interacts with
ments.' It has been suggested that it may be other systems to learn mo\ement and how it
difficult to find experimental tasks at \\hich aids to control that movement.

adults don't already have significant variable Another challenge to the schema theon,'
practice during normal activities, while chil- has been its inability' to account for the im-
dren, with much less experience, are more na- mediate acquisition of new U'pes of coordi-
ive subjects (12). Therefore, the experiments nation or new forms of movement. For ex-
may be more valid in children. ample, researchers have shown that if all of a
Chapter Two Motor Le.\rning .^nd Reconêiq" of Fc^nction 31
two f>airs are re-

centif>ede's limbs except for
moved, the centipede will immediateh" pro-
duce a quadrupedal gait 13 It has been ar-( 1.
gued that findings such as these cannot be

accovmted tor by schema theon." (2 >.
Fitts and Posner: Stages of Motor

Learning
Fitts and Posner 14 , two researchers Cognitive Associative Ajtooomous
from the field of ps\cholog\', described a the- 33ges of Learning

OT}- of motor learning related to the stages
that people go through in learning a new skill. Figure 23. The chai^ng anenbanal demands associ-
ated with the thfee stages of motor sidll acquisition out-
They suggest there are three main phases in-
lined by Fitts and Fôsner.
vohed in skill learning. In the first stage the
learner is concerned with understanding the
nature of the task, developing strategies that been described as the autonomous stage.
could be used to cany out the task, and de- Fitts and Posner define this stage by the au-
termining how the task should be evaluated. tomaridt)' of the skill, and the low degree of
These eflbrts require a high degree of cogni- attention required for its performance, as
tive acti%it\' such as attention. Accordingly, showTi in Figure 2.3. Thus, in this stage the
this stage is referred to as the cognitive stage person can begin to devote his or her atten-
of learning. tion to other aspects of the skill in general,
In this stage the person experiments like scanning the en\iromnent for obstacles
with a variety- of strategies, abandoning those that might impede performance, or one may
that don't work w hile keeping those that do. choose to focus on a secondary- task ( like talk-
Performance tends to be quite variable, per- ing to a friend w hile pertbrming the task i, or
haps because many strategies are being sam- save one's energ\, so that one does not be-
pled tor fjerfbrming the task. However, im- come fatigued.
provements in f>ertbrmance are also quite Using this theor\- of motor learning we
large in this first stage, p>erhaps as a result of would learn to reach tbr a glass in the follow-
selecting the most elective strateg\- tor the ing way. Your first experience of using the
task. glass w ould require a great deal of attention
The second stage in skill acquisition is and conscious thought. You might make a lot
described by Fitts and Posner as the associa- of errors and spill a lot of water, while you
tive stage. By this time the person has se- experimented with different movement strat-
lected the best strategy" for the task and now egies to accomplish the task. ^Mien mo\ing
begins to refine the skill. Thus, during this into the second stage, however, the move-
stage there is less variability' in f»ertbrmance, ment toward the glass would be refined and
and improvement also occurs more slowi)-. It \"ouwould use an optimal strategy At this .
is proposed that \ erbal-cognirive aspects of

point the task wouldn't require your full at-
learning are not as important at this stage be- tention. In the third autonomous stage, you
cause the f)erson focuses more on refining a w ould be able to reach for the glass while ear-
particular partem rather than on selecting ning on a conversation or being engaged in
among alteman\"e strategies i 1 >. This stage other tasks.
may last fi-om da« to weeks or months, de-
pending on the f>ertbrmer and the intensity of LL\UTATIONS
practice. This stage is equivalent to the motor
stage described by Adams. Schmidt ( 1 • notes that ver\' Utrie re-
The third stage of skill acquisition has search has been focused on the autonomous
Figure 2.4. A diagram showing the process pro- Exploration of perceptual and motor workspaces
posed by Newell of exploring the sensory and mo-
torworkspace (A) in order to find optimal solutions
to movement tasks (B).
. Perceptual
A workspace
Mapping perceptual-motor workspaces to

create optimal solutions
stage of learning, partly because it would take dination between perception and action in a
months or years to bring man\' subjects to this way consistent with the task and environmen-
skill level on a laborator\' task. Thus, he states tal constraints. What does he mean b\- this.'
that the principles that govern motor learning He proposes that, during practice, there is a
in this stage are largeh- unknown. search for optimal strategies to solve the task,
given the constraints. Part of the search for
Newell's Theory of Learning as optimal strategies involves finding the most
Exploration appropriate perceptual cues and motor re-
sponses for the task. Thus, the perception and

Karl Newell drew hea\il\' from both sys- action systems can be considered to be incor-
tems and ecological motor control theories to porated or mapped into an optimal task so-
create a theory of motor skill acquisition lution.
based on search strategies (2 ). In the previous Critical to the search for optimal strat-
learning theories proposed by Adams and egies is the exploration of the perceptual-mo-
Schmidt, practice produced a cumulative con- tor workspace. This process of exploring the
tinuous change in behavior due to a gradual sensor)' and motor workspace in order to find
buildup of the strength of motor programs. optimal solutions is shown diagrammatically
It was proposed that, with practice, a more in Figure 2.4. Newell believes that one usefiil
appropriate representation of action is devel- outcome of his theor\' will be the impetus to
oped. identifi" critical perceptual variables essendal
In contrast. New ell suggests that motor to optimal task-relevant soludons. These crit-
learning is a process that increases the coor- ical variables will be usetiil in designing search
Chapter Two Motor Le.\rning .vnd Recox'ery of Function 33
strategies that produce efficient mapping of cues to find the information we need to op-
perceptual information and mo\ement pa- timally solve the task problem.
rameters. This idea of a search of the workspace is
Newell believes that perceptual infor- similar to the concept of discoven- learning,
mation has a number of roles in motor learn- yet discoven,' learning concepts do not address
ing. In a prescriptive role, perceptual infor- die best ways for the learner to channel the
mation of
relates to understantiing the goal search through the workspace.
the and the movements to be learned.
task, Newel! discusses ways to augment skill
This information has t\pically been given to learning. The first is to help the learner un-
learners through demonstrations. derstand the nature of the perceptual motor
Another role of perceptual information workspace. The second is to understand the
is as feedback^ both during the movement natural search strategies used by performers in
(concurrent feedback, sometimes called exploring space. And the third is that of pro-
knowledge of performance) and on comple- viding augmented information to facilitate
tion of the movement (knowledge of results). the search.
Finally, he proposes that perceptual informa- One central prediction of this theor\' is
tion can be used to structure the search for a of motor skills will be depen-
that the transfer
perceptual-motor solution that is appropriate dent on the similarity' between the two tasks
for the demands of the task. Thus, in this ap- of the optimal perceptual -motor strategies
proach, motor learning is characterized bv op- and relatively independent of the muscles
timal task-rele\'ant mapping of perception and used or the objects manipulated in the task.
action, not by a rale-based representation of In summan', this new approach to mo-
action. tor learning emphasizes of
skill as a reflection
In NewelFs approach, during the course a dynamic exploraton' activir\', involved in
of learning to reach for a glass, repeated prac- mapping the perceptual-motor workspace to
tice with reaching for a variet\' of glasses that create optimal strategies for performing a
contain a variet\' of substances within them, task.
results in learning to match the appropriate
movement dynamics for the task of reaching. LIMITATIONS
But in addition, we learn to distinguish what
characteristics of the task we need to know to
This is a ven,' new theon-. One of its ma-
organize our actions. Such characteristics as
jor limitations is that it has yet to be applied
the size of the glass, how slipper\' the surface
to specific examples of motor skill acquisition
is, how full it is, are essential perceptual
in any systematic way. As a result, it is an un-
cues that help us develop optimal movement tested theon'.
strategies for grasping any variation of glas-
ses.
Various sensor)' cues help us to create

Factors Contributing to
optimal motor strategies. If a perceptual cue Motor Learning
suggests a hea\-A' glass, we grasp with more
force. If the glass is full, we modulate our Ver\' often therapists ask themselves
speed and trajector\- to accommodate the sit- questions like: Is the tv'pe of feedback that I
uation. If we lack preciseness in these sensor\' am giving to my patients concerning die qual-
cues, we can still create a motor strategy', but ity' of their movements really effective? Could
it might be less than optimal. That is, the fluid I form of feedback that might
give a different
may spill, or the glass may slip. Knowledge be better.* Should I give feedback with ever\'
about the critical perceptual cues associated trial that the patient makes, or would it be
with a task is essential in dealing with a new- better to withhold feedback occasionally and
variation of the task. When faced with a no\el make the patients tr\' to discern bv themselves
variant, we acti\'ely explore the perceptual if their movement is accurate or efficient?
What is the best timing for feedback? In the rently with the taskand in addition, at the end
following section we discuss research in motor of the which case it is called terminal
task, in
learning that has attempted to answer these feedback. An example of concurrent feedback
questions. We review the research in relation would be verbal or manual guidance to the
to the different motor learning factors that are hand of a patient learning to reach for objects.
important to consider when retraining pa- An example of terminal feedback would be
tients with motor control problems, including telling a patient after a first unsuccessful at-
feedback, practice conditions, and variabilit\- tempt to rise from a chair, to push harder the
of practice. next time, using the arms to create more force
to stand up (1 ).
Feedback
KNOWLEDGE OF RESULTS
We have already discussed the impor-
tance of feedback in relation to motor learn- Knowledge of results (KR) is one im-
ing. Clearly, some form of feedback is essen- portant form of extrinsic feedback. It has been
tial for learning to take place. In the following defined as terminal feedback about the out-
section we describe the t\'pes of feedback that come of the movement, in terms of the mov-
are available to the performer, and the con- ement's goal ( 1 ). This is in contrast to knowl-
tributions of these different t)pes of feedback edge of performance KP), which ( is feedback
to motor learning. relating to the movement pattern used to
The broadest definition of feedback in- achieve the goal.
cludes all the sensory information that is avail- Research has been performed to deter-
able as the result of a movement that a person mine the tvpes of feedback that are the best
has produced. This is t\'pically called re- to give a subject. Almost all of the research
sponse-produced feedback (I). This feed- that has been performed involves studying the
back is usually further divided into two sub- efficacy of different t\'pes of knowledge of re-
classes, that of intrinsic feedback and sults. Typically, research has shown that
extrinsic feedback. knowledge of results is an important learning
variable, that is, it is important for learning
INTRINSIC FEEDBACK motor tasks (15). However, there are certain
tv'pes of tasks where intrinsic feedback (for ex-
Intrinsic feedback is feedback that ample, visual or kinaesthetic ) is sufficient to
comes to the individual simply through the provide most error information, and KR has
various sensory systems as a result of the nor- only minimal For example, in learn-
effects.
mal production of the movement. This ining tracking tasks KR only minimally im-
cludes such things as visual information con- proves the performance and learning of a
cerning whether a movement was accurate, as subject (1).
well as somatosensory information concern- It has also been shown that KR is a per-
ing the position of the limbs as one was mov- formance variable, that is, it has temporary' ef-
mg(l). fects on the abilit>' of the subject to perform
a task. This may be due to motivational or
EXTRINSIC FEEDBACK alerting effects on the performer, as well as
guidance effects (that is, it tells the subject

E.xtrinsic feedback is information that how to perform the task better in the next
supplements For example,
intrinsic feedback. trial).
when you tell a patient that he/she needs to When should KR be given for optimal
pick up his/her foot higher to clear an object results.* Should it be given right after a move-
while walking, you are offering extrinsic feed- ment? What delay is best before the next
back. movement is made, to ensure maximum
Extrinsic feedback can be given concur- learning efficiency? Should KR be given after
Chapter Two Motor LEARNING and Rec;overy of Function 35
every movement? These are important ques- These results suggest that summary KR
tions for the therapist who wants to optimize is the best feedback, but if thiswere so, group
the learning or relearning of motor skills in 3 should have been as good as group 2, and
patients with motor disorders. this was not the case. It has thus been con-
Experiments attempting to determine cluded that immediate KR is detrimental to

the optimum KR delay interval have found learning, because it provides too much infor-
very little effect of KR delay on motor learn- mation, and allows the subject to rely on the
ing efficacy. The same is true of the post-KR information too strongly ( 1 ).
delay inter\'al. There may be a slight reduction What is the best number of trials to
in learning if the KR delay is ver)' short, but complete before giving KR? This appears to
any effects are very small. However, it has vary depending on theFor very simple
task.
been shown that it is good not to fill the KR- movement dming tasks, in which KR was
delay interval with other movements, since given after one trial, five trials, 10 trials, or 15
these appear to interfere with the learning of trials, the performance on acquisition trials
the target movements. Research on the effects was best for the most frequent feedback, but
of filling the post-KR delay interval with ex- when a transfer test was given, the perfor-
traneous activities is less clear. Apparently, this mance was best for the 15-triaI summary
interval is not as important as the KR-delay group. In a more complex task, where a pat-
interval for the integration of KR information. tern of moving lights had to be intercepted by
It has also been recommended that the inter- an arm movement (like intercepting a ball
trial intcnml should not be excessively short, with a bat),most effective summary
the
but the literature in this area shows conflicting length for learning was five trials, and any-
results (1, 16) concerning the effects of dif- thing more or less was less efficient (1).
ferent lengths of intertrial intervals on learn- How precise must KR be in order to be
ing. most effective? The answer varies for adults vs.
What happens to learning eflficacy if KR children. For adults, quantitative KR appears
is not given every trial.' For example, if you to be best, with the more precise KR giving
ask a patient to practice a reaching movement more accurate performance, up to a point, be-
and only give the patient feedback on the ac- yond which there is no fiirther improvement.
curacy of the movement every five or 10 trials, For adults, units of measure (for example,
what do you think might happen? One might inches, centimeters, feet, miles) do not seem
assume that decreasing the amount of KR to be important, with nonsense units even be-
given would have a detrimental effect on ing effective. However, in children, unfamiliar
learning. However, experiments in this area units or ver\' precise KR can be confiising and
have shown surprising results. degrade learning (1, 18).
Experimenters compared the perfor-
mance of (fl) subjects who had KR feedback Practice Conditions
on every trial; (b) subjects who had summary
KR, that is KR for each of the trials only at We have already discussed the impor-
the end of an entire block of 20 trials; and (c) tance of KR to learning. A second variable
subjects who had both types of feedback. It that is also very important is practice. Typi-
was found that at the end of the acquisition cally, the more practice you can give a patient,
trials, performance was best if KR was given the more the patient learns, with other things
after every trial (groups I and 3 were far better being equal. Thus, in creating a therapy ses-
than group 2). However, when performance sion, the number of practice attempts should
was then compared for the groups on transfer be maximized. But what about fatigue? How
tests, where no KR was given at any dme, the should the therapist schedule practice periods
group that was originally the least accurate, vs. rest periods? Research to answer these
the summary KR only group (group 2), was questions is summarized in the following sec-
now the most accurate (17). tions.
MASSED AND DISTRIBUTED variable than for the constant practice group
PRACTICE (19). Thus, in general, variable practice ap-
pears to allow a person to perform signifi-
To answer these questions researchers cantiy better on novel variations of the task.
have performed experiments comparing n\o
types ot practice sessions: massed and distrib- CONTEXTUAL INTERFERENCE
uted. Massed practice is defined as a session
in which the amount of practice time in a trial
Surprisingly, it has also been found that
is greater than the amount of rest between

factors that make performing a task more dif-
ficult initially ven,' often make learning more
trials. This may lead to fatigue in some tasks.
Distributed practice is defined as a session in effective in thelong run. These t^'pes of fac-
tors have been called context effects (1). For
which the amount of rest between trials equals
example, if you were to ask a person to prac-
or is amount of time for a
greater than the
trial (1).For continuous tasks, massed practice five different tasks in random order, vs.
blocking the trials for each task into individual
tice has been proven to decrease performance
groups, you might presume that it would be
markedly while it is present, but afl:ects learn-
easier to learn each task in a blocked design.
ing only slightly when learning is measured on
a transfer task in distributed conditions. In
However, this is not the case. While perfor-
this case fatigue may mask the original learn-
mance is better during the acquisition phase,
ing effects during massed practice, but they
when tested on a transfer task, performance is
actually better in the randomly ordered con-
become apparent on the transfer tasks. For
ditions.
discrete tasks, the research results are not as
It has been concluded that the critical
clear, and appear to depend considerably on
factor in improving learning is that the subject
the task (1).
Keep in mind that in the therapy setting
has to do something different on consecutive
of injur)' due to fatigue will increase dur-
a risk
trials (1). What are the implications of these
ing massed practice for tasks that may be
results.' Clearly, traditional methods for re-
somewhat dangerous for the patient, such as

training by practicingone skill repeatedly are
which probably not the most effective. On the con-
tasks in a fall could result. In this case,
trar}', encouraging the patient to practice a
it is best not to overly fatigue the patient and
risk injun'.
number of tasks in random order would prob-
ably be more successftil for long-term reten-
tion (1).
VARJABLE PRACTICE
WHOLE VS. PART TRAINING
As we mentioned when discussing
Schmidt's schema theory, generalizabilitii' of One approach to retraining fijnction is
practice is considered a ver\' important vari- to break the task down into interim steps,
able in motor learning. In general, research helping the patient to master each step prior
has shown that variable practice increases the to learning the entire task. This has been
adaptability of learning. For example, in one called task analysis and is defined as the pro-
experiment one group of subjects practiced a cess of identifying the components of a skill
timing task (they had to press a button when or mo\'ement and then ordering them into a
a moving pattern of lights arrived at a partic- sequence. How are the components of a task
ular point) at \ariable speeds of 5, 7, 9, and defined.' They are defined in relationship to
11 miles/hr, while a second group (constant the goals of the task. So, for example, a task
practice) practiced one of those
at only analysis approach to retraining mobilit)'
speeds. Then, all subjects performed a transfer would be to break down the locomotor pat-
test, in which they performed at a novel light tern into naturally occurring components
speed outside their previous range of experi- such as step initiation, stability' during stance,
ence. The absolute errors were smaller for the or push-oft' to achieve progression. During
Chapter Two Motor Le.\rning .•VXD Recox-ery of Function 37
mobilin' retraining, the patient would practice pursuit task. On the first day, all subjects
each of these components in isolation, before practiced 25 trials. On days 2-9, one group
combining them into the whole gait pattern. of subjects continued with physical practice,
But each of these components must be prac- while a second group performed only mental
ticed within the overall context of gait. For practice, and a third group was given no prac-
example, ha\ing a patient practice hip exten- tice. On day 10, all subjects were retested, and
sion while prone will not necessarily increase the mental practice group had improved al-
the patient's abilit\- to achieve the goal of most as much as the physical practice group,
stance stability', even though both require hip while the no-practice group showed little im-
extension. Thus part-task training can be an provement.
etfecti\e way to retrain some tasks, if the task \\Tiy is this the case? One hypothesis is
itself can be naturally di\'ided into units that that the neural circuits underhing the motor
reflect the inherent goals of the task (20, 21 ). programs for the moxements are actualh' trig-
gered during mental practice, and the subject
TRANSFER either does not activate the final muscle re-
sponse at all, or activates responses at ver\- low-
A critical issue in rehabilitation is how levelswhich do not produce movement. In

new task, or to
training transfers, either to a Chapter 3, we discussed experiments showing
a new emironment. For example, wiU learn- that one part of the brain, the supplementarv'
ing a task in a clinical en\ironment transfer to motor cortex, is activated during mental prac-
a home environment? Or does practicing tice.
standing balance transfer to a dynamic bal-

ance task such as walking around the house.' GUID.ANCE
What determines how well a task
learned in one condition will transfer to an- One technique often used in therapy is
other.- Researchers have determined that the guidance, that is, the learner is physically
amount of transfer depends on the similarirv' guided through the task to be learned. Re-
between the two tasks or the two environ- search has again explored die efliciencA' of this
ments (22, 23). A critical aspect in both ap- form of learning vs. other forms of learning
pears to be whether the neural processing de- that involve and error discover)- proce-
trial
mands in the two situations are similar. For dures. In one set of experiments 1 ), various (
example, training a patient to maintain stand- forms of physical guidance were used in teach-
ing balance in a well-controlled environment, ing a complex elbow movement task. \\Tien
such as on a firm, flat surface, in a well-lit performance was measured on a no-guidance
clinic, will not necessarily enable the patient transfer test, phvsical guidance was no more
to balance in a home environment that con- effective than simply practicing the task under
tains thick carpets, uneven surfaces, and \isual unguided conditions. In other experiments
distractions. The more closely the demands in (25 ), practice under unguided conditions was
the practice environment resemble those in found less effective for acquisition of the skiU,
the actual emironment, the better die transfer but was more effective for later retention and
(20,21). transfer. This is similar to the results just cited,
which showed that conditions that made the
MENT.AL PR.-\CTICE performance acquisition more difficult en-
hanced performance in transfer tests.
It has been shown that mentally practic- This doesn't mean that we should never
ing a skill ( the act of performing the skill in use guidance in teaching skills, but it implies
one's imagination, without any action in- that if guidance is used, it should be used only
volved ) can produce large positive efl:"ects on at the outset of teaching a task, to acquaint
the performance of the task. For example, the performer with the characteristics of the
Rawlings et al. (24) taught subjects a rotary task to be learned.
Recovery of Function ble. Upon maturation, function was localized

to various parts of the CNS. Research at the
Concepts Related to Recovery of time suggested that regeneration and reor-
Function ganization was not possible within the adult
CNS. This view of the CNS naturally led to
To understand concepts related to re- therapy directed at compensation, since re-
covery of function it is necessary first to define covery in the strict sense of the word was not
terms such as fianction and recovery. possible. More recent research in the field of
neuroscience has begun to show that the
FUNCTION adult CNS has great plasticity and retains an
incredible capacity for reorganization. Studies
Function is defined here as the complex
on neural mechanisms underlying recover)' of
activity of the whole organism that is directed
function are covered in Chapter 4 of this text.
at performing a behavioral task (26). Optimal
function is characterized by behaviors that are
efficient in accomplishing a task goal in a rel- SPARING OF FUNCTION
evant environment.
When a function is not lost, despite a
RECOVERY brain injury, it is referred to as a spared func-
tion (26). For example, when language de-
The term recovery has a number of dif- velops normally in children who have suffered
ferent meanings pertaining to regaining fianc- brain damage early in life, retained language
tion that has been lost following an injury. A function is said to be spared.
stringent definition of recovery requires
achieving the functional goal in the same way
itwas performed premorbidly, that is, using STAGES OF RECOVERY
the same processes utilized prior to the injur)'
Stages of recovery from neural injury
(27). Less stringent definitions define recov-
ery as the ability to achieve task goals using
have been described by several authors. Stages
effective and efficient means, but not neces- of recovery are based on the assumption that
the process of recovery can be broken down
sarily those used premorbidly (28).
into discrete stages. Classically, recovery is di-
RECOVERY VS. COMPENSATION vided into spontaneous recovery and forced

recovery. Forced recovery is recovery ob-
Is recovery the same or different from tained through specific interventions de-
compensation? Compensation is defined as signed to impact neural mechanisms (30).
behavioral substitution, that is, alternative be- The presumption is that different neural
havioral strategies are adopted to complete a mechanisms underlie these relatively discrete
task. Recovery is achieving function through stages of recovery. Chapter 4 describes how
original processes, while compensation is research on neural mechanisms might con-
achieving fiinction through alternative pro- tribute to various stages of recovery.
cesses. Thus function returns, but not in its
identical premorbid form. Factors Contributing to Recovery

A question of concern to many thera- of Function
pists is: Should therapy be directed at recovery
of fiinction or compensation? The response to Jean Field (31 ), a physical therapist who
this question has changed over the years as has written extensively on the neural basis for
our knowledge about the plasticity and mal- recovery of function, summarizes a number of
leability of the adult CNS has changed (29). factors that affect the outcome of damage to
For many years, the adult mammalian CNS the nervous system as well as the extent of
was characterized as both rigid and unaltera- subsequent recovery.
EFFECT OF AGE ON QUALITY OF THE LESION AND

RECOVERY RECOVERY
Dr. Held notes that the age of the in- Held (31) notes that there are a number
dividual at the time of the lesion affects re- of characteristics of lesions that affect the ex-
covery of fiinction, but in a complex manner. tent of recovery from injur}'. For example, a
Earl\' views on age-related effects on recover}' small lesion has a greater chance of recovery,
of brain fijnction proposed that injur\' during as long as a fiinctional area hasn't been en-
infancy caused fewer deficits than damage in tirely removed. In addition, slowly developing
the adult years. For example, in the 1940s, lesions appear to cause less fiinctional loss
Kennard (32, 33) performed experiments in than lesions that happen quickly. For exam-
which she removed the motor cortex of infant ple, case studies have shown that a person
vs. adult monkeys and found that infants were who fiinctioned well until near death, upon
able to feed, climb, walk, and grasp objects, autopsy, had a large lesion in the brain tissue.
while adults were not. In humans, this effect This phenomenon has been explored experi-
has been noted in language fiinction, where mentally, by making serial lesions in animals,
damage to the dominant hemisphere shows in which the animal is allowed to recover be-
litde or no effect on speech in infants, tween lesions (26). If a single large lesion is
but causes different degrees of aphasia in made in the motor cortex (Brodmann's areas
adults. 4 & 6), animals become immobilized, where
However, as we understand more about similar lesions produced serially over a period
the fiinction of different brain areas, research- of time allow the animal to walk, feed, and
ers are concluding that not all areas show the right itself with no difficulty (35).
same capacity For example,

for regeneration.
injur)' to some areas of the brain shows similar EFFECT OF EXPERIENCE ON
deficits whether it occurs in the infant or RECOVERY
adult, while damage to other areas may show
litde effect in infancy, yet problems develop Held (31) notes that studies in which
later with fijrther maturation. ratswere raised in enriched environments
Why is this? It has been hypothesized show many resultant changes in the brain
that if an area is mature, injury will cause sim- morpholog}' and biochemistry, including in-
ilar damage in infants and adults. But, if an- creased brain weight, dendritic branching,
other area that is fiinctionally related is not yet and enzyme activity. As a result of these find-
mature, it may assume the function of the in- ings, researchers wondered if this enrichment
jured area. In addition, if an immature area is would improve responses to brain injury. Ex-
damaged and no other area assumes its fiinc- periments showed that preinjury environmen-
tion, no problems may be seen in infancy, tal enrichment protects animals against cer-
but in later years, deficits may become ap- tain deficits after brain lesions.For example,
parent. of rats received lesions of the cortex,
rvvo sets
In addition, when children have brain one group with preoperati\'e enrichment and
injuries in thespeech areas, there is probably a control group. After surger}', the enriched
lossof other fiinctions to spare the fiinction animals made fewermistakes during maze
of speech. Researchers have found that the learning, and performed better than
in fact
IQ's of children with spared speech following control animals without brain damage.
early brain injury were consistendy lower than In a second study by Held et al. (36) the
those of children who had a when
brain injur}' effect of pre- and postoperative enrichment
they were older (34). This implies that when was compared for a locomotor task following
a fiinction is spared, a crowding effect may removal of sensorimotor cortex. They found
occur, and thus it occurs at the cost of com- that preoperatively enriched rats were no dif-
promising another behavior (26). ferent from enriched sham-lesioned controls
on both behavioral and fine-grained move- ately afi:er the lesion and is specific to the
ment analysis. The group that was only post- involved limb.
operatively enriched was mildly impaired in
locomotor skills, but recovered more quickly Clinical Overview
than the lesioned controls, though they never
regained flill locomotor function. Thus, post- By now it should be clear that the field
operative enrichment is effective, but doesn't of rehabilitation has much in common with
allow the same extent of recover\^ as preop- the field of motor learning, defined as the
erative enrichment. study of the acquisition of movement. More
Held suggests that enriched subjects accurately, therapists involved in treating the
may have developed functional neural cir- adult neurological patient are concerned with
cuitry that is more varied than that of re- issues related to motor relearning, or the reac-
stricted subjects, and this could provide them quisition of movement. The pediatric patient
with a greater abilit>' to reorganize the ner- who is born with a CNS deficit, or experiences
vous system after a lesion, or simply to use injur)' early in life, faces the task of acquisition
alternate pathways to perform a task. of movement in the faceof unknown mus-
culoskeletal and neural constraints. In either
EFFECT OF TRAINING ON RECOVERY case, the therapist is concerned with structur-
ing therapy in ways to maximize acquisition
According to Held, training is a differ- and/or recovery of function.
ent form of exposure to enriched environ- Remember Mr. Smith from the begin-
ments in that activities used are specific rather ning of this chapter.' Mr. Smith had been re-
than generalized. Ogden and Franz [i7) perceiving therapy for 5 weeks and had recovered
formed an interesting study in which they much of his abilit>' to function. We wanted to
produced hemiplegia in monkeys by making know more about why this happened. What
lesions in the motor cortex. They then gave is the cause of Mr. Smith's recover)' of motor
four types of postoperative training: (a) no function? How much of his recover)' may be
treatment, {b) general massage of the involved attributed to therapeutic inter\'entions? How
arm, c) restraint of the noninvolved limb, and
(
manv of his reacquired motor skills will he be
(
d) restraint of the noninvolved limb coupled able to retainand use when he leaves the re-
with stimulation of the involved limb to habilitation and returns home?
facilit)'
move, along with forced active movement of Mr. Smith's reacquisition of flinction
the animal. The last condition was the only cannot be attributed to any one factor. Some
one to show recovery, and in this condition it of Mr. Smith's functional return will be due
occurred within 3 weeks. to recover)', that is, regaining original control
A second study by Black et al. (38) ex- of original mechanisms; some will be due to
amined recovery from a motor cortex fore- compensator)' processes. In addition, age,
limb area lesion. They initiated training im- premorbid function, site and size of lesion,
mediately afiier 4 months, with
surgery' or at and the effect of inten'entions all interact
training lasting 6 months. They found that to determine the degree of function
training of the involved hand alone, or train- regained.
ing of the involved and normal hand together, Mr. Smith has had excellent therapy as
was more effective than training the normal well! Mr. Smith has been in\'ol\'ed in carefully
hand alone. When training was delayed, re- organized therapy sessions that have contrib-
covery was worse than when it was initiated uted to his reacquisition of task-relevant be-
immediately following the lesion. haviors. Both associative and nonassociative
Held concludes that recovery' is affected forms of learning may have played a role in
by the state of the system at the time of a le- his recovery. Habituation was used to de-
sion, and that training after the lesion im- crease complaints of dizziness associated with
proves recovery best when it occurs immedi- inner ear problems.
Chapter Two Motor Lf.,\rning and PIecovery of Function 41
Trial and error learning (operant con- or the relationship of one's behavior to a
ditioning) was used to help him discover op- consequence (operant conditioning).
timal solutions to many Rmctional tasks. His 6. Classical conditioning consists of learning to
therapist carefiilly structured his en\'ironment

pair two stimuli. During operant condition-
ing we learn to associate a certain response,
so that optimal strategies were reinforced. For
from among many that we have made, with
example, biofeedback was used to help him
a consequence.
develop better foot control during locomo-
7. Procedural learning refers to learning tasks
tion.
that can be performed automatically without
Functionally rele\ant tasks were prac- attention or conscious thought, like a habit.
ticed under wide-ranging conditions. Under 8. Declarative learning results in knowledge
optimal conditions, this would lead to pro- that can be consciously recalled, and thus
cedural learning, ensuring that Mr. Smith requires processes such as awareness, atten-
would be able to transfer many of his newly tion, and reflection.
gained skills to his home environment. Prac- 9. Different theories of motor control include
ticing tasksunder varied conditions was aimed Adams' closed loop theory of motor control,
Schmidt's schema theory, Fitts' and Posner's
at the development of rule-governed action or
theory on the stages of motor learning, and
schemas. Recognizing the importance of de-
Newell's theory of learning as exploration.
veloping optimal perceptual and motor strat-
10. Classical recovery
is divided into spontane-
egies, his therapist structured his therapy ses-
ous recovery and forced recovery, that is,
sions so Mr. Smith explored the
that recovery obtained through specific inter-
perceptual en\ironment. This was designed to ventions designed to impact neural
facilitate the optimal mapping of perceptual mechanisms.
and motor strategies for achie\ing functional 11. Experiments show that preinjury environ-
goals. Finally, therapy was ciirected at helping mental enrichment protects animals against
Mr. Smith repeatedly solve the sensoni'-motor certain deficits after brain lesions.
12. Training after the lesion improves recovery
problems inherent in various fimctional tasks,
best when occurs immediately after the le-
it
rather than teaching him to repeat a single
sion and when is specific to the involved
it
solution.
limb.
Summary
References
1. Motor learning, like motor control, emerges
from a complex set of processes including 1. Schmidt RA. Motor control and learning.
perception, cognition, and action. 2nd ed. Champaign, IL: Human Kinetics,
2. Motor learning results from an interaction of 1988.
the individual with the task and environ- 2. Newell KM. Motor skill acquisition. Ânnu
ment. Rev Psychol 1991;42:213-237.
3. Nonassociative learning occurs when an or- 3. Schmidt RA. Motor learning principles for
ganism is given a single stimulus repeatedly. physical therapy. In: Contemporary- manage-
As a result, the nervous system learns about ment of motor control problems. Proceed-
the characteristics of that stimulus. ings of the II Step Conference. Alexandria,
4. Habituation and sensitization are two ver)' VA: .\PTA, 1992:49-62.
simple forms of nonassociative learning. Ha- 4. Kupfermann 1. Learning and memor\'. In:
bituation is a decrease in responsiveness that Kandel ER, Schwartz JH, lessell TM, eds.
occurs as a result of repeated exposure to a Principles of neuroscience. 3rd ed. New York:
nonpainful stimulus. Sensitization is an in- Elsevier, 1991:997-1008.
creased responsiveness following a threat- 5. Adams A closed-loop theor\' of motor
JA.
ening or noxious stimulus. learning. J Motor Behav 1971;3:111-150.
5. In associative learning a person learns to pre- 6. Ivr>' R Representational issues in motor
dict relationships, either relationships of one learning: phenomena and theor\'. In: Keele S,
stimulus to another (classical conditioning) Heuer H, eds. Handbook of perception and
action: motor skills. New York: Academic management of motor control problems.
Press, in press. Proceedings of the II Step Conference. Al-
7. Taub E, Herman AJ. Movement and learning e.\andria, VA: .\PTA, 1991.
in the absence of sensor\' feedback. In: Freed- 21. Schmidt R.\.Motor learning principles for
man SJ, ed. The neuropsycholog\' of spatially physical therapy. Contemporar\' manage-
oriented behavior. Homewood, IL: Dorsey ment of motor control problems. Proceed-
Press, 1968. ings of the II Step Conference. Alexandria,
8. Rothweil JC, Traub MM, Day BL, Obeso JA, VA:.\PTA, 1991.
Marsden CD. Manual motor performance in 22. Schmidt Rj\, Young DE. Augmented kine-
deafFerented man. Brain 1982;105:515-542. matic information feedback for skill learning:
9. Fentress JC. Development of grooming in a new research paradigm. J Mot Behav 1987.
mice with amputated forelimbs. Science 23. Lee TD. Transfer-appropriate processing: a
1973;179:704. framework for conceptualizing practice ef-
10. Schmidt RA. A schema theory of discrete mo- fects in motor learning. In: Meijer OG, Roth
tor skill learning. Psychol Rev 1975;82:225- K, eds. Complex movement beha\ior: the
260. motor-action controversy. Amsterdam:
11. Kerr R, Booth B. Skill acquisition in elemen- North Holland, 1988.
tary school children and schema theor\-. In: 24. Rawlings EI, Rawlings IL, Chen CS, Yilk
Landers DM, Christina R\V, eds. Psycholog\- MD. The facilitating effects of mental re-
of motor behavior and sport, vol. 2. Cham- hearsal in the acquisition of rotar\- pursuit
paign, IL: Human Kinetics, 1977. tracking. Psychonomic Science 1972;26:71-

12. Shapiro DC, Schmidt RA. The schema the- 73.
ory': recent e\idence and developmental im- 25. Singer RN. Motor learning and human per-
plications. In: Kelso FAS, Clark JE, eds. The formance. 3rd ed. New York: Macmillan,
development of movement control and co- 1980.
ordination. New York: John Wiley & Sons, 26. Craik RL. Recover)' processes: maximizing
1982:113-173. fimction. In: Contemporar\'management of
13. Kugler PN, Kelso JAS, Tur\-ey MT. On the motor control problems. Proceedings of the
concept of coordinative structures as dissipa- II Step Conference. Alexandria, \'A: .APTA,
tive structures: I. Theoretical line. In: Stel- 1992:165-173.
mach GE, Requin J, eds. Tutorials in motor 27. Almli RB, Finger S. Toward
a definition of
behavior. .\msterdam: North-Holland, recovepi- of function. Le Vere TE, Almli

In:
1980:3-37. RB, Stein DG, eds. Brain injury and recover)-:
14. Fitts PM, Posner MI. Human performance. theoretical and controversial issues. New-
Belmont, CA: Brooks/Cole, 1967. York: Plenum, 1988:1^.
15. Bilodeau EA, Bilodeau IM, Schumsky DA. 28. Slavin MD, Laurence S, Stein DG. Another
Some effects of introducing and withdrawing look at vicariation. In: Le Vere TE, .-Mmli RB,
knowledge of results early and late in practice. Stein DG, eds. Brain injur\- and recovery: the-
J Exper Psych 1959;58:142-144. oretical and controversial issues. New York:
16. Saknoni AW, Schmidt R.\, Walter CB. Plenum, 1988:165-179.
Knowledge of results and motor learning: a 29. Gordon J. Assumptions underlying physical
review and critical reappraisal. Psychol Bull therapy intenention: theoretical and histori-
1984;95:355-386. cal perspectives. In: Carr JH, Shepherd, RB,
17. Laven- JJ. Retention of simple motor skills as Gordon J et al., eds. Movement sciences:
a fiincrion of D,pe of knouledge of results. foundations for physical therapy in rehabili-
Can J Psych 1962;16:300-311. tation. Rockville, x\ID: Aspen Systems,

18. Newell KM, Kennedy JA. Knowledge of re- 1987:1-30.
sults and children's motor learning. Dev 30. Bach-y-Rita P, Balliet R. Recover, from
Psych 1978;14:531-536. stroke. In: Duncan P,Badke MB, eds. Stroke
19. Catalano JF, Kleiner BM. Distant transfer and rehabilitation: the recover\- of motor control.
practice variability. Percept Mot Skills 1984; Chicago: Year Book Medical Publishers,
58:851-856. 1987:79-107.
20. Winstein CJ. Designing practice for motor 31. Held JM. Recover)- of function after brain
learning: clinical implications. Contemporary- damage: theoretical implications for thera-

Chapter Two Motor Learning .\nd Reco\-ery of Function 43
peutic inter\'ention. In: Carr JH, Shepherd, 35. TraWs AM, Woolsey CN. Motor performance
RB, Gordon J, et al., eds. Movement sci- of monkeys after bilateral partial and total ce-
ences: foundations for physical therapy in re- rebral decortication. Am J Phys Med 1956;
habilitation. Rock\ille, MD: Aspen S\stems, 35:273-310.
1987:155-177. 36. Held JM, Gordon F, Gentile AM. Environ-
32. Kennard MA. Relation of age to motor im- mental influences on locomotor recover)' fol-
pairment in man and in sub-human pri- lowing cortical lesions in rats. Behav Neurosci
mates. Arch Neurol Psychiatr\^ 1940;44:377- 1985;99:678-690.
398. 37. Ogden R, Franz SI. On cerebral motor con-
33. Kennard MA. Cortical reorganization of mo- trol: the recover^' from experimentallv pro-
tor function: studies on a series of monkeys duced hemiplegia. Psychobiolog\' 1917;
of various ages from infancy to maturitw Arch l:33-i9,
Neurol Psychiatr 1942;48:27-240. 38. Black P, Markowitz RS, Cianci SN. Recoven,-
34. Woods BT. The restricted effects of right of motor fiinction after lesions in motor cor-
hemispheric lesions after age one: Wechsler tex of monkeys. Ciba Found Symp 1975;
test data. Neuropsychologia 1980;18:65-70. 34:65-83.
Chapter 3
Physiology of Motor Control

Introduction and Overview
Motor Control Theories and Physi^^^
mpH Central Visual Pathways
Lateral Geniculate Nucleus
Overview of Brain Function Superior Coliiculus
Spinal Cord Pretectal Region
Brainstem Primary Visual Cortex
Cerebellum Higher-Order Visual Cortex
Diencephalon Vestibular System
Cerebral Hemispheres Peripheral Receptors
—
Neuron Basic Unit of the CNS Semicircular Canals
Sensory/Perceptual Systems Utricle and Saccule
Somatosensory System Central Connections
Peripheral Receptors Vestibular Nuclei
Muscle Spindle Action Systems
Stretch Reflex Loop Motor Cortex
Colgi Tendon Organs Corticospinal Tract
Joint Receptors Function
Cutaneous Receptors Motor Cortex
Role of Somatosensation at the Spinal Supplementary and Premotor
Cord Level Cortex
Ascending Pathways Higher-Level Association Areas
Dorsal Column-Media Lemniscal
1 Association Areas of the Frontal Region
System Cerebellum
Anterolateral System Anatomy of the Cerebellum
Thalamus Flocculonodular Lobe
Somatosensory Cortex Vermis and Intermediate
Association Cortices Hemispheres
Visual System Lateral Hemispheres
Peripheral Visual System Basal Ganglia
Photoreceptor Cells Anatomy of the Basal Ganglia
Vertical Cells Role of the Basal Ganglia
Horizontal Cells Summary
Introduction and the interaction of both perceptual and action
Overview systems, with cognition affecting both sys-

tems at many different levels. Within each of
Motor Control Theories and these systems are many levels of processing,
Physiology which are illustrated in Figure 3.1. For ex-
ample, perception can be thought of as pro-
As we mentioned in Chapter 1, theories gressing through various processing stages.
of motor control are not simply a collection Each stage reflects specific brain structures
of concepts regarding the nature and cause of that process sensor)' information at different
movement. They must take into considera- levels, from initial stages of sensory processing
tion current research findings about the struc- to increasingly abstract levels of interpretation
ture and fiinction of the nen'ous system. Re- anci integration inhigher levels of the brain.
member motor control
that is about the Recent neuroscience research suggests
nature and cause of movement. It arises from that movement control is achieved through
45
46 Section I THEORETICAL FR.\MEVVORK
PERCEPTION ACTION
Sensing Perceiving Interpreting Strategy/ Activation

plan
Periphery 1 = and 2° Higher-level Supplementary 1 ° Motor Motor

Sensory sensory motor cortex cortex neurons
cortices processing BG/CB and
BG/CB
areas in muscles/
the parietal. joints
occipital,
and temporal
lobes
Figure 3.1. Model ot the interaction between perceptual, action and cognitive processes involved in motor control.
BC = basal ganglia; CB = cerebellum.
the cooperative effort of many brain struc- In parallel distributed processing, the
tures, which are organized both hierarchically same signal is processed simultaneously
and in parallel. This means that a signal may among many different brain structures,
be processed in t\\ô ways. A signal may be though for different purposes. For example,
processed hierarchically, within ascending the cerebellum and the basal ganglia process
levels of the central ner\ous system (CNS). In higher level motor information simulta-
addition, the same may be processed
signal neously, before sending it back to the motor
simultaneously among many different brain cortex for action.
structures, showing parallel distributed pro- This chapter re\iews the processes un-
cessing. Hierarchical processing, in conjunc- derlying the production of human movement.
tion with distributed processing, occurs in The first section of this chapter presents an
both the perceptual and action systems of over\iew of the major components of the
movement control. CNS and the structure and fiinction of a neu-
When we talk about "hierarchical" pro- ron, the basic unit of the CNS. The remaining
cessing in this chapter, we are describing a sys- sections of this chapter discuss in more detail
tem which higher levels of the brain are
in the neural anatomy (the basic circuits), and
concerned with issues of abstraction of infor- the physiology- (the fiinction) of the systems
mation. For example, within the perceptual involved in the production and control of
system, hierarchical processing means that movement. The chapter follows the neural
higher brain centers ititejjrate inputs from anatomy and physiolog\' of movement con-
many senses, and interpret incoming sensor,' trol from perception into action, recognizing
information. On the action side of movement that it is often difhcult to distinguish where
control, higher levels of brain function form one ends and the other begins.
motor plans and strategies for action. Thus,
higher levels might select the specific response Overview of Brain Function
to accomplish a particular task. Lower levels
of processing would then carr\' out the de- Brain fiinction underlying motor con-
tailed monitoring and regulation of the re- trol is t\'pically divided into multiple process-
sponse execution, making it appropriate for ing levels, including the spinal cord, the brain-
the context in which it is carried out. stem, the cerebellum, the diencephalon, and
Chapter Three PmsioLOGY OF Motor Control 47
the cerebral hemispheres, including the cere- mation, which regulates our arousal and
bral cortex and basal ganglia (1, 2). awareness, is also found within the brain-
stem (1).
SPINAL CORD The anatomist's view of the brainstem
(Fig. 3. 2 A) shows di\isions from caudal to
At the lowest level of the perception-ac-
rostral into the medulla, pons, and midbrain,
tion hierarchy is the spinal cord, and the sen- shows
while the abstract model Fig. ( 3. 25) its
son- receptors and muscles that it innerâtes.
input connections from the spinal cord and
The circuitA- of the spinal cord is in\ol\ed in
higher centers (the cerebellum and motor
the initial reception and processing of soma-
cortex) and its motor pathways back to the
tosenson- information (fi-om the muscles,
spinal cord.
joints, and skin) contributing to the control
of posture and movement. At the level of spi- CEREBELLUM
nal cord processing, we can expect to see a
fairlv simple relationship between the sensor\- The cerebellum lies behind the brain-
input and motor output. At the spinal cord stem and is connected to it by tracts called
level, we see the organization of reflexes, the "peduncles" (Fig. 3.2^1). .-^s you can see from
most stereotyped responses to senson.- stimuli, Figure 3.25, the cerebellum receives inputs
and the basic flexion and extension patterns from the spinal cord giving it feedback about
(
of the muscles involved in leg movements, movements) and from the cerebral cortex
such as kicking and locomotion (1). (gi\ing it information on the planning ot
Sherrington called the motor neurons of movements), and it has outputs to the brain-
the spinal cord the "final common pathway," stem.The cerebellum has many important
since they are the last processing level before motor control. One is to adjust
frinctions in
muscle activation occurs. Figure 3. 2 A shows our motor responses by comparing the in-
the anatomist's \iew of the ner\ous system tended output with sensor\- signals, and then
with the spinal cord positioned caudally. Fig- to update the movement commands if they
ure 3.25 shows an abstract model of the ner- deviate from the intended trajector\-. The cer-
vous svstem with the spinal cord positioned ebellum also modulates the force and range
at the bottom of the hierarchy, with its many of our movements and is involved in motor
parallel pathwa\s. In this view , the sensor\' relearning.
ceptors are represented by input arrows and
the muscles by output arrows. DIENCEPHALON
BR-AINSTEM ,\s we mo\'e rostralh' in die brain, we
next find the diencephalon, which contains
The spinal cord extends rostrally to join the thalamus (Fig. 3.2A). The thalamus pro-
the next neural processing level, the brain- cesses most of the information coming to the
stem. The brainstem contains important nu- cortex from the man\- parallel input pathways
clei involved in postural control and loco- (from the spinal cord, cerebellum, and brain-
motion, including the vestibular nuclei, the stem) (Fig. 3.25). These pathways stay seg-
red nucleus, and the reticular nuclei. The regated during the thalamic processing, and
brainstem receives somatosenson,* input from during the subsequent output to the different
the skin and muscles of the head, as well as parts of the cortex ( I ).
sensoPi' input from the vestibular and visual
svstems. In addition, nuclei in the brainstem CEREBR.\L HEMISPHERES

control the output to the neck, face, and eyes, (CEREBRAL CORTEX .AND BAS.AL
and are critical to the function of hearing and GANGLL\)
taste. In fact, all the descending motor path-
wavs except the corticospinal tract originate .\s we move higher, we find the cerebral
in the brainstem. Finally, the reticular for- hemispheres, which include the cerebral cor-
Postcentral
gyrus
Temporal
lobe
Forebrain
1 -Telencephalon
2-Diencephalon:
thalamus, hypothalamus
Midbrain
3-Mesencephalon
Hindbrain
Brainstem: 4-Metencephalon:
midbrain, pons, cerebellum
^ pons,
medulla oblongata
5-Metencephalon
medulla oblongata
Figure 3.2. A, Illustration of the nervous system from an anatomist's view. B, An abstract model of the nervous system.
(Adapted from Kandel E, Schwartz |H, Jessell TM, eds. Principles of neuroscience. 3rd ed. NY: Elsevier; 1991 :8.)
Chapter Three PHisioLOGY OF MOTOR Control 49
Nonmotor Premotor
cortical areas areas
cortical
f±
Basal
ganglia
Descending
pathways
Ascending
pathways
Independent
Segmental
sensory
events
Afferent
input (spinal) networks ^ Muscles
Muscles
Sensory Displacement
consequences
of movement
Figure 3.2. continued
tex and basal ganglia. Lying at the base of the tract and the corticobulbar system (Fig.
cerebral cortex, the basal ganglia (Fig. 3.2) 3.2^).
receive input from most areas of the cerebral In light of these various subsystems in-
cortex,and send their output back to the mo- volved in motor control, clearly, the nenous
tor cortex, via the thalamus. Some of the s\stem is organized both hierarchically and
functions of the basal ganglia involve higher- "in parallel." Thus, the highest levels of con-
order, cognitive aspects of motor control, trol not onh' affect the next le\els down, they
such as the planning of motor strategies (1). also can act independently on the spinal mo-
The cerebral cortex (Fig. 3.2^1) is often tor neurons. This combination of parallel and
considered the highest level of the motor con- hierarchical control allows a certain oxerlap of
trol hierarchy. The parietal and premotor ar- functions, so that one system is able to take
eas, along with other parts of the ner\'ous sys- over from another when en\'ironmental or
tem, are imoh'ed in identit\'ing targets in task conditions require it. This also allows a
and pro-
space, choosing a course of action, certain amount of recoveni' from traumatic in-
gramming movements. The premotor areas jur\', by the use of alternative pathways.
send outputs mainly to the motor cortex, To better understand the flinction of
which sends its commands on to the brain- the difterent levels of the ner\ous system, let's
stem and spinal cord v\z the corticospinal examine a specific action and walk through
the pathways of the nervous system that con- record from a neuron intracellularly with an
tribute to its planning and execution. For ex- electrode, they discover that the inside of the
ample, perhaps you're thirsty and want to ceU has a resting potential of about — 70 mv
pour some milk from the milk carton in front with respect to the outside (Fig. 3.3). This
of you into a glass. Sensory inputs come in electrical potential is caused by an unequal
from the periphery to tell you what is hap- concentration of chemical ions on the inside
pening around you, where you are in space, vs. the outside of the cell. Thus, K* ions are
and where your joints are relative to each high on the inside of the celland Na* ions are
other: they give you a map of your body in high on the outside of the cell, and an elec-
space. Higher centers in the cortex make a trical pump within the cell membrane keeps
plan to act on this information in relation to the ions in their appropriate concentrations.
the goal: reaching for the carton of milk. When the neuron is at rest, K* channels are
From your sensory map, you make a open and keep the neuron at this negative po-
movement plan (using, possibly, the parietal tential (2-4).
lobes and premotor cortex). You're going to When a neuron is excited, one sees a se-
reach over the box of corn flakes in front of ries of dramatic jumps in voltage across the
you. This plan is sent to the motor cortex, and cell membrane. These are the action poten-
muscle groups are specified. The plan is also tials, ner\'e impulses, or spikes. They don't go
sent to the cerebellum and basal ganglia, and to zero voltage, but to -1-30 mv (as shown in
they modify it to refine the movement. The Fig. 3.3). That is, the inside of the neuron
cerebellum sends an update of the movement becomes positive. Action potentials are also
output plan to the motor cortex and brain- about 1 msec in duration and quickly repo-
stem. Descending pathways from the motor larize. The height of the action potential is
cortex and brainstem then activate spinal cord always about the same: —70 -I- 30 mv = ca.
networks, spinal motor neurons activate the 100 mv.
muscles, and you reach for the milk. If the How does the neuron communicate this
milk carton is fiill, when you thought it was information to the nextcell in line.' It does
almost empty, spinal reflex pathways will thisthrough the process of synaptic trans-
compensate for the extra weight that you mission. A cleft 200A wide separates neu-
didn't expect and activate more motor neurons. Each action potential in a neuron re-
rons.Then, the sensory consequences of your leases a small amount of transmitter

reach will be evaluated, and the cerebellum substance. It diflfuses across the cleft and at-
will update the movement — in this case, to taches to receptors on the next cell, which
accommodate a heavier milk carton. open up channels in the membrane and de-
polarize the cell. One action potential makes
Neuron —Basic Unit of the CNS only a small depolarization, called an excit-
atory postsynaptic potential, the EPSP. The
The lowest level in the hierarchy is the EPSP normally dies a\\'ay after 3 to 4 msec,
single neuron in the spinal cord. How does it and as a result, the next cell is not acti-
fiinction? What is its structure? To explore vated (2).

more fiilly the ways that neurons communi- But if the first cell fires enough action
catebetween the levels of the hierarchy of the potentials, there is a series of EPSPs, and they
nervous system, we need to review some of continue to build up depolarization to the
the simple properties of the neuron, including threshold voltage for the action potential in
the resting potential, the action potential, and the next neuron. This is called summation.
synaptic transmission. There are two kinds of summation, temporal
Remember that the neuron, when it is and spatial, and these are illustrated in Figure
at rest, always has a negative electrical charge 3.3. Temporal summation results in depo-
or potential on the inside of the cell, with re- larization because of synaptic potentials that
spect to the outside. Thus, when physiologists occur close together in time. Spatial sum-
Chapter Three Physiology OF Motor Control 51
+3U
^ J\.TJ V-
-50 1
-70 _.^
f ^
1
f
+30
)1
/
-50 -
-70
-
_^
t
--
_^saa/^
tttt
^ tttt
i
\;__
Figure 3.3. A schematic drawing illustrating important aspects of neuron physiology including the resting potential
(RP) of -70 mv and changes during an action potential, and the spatial {{op) and temporal (bottom) summation
properties of a neuron.
mation produces depolarization because of istp,' and the stimuli activating the neuron will
the action of multiple cells synapsing on the determine how it will respond to these signals
postsynaptic neuron. Spatial summation is re- inone mode or another (3).
ally an example of parallel distributed pro- With this overview of the essential ele-
cessing, since multiple pathways are affecting ments of the ner\'ous system, we can now turn
the same neuron (2). our attention to the heart of this chapter, an
The effectiveness of a gi\'en synapse in-depth discussion of the sensory-motor pro-
changes with experience. For example, if a cesses underlving motor control.
gi\'en neuron is activated over a short period
of time, it may show synaptic facilitation, in Sensory/Perceptual
which it releases more transmitter and there- Systems
fore more easily depolarizes the next cell. Al-
ternatively, a cell may also show defacilita- What is the role of sensation in the pro-
tion, or habituation. In this case, the cell is duction and control of movement? In the
depleted of transmitter, and thus is less effec- chapter on motor control theories, there were
tive in influencing the next cell. Manv mech- divergent views about the importance of sen-
anisms can cause synaptic facilitation or ha- sor\' input in motor control. Current neuro-
bituation in different parts of the nervous science research suggests that sensory infor-
system. Increased use of a given pathway can mation plays many different roles in the
result in synaptic facilitation. However, in a control of movement.
different pathway, increased use could result Sensor*' inputs ser\'e as the stimuli for
in defacilitation or habituation. Variations in reflexive movement organized at the spinal
the coding within the neuron's internal chem- cord level of the nervous system. In addition.
sensory information has a vital role in modu- The bag fiber is thicker than the chain fiber,
lating the output of movement that results and projects beyond the capsule, attaching to
from the of pattern generators in the
activity the connective tissue surrounding the extra-
spinal cord. An
example of this t>'pe of move- fijsal fiber fascicle. The chain fibers attach to
ment might be locomotor output from pat- the spindle capsule or to the bag fiber (Fig.
tern generators in the spinal cord. Likewise, 3.4^4). Each fiber type can be divided into
at the spinal cord level, sensor^' information equatorial, juxtaequatorial, and polar regions.
can modulate movement that results from The nuclear bag fiber has many spherical nu-
commands originating in higher centers of clei at the equatorial region, and gives a slow
the nervous system. The reason that sensation twitch contraction, while the nuclear chain fi-
can modulate all these t}'pes of movement is ber has a singlerow of nuclei, and gives a fast
that sensory receptors converge on the motor twitch contraction. The equatorial region is
neurons, considered the final common path- ver)' elastic, like a balloon fiiU of water.
way. But anodier role of sensory information The muscle spindle sends fibers into the
in movement control is accomplished via as- ner\'ous system via afferent fibers, and it is
cending pathways, which contribute to the controlled by the CNS via efferent fibers.
control of movement in much more complex Let's consider the afferent endings. The mus-
ways. cle spindle sends information to the nervous
system via two kinds of afferent fibers, the
Somatosensory System Piroup la and the ^roup 11 afferents. The la
fiber sensory endings wrap around the equa-
The somatosensory system, from the torial region, while the group II endings are
lowest to the highest level of the CNS hier-
on the juxtaequatorial region. The la afferents
archy, going from the reception of signals in
go to both bag and chain fibers, while the
the periphery to the integration and interpre-
group II afferents go to mainly to the chain
tation of those signals relative to other sensory
fibers(Fig. 3.4yl) (2, 5).
systems, is described in diis section. Pay close Both bag and chain muscle fibers are in-
attention to how hierarchical and parallel dis-
nervated by 7-motor neurons. Thecell bod-
tributed processing contribute to the analysis of the 7-motor neurons are inside the ven-
ies
of somatosensory signals.
tral horn of the spinal cord, intermingled with
the a-motor neurons, inner\'ating the extra-

PERIPHERAL RECEPTORS The 7-motor neuron endings are
fijsal fibers.
Muscle Spindle at the polar, striated region of the bag and the
chain muscle fibers, as shown in Figure 3.4^4.
Most muscle spindles are located in the There are two types of 7-fibers: {a) the 7-dy-
muscle belly of skeletal muscles. They consist namic, innervating the bag fiber, and [b) the
of specialized muscle fibers, called intrafusal 7-static, innervating the chain fiber.
fibers, surrounded by a connective tissue cap- Passive muscle stretch causes stretch of
sule (extrafijsal fibers are the regular muscle the equator of intrafiasal fibers. The equator
fibers). In humans, the muscles with the high- of the bag fiber is easily stretched, because it
est spindle density (spindles per muscle) are is so elastic, while the chain fiber equator
the extraocular, hand, and neck muscles. Is it stretches less rapidly because it is stiffer, with
surprising that neck muscles have such a high less nuclei. Remember, the la's are on the
spindle densit)'.* This is because we use these equator of the bag and chain fibers; thus, they
muscles in eye-head coordination as we reach ha\'e alow threshold to stretch and will follow
for objects and move about in the environ- changes in length easily. This means that the
ment (5). la afferents code the rate of stretch (a dynamic
Intrafrisal muscle fibers are much response) and the length of the muscle at the
smaller than extrafrisal fibers. There are t\vo end of stretch (static response) (5).
t)'pes: nuclear baq and nuclear chain fibers. The group II afferents end on the jux-
Chapter Three Physiology of Motor Control 53
Primary afferent (la)
Efferent
Efferent
Ks)
Tendon
Muscle
Figure 3.4. Anatomy of the muscle receptors: muscle spindle and Colgi tendon organ. A, The contents of the muscle
spindle showing the nuclear bag and chain fibers. B, The spindle-shaped Colgi tendon organ, located at the muscle-
tendon junction and connected to 1 5 to 20 muscle fibers.
taequatorial region of the chain fiber. This is How is information from the muscle
a stifFer region, and as a result, the group II spindle utilized during motor control.* Muscle
afFerents have a higher threshold than do the spindle information employed at many lev-
is
la's. The group II afferents code only muscle els of the CNS hierarchy. At the lowest level,
length and have no dvnamic response. Static it is involved in reflex acti\'ation of inuscles.
responses are linearly correlated with the However, as the information ascends the CNS
length of the muscle. Thus, the la afferents hierarchy, it is used in increasingly complex
respond well to slight tendon taps, sinusoidal and abstract ways. For example, it may con-
stretches,and even \ibration of the muscle tribute to our perception of our sense of ef-
tendon, while group II afFerents do not re- fort. In, addition, it is carried over different
spond to these stimuli (5). pathw ays to different parts of the brain, in this
way contributing to tiie parallel distributed aptic reflex, inhibiting its own muscle and ex-
nature of brain processing. citing its antagonist.
Stretch reflex loop. When a muscle is Researchers used to think that the GTO
stretched, it stretches the muscle spindle, ex- was only active in response to large amounts
citing the la afferents. They have excitatory of tension. So they hypothesized that the role
monosynaptic connections to the a motor of the GTO was to protect the muscle fi-om
neurons, which activate their own muscle and injur)'. Current research has shown that these
synergistic muscles. They also excite la inhib- receptors constantly monitor muscle tension
itory interneurons, which then inhibit the a and are very sensitive to even small amounts
motor neurons to the antagonist muscles. For of tension changes caused by muscle contrac-
example, if the gastrocnemius muscle is tion. A newly hypothesized fiinction of the
stretched, the muscle spindle la afferents in GTO is that it modulates muscle output in
the muscle are excited, and, in turn, excite the response to fatigue. Thus, when muscle ten-
a motor neurons of the gastrocnemius, which sion is reduced due to fatigue, the GTO out-
cause it to contract. The la afferent also ex- put is reduced, lowering its inhibitory effect
cites the la inhibitorv' interneuron, which in- on its own muscle (2, 5).
hibits motor neurons to the antagonist mus- It has also been shown that the GTOs
cle, the tibialis anterior, so that, if this muscle of the extensor muscles of the leg are active
was contracting, it now relaxes. The group II during the stance phase of locomotion and act
afferents also excite their own muscle, but di- to excite the extensor muscles and inhibit the
synaptically (2, 5). flexor muscles until the GTO is unloaded (6).
What is the purpose of 7 fiber activity, This is exactly the opposite of what would be
and when are these fibers active.' Whenever expected from the reflex when it is activated
there is a voluntary' contraction, there is a-y when the animal is in a passive state. Thus,
coactivation. Without this coactivation, spin- the reflex appears to have different properties
dle afferents would be silent during muscle under different task conditions.
contraction. With it, the nuclear bag and Researchers have hypothesized that the
chain fibers contract as well as the regular ex- fiinction of the muscle spindles and GTOs to-
trafusal fibers of the muscle, and thus the po- gether may be that of muscle stiffness regu-
lar region of the muscle spindle can't go slack. lation. Muscle stiffness may be defined as the
Because of this coactivation, if there is unex- force/unit length of a muscle. This is exacdy
pected stretch during the contraction, the what the GTO and muscle spindle are recip-
group la and II afferents will be able to sense rocally controlling: Force (GTO)/unit length
it, and compensate. (muscle spindle) (5).
Golgi Tendon Organs Joint Receptors
Golgi tendon organs (GTOs) are spin- How do joint receptors work and what
dle-shaped and located at the muscle-tendon is their function.' There are a number of dif-
junction (Fig. 3AB). They connect to 15 to ferent tv'pes of receptors within the joint itself,
20 muscle fibers. Afferent information from including Rufhni-t^'pe endings or spray end-
the GTO is carried to the nerx'ous system via ings, paciniform endings, ligament receptors,
the lb afferent fibers. Unlike the muscle spin- and free nen'e endings. They are located in
dles, they have no efferent connections, and different portions of the Mor-
joint capsule.
thus are not subject to CNS modulation. phologically, they share the same character-
This is how GTOs fiinction. The GTO istics as many of the other receptors found in
is sensitive to tension changes that result from the nen'ous system. For example, the liga-
either stretch or contraction of the muscle. ment receptors are almost identical to GTOs,
The GTO responds to as little as 2 to 25 g while the paciniform endings are idendcal to
force. The GTO reflex is an inhibitor^' disyn- pacinian corpuscles in the skin.
Chapter Three PmsioLOGY of Motor Control 55
Joint function has many intriguing as- pad of a cat's foot lighdy, it w ill extend it. This
pects. The joint receptor information is used is called the placing reaction, and it is found
of the hierarchy of season'
at several levels in human infants as well. In contrast, a sharp
processing. Some researchers have found that focal stimulus tends to produce withdrawal,
joint receptors appear to be sensitive onlv to or flexion, even when it is applied to exactiy
extreme joint angles (7). Because of this, the the same area of the foot. This is called the
joint receptors may provide a danger signal flexor withdrawal reflex, and it is used to pro-
about extreme joint motion. tect us from injur}'. The t\'pical pattern of re-
Other researchers have reported that sponse in the cutaneous reflex is ipsilateral
many indixidual joint receptors respond to a flexion,and contralateral extension, which al-
limited range of joint motion. This phenom- lows you to support your weight on the op-
enon has been called rntijie fractionation^ posite limb (mediated by group III and IV
with multiple receptors being activated in afFerents).
overlapping ranges. Afferent information It is important to remember that even
from joint receptors ascends to the cerebral though we consider reflexes to be stereo-
cortex and contributes to our perception of typed, they are modulated by higher centers,
our position in space. The CNS determines depending on the task and the context. Re-
joint position by monitoring which receptors member our example of the flexor reflex,
are actixated at the same time, and this allows which topically causes withdrawal of a limb
the determination of exact joint position. from a noxious stimulus. However, if there is
more at stake than not hurting yourself, such
Cutaneous Receptors as saving tiie life of your child, the CNS in-
hibits the activation of this reflex movement
There are also several tx'pes of cutaneous in fa\'or of actions more appropriate to the
receptors: (a) mechanoreceptors, including situation.
pacinian corpuscles, MerkeFs discs, Meis-
sner's corpuscles, Rufhni endings, and lanc- ROLE OF SOAIATOSENSATION AT
eolate endings around hair follicles, cfetecting THE SPINAL CORD LE\TL
mechanical stimuli; (b) thermoreceptors, de-
tecting temperature changes; and (c) nocicep- Information from cutaneous, muscle,
tors, detecting potential damage to the sidn. and joint receptors modifies the output of cir-
The number of receptors within the sensiti\e cuits at the spinal cord level that control such
areas of the skin, such as the tips of the fingers, basic activities as locomotion. In the late
is ver\' high, on the order of 2500 per square 1960s, Grillner performed experiments in
centimeter (8). which he cut the dorsal roots to the cat spinal
Information from the cutaneous system cord to eliminate sensor\' feedback from the
is also used in hierarchical processing in sev- periphery' (9). He stimulated the spinal cord
eral dift'erent ways. At lower levels of the CNS and was able to acti\ate the neural pattern
hierarchy, cutaneous information gives rise to generator for locomotor patterns. He found
reflex movements. Information from the cu- that low rates of repetiti\e stimulation gave
taneous system also ascends and provides in- rise to a walk, higher rates to a trot, and then
formation concerning body position essential a gallop. This suggests that complex move-
for orientation within the immediate emironments, such as locomotion, can be generated
ment. at the spinal cord level without supraspinal in-
The nervous system uses cutaneous influences or inputs from the peripher\'.
formation for reflex responses in various ways, If we don't need senson,' information to
depending on the extent and t\'pe of cutane- generate complex movement, does that mean
ous input. A light diffijse stimulus to the bot- there is no role for sensor)' information in its
tom of the foot tends to produce extension in execution? No. Hans Forssberg and his col-
the limb, as for example, when \ou touch the leagues have shown that senson' information
Fasciculus gracilis,
cuneatus
Free nen/e endings,

Merkel's disk
Upper
spinal
cord
eissner's corpuscle,
Pacinian corpuscle.
Muscle spindle
Lower
spinal
cord
Figure 3.5. Ascending sensory systems: the dorsal-column pathway containing information from touch and pressure
receptors.
Chapter Three Ph^'siology of Motor Control 57
modulates locomotor output in a ver>' elegant in the brainstem. The D-C pathway also con-
way ( 10). When he brushed the paw ofa spin- tains information from touch and pressure re-
alized cat with a stick during the swing phase ceptors, and codes especially for discrimina-
of walking, it caused the paw to flex more tive fine touch. This path\\ay is shown in
strongly and get out of the way of the stick. Figure 3.5 (11).
But during stance, the very same stimulation Where does this information go, and
caused stronger extension, in order to push how is it processed? The pathways synapse at
off more quickly and avoid the stick in this multiple levels in the nervous system, includ-
way. Thus, he found that the same cutaneous ing the medulla, where second-order neurons
input could modulate the step cycle in differ- become the medial lemniscal pathway and
ent fiinctional ways, depending on the con- cross over to the thalamus, synapsing with
text in which it was used. third-order neurons, which proceed to the so-
matosensory cortex. Every of the hier-
level
ASCENDING PATHWAYS archy has the modulate the

ability to
information coming into it from below.
Information from the trunk and limbs is Through synaptic excitation and inhibition,
also carried to the sensor\' cortex and cere- higher centers have the ability to shut oft" or
bellum. Two systems ascend to the cerebral enhance ascending information. This allows
cortex: the dorsal column-medial lemniscal higher centers to selectively tune (up or
(DC-ML) system and the anterolateral sys- down) the information coming from lower
tem. (Systems that ascend to the cerebellum centers.
are discussed later in the chapter.) These are As the neurons ascend through each
shown in Figures 3.5 and 3.6. They are ex- level to the brain, theinformation from die
amples of parallel ascending systems. Each re- receptors is increasingly processed to allow
lays information about somewhat different meaningflil interpretation of the information.
fijnctions, but there is some redundancy be- This is done by selectively enlarging the re-
tween the two pathways. What is the advan- ceptive field of each successive neuron.
tage of parallel systems? They give extra sub-
tlety' and richness to perception, by using Anterolateral System
multiple modes of processing information.
They also give a measure of insurance of con- The second ascending system, shown in
tinued hinction in case of injury (2, II). Figure 3.6, is the anterolateral (AL) system.
It consistsof the spinothalamic, spinoreticu-
Dorsal Column-Medial Lemniscal System lar, and spinomesencephalic tracts. These fi-
bers cross over upon entering the spinal cord
The formed mainly
dorsal columns are and then ascend to brainstem centers. The an-
by dorsal root neurons. These are thus first- terolateral system has a dual fianction. First, it
order neurons. The majority of the fibers transmits information on crude touch and
branch on entering the spinal cord, synapsing pressure, and thus contributes in a minor way
on interneurons and motor neurons to mod- to touch and limb proprioception. It also
ulate spinal activit)', and send branches to as- plays a major role in relaying information re-
cend in the dorsal column pathway toward the lated to thermal and nociception to higher
brain. What are the hinctions of the dorsal brain centers. All levels of the sensory pro-
column neurons? They send information on cessing hierarchy act on the AL system in the
muscle, tendon, and joint sensibilit}' up to the same manner as for the DC-ML system (II).
somatosensory cortex and other higher brain There is a redundancy of information in
centers. There is an interesting exception, both tracts. A lesion in one tract doesn't cause
however. Leg proprioceptors have their own complete loss of discrimination in any of these
private pathway to the brainstem, the lateral senses. However, a lesion in both tracts causes
column. They join the dorsal column pathway severe loss. Hemisection of the spinal cord
58 Section! THEORETICAL FRAMEWORK
Reticular
formation
Superficial end organs,

Ruffini corpuscle.
Free ending,
Krause's end bulb
Spinal
cord
Vessel wall,
free ending,
deep visceral
pain and temperature.

system, containing information on
Figure 3.6. Ascending sensory systems: the anterolateral
Chapter Three Pfhsiology of Motor Control 59
(caused by a serious accident, for example) body relati\'e to your en\ironment and the
would cause tactile sensation and propriocep- position of one bodv segment relative to an-
tion in the arms to be lost on the ipsilateral other (11, 12).
side (fibers haven't crossed yet), while pain Contrast sensitivity' is \er\' important to
and temperature sensation would be lost movement control, since it allows the detec-
on the contralateral side (fibers have al- tion of the shape and edges of objects. The
readv crossed upon entering the spinal cord) somatosensorx' cortex processes incoming in-
(in'. formation to increase contrast sensitivit\' so
that we can more easily identify' and discrim-
TH.\L\MUS inate between different objects through
touch. How does it do this.- It has been shown
Information from both the ascending that the receptive fields of the somatosensory'
somatosenson,' tracts, like information from neurons have an excitaton.' center and inhib-
\'irtually all senson' systems, goes through the itoPi' surround. This inhibitorx' surround aids
thalamus. This is a major processing center of in two-point discrimination through lateral
the brain, and a lesion in this area will cause inhibition.
severe senson,- (and motor) problems. How does lateral inhibition work.* The
cell that is excited inhibits the cells next to it,
SOMATOSENSORY CORTEX thus enhancing contrast bet\\'een excited and

nonexcited regions of the body. The receptors
The somatosensory cortex is a major pro- don't have lateral inhibition. But it comes in
cessing area for all the somatosensorx- modal- at the level of tiie dorsal columns, and at each
ities, and marks the beginning of conscious subsequent step in the relay. In fact, humans
awareness of somatosensation. Somatosen- ha\e a sufficiently sensitive somatosensor\' sys-
sor\- cortex is divided into t^vo major areas: tem to perceive the activation of a single tac-
primary somatoscnsoiy cortex (SI) (also called tile receptor in the hand (11, 12).
Brodmann's area 1, 2, 3a, and 3b); and sec- There are also special cells within the so-
otidaiy somatoscnsoiy coitcx (S\l) (Fig. 3.7^4). matosensorx' cortex that respond best to mov-
In SI, kinesthetic and touch information from ing stimuli and are directionally sensitive. One
body is organized
the contralateral side of the does not find this feature in the dorsal col-
in a somatotopic manner and spans four cy- umns, nor in the thalamus. These higher level
toarchitectural areas, Brodmann's areas 1,2, processing cells also have larger receptive
3a, and 3b (11). fields than the npical ceUs in SS, often encom-
It is in this area that we begin to see passing a number of fingers. These cells ap-
cross-modalir*' processing. That means that pear to respond preferentially when neigh-
joint receptors, muscle spindles, and cutane- boring fingers are stimulated. This could
ous information are now integrated to give us indicate their participation in such functions
information about movement in a given body as the grasp of objects.
area. This information is laid on top of a map It has recently been found that the re-
of the entire body, which is distorted to reflect ceptive fields of neurons in the somatosensory
the relative weight given sensor\' information cortex are not fixed in size. Both injun,- and
from certain areas, as you see in Figure iJB. experience can change their dimensions con-
For example, the throat, mouth, and hands siderably. The implications of these studies
are heavily represented because we need more are considered in the motor learning sections
detailed information to support the move- of this book (8).
ments that are executed by these structures. Somatosensorx- conex also has descend-
This is the beginning of the spatial processing ing connections to the thalamus, dorsal col-
that is essential to the coordination of mo\e- umn nucleus, and the spinal cord, and thus
ments in space. Coordinated movement re- has the abilit\' to modulate ascending infor-
quires information about the position of the mation coming through these structures.
Postcentral
Posterior
parietal lobe
SI primary
Lateral Medial
Figure 3.7. Somatosensory cortex and association areas. A, Located in the parietal lobe, the somatosensory cortex
contains three major divisions: the primary (SI), secondary (Sll), and the posterior parietal cortex. B, Sensory homunculus
showing the somatic sensory projections from the body surface. (Adapted from Kandel E, Schwartz )H, Jessell TM, eds.
Principles of neuroscience. 3rd ed. NY: Elsevier, 1 991 :368, 372.)
Chapter Three Ph^'Siology of Motor Control 61
Primary somatosensory cortex
Parietal-temporal-occipital
association cortex
Limbic association
cortex
Figure 3.8. Schematic drawing showing the locations of primary sensory areas, higher-level sensory association areas,
and higher-level cognitive (abstract) association cortices. (Adapted from Kandel E, Schwartz )H, Jessell TM, eds. Prin-
ciples of neuroscience. 3rd ed. NY: Elsevier, 1991:825.)
ASSOCIATION CORTICES or other animals cause problems with the

learning of skills that use information regard-
It is in the many association cortices that ing the position of the body in space. In ad-
we begin to see the transition from perception dition, certain cells in these areas appear to be
to action. It is here too that we see the inter- activated during visually guided movements,
play between cognitive and perceptual pro- with their activity becoming more intense
cessing. The association cortices, found in pa- when the animal attends to the movement.
rietal, temporal, and occipital lobes, include These findings support the hypothesis
centers for higher level sensor)' processing and that the parietal lobe participates in processes
higher level abstract cognitive processing. involving attention to the position of and ma-
The locations of these various areas are shown nipulation of objects in space (13).
in Figure 3.8. These e.xperimental results are fiirther
Within the parietal, temporal, and oc- supported by observations of patients with
cipital cortices are association areas, which are damage to the parietal lobes. Their deficits in-
hypothesized to link information from several clude problems with body image and percep-
senses. Area 5 of the parietal cortex is a thin tion of spatial relations, which may be ver)'
strip posterior to the postcentral g\'rus. After important in both postural control and vol-
intermodalit)' processing has taken place untary' movements. Clearly, lesions to this
within area SI, outputs are sent to area 5, area don't simply reduce the ability' to per-
which integrates information between body ceive information coming in from one part of
parts. Area 5 connects to area 7 of the parietal the body; in addition, they can affect the abil-
lobe. Area 7 also receives processed visual in- ity' to interpret this information.
formation. Thus, area 7 probably combines For example, people with lesions in the
eye-limb processing in most visually triggered right angular gyrus (the nondominant hemi-
or guided activities. sphere), just behind area 7, show complete
Lesions in areas 5 or 7 in either humans neglect of the contralateral side of body, ob-
jects, and drawings. This is called agnosia or other, and the motion of our body. When vi-
the inabilitA' to recognize. When their own sion plays this role, it is referred to as visual-
arm or leg is passively moved into their visual proprioception, which means that it gives us
field, they may claim that it isn't theirs. In cer- information not only about the environment,
tain cases, patients may be totally unaware of but about our own body. Later chapters show
the hemiplegia that accompanies the lesion how vision plays a key role in the control of
and may thus desire to leave the hospital early posture, locomotion, and manipulator\' fiinc-
since they are unaware that they have any tion. In the following sections, we consider
problem (13). Many of these same patients the anatoni}' and physiolog)' of the visual sys-
show problems when asked to cop\' drawn fig- tem to show how it supports these roles in
ures. They may draw objects in which one- motor control.
half of it is missing. This is called constrtic-
tional apraxia. Larger lesions may cause the PERIPHERAL VISUAL SYSTEM
inability' to operate and orient in space or the
inability' to perform complex sequential Photoreceptors
tasks.
When right-handed patients have le- Let's fiirst look at an overall view of the
sions in the left angular g>'rus (the dominant e\'e. The eye is a great instrument, designed
hemisphere), they show such svmptoms as to focus the image of the world on the retina
confiision bet\veen and right, difficult)' in
left with great precision. As illustrated in Figure
naming though they can sense
their fingers, 3.9, light enters the eye through the cornea
touch, and difficult)' in writing, though their and is focused by the cornea and lens on the
motor and sensor}' fimctions are normal for retina at the back of the eye. An interesting
the hands. Alternatively, when patients have feature of the retina is that light must travel
lesions to both sides of these areas, they often through all the layers of the e)'e and the neural
have problems attending to \'isual stimuli, in layers of the retina before it hits the photo-
using vision to grasp an object, and in making receptors, which are at the back of the retina,
\oluntar\' e\'e mo\'ements to a point in space facing away from the light source. Luckily,
(13). these la\'ers are nearly transparent.
We have just taken one sensory system, There are X\\o t\pes of photoreceptor
the somatosenson,' system, from the lowest to cells: the rods and the cones. The cones are
the highest level of the CNS hierarchv, going flinctional for vision in normal daylight and
from the reception of signals in the periphen,' are responsible for color vision. The rods are
to the integration anci interpretation of those responsible for x-ision at when the
night
signals relative to other sensor\- systems. We amount of light is ver\' low and too weak to
have also looked at how hierarchical and par- activate the cones. Right at the fovea, the rest
allel distributed processing have contributed of the layers are pushed aside so the cones can
to the analysis of these signals. We are now- receive the light in its clearest form. The blind
going to look at a second sensor)- system, the spot (where the optic nene leaves the retina)
\isual system, in the same wa\'. has no photoreceptors, and therefore we are
blind in this one part of the retina. Except for
Visual System 20 times more rods than
the fovea, there are
cones However, cones are more
in the retina.
Vision ser\'es motor control in a number important than rods for normal vision, be-
of ways. Vision allows us to identify' objects in cause their loss causes legal blindness, while
space, and to determine their movement. total loss of rods causes onlv night blindness
When vision plays this role, it is considered an (14).
exteroceptive sense. But vision also gives us Remember that sensor)' dift'erentiation
information about where our body is in space, is a key aspect of senson' processing that sup-
about the relation of one body part to an- ports motor control. To accomplish this, the
Lateral geniculate
nucleus
Calcarine
fissure
Ganglion
neuron
Figure 3.9. Illustration of the eye, its relationship to the horizontal and vertical cells (insert), and the visual pathways
from the retina to the thalamus, midbrain, and area 1 7 of the cerebral cortex. (Adapted from Kandel E, Schwartz |H,
Jessell TM, eds. Principles of neuroscience. 3rd ed. NY: Elsevier, 1 991 :401 ,415, 423.)
64 Section I THEORETICAL FRAA-IEWORK
visual system has to identify objects and de- reached, the pathways and fiinctions of the
termine if they are moving. So how are object difterent classes of cells are straightforward.
identification and motion sense accomplished Let's first look at the bipolar cell path-
in the visual system? There are two separate way. There are two types of pathways that in-
pathways to process them. We will follow volve bipolar cells, a "direct" pathway and a
these pathways from the retina all the way up "lateral" pathway. In the direct pathway, a
to the visual cortex. In addition, contrast sen- cone, for example, makes a direct connection
sitivity is used in both pathways to accomplish with a bipolar cell, which makes a direct con-
the goal of object identification and motion nection with a ganglion cell. In the lateral
sense. Contrast sensitivity' enhances the edges pathway, activity of cones is transmitted to the
of objects, giving us greater precision in per- ganglion cells lateral to them through hori-
ception. As in the somatosensor)' system, all zontal cells or amacrine cells. If you look at
three processes are used extensively in the vi- Figure 3.9, you will see these organizational
sual system. This processing begins in the ret- possibilities (14, 15).
ina. So let's first look at the cells of the retina, In the direct pathway, cones (or rods)
so that we can understand how they work to- connect directly to bipolar cells with either
gether to process information (14). "on-center" or "off-center" receptive fields.
The receptive field of a cell is the specific area
Vertical Cells
of the retina to which the cell is sensitive,
when that part of the redna is illuminated.
In addition to the rods and cones, the The receptive field can be either excitatory' or
retina contains bipolar ccWs und jjanjilion cells, inhibitory, increasing or decreasing the cell's
which you might consider "vertical" cells, membrane potendal. The receptive fields of
since they connect in series to one another but bipolar cells (and ganglion cells) is circular. At
have no lateral connections (Fig. 3.9). For ex- the center of the retina, the receptive fields are
ample, die rods and cones make direct syn- small, while in the periphen,', receptive fields
aptic contact with bipolar cells. The bipolar are large. The term "on-center" means that
cells in turn connect to the ganglion cells. the cell has an excitatory' central portion of the
And the ganglion cells then relay visual infor- receptive field, with an inhibitor)' surrounding
mation to the CNS, by sending axons to the area. "Off-center" refers to the opposite case
lateral geniculate nucleus and superior coUic- of an inhibitor)' center and excitator\' sur-
ulus as well as to brainstem nuclei (14-15). round (15).

How do the cells take on their antago-
Horizontal Cells nistic surround characteristics.^ It appears that

horizontal cells in the surround area of the
There is another class of neurons in the bipolar cell receptive field (RF) make connec-
retina, which we are calling "'horizontal''' tions onto cones in the center of the field.
cells. These neurons modulate the flow of in- When light shines on the peripher\' of the re-
formation within the retina by connecting the ceptive field, the horizontal cells inhibit the
"vertical" cells together laterally. These are cones adjacent to them.

called the horizontal and amacrine cells. The Each t\'pe of bipolar cell then synapses
horizontal cells mediate interactions bet\veen with a corresponding of ganglion cell:
t)'pe
the receptors and bipolar cells, while the am- on-center and off-center, and makes excit-
acrine cells mediate interactions between bi- atory' connections with that ganglion cell (14,
polar and ganglion cells. The horizontal cells 15).
and amacrine cells are critical for achieving On-center cells giveveiy few action po-
contrast sensitivity. Though it may appear tentials in the dark, and are activated when
that there are complex interconnections be- their RF is illuminated. When the periphen'
tween the receptor cells and other neurons of their RF is illuminated, it inhibits the effect
before the final output of the ganglion cells is of stimulating the center. Off-center ganglion
Chapter Three Pi-nsiOLOGY of Motor Control 65
cells likewise show inhibition when light is ap- Thus, the optic ner\'es from the left and
plied to the center of their RP, and they fire right eyes leave the retina at the optic disc, in
at the fastest rate just after the light is turned the back. They travel to the optic chiasm
off. They also are activated if light is applied where the nerves from each eye come to-
only to the penpher\' of their RF. gether, and axons from the nasal side of the
Ganglion cells are also influenced by the eyes cross, while those from the temporal side
activin' of amacrine cells. Many of the ama- do not cross. At this point, die optic ner\'e
crine cells function in a similar manner to hor- becomes the optic tract. Because of this re-
izontal cells, transmitting inhibitor\' inputs sorting of tlie optic nen'es, the left optic tract
from nearbv bipolar cells to the ganglion cell, has a map of the right visual field. This is sim-
increasing contrast sensitivity. ilar to what we found for the somatosensory
These two t}'pes of pathways (on- and system, in which information from the op-
off-center) for processing retinal information posite side of the body was represented in the
are two examples of parallel distributed pro- thalamus and cortex.

of similar information within the ner-
CMJz'w^/T One of die targets of cells in the opdc
vous system.We talked about a similar center- tract is the lateral geniculate nucleus (LGN)
mrroiind inhibition in cutaneous receptor of the thalamus. The lateral geniculate nu-
receptive fields. What is the purpose of this cleus has six layers of cells, which map the
t^'pe of inhibition? It appears to be ver>' im- contralateral \isual field. The ganglion cells
portant in detecting contrasts betsveen ob- from different areas project onto specific
jects, rather than the absolute intensity of points in the LGN, but just as we find for
light prociuced or reflected by an object. This somatosensory maps of the body, certain areas
inhibition allows us to detect edges of objects are represented much more strongly than oth-
ver\' easily. It is ver\' important in locomotion, ers. The fovea of the retina, which we use for
when we are walking down stairs and need to high acuit)' vision, is represented to a far
see the edge of the step. It is also important greater degree than the peripheral area. Each
in manipulator)' fiinction in being able to de- layer of the LGN gets input from only one
termine the exact shape of an object for grasp- eye. The first two layers (most ventral) are
ing. called the majjnocellular (large cells) layers,
The ganglion cells send their axons, via and layers four through six are called the par-
the optic nerve, to three different regions in vocelhilar (small cells) layers. The projection
the brain, the lateral geniculate nucleus, the cells of each layer send axons to the visual cor-
pretectum, and the superior colliculus (16) tex (16).
(Fig. 3.9). The receptive fields of neurons in the
LGN are very similar to those found in the
CENTRAL VISUAL PATHWAYS ganglion cells of the retina. There are separate
on-center and off-center receptive field padi-
Lateral Geniculate Nucleus ways. The magnocellular layers appear to be
involved in the analysis of mo\'ement of the
To understand what part of the retina visual image, and the coarse details of an ob-
and visual field are represented in these dif- ject, while the parvocellular layers function in
ferent areas of the brain, let's first discuss the color vision and a more detailed structural
configuration of the \isual fields and hemiret- analysis. Thus, magnocellular layers will be
ina. The left half of the visual field projects on more important motor functions like bal-
in
the nasal ( medial — next to the nose ) half of ance control, where movement of the visual
the retina of the left eye and the temporal (lat- field gives us information about our body
eral) half of the retina of the right eye. The sway, and in reaching for moving objects. The
right visual field projects on the nasal half of panocellular layers will be more important in
the retina of the right eye and the temporal the final phases of reaching for an object,
half of the retina of the left eye (16). when we need to grasp it accurately.
Superior CoUiculus Output ceUs from priman' visual cortex then

project to Brodmann's area 18. From area 18,
Ganglion cell axons in the optic tract neurons project to the medial temporal cortex
also terminate in the superior colliculus(m ad- (area 19), the inferotemporal cortex (areas 20,
dition to indirect visual inputs coming from 21 ), and the posterior parietal cortex (area 7).
the visual cortex). It has been hypothesized In addition, outputs go to the superior coUic-
that the superior coUiculus maps the visual ulus and also project back to the LGN (feed-
space around us in terms of not only visual, back control). The primar\' visual cortex con-
but also auditor)' and somatosenson,' cues. tains a map of the retina with topographic
The three senson,' maps in the superior coUic- mapping. There are six additional represen-
ulus are different from those seen in the sentations of the retina in the occipital lobe alone
sor)' cortex. Body areas here are not mapped (16).
in terms of density' of receptor cells in a par- The receptive fields of ceUs in the visual
ticular area, but in terms of their relationship cortex are not circular anymore, but linear:
to the retina. Areas close to the retina (the the light must be in the shape of a line, a bar,
nose) are given more representation than or an edge to excite them. These cells are clas-
away (the hand). For any part of the
areas far sified as simple or complex cells. Simple ceUs
body, the visual, auditor}', and somatosensory respond to bars, with an excitaton,' center and
maps are aligned, in the different layers of the an inhibitor,' surround, or vice versa. They
coUiculus (16). also have a specific axis of orientation, for
In addition to these three maps, located which the bar is most effective in exciting the
in the upper and middle of the seven lasers of cell. All axes of orientation for all parts of the
the coUiculus, there motor map in the
is a retina are represented in the visual cortex. Re-
deeper layers of the coUiculus. Through these sults of experiments by Flubel and VViesel (18)
output neurons, the coUiculus controls sac- suggest that this bar-shaped receptive field is
cadic eye movements that cause the eye to created from many geniculate neurons with
move toward a specific stimulus. The superior partially overlapping circular receptive fields
coUiculus then sends outputs to (a) regions of in one line, comerging onto a simple cortical
the brainstem that control eye movements, ceU. It has been suggested that complex ceUs
[b) the tectospinal tract, mediating the reflex have convergent input from many simple
control of the neck and head, and (c) the tec- ceUs. Thus, their receptive fields are larger
topontine tract, which projects to the cere- than simple ceUs, and have a critical axis of
beUum, for further processing of eye-head orientation. For many complex ceUs, the most
control (16). useful stimulus is movement across the
field.
Pretectal region The visual cortex is di\ided into col-
umns, with each column consisting of ceUs
Ganglion cells also terminate in the pre-
with one axis of orientation, and neighboring
tectal region. The pretectal region is an im-
columns receiving input from the left vs. the
portant x-isual reflex center involved in pupil-
right eve. Hubel and Wiesel used the name
lary' eye reflexes, in which the pupil constricts
hypcrcolumn to describe a set of columns from
in response to light shining on the retina.
one part of the retina, including aU orienta-
tion angles for the two eyes (17).

PRIMARY VISUAL CORTEX
From the LGN, axons project to the vi- HIGHER-ORDER \TSUAL CORTEX
sual cortex also called striate cortex to Brod-
(
)
mann's area 17, which is in the occipital lobe Central visual processing pathways con-
(Fig. 3.9). The inputs from the two eyes al- tinue on to include ceUs in the priman' visual
ternate throughout the striate cortex, produc- cortex, located in the occipital lobe, and ceUs
ing what are caUed ocular dominance columns. in the higher-order visual cortices, located in
the temporal and parietal cortex as well. These this information appears to require attention,
areas are shown in Figure 3.9. Higher order- which may be mediated by subcortical struc-
cortices are involved in the integration of so- tures such as the superior colliculus, as well as
matosensory and visual information underly- cortical areas, such as the posterior parietal
ing spatial orientation, an essential part of all and prefrontal cortex. It has been hypothe-
actions. This interaction between visual anci sized that the CNS takes information related
somatosensory inputs within higher-order as- to color, size, distance, and orientation and
sociation cortices was previously discussed in organizes it into a "master map" of the image
the somatosensory section of this chapter. (21). Our attentional systems allow us to fo-
It has been suggested that the cells cus on one small part of the master map as we
within the visual pathways contribute to a hi- identif}' objects or move through space.
erarchy within the visual system, with each
level of the hierarchy increasing the visual ab- Vestibular System
straction (19). In addition, there are parallel
pathways through which this information is The vestibular system is sensitive to two
processed. These pathways involve the mag- types of information: the position of the head
nocellular layers (processing movement and in space and sudden changes in the direction
coarse detail —processing "where") and

the of movement of the head. Although we aren't
parvocellular layers (processing fine detail and consciously aware of vestibular sensation, as
—
color processing the "what") of the lateral we are of the other senses, vestibular inputs
geniculate nucleus (20). are important for the coordination of many
There is interesting clinical evidence to motor responses and help to stabilize the eyes
support the existence of these parallel pro- and to maintain postural stability during
cessing pathways. A perceptual deficit called stance and walking. Abnormalities within the
"movement agnosia" occurs after damage to vestibular system result in sensations such as
the medial temporal (MT) area or the medial dizziness or unsteadiness, which do reach our
superior temporal (MST) regions of the cor- awareness, as well as problems with focusing
tex. Patients show a specific loss of motion our eyes and in keeping our balance.
perception without any other perceptual Like other sensor^' systems, the vestib-
problems. Other patients with damage to ular system can be di\'ided into two parts, a
Brodmann's areas 18 or 37 lose only color peripheral and a central component. The pe-
vision, but can still identify form (achroma- ripheral component consists of the sensory re-
topsia). Still other patients lose the ability to ceptors and 8th cranial nerve, while the cen-
identify forms (with damage to areas 18, 20, tral part consists of the four vestibular nuclei
21) (20). as well as the ascending and descending tracts.

How do we sense motion? The mag-
nocellular pathway continues to area and MT PERIPHERAL RECEPTORS
MST and the visual motor area of the parietal
lobe. In MT, the activity in the neurons is re- Let's first look anatomy of the
at the
lated to the velocity and movement direction vestibular system (Fig. 3.10). The vestibular
of objects. This information is then flirther system is part of the membranous labyrinth of
processed in MST for visual perception, pur- the inner ear. The other part of the labyrinth
suit eye movements, and guiding the move- is the cochlea, which is concerned with hear-
ments of the body through space. ing. The membranous labyrinth consists of a
How do we take the information pro- continuous series of tubes and sacs located in
cessed by these parallel pathways and organize the temporal bone of the skull. The membra-
it into a perceptual whole? This process by nous labyrinth is surrounded by a fluid called
which the brain recombines information pro- the perilymph, and filled with a fluid called the
cessed in its different regions is called the endolymph. The endolymph has a densit)'
"binding problem." The recombination of greater than water, giving it inertial charac-
Ascending tract
Superior
Medial
vestibular
vestibular
nucleus
Lateral
vestibular
Anterior semicircular canal
nucleus
Inferior
vestibular Posterior semicircular canal
nucleus
Lateral semicircular canal
Ampulla
Right AC
Figure 3.10. Vestibular system. A, Schematic drawing of the membranous labyrinth (otoliths and semicircular canals)
and the central connections of the vestibular system. Shown are the ascending vestibular inputs to the oculomotor
complex, important and the descending vestibulospinal system, important for posture and balance.
for stabilizing gaze,
B, Location of the paired semicircular canals within the temporal boneof the skull. AC = anterior canal; PC = posterior
canal.
teristics, which arc important to the way the 100 Hz, so they have a wide range
cells fire at
vestibular system functions. of frequencies for modulation. Thus, changes
The vestibular portion of the labmnth in firing fi-equenq,'of the neurons either up
includes five receptors: three semicircular ca- or down of this tonic
are possible because
nals, the utricle, and the saccule. resting discharge, which occurs in the absence
of head motion (22, 23).
Semicircular Canals Because canals on each side of the head
are appro.vimately parallel to one another,
The semicircular canals function as an- they work together in a reciprocal fashion.
gular accelerometers. They lie at right angles The two horizontal canals work together,
to each other, and are named the anterior, while each anterior canal is paired with a pos-
posterior, and horizontal canals on either side terior canal on the opposite side of the head.
of the head (Fig. 3.10). At least one pair is When head motion occurs in a plane specific
affected b\' an)' given angular acceleration of to a pair of canals, one canal will be excited,
the head or body. The sensor\- endings of the while its paired opposite canal will be h\per-
semicircular canals are in the enlarged end of polarized.
each canal, which is called the aiiipiilla, near Thus, angular motion of the head, ei-
its junction with the utricle. Each ampulla has ther horizontal or vertical, results in either an
an anipiillajy crest, \\hich contain the vestib- increase or decrease in hair cell acti\it\", which
ular hair cells. The upward
hair cells project produces a parallel change in the frequenc)" of
into the cupula Latin for small inverted cup I,
(
neuronal acti\it\- in paired canals. Receptors
made of gelatinous material, and extending to in the semicircular canal are \cr\ sensitive:
the top of the ampulla, preventing movement they respond to angular accelerations of .1°/
of the endol\Tnph past the cupula. The hair sec-, but do not respond to steady-state mo-
cells are the vestibular receptors, and are in- tion of the head. During prolonged motion
ner\"ated by bipolar sensor\' neurons, which of the head, the cupula returns to its resting
are part of the 8th nerve. Their cell bodies are position, and firing frequencx' in the neurons
located in the vestibular ganglion (22, 23). returns to its stead\" state.
How
do the semicircular canals signal
head motion to the nervous system.- UTien the Utricle and Saccule
head starts to rotate, the fluid in the canals
doesn't move initially, due to its inertial char- The utricle and saccule pro\ide infor-
acteristics. As a result, the cupula, along with mation about body position with reference to
its hair cells, bends in the opposite direction the force of gra\it}- and linear acceleration or
to head movement. \Mien head motion stops, movement of the head in a straight line. On
the cupula and hair ceUs are deflected m the the wall of these structures is a thickening
opposite direction, that is, the direction in where the epithelium contains hair cells. This
which the head had been mo\ing. area is called the macula Latin for spot), and
(
WTien the hair cells bend, the\- cause a is where the receptor cells are located. The
change in the firing frequence of the ner\"e, hair cells project tufts or processes up into a
depending on which way the hair cells are gelatinous membrane: the otolith orân
bent. For each hair ceU, there is a kinocilium —
(Greek, from "lithos" stone). The otolith
(the tallest tuft) and 40 to 70 stereocilia, organ has many calcium carbonate crystals
which increase in length as they get closer to called otoconia, or otoliths 12 ( ).
the kinocilium. Bending the hair cell toward The macula of the utricle lies in the hor-
the kinocilium causes a depolarization of the izontal plane when the head is held horizon-
hair cell and an increase in firing rate of the tally, so the otoliths rests upon
it. But if the
bipolar cells of the 8th ner\e, and bending head is tilted, or accelerates, the hair cells are
awa\' causes h\perpolarization and a decrease bent by the movement of the gelatinous mass.
in firing rate of bipolar cells. At rest, the hair The macula of the saccule lies in the vertical
plane when the head is positioned normally, Ascending information from the vestib-
so it responds selectively to vertically directed ular system to the oculomotor complex is re-
linear forces. As in the semicircular canals, hair sponsible for the vestibulo-oculomotor i-eflex,
cells in the otoliths respond to bending in a which rotates the eyes opposite to head move-
directional manner. ment, allowing the gaze to remain steady on
an image even when the head is moving (22,
CENTRAL CONNECTIONS 23).
Vestibular nystagmus is the rapid alter-
Vestibular Nuclei nating movement of the eyes in response to
continued rotation of the body. One can cre-
Neurons from both the otoliths and the nystagmus in a subject by rotat-
ate vestibular
semicircular canals go through the 8th nerve, ing the person seated on a stool to the left:
and have their cell bodies in the vestibular when the acceleration first begins, the eyes go
ganglion (Scarpa's ganglion). The axons then slowly to the right, to keep the eyes on a single
enter the brain in the pons, and most go to point in space. When the eyes reach the end
the floor of the medulla, where the vestibular of the orbit, they "reset" by moving rapidly
nuclei are located. There are four nuclei in the to the left; then they move again slowly to the
complex: the lateral vestibular nucleus (Dei- right.
ters'), the medial vestibular nucleus, the su- This alternating slow movement of the
perior vestibular nucleus, and the inferior, or eyes in the direction opposite head move-
descending vestibular nucleus. A certain por- ment, and rapid resetting of the eyes in the
tion of the vestibular neurons go from the direction of head movement, is called nystag-
sensory receptors to the cerebellum, the retic- mus. It is a normal consequence of accelera-
ular formation, the thalamus, and the cerebral tion of the head. However, when nystagmus
cortex. The central connections of the vestib- occurs without head movement it is usually
ular system are pictured in Figure 3.10. an indication of dysfijnction in the peripheral
The lateral vestibular nucleus receives or central nervous system.
input from the utricle, semicircular canals, Postrotatory nystagmus is a reversal in
cerebellum, and spinal cord. The output con- the direction of nystagmus, and occurs when
tributes to vestibulo-ocular tractsand to the a person who is spinning stops abrupdy. Pos-
lateral vestibulospinal tract, which activates trotatory nystagmus has been used clinically
antigravity muscles in the neck, trunk, and to evaluate the ftinction of the vestibular sys-
limbs. tem (24).
Inputs to the medial and superior nuclei The vestibular apparatus has both static
are from the semicircular canals. The outputs and dynamic ftinctions. The dynamic fimc-
of the medial nucleus are to the medial ves- tions are controlled mainly by the semicircular
tibulospinal tract (MVST), with connections canals, allowing us to sense head rotation and
to the cervical spinal cord, controlling the angular accelerations, and allowing the con-
neck muscles. Information in the MVST plays trol of the eyes through the vestibulo-ocular
an important role in coordinating interactions reflexes. The static functions are controlled by
between head and eye movements. In addi- the utricle and saccule, allowing us to monitor
tion, neurons from the medial and superior absolute position of the head in space, and are
nuclei ascend to motor nuclei of the eye mus- important in posture. (The utricle and saccule
cles, and aid in stabilizing gaze during head also detect linear acceleration, a dynamic
motions. ftmction.)
The inputs to the inferior vestibular nu-
cleus include neurons from the semicircular Action Systems
canals, utricle, saccule, and cerebellar vermis,
while the outputs are part of the vestibulo- The action system includes areas of the
spinal tract and vestibuloreticular tracts. ner\'ous system such as motor cortex, cere-
bellum, and basal ganglia, which perform pro- most detailed control (the
that require the
cessing essential to the coordination of move- mouth, and hand), allowing finely
throat,
ment. graded movements, are most highly repre-
Remember our example presented in sented (26).
the beginning of this chapter. You're thirst\' Inputs to the motor areas come from
and want to pour some milk from the milk the basal ganglia, the cerebellum, and from
carton in front of you into a glass. We've al- sensorN' areas, including the peripher)' (via the
ready seen how sensory structures help you thalamus), SI, and sensory association areas in
form the map of your body in space and locate the parietal lobe. Interestingly, MI neurons
the milk carton relative to your arm. Now you receive senson' inputsfrom their own muscles
need to generate the movement that will al- and also from the skin above the muscles. It
low you to pick up the carton and pour the has been suggested that this transcortical
milk. You will need a plan to move, you will pathway might be used in parallel with the
need to specify specific muscles (both timing spinal reflex pathway to give additional force
and force), and you will need a way to modify output in the muscles when an unexpected
and refine the movement. So let's look at the load is encountered during a movement {27).
structures that allow you to do that. This pathway has also been hypothesized to
be an important proprioceptive pathway fianc-
Motor Cortex tioning in postural control (25).
The motor cortex is situated in the fron- CORTICOSPINAL TRACT

tal lobe and consists of a number of different
processing areas, including the primary motor Outputs from the motor cortex contrib-
cortex (MI) the supplementary motor area ute to the corticospinal tract (also called the
(SM), (occasionally called Mil), and the pre- pyramidal tract) and often make excitatory
motor cortex (Fig. 3.11^4). These areas inter- monosynaptic connections onto a motor
act with sensory processing areas in the pari- neurons, in addition to polysynaptic connec-
etal lobe and also with basal ganglia and tions to 7 motor neurons, which control mus-
where we want to
cerebellar areas to identify cle spindle length. In addition to their mono-
move, to plan the movement, and finally, to synaptic connections, corticospinal neurons
execute our actions (25). make many polysynaptic connections through
All three of these areas have their own interneurons within the spinal cord.
somatotopic maps of the body, so that if dif- The corticospinal tract includes neurons
ferent regions are stimulated, differentmus- from primary motor cortex (about 50%), sup-
cles and body parts move. The primary motor plementary motor cortex, premotor areas,
cortex (Brodmann's area 4) contains a very and even somatosensory' cortex (Fig. 3.12).
complex map of the body. There is often a The fibers descend ipsilaterally from the cor-
one-to-one correspondence between cells tex through the internal capsule, the mid-
stimulated and the activation of individual a brain, and the medulla. In the medulla, the
motor neurons in the spinal cord. In contrast fibers concentrate to form "pyramids" and
to a one-to-one activation pattern typical of near the junction of the medulla and the spi-
neurons in the primar}' motor cortex, stimu- nal cord, most (90%) cross to form the lateral
lation of neurons in the premotor and supple- corticospinal tract. The remaining 10% con-
mentary' motor areas (Brodmann's area 6) tinue uncrossed to form the anterior cortico-
typically activates multiple muscles at multiple spinal tract. The majorit)' of the anterior cor-
joints, giving coordinated actions. ticospinal neurons cross just before they
The motor map, or motor homunculus terminate in the ventral horn of the spinal
(shown in Fig. 3.115), is similar to the sen- cord. Most axons enter the ventral horn and
son,' map in the way it distorts the represen- terminate in the intermediate and ventral ar-
tations of the body. In both cases, the areas eas on interneurons and motor neurons.
Primary motor cortex

Supplementary
motor area Primary somatosensory cortex
Medial Lateral
Figure 3.11. Motor cortex. A, Lateral view ot the brain showing the location ot the primary motor cortex, supple-
mentary motor area, and premotor cortex. B, Motor homunculus. (Adapted from Kandel E, Schwartz )H, Jessell TM,
eds. Principles of neuroscience. 3rd ed. NY: Elsevier, 1 991 :61 0, 613.)
Chapter Three PmsiOLOGY of Motor Control 73
Trunk Knee Toes Figure 3.12. Pyramidal or (corticospinal) tract.
Corona yA
radiata /
Anterior corticospinal Lateral corticospinal

tract (ventral. tract fibers (posterior,
uncrossed fibers crossed)
Efferent segmental
nerve
Lower
cord
FUNCTION extension movements. He found that the fir-
ing rate of the corticospinal neurons codes («)

Motor Corte.x the force used to move a limb, and h) in some
(
cases, the rate of change of force. Thus, both

What is the specific function of motor absolute force and the speed of a movement
cortex in movement control.' Evarts (28) re- are controlled by the primar\- motor cortex.
corded the activity- of corticospinal neurons in No\\', think about a tvpical movement
monkeys while thev made wrist flexion and that we make — reaching for the carton of
74 Section I THEORETICAL FR.'VMEWORK
milk, for example. How does the motor cor- Simple finger flexion
tex encode the execution of such a complex (performance)
Somatosensory
movement? Researchers performed experi- cortex
ments in which a monkey made arm move-
ments to many different targets around a cen-
tral starting point (29). They found that there
were specific movement directions where each

neuron was activated maximally, yet each re-
sponded for a wide range of movement direc-
tions. To explain how movements could be
finely controlled when neurons are so broadly
timed, these researchers suggested that ac-
tions are controlled by a population of neu-
rons. The activity of each of the neurons can Complex finger movement
(performance)
be represented as a vector, whose length rep- Supplementary
resents the degree of activity' in any direction. motor area
The sum of the vectors of all of the neurons

\\ould then predict the movement direction
and amplitude.
If this is the case, does it mean that
whenever we make a movement, for example,
with our ankle, the exact same neurons are
activated in the priman' motor cortex.' No. It
has been shown that specific neurons in the
cortex, activated when we pick up an object,
may remain totally silent when we make a sim-
movement such as a gesture in anger. This
ilar
is a ven- important point to understand be-

cause it implies that there are many parallel
motor pathways out an action se-
for carrying
quence, just as there are parallel pathways for
senson,' processing. Thus, simply by training
a patient in one situation, we can't automat-
ically assume that the training will transfer to
all other activities requiring the same set of
muscles (25).
Supplementan,- and Premotor Cortex
Figure 3.13. Changes in blood flow during different

What are the fianctions of the supple- motor behaviors indicating the areas of the motor cortex
mentan,' and premotor areas.' Roland and his involved in the behavior. (Adapted from Roland PE, Lar-
colleagues (30) performed some interesting sen B, Lassen NA, Skinhof E. Supplementary motor area
and other cortical areas in organization of voluntary
experiments with humans, which have begun
movements in man. J Neurophysiol 1980;43:1 18-136.)
to clarify their fijnctions. He asked subjects to
perform tasks ranging ft-om ver^' simple to
complex mo\'ements, and while they were
making the movements, he assessed the
amount of cerebral blood flow in different ar-
eas of the brain. (To measure blood flow, one
Chapter Three PHYSIOLOGY OF MOTOR Control 75
injects short-lived raciioacti\e tracer into the Therefore, the premotor area probably has a
blood, then measures the radioactivity' in dif- role in motor learning (30).
ferent brain areas with detectors on the scalp.
Higher- Level Association Areas
As shown in Figure 3.13, when subjects
were asked to perform a simple task (simple ASSOCIATION .\REAS OF THE
repetitive movements of the index finger or FRONTAL REGION
pressing a spring between die thumb and in- The association areas of the frontal
dex finger), the blood flow increase was only regions (areas rostral to Brodmann's area 6)
in the motor and sensor\' cortex. In contrast, are important for motor planning and other
when they were asked to perform a complex cognitive behaviors. For example, these areas
task (a sequence of movements insohing all probably integrate senson,' information and
four fingers, touching the thumb in different then select the appropriate motor response
orders), subjects showed a blood flow increase from the many possible responses (31).
in the supplementar*' motor area, bilaterally, The prefrontal cortex can be divided
and in the priman- motor and senson' areas. into the principal sulcus and the prefrontal
Finally, when they were asked to rehearse the convexities (refer back to Fig. 3.8). Experi-
task, but not perform it, the blood flow in- ments have indicated that the neurons of the
crease was only in the supplementar\' motor principal sulcus are invoh'ed in the strategic
area, not the primar\' senson,' or motor cortex. planning of higher motor frmctions. For ex-
Roland concluded that the supplementary ample, experiments on monkeys in which this
area is active when a sequence of simple bal- area was lesioned showed that they had diffi-
listic movements is planned. Thus, it partici- cult)' with performing spatial tasks in which
pates in the assembly of the central motor information had to be stored in working
program or forms a motor subroutine. memon,' in order to guide fiiture action. In
These experiments were performed by other experiments, neurons in this area were
the subjects in intrapersonal space (mo\ing shown to be acti\'e as soon as a cue was pre-
one body part in relation to another, where sented and to remain active throughout a de-
only a body reference system is needed). Mov- lay period, when the cue wasn't present, but
ing the limbs in extrapersonal space requires the monkey had to keep the cue in working
a different reference system, which is three- memoPi', before performing the task (13).
dimensional, and fixed by points in the envi- This area is densely interconnected with
ronment. For example, picking up a milk car- the posterior parietal areas. The prefrontal
ton fi-om a table requires this fi-amework. and parietal areas are hypothesized to work
two t\pes of movements are
Since these closely together in spatial tasks that require
different,one might expect the cerebral or- attention.
ganization to be different. To test this, Ro- By contrast, lesions in the prefrontal
land had the subjects perform a new move- conve.xit)' cause problems in performing any
ment. They were asked to make a spiraling kind of delayed response task, .\nimals w ith
movement in the air or move their fingers ( us- these lesions have problems with tasks where
ing no \ision ) over a grid according to specific they ha\'e to inhibit certain motor responses
cues (maze test). At that point, the superior at specific moments. Lesions in adjacent areas
parietal region was active along with the other cause problems with a monkey's abilit>' to se-
regions. So this region must be necessar\' for lect from a variet)' of motor responses when
the planning of voluntan,' movements in ex- given difterent sensor\' cues (13).
trapersonal space. Lesions in other prefrontal regions
Roland noted that the premotor area cause patients to have ditfrcult)' with changing
was activated only when a new motor pro- strategies when they are asked to. Even when
gram was established, or alternatively when a they are shown their errors, they fail to correct
pre\iously learned program was modulated. them.
Inputs
Spinocerebellum
Inputs
I Corticopontine
I
Spinal and trigeminal

I I
Visual and auditory

Vermis ^^
Vestibular
Intermediate |;
I
hemisphere
Lateral hemisphere
(cerebrocerebellum)
Vestibulocerebellum
Outputs
To medial descending
Fastigial nucleus
systems Motor
"
execution
To lateral descending
systems
Motor
To motor and U pianning
premotor cortices
Balance and
> To vestibular
nuclei > eye movements
Figure 3.14. A schematic drawing showing the basic anatomy of the cerebellum, including A, its inputs, and B, its
outputs. (Adapted from Chez C. The cerebellum. In: Kandel E, Schwartz )H, )essell TM, eds. Principles of neuroscience.
3rded. NY: Elsevier, 1991:633.)
Cerebellvim system, consistent with its role as a regulator

of motor output (32, 33).
The cerebellum is considered one of How does the cerebellum adjust the
three important brain areas contributing to output of the motor systems? Its fiinction is
coordination of movement, in addition to the related to its neuronal circuitry. Through this
motor cortex and the basal ganglia complex. circuitr)' and its input and output connec-
Yet, despite its important role in the coordi- tions, it appears to act as a comparator, a sys-
nation of movement, the cerebellum doesn't tem that compensates for errors by comparing
play a primary role in either sensory or motor intention with performance.
fLmction. If the cerebellum is destroyed, we The cerebellum's input and output con-
don't lose sensation or become paralyzed. nections are vital to its role as error detector,
However, of the cerebellum do pro-
lesions and summarized in Figure 3.14. Its inputs
are
duce devastating changes in our abilit\' to per- (Fig. 3.14^) include information from odicr
form movements, from the vet}' simple to the modules of the brain related to the program-
most elegant. The cerebellum receives affer- ming and execution of movements. This in-
ent information from almost ever)' sensor}' formation is often referred to as "efterence
copy" or "corollaPi' discharge" when it comes Vermis and Intermediate

from the priman,' motor cortex, since it is hy- Hemispheres
pothesized to be a direct copy of the motor
cortex output to the spinal cord. The cere- The \'ermis and intermediate hemi-
bellum also receives sensor\' feedback infor- spheres receive proprioceptive and cutaneous
mation (reatferencc) from the receptors about inputs from the spinal cord (via the spinocer-
the movements as they are being made. After ebellar tracts) in addition to visual, vestibular,
processing this information, outputs (Fig. and auditor)' information. Researchers used
3.14B) from the cerebellum go to the motor maps of the com-
to think that there were t\vo
cortex and other systems within the brainstem pletebody in the cerebellum, but now it has
to refine the movement. been shown that the maps are much more
complex and can be divicied into many smaller
ANATOMY OF THE CEREBELLUM maps. This has been called fractured soninto-
topy. These smaller maps appear to be related
An understanding of the anatomy of the to functional activities: thus, in the rat, the
cerebellum is helpful in explaining its ftinc- mouth and paw recepti\e fields are positioned
tion. The cerebellum consists of an outer layer closely together, possibly to contribute to the
of gray matter (the cortex), internal white control of grooming behax'ior. Inputs to this
matter (input and output fibers), and three part of the cerebellum go through the fastigial
pairs oi deep nuclei: the faitijjial Jiiiclcus, the nucleus (vermis) and interposed nucleus (in-
interposed nucleus^ and the dentate nucleus. termediate lobes) (34).
cerebellum go first to one
All the inputs to the There are four spinocerebellar tracts
of these three deep cerebellar nuclei and then that relay information from the spinal cord to
proceed to the cortex. All the outputs of the the cerebellum. Two tracts relay information
cerebellum go back to the deep nuclei, before from die arms and the neck, and two rela\'
going on to the cerebral cortex or the brain- information from the trunk and legs. Inputs
stem (32, 33). are also from the spino-olivo-cerebellar tract,
The cerebellum can be divided into through the inferior olivary nucleus (climbing
three zones, phylogenetically (refer back to fibers). These latter inputs are important in
Fig. 3.14). The oldest zone corresponds to learning and are discussed later.
the flocculonodular lobe. It is liinctionalh' re- What are the output pathways of this
lated to the vestibular system. The phyloge- part of the cerebellum? The outputs go to the
netically more recent areas to develop are the {a) brainstem reticular formation, (b) vestib-
vermis and intermediate part of the hemi- ular nuclei, (c) thalamus and motor cortex,
spheres and the lateral hemispheres^ respec- and {d) red nucleus in the midbrain.
tively. These three parts of the cerebellum What are the ftmctions of the vermis and
have distinct ftmctions and input cûtput con- intermediate lobes.' First, they appear to fiinc-
nections. tion in the control of the actual execution of

movement: they correct for dexiations from
Flocculonodular Lobe an intended movement through comparing
feedback from the spinal cord with the in-
The flocculonodular lobe receixes intended motor command. They also modulate
puts from both the \isual system and the ves- muscle tone. This occurs through the contin-
tibular system, and its outputs return to the uous output of excitatoiy activit)' from the
vestibular nuclei. It ftmctions in the control and interpositus nucleus, which mod-
fastigial
of the a.xial muscles that are used in equilib- of the 7-motor neurons to
ulates the activitx'
rium control. If a patient experiences dys- the muscle spindles. When there are lesions in
ftmction in this system, one observes an ataxic these nuclei, there is a significant drop in mus-
gait, wide-based stance and nystagmus. cle tone (hypotonia) (32).
Lateral Hemispheres lular circuitry of the cerebellum has been

shown to be perfect for the long-term modi-
The last part of the cerebellum and the fication of motor responses. Experiments have
newest phylogenetically is the lateral zone of shown that as animals learn a new task, the
the cerebellar hemispheres (Fig. 3.14). It re- climbing fiber, a type of neuron that detects
ceives inputs from the pontine nuclei in the movement error, alters the effectiveness of the
brainstem that relay information from wide synapse of a second fiber, the granule cell par-
areas of the cerebral cortex (sensory, motor, allel fiber, onto the main output cells of the
premotor, and posterior parietal). Its outputs cerebellum, the Purkinje cells (37).
are to the thalamus, motor, and premotor This type of cerebellar learning also oc-
cortex. curs in vestibulo-ocular reflex circuitry, which
What is the function of the lateral hem- includes cerebellar pathways. This reflex keeps
ispheres.* This part of the cerebellum func- the eyes fixed on an object when the head
tions in the preparation of movement, turns. In experiments in which humans wore
whereas the intermediate lobes fianction in prismatic lenses that reversed the image on
movement execution and fine-tuning of on- the eye, the gain of the vestibulo-ocular reflex
going movement via feedback information. It was altered over time. This modification of
appears that the lateral hemispheres of the the reflex did not occur in patients with cer-
cerebellum participate in programming the ebellar lesions (38).
motor cortex for the execution of movement.
The cerebellar pathways are one of many par- Basal Ganglia
allel pathways affecting the motor cortex. The
others probably include the supplementary The basal ganglia complex consists of a
and premotor areas. set of nuclei at the base of the cerebral cortex,
The lateral hemispheres also appear to including the putamen^ caudate nucleus, glo-
fijnction in the coordination of ongoing bus pallidus, subthalamic nucleus, and sub-
movements. It has been shown that cooling stantia nigra. Basal means "at the base," or
parts of the cerebellum disturbs the timing of in other words, "just below the cortex." As
agonist and antagonist muscle responses dur- with patients with cerebellar lesions, patients
ing rapid movements (35). The antagonist ac- with basal ganglia damage are not paralyzed,
tivity becomes delayed, giving a hypermetric but have problems with the coordination of
or "overshootin g" movement. As corrections movement. Advancement in our understand-
are attempted in cerebellar patients, one sees ing of basal ganglia fimction first came from
iiniri fendeH movements in the opposite dire c- clinicians, especially from James Parkinson,
tion, giving in tention tremo r. who in 18 1 7 first described Parkinson's dis-
In addition, the lateral cerebellum may ease as "the shaking palsy" (39).
contribute to a more general timing fiinction The basal ganglia were once believed to
that affects perception as well as action. Pa- be part of the extrapyramidal motor system,
tients with cerebellar lesions often make tim- which was believed to act in parallel with the
ing errors during movement. Those with lat- pyramidal system (the corticospinal tract) in
eral hemisphere lesions show errors in timing movement control. Thus, clinicians defined
related to perceptual abilities, which research- pyramidal problems as relating to spasticity
ers think may be related to problems with a and paralysis, while extrapyramidal problems
central clock-like mechanism (36). In con- were defined as involuntary' movements and
trast, patients with intermediate lobe lesions rigidity.As we have seen in this chapter, this
make errors related to movement execution distinction is no longer valid since many other
(36). brain systems also control movement. In ad-

Finally, many parts of the cerebellum, dition, the pyramidal and extrapyramidal sys-
including the lateral cerebellum, seem to be tems are not independent, but work together
important in motor learning. The unique cel- in controlling movements.
Chapter Three PHYSIOLOGY of Motor Control 79
ANATOMY OF THE BASAL GANGLIA plex is the termination site for tracts from the
entire cerebral cortex, but not the spinal cord
The major connections of the basal gan- (39).

glia are summarized in Figure 3.15, including Their outputs also influence different
the major afferent (3.15yl), central (3.155), motor system. The basal ganglia
parts of the
and efferent (3.15C) connections. The main complex sends its outputs to the prefrontal
input nuclei of the basal ganglia complex are and premotor cortex areas, involved in higher
the caudate and the putamen. The caudate level processing of movement, while the cer-
and the putamen develop Irom the same ebellar output goes back to the motor cortex,
structure and are often discussed as a single and also to the spinal cord via brainstem path-
unit, the striatum. Their primani' inputs are ways. This suggests that the cerebellum is
from widespread areas of the neocortex, in- more directly involved in the on-line control
cluding sensor\', motor, and association areas of mo\ement (correcting errors), while the
(39,40). basal ganglia fijnction may be more complex,
The globus pallidus has two segments, including ftanctions related to the planning
internal and external, and is situated next to and control of more complex motor behavior.
the putamen, while the substantia nigra is sit- The basal gangliamay plav a role in se-
uated a little more caudaily, in the midbrain. lectively activating some movements as they
The internal segment of the globus pallidus suppress others (39, 40). Diseases of the basal
and the substantia nigra are the major output ganglia ripically produce involuntar\' move-
areas of the basal ganglia. Their outputs ter- ments (dyskinesia), poverrs' and slowness of
minate in the prefrontal and premotor cortex movement, and disorders of muscle tone and
areas, by way of the thalamus. The final nu- postural reflexes. Parkinson's disease symp-
cleus, the subthalamic nucleus, is situated just toms include resting tremor, increased muscle
below the thalamus. tone or rigidit\', slowness in the initiation of
The connections within the basal gan- movement (akinesia) as well as in the execu-
glia complex are as follows: Cells in both the movement (bradykinesia). The site of
tion of
caudate and putamen terminate in the globus the lesion is in the dopaminergic pathway
pallidus and the substantia nigra in a soma- from the substantia nigra to the striatum. The
totopic manner, as seen for other pathways in tremor and rigidit}' may be due to loss of in-
the brain. Cells from the external segment of hibiton' influences widiin the basal ganglia.
the globus pallidus terminate in the subtha- Huntington's disease characteristics include
lamic nucleus, while the subthalamic nucleus chorea and dementia. Symptoms appear to be
and substantia
projects to the globus pallidus caused by loss of cholinergic neurons and
nigra. Other inputs to the subthalamic nu- GABA-ergic neurons in the striatum (39, 40).
cleus include direct inputs from the motor This concludes our review of the phys-
and premotor cortex. iological basis for motor control. In this chap-
ter we ha\e tried toshow you the substrates
for mo\ement. This has involved a review of
ROLE OF THE BASAL GANGLL\ the perception and action systems, and the
higher-level cogniti\e processes that play a
The basal ganglia and cerebellum ha\e part in their elaboration. We have tried to
many similarities in the way they interact with show the importance of both the hierarchical
the rest of the elements of the motor system. and distributed nature of these svstems. The
But what are their differences? First, their in- presentation of the perception and action sys-
put connections are different. The cerebellum tems separately is somewhat misleading. In
receives input only from the sensor)' and mo- real life, as movements are generated to ac-
tor areas of the cerebral cortex. It also receives complish tasks in varied environments, the
somatosenson,- information directly from the boundaries between perception, action, and
spinal cord. However, the basal ganglia com- cognition are blurred.
80 Section I THEORETICAL FIL\MEWORK
Caudate nucleus
Thalamus
Substantia nigra,
pars compacta
Substantia nigra,
pars reticulata
Internal segment
Globus pallidus external segment
From cerebral
cortex
Afferent connections
Basal ganglia connections Efferent connections
Figure 3.15. summarizing A, the major afferent, B, the central, and C, the efferent connections of the basal
Illustration
ganglia. (Adapted from Cote L, Crutcher MD. The basal ganglia. In: Kandel E, Schwartz JH, )essell TM, eds. Principles
of neuroscience. 3rd ed. NY: Elsevier, 1991 :649.)
Chapter Three PKisiOLOGY OF Motor CONTROL 81
identify and discriminate between different
Summary objects. This is done through lateral inhibi-
tion, in which the cell that is excited inhibits
1 Movement control is achieved through the the cells next to it, thus enhancing contrast
cooperative effort of many brain structures, between excited and nonexcited regions of
which are organized both hierarchically and the body or visual field.
in parallel. 9. There are also special cells within the SS and
2. Sensory inputs perform many functions in visual systems that respond best to moving
the control of movement. They (a) serve as stimuli and are directionally sensitive.
the stimuli for reflexive movement organized 10. In the association cortices, we begin to see
at the spinal cord level of the nervous sys- the transition from perception to action. The
tem; (b)modulate the output of movement parietal lobe participates in processes in-
that results from the activity of pattern gen- volving attention to the position of and ma-
erators in the spinal cord: (c) modulate com- nipulation of objects in space.
mands that originate in higher centers of the 1 1 The action system includes areas of the ner-
ner\'ous system; and {di contribute to the per- vous system such as motor cortex, cerebel-
ception and control of movement through lum, and basal ganglia.
ascending pathways in much more complex 1 2. The motor cortex interacts with sensory pro-
ways. cessing areas in the parietal lobe and with
3. In the somatosensory system, muscle spin- basal ganglia and cerebellar areas to identify
dles, Golgi tendon organs, joint receptors, where we want to move, to plan the move-
and cutaneous receptors contribute to spinal ment, and finally, to execute our actions.
reflex control, modulate spinal pattern gen- 13. The cerebellum appears to act as a com-
erator output, modulate descending comparator, a system that compensates for errors
mands, and contribute to perception and by comparing intention with performance. In
control of movement through ascending addition, modulates muscle tone, partici-
it
pathways. pates in programming of the motor cortex

the
4. Vision (a) allows us to identify objects in for the execution of movement, contributes
space, and to determine their movement (ex- to the timing of movement, and to motor
teroceptive sensation) and (b) gives us infor- learning.
mation about where our body is in space, 14. Basal ganglia function is related to the plan-
about the relation of one body part to an- ning and control of complex motor behavior.
other, and the motion of our body (visual- In addition, it may play a role in selectively
proprioception). activating some movements and suppressing

5. The vestibular system is sensitive to Uvo others.
types of information: the position of the head
in space and sudden changes in the direction
References
of movement of the head.
6. As sensory information ascends to higher L Kandel E. Brain and behavior. Kandei E, In:
levels of processing, every level of the hier- Schwartz JH, Jessell ofTM, eds. Principles
archy has the ability to modulate the infor- neuroscience. 3rd ed. NT: Elsevier, 1991:5-
mation coming into it from below, allowing 17.
higher centers to selectively tune (up or 2. Patron HD, Fuchs .A, HiUe B, Scher A,
down) the information coming from lower Steiner R. Textbook of physiolog\', vol 1.
centers. 21st ed. Philadelphia: W^ Saunders, 1989.

7. Information from sensory receptors is in- 3 Koester J . Passive membrane properties of the
creasingly processed as it ascends the neural neuron. In: Kandel E, Schwartz JH, Jessell
hierarchy, enabling meaningful interpreta- TM, eds. Principles of neuroscience. 3rd ed.
tion of the information. This is done by se- NT: Elsexier, 1991:95-103.
lectively enlarging the receptive field of each 4. Kandel ER. Cellular basis of behavior: an in-
successively higher neuron. troduction to beha\ioral neurobiology. San

8. The somatosensory and visual systems pro- Francisco: Freeman, 1976.
cess incoming information to increase con- 5. Gordon J, Ghez C. Muscle receptors and spi-
trast sensitivity so that we can more easily nal reflexes: the stretch reflex. In: Kandel E,
Schwartz JH, Jessell TM, eds. Principles of 19. Hubel DH. Eye, brain and vision. NY: Sci-
neuroscience. 3rd ed. NY: Elsevier, entific American Library, 1988.
1991:564-580. 20. Kandel ER. Perception of motion, depth and
6. Pearson KG, Ramirez JM, Jiang W. Entrain- form. In; Kandel E, Schwartz JH, Jessell TM,
ment of the locomotor rhythm by group lb eds. Principles of neuroscience. 3rd ed. NY:
afferents from ankle extensor muscles in spi- Elsevier, 1991:440^66.
nal cats. Exp Brain Res 1992;90:557-566. 21. Treisman A. Features and objects: the four-
7. Burgess PR, Clark FJ. Characteristics of knee- teenth Bardett memorial lecture. J Exp Psy-
(Lond)
joint receptors in the cat. J Physiol chol I988;40A:20I-237.
1969;203:317-325. 22. Kelly JP. The sense of balance. In: Kandel E,
8. Kandel E, Jessell TM. Touch. In: Kandel E, Schwartz JH, Jessell TM, of
eds. Principles
Schwartz JH, Jessell TM, eds. Principles of neuroscience. NY:
3rd ed.
Elsevier,
neuroscience. 3rd ed. NY: Elsevier, 1991:500-511.
1991:367-384. 23. Baloh RW: Dizziness, hearing loss and tin-
9. Grillner S, Wallen P. Central pattern gener- nutus: the essentials of neurotology. Phila-
ators for locomotion, with special reference delphia: FA Da\is, 1984.
to vertebrates. Armu Rev Neurosci 1985; 24. Ayres J. Sensory integration and learning dis-
8:233-261. orders. Los Angeles: Western Psychological
10. Forssberg H, Grillner S, Rossignol S. Phasic Ser\'ices; 1972.
gain control of reflexes from the dorsum of 25. Ghez C. Voluntary movement. In: Kandel E,
the paw during spinal locomotion. Brain Res Schwartz JH, Jessell TM, eds. Principles of
1977;132:121-139. neuroscience. 3rd ed. NY: Elsevier,
11. Martin JH, Jessell TM. Anatomy of the so- 1991:609-625.
matic sensory system. In: Kandel E, Schwartz 26. Penfield W, Rasmussen T. The cerebral cor-
JH, Jessell TM, eds. Principles of neurosci- tex of man: a clinical study of localization of
ence. 3rd ed. NY: Elsevier, 1991:353-366. function. NY: Macmillan, 1950.
12. Martin J. Coding and processing of sensory 27. Conrad B, Matsunami K, Meyer- Lohmann J,
information. In: Kindel E, Schwartz JH, Jes- Wiesendanger M, Brooks VB. Cortical load
sell TM, eds. Principles of neuroscience. 3rd compensation during voluntary elbow move-
ed. NY: Elsevier, 1991:329-340. ments. Brain Res 1974;71:507-514.
13. Kupfermann I. Localization of higher cog- 28. Evarts EV. Relation of pyramidal tract activit)'
nitive and affective functions: the association to force exerted during voluntary movement.
cortices. In: Kandel E, Schwartz JH, Jessell J Neurophysiol 1968;31:14-27.
TM, eds. Principles of neuroscience. 3rd ed. 29. Georgopoulos AP, Kalaska JF, Caminiti R,
NY: Elsevier, 1991:823-838. Massey JT. On the relations between the di-
14. Tessier-Lavigne M. Phototransduction and rection of t^vo-dimensional arm movements
information processing in the retina. In: Kan- and cell discharge in primate motor cortex. J
del E, Schwartz JH, Jessell TM, eds. Princi- Neurosci 1982;2:1527-1537.
ples of neuroscience. 3rd ed. NY: Elsevier, 30. Roland PE, Larsen B, Lassen NA, Skinhof E.
1991:400-417. Supplementary motor area and other cortical
15. DowUng JE. The retina: an approachable part areas in organization of voluntary move-
of the brain. Cambridge, MA: Belknap Press, ments in man. J Neurophysiol 1980;43:
1987. 118-136.
16. Mason C, iCandel ER. Central visual path- 31. Fuster JM. The prefrontal cortex: anato-
ways. In: Kandel E, Schwartz JH, Jessell TM, my, physiology and neuropsychology of the
eds. Principles of neuroscience. 3rd ed. NY: frontal lobe. 2nd ed. NY: Raven Press.
1991:420^39.
Elsevier, 1989.
17. Hubel DH, Wiesel TN. Receptive fields of 32. Ghez C. The cerebellum. In: Kandel E,
single neurones in the cat's striate cortex. J Schwartz JH, Jessell TM, eds. Principles of
Physiol (Lond) 1959;148:574-591. neuroscience. 3rd ed. NY: Elsevier,
18. Hubel DH, Wiesel TN. Receptive fields, bin- 1991:627-646.
ocular interaction and fimctional architecture 33. Ito M. The cerebellum and neural control.
in the cat's visual cortex. J Physiol (Lond) New York: Raven Press, 1984.
1962;160:106-154. 34. Shambes GM, Gibson JM, Wclker W. Frac-
Chapter Three PHYSIOLOGY OF Motor Control 83
cured somatotopy in granule cell tactile areas during motor learning. Brain Res 1977;
of rat cerebellar hemispheres revealed by mi- 128:309-328.
cromapping. Brain Behav Evol 1978;15:94— 38. Gonshor A, Melvill Hones G. Short-term
140. adaptive changes in the human vestibulo-oc-
35. Brooks VB, Thatch VVT. Cerebellar control ular reflex arc. J Physiol (Ixjnd) 1976;
ot posture and movement. In: Brooks VB, ed. 256:361-379.
Handbook of physiology, section 1: ner\ous 39. Cote L, Outcher MD. The basal ganglia. In:
system, vol 2,Motor control, part 2. Be- Kandel E, Schwartz JH, Icssell TM, eds. Prin-
thesda, MD:American Physiological Societs', ciples of neuroscience. 3rd cd. NY: Elsevier,
1981:877-946. 1991:647-659.
36. U'ry RB, Keele SW. Timing functions of the 40. Alexander GE, Crutcher MD. Eunctional ar-
cerebellum. J Cogn Neurosci 1989;1:136- chitecture of basal ganglia circuits: neural
152. substrates of parallel processing. Trends Neu-
37. Gilben PFC:, Thach \VT. Purkinje cell activity rosci 1990;13:266-271.
Chapter 4
Physiological Basis of Motor

Learning and Recovery of
Function
Introduction Declarative Learning

Defining Neural Plasticity Procedural Learning
Learning and Memory Perceptual Learning
Localization of Learning and Memory Plasticity and Recovery of Function
How Does Learning Change the Structure and Cellular Responses to Injury
Function of Neurons within the Neuronal Shock
, Brain? Synaptic Effectiveness
\
Physiological Basis of Nonassociative Denervation Supersensitivity
. Forms of Learning SilentSynapses
'
Habituation Regenerative and Reactive
• Sensitization Synaptogenesis
Neural Plasticity and Associative Learning Global Aspects of Plasticity
Classical Conditioning Summary
— |, Operant Conditioning
^j^g^HBIg
Introduction This chapter focuses on the physiologi-

cal basis of motor learning and recover^' of
In Chapter 2 we defined learning zs the fimction, showing the similarities and differ-
process of acquiring knowledge about the ences bet^veen these important functions. The
world, and motor Icnrnhijj as the process of material in this chapter builds on material pre-
the acquisition and/or modification of move- sented in the chapter on the physiological ba-
ment. We also mentioned that, just as motor sis of motor control. Since we assume that the
control must be seen in light of the interaction reader has a basic familiarit\' with the concepts
between the individual, the task, and the en- presented in Chapter 3, these concepts will
vironment, this also apphes to motor learning. not be reviewed again in this chapter.
In this chapter we extend our knowl- Integra] to a discussion on the physio-
edge of the physiological basis of motor con- logical basis of motor learning are issues re-
trol to include motor learning. This chapter lated to the phvsiological basis for memor\'. A
demonstrates that the physiological basis for fundamental question addressed in this chap-
motor learning, like motor control, is distrib- ter is: how does learning change the structure
uted among many brain structures and pro- and fiinction of neurons in the brain? Of equal
cessing levels, rather than being localized to a concern is the question: what changes in the
particular learning site of the brain. Likewise, stnicture and fimction of neurons underlie
it illustrates that the physiological basis for the the recover}' of fimction following injur\'.' We
recover)- of function is similar to learning, in also explore whether physiological plasticit\'
that reco\'er\' involves processes occurring associated with recover)' of fiinction is the
throughout the ner\'ous system and not just same or different from that inxohed with
at the lesioned site. These processes have learning. Previous views have t>'pically held
many common properties with those occur- that recovery and learning are different, but
ring during learning. physiological studies suggest that they are
85
86 Section I THEORETICAL FR.\ME\VORK
similar in that many of the same neural mech- and storage of that knowledge or abUit)' (1).
anisms underlie both t\pes of change. Learning reflects the process by \\hich we ac-
quire knowledge; memor}' is the product of
Defining Neural Plasticity that process. Memor\' is often divided into
short- and long-term storage. Short-tenn
We define plasticit\' as the abilit\- to memor\' refers to workinjj memon,-, which has
show modification. Throughout this book we a limited capacity' for information and lasts for
use the term plasticit>- as it relates to neural only a few moments. Short-term memon- re-
modifiabilit)-. Plasticin., or neural modifiabil- flects amomentan' attention to something,

ity,may be seen as a continuum from short- such as when we remember a phone num-
term functional changes to long-term struc- ber only long enough to dial it and then it's
tural changes. Short-term functional plasticity gone.
refers to changes in the efficienc)' or strength Long-term memon' is intimately re-
of synaptic connections. In contrast, struc- lated to the process of learning. Long-term
tural plasticity refers to changes in the orga- memor\' can also be seen as a continuum. Ini-
nization and numbers of connections among tial stages of long-term memon- formation
neurons. would reflect fijnctional changes in the efli-

Similarly, learning can be seen as a con- cienc\' of smapses. Later stages of memory-
tinuum of short-term to long-term changes in formation reflect structural changes in s\ti-

the capabilit\- to produce skilled actions. The aptic connections. These memories are less
gradual shift fi-om short-term to long-term subject to disruption.
learning reflects a move along the continuum
of neural modifiabilit\% as increased synaptic Localization of Learning and
efficienc\- gradually gives way to structtiral Memon'
changes, which are the underpinning of long-
term modification of beha\ior. This relation- Are learning and memor\- localized in a
ship is shown in Figure 4.1. specific brain structure.- It appears that they
are not. In fact, learning can occur in all parts
Learning and Memory of the brain. Learning and the storage of that
learning, memon.-, appear to involve both par-
Learning is defined as the acquisition of allel and hierarchical processing within the
knowledge or abifit\'; memor\' is the retention CNS. Even for relatively simple learning tasks.
Figure 4.1. A diagram showing the gradual Neural modifiability

shift from short-term to long-term learning is
reflected in a move along the continuum of

neural modifiabilit>'. Short-term changes, as-
sociated with an increased synaptic effi- Changes in
Changes in synetptic Persisting
and gradually give way synaptic
ciency, persist to efficiency changes
connections
structural changes, the underpinning of long-
term learning.
Parallel continuum of learning
Short-term 1 Persisting Long-term

changes changes changes
Chapter Four Physiology of Motor Learning and Recovery of Function 87
multiple parallel channels of information Physiological Basis of

are used. In addition, the information Nonassociative Forms of Learning
can be stored in many different areas of the
brain. Remember that in nonassociative forms
mechanisms underlying
Apparently, of learning, the person is learning about prop-
learning and memon,- are the same whether erties of a stimulus that is repeated. The
the learning is occurring in fairly simple cir- learned suppression of a response to a non-
cuits, or involves ver\' complex circuits incor- noxious stimulus is called habituation. In
porating many aspects of the CNS hierarchy. contrast, an increased response to one stim-
Thus, current neuronal models of memory- ulus that is consistendy preceded by a noxious
suggest that a memor\- consists of a pattern of stimulus is called sensitization. Keep in mind
changes in synaptic connections among net- that nonassociative forms of learning can be
works of neurons distributed throughout the short-term or long-lasting. What are the neu-
brain (1). ralmechanisms underlying these simple forms
This chapter describes the continuum of of learning, and do the same neural mecha-
plasticit)' within the ner%'ous system which nisms underlie both short- and long-term
represents learning, and specifically, motor changes?
learning. The processes underlying learning in
the ner\ous system, as well as those that un- H.\BITUATION
derlie of function, are described.
reco\er\'
Once understood, principles of plasticity' re- Habituation was first stucfied by Sher-
lated to learning and recoven' of function can rington, who found that the flexion reflex ha-
be derived. Then, in later chapters, these prin- bituated with many stimulus repetitions.
ciples are applied to therapy settings. More recent research examining habituation
in relatively simple net\vorks of neurons in in-
vertebrate animals has shown that habituation
How Does Learning is related to a decrease in synaptic acti\ity

between senson,' neurons and their con-
Change the Structure nections to interneurons and motor neurons
AND Function of Neurons (2,3,).
WITHIN THE Brain? During habituation, there is a reduction

in the amplitude of synaptic potentials (a de-
Many factors potentially modifx' synap- creased excitator\' postsynaptic potential
tic connections. During development, syn- [EPSP]) produced by the sensor>' neuron on
aptic connectivir\' develops under the control the interneuron and motor neuron. This
of genetic and developmental processes. short-term change in EPSP amplitude during
These connections are fine-tuned during var- habituation is illustrated in Figure A.2A. Dur-
ious critical periods of development due to ining initial stages of learning, the decreased
teracting environmental and genetic factors. size of the EPSP may last for only several min-
We are concerned in this chapter with nctir- utes. With continued presentation of the
in-depeiident modifications of synaptic con- stimulus, persisting changes in synaptic eflfi-
nections, that is, both the transient and lonjj- cacv occur, representing longer-term memory'
tcrm modulation of synapses resulting fi-om for habituation.
experience. Learning alters our capabilit\' for During the course of learning, contin-
acting by changing both the etfectix eness and ued presentation of the stimulus results in
anatomic connections of neural pathways. We structural changes in the sensor*' cells them-
discuss modifications of synaptic connections selves. Structural changes include a decrease
at both the cellular level and at the level of in the number of synaptic connections be-
whole nersvorks of neurons. rvveen the sensor\' neuron and interneurons
Section I THEORETICAL FRAMEWORK
Short-term habituation Long-term habituation
g EPSP g o
o- amplitude
a
Control Decreased connections
Long-term sensitization
Q Q
o
C Control Increased connections
Figure 4.2. Neuronal moditications underlying short- and long-term nonassociative learning. A, Short-term habitua-
tion results from a decrease in EPSP amplitude at the synapse between the sensory and motor neuron. B, Long-term
habituation results in a decrease in numbers of connections. C, Long-term sensitization results in an increase in numbers
of connections. (Adapted from Kandel ER. Cellular mechanisms of learning and the biological basis of individuality.
In: Kandel ER, Schwartz JH, Jessell TM, eds. Principles of neuroscience. 3rd ed. New York: Elsevier, 1991:1009-1031.)
and motor neurons, sho\\n diagrammatically with exercise there is a temporary' decrease in
in Figure 4.28. In addition, the number of the synaptic effectiveness of certain vestibular
active transmitting zones within existing con- neurons and their connections, due to a de-
nections decreases. As a result of these struc- crease in the size of the EPSPs. With contin-
tural changes, habituation persists over weeks ued exercise, changes in synaptic effectiveness
and months, representing long-term memor\' become more permanent. In addition, struc-
for habituation. Thus, the process of habitu- tural changes, including a reciuction in the
ation does not involve specific m em oiy storage number of vestibular neuron synapses con-
neurons found in specialized parts of the necting to interneurons, occurs. With the ad-
CNS. Rather, memory (retention of habitua- vent of structural changes, the decline in diz-
tion) results from a change in the neurons that ziness in response to the repeated head
are normal components of the response path- movement persists, allowing the patient to
way. discontinue the exercise without reexperienc-
How might this research apply to treating symptoms of dizziness. It is possible that
ment strategies used by therapists in the if exercises are discontinued too soon, before
clinic? As we mentioned earlier, habituation structural changes have occurred in the sen-
exercises are given to patients who have cer- sor}' connections, dizziness symptoms will re-
tain t)'pesof inner ear disorders resulting in cur due to the loss of habituation.
complaints of dizziness when they move their
head in certain ways (4). When patients begin SENSITIZATION
therapy, they may experience an initial decline
in the intensit)' of their dizziness symptoms As we mentioned in Chapter 2, sensiti-
during the course of one session of exercise. zation is caused by a strengthening of re-
But the next day, dizziness is back at the same sponses to potentially injurious stimuli. Sen-
level. Gradually, over days and weeks of prac- sitization may also be short- or long-term,
ticing the exercises, the patient begins to see and itmay involve the exact set of synapses
that decreases in dizziness persist across ses- that show habituation. However, the mecha-
sions (4). nisms involved in sensitization are a httie
KandePs research suggests that initially more complex than those involved in habit-
Chapter Four Phvsioi.ogv ov MoroR I.karninc; anh Rk(:()\ery of Function 89
nation. One way tliat sensitization may occur (operant conditioninjf). Through associative
is by prolonging the action potential through learning we learn to form key relationships
changes in potassium conductance. This al- that help us adapt our actions to the en\iron-
lows more transmitter to be releascti from the ment.
terminals, giving an increased HPSP. It also Researchers examining the physiologi-
appears to improve the mobilization of trans- cal basis for associative learning have found
mitter, making it more available for release that it can take place through simple changes
(2). in synaptic efficiency without requiring com-
Sensitization, like habituation, can be plex learning networks. Associative learning,
short- or long-term. Mechanisms for long- whether short-term or long-term, utilizes
term memor\' of sensitization involve the common cellular processes. Initially, when
same cells as short-term memon,', but now re- nvo neurons fire at the same time (that is, in
flect structural changes in these cells (3, 5). association), there is a modification of existing
Kandel (6) has shown that in invertebrates proteins within these two neurons that pro-
short-term sensitization involves changes in duces a change in synaptic efficiency. Long-
preexisting protein structures, while long- term association results in the synthesis of new
term sensitization involves the synthesis of proteins and the subsequent formation of new
new protein. This synthesis of new protein at synaptic connections benveen the neurons.
the synapse implies that long-term sensitiza-
tion involves changes that are genetically in- CLASSICAL CONDITIONING : ifi/'<^t^/i^
fluenced
During classical conditioning, an ini-
This genetic influence also encompasses
tiallyweak stimulus (the conditioned stimu-
the growth of new synaptic connections, as
lus) becomes highly efTecti\'e in producing a
illustrated in Figure 4.2C. Animals who
response when it becomes associated v\'ith an-
showed long-term sensitization were founci to
other stronger stimulus (the unconditioned
have twice as many synaptic terminals as un-
stimulus). It is similar to, though more com-
trained animals, increased dendrites in the
plex than, sensitization. In fact, it may be that
postsynaptic cells, and an increase in numbers
classical conditioning is simply an extension
of active zones at synaptic terminals, from 40
of the processes involved in sensitization.
to 65% (7).
Remember that in classical condition-
In summar)', the research on habitua-
ing, timing is critical. Wlien conciitioned and
tion and sensitization suggests that short-
unconditioned stimuli converge on the same
term and long-term memor\' may not be sep-
neurons, facilitation occurs if the conditioned
arate categories, but may be part of a single
stimulus causes action potentials in the neu-
graded memon,' fimction. With sensitization,
rons just before the unconditioned stimulus
as with habituation, long-term and short-
arrives. This is because action potentials allow
term memon' involve changes at the same
Ca* to move into the presynaptic neuron and
synapses. While short-term changes reflect
this Ca* acti\'ates special modulator\' trans-
relatively temporary' changes in synaptic effec-
mitters involved in classical conditioning. If
tiveness, structural changes are the hallmark
the acti\'itv' occurs after the unconditioned
of long-term memon' (2).
stimulus, Ca* is not released at the right time
and the stimulus has no effect (2, 8).

Neural Plasticity and Associative
Learning OPERANT C:ONDITIONING
Remember that during nssocintirc Although operant conditioning and
Irnruintj a person learns to predict relation- classical conditioning may seem like nvo dif-
ships, either relationships of one stimulus to ferent proces.ses, in fact, the laws that govern
another (classical conditioriiujf), or the rela- the nvo are similar, indicating that the same
tionship of one's behavior to a consequence neural mechanisms may control them. In each
tvpe of conditioning, learning involves the de- poral lobe areas removed due to epileps\'. Af-
velopment of predictive relationships. In clas- were no longer able
ter surger}', the patients
sical conditioning, a specific stimulus predicts to acquire long-term declarative memories,
a specific response. In operant conditioning, though they remembered old memories.
we learn to predict the outcome of specific Their short-term memor\' was normal, but if
behaviors. Howe\er, the same cellular mech- their attention was distracted ft-om an item
anisms that underlie classical conditioning are held in short-term memor)', they forgot it
also responsible for operant conditioning. completely. However, skill learning was un-
afi:ected in these patients. They would ofi:en
DECLARATIVE LEARNING learn a complex task but be unable to remem-
ber the procedures that made up the task
Remember that associative learning can or the events surrounding learning the task
also be thought of in terms of the t^'pe of (10).
knowledge acquired. Procedural learning re- This work suggests that the temporal
fers to learning tasks that can be performed lobes and hippocampus may be important to
automatically without attention or conscious the establishment of memon-, but are not a
thought. In contrast, declarative learninjj re- part of the memor\' storage area.
quires conscious processes such as awareness The hippocampus, which is a subcortical
and attention, and results in knowledge that structure, and part of the temporal lobe cir-
can be expressed consciously. Procedural cuitry', is critical for declarative learning. Re-
learning is e.xpressed through improved per- search hasshown evidence of plastic changes
formance of the task learned, while declarative in hippocampal neurons similar to those
learning can be expressed in a form other than found in neural circuits of simpler animals
that in which it was learned. when learning takes place.
Consistent with the vxo t\'pes of asso- Researchers have shown that pathways
ciative learning described, scientists believe in thehippocampus sho\\' a facilitation that
that the circuits involved in the storage of has been called long-term potentdation
these two r\'pes of learning are different. Pro- (LTP), which is similar to the mechanisms
cedural memory involves primarily cerebellar causing sensitization (2, II). For example, in
circuitr^', while declarative memory involves one region of the hippocampus LTP occurs
temporal lobe circuitr\' (1). when a weak and an excitatory input arrive at
Wilder Penfield, a neurosurgeon, was thesame region of a neuron's dendrite. The

one of the first researchers to understand the weak input will be enhanced if it is activated
important role of the temporal lobes in mem- in association with the strong one. This pro-
ory' function. While performing temporal lobe cess is shown in Figure 4.3. LTP appears to
surger}' in patients with epilepsy, he stimu- require the simultaneous firing of both pre-
lated the temporal lobes of the conscious pa- and postsynaptic cells. After this occurs, LTP
tients, in order to determine the location of is maintained through an increase in presyn-
the diseased vs. normal tissue. The patients aptic transmitter release.
experienced memories ftôm the past as if they Long-term potentiation has been found
were happening again. For example, one pain many areas of the brain, in addition to the
tient heard music fi"om an event long ago, and hippocampus, and it has been shown that it is
saw the situation and felt the emotions that involved in spatial memory (2). For example,
surrounded the singing of that music, with ev- Morris et al. 12 performed an experiment in
( )
er)T:hing happening in real time (9). which rats swam a water maze to find a plat-
In humans, lesions in the temporal lobe form under the water. The water was made
of the cortex and the hippocampus may in- opaque in order to block the use of vision in
terfere with the laying down of declarative finding the target. The rats were released in
memory. A few patients have been studied af- difiêrent parts of the maze and were required
ter having the hippocampus and related tem- to use spatial cues related to the position of
Chapter Four Physiology of Motor Learning and Rec;ov'hry of Function 91
Long-term potentiation
Before association
Weak
stimulus
Strong EPSP Weak EPSP
Strong Strong
stimulus
After association
Weak
stimulus
Strong
stimulus
Increased EPSP in weak stimulus

when paired with strong
B
Figure 4.3. Schematic of the cellular basis for long-term potentiation. A, Prior to pairing with a strong stimulus, the
weak stimulus produces only a weak EPSP. B, After association, there is an increased EPSP produced by the weak
stimulus when paired with the strong. (Adapted from Kandel ER. Cellular mechanisms of learning and the biological
basis of individuality. In: Kandel ER, Schwartz )H, Jesseli TM, eds. Principles of neuroscience. 3rd ed. New York:
Elsevier, 1991:1009-1031.)
the walls to find the target. They also per- t^pes of input fibers, the climbing fibers and
formed a nonspatial task where the platform the mossy fibers, and one t\pe of output fiber,
was above the water and the rat could simply the Purkinje cells. Climbing fiber input to the
use visual cues to swim to the target. Purkinje cells t>pically signals error and is im-
These experimenters showed that portant in the correction of ongoing move-
blocking special receptors in hippocampal ments. In contrast, mossy fiber input to the
neurons caused the rats to fail to learn the spa- Purkinje cells pro\ides kinesthetic informa-
tial version of the task. This finding suggests tion about ongoing movements, important in
that certain hippocampal neurons are in- the control of those movements. Figure 4.4
volved in spatial learning through LTP. reviews the relationship of these fibers.
It has been shown that the climbing fi-
PROCEDURAL LEARNING ber inputs signaling error to the Purkinje cells
may increase or decrease of

the strength
Procedural learning appears to involve mossy onto the same Purkinje
fiber synapses
the cerebellum. cells. This produces a long-term change in
The unique cellular circuitn,' of the cer- Purkinje cell output, which contributes to
ebellum has been shown to be perfect for the motor learning.
long-term modification of motor responses. Gilbert and Thach (13) examined the
You will recall that the cerebellum has two role of the cerebellum in motor learning dur-
Cortex I
Input
Input . . , Spinal cord, external cuneate nucleus,
Inferior olive Output
Reticular nuclei, pontine nuclei
Figure 4.4. A diagram of the cerebellum showing the relationship between mossy and climbing fiber input important
to learning.
ing experiments in which monkeys were was a sudden increase in activity' in the climb-
trained to return a handle to a central position ing fibers, signaling the error.
whenever it was moved to the left or right. This increase in climbing fiber activin,'
During die sessions, they recorded the activit)' was associated with a reduction in the effi-
of Purkinje neurons in the arm area of the an- ciency of the mossy fiber connections to the
terior lobe of the cerebellum. Once the task Purkinje cells. The reduction in Purkinje cell
was learned and repeatedly performed in the output then was associated with an increase in
same way, the arm mosement was accompa- fiDrce generation, allowing the monkey to
nied by predictable changes occurring pri- now successfiilly complete the task. Thus, it
marily in mo.ssy fiber inputs reporting the ki- appears that changes in synaptic efficiency be-
nesthetics of the movement, with an tween these neurons in the cerebellum are an
occasional climbing fiber input. important link in the modification of move-
Then the experimenters modified the ments through procedural learning.
task, requiring the monkeys to use more force This t\'pe of cerebellar learning may also
to return the handle to the original position. occur in the vestibulo-ocular reflex circuitr\',
At first the animal wasn't able to return the which includes cerebellar pathways. This re-
handle in one simple movement. But gradu- flex keeps the eyes fixed on an object when
ally, the animal learned to respond correctlv. the head turns. In experiments in which hu-
On the first few trials of the new task, there mans wore prismatic lenses that reversed the
Chapter Hour Physiology of Motor Learning and Recovery of Function 93
image on the eye, the vestibulo-ocular reflex sor\' memories, emotional memories, spatial
was reversed over time. This modification of memories, or motor memories. Thus, the first
the reflex did not occur with cerebellar lesions pathway can arouse the other pathways or be
(14). aroused by them through reciprocal connec-
tions between these difterent parts of the cor-
PERCEPTUAL LEARNING tex.
Perceptual learning, or the formation of Plasticity and Recovery

sensor\' memories, is actually a form of non- OF Function
associative learning ( 1 ). It is a more complex
form of nonassociative learning than either In the early part of this centurx', Ramon
habituation or sensitization, so it is presented y Cajal performed experiments which sug-
separately. How
does perceptual learning ac- gested that growth was not possible in neu-
For example, when you are first
tually occur? rons in the adult mammalian CNS. This led
introduced to a new skill, and see someone to a view of the CXS as a static structure with
perform it, you are often able to remember rigid and unalterable connections (16). This
the essence of the skill after only one expo- view persisted until the late 1960s and 1970s,
sure. One hypothesis is that, in the process of when researchers began to disco\er growth
\iewing a new scene, the brain stores a coded and reorganization of neurons in the adult
representation of it in our visual cortex, and CNS after injur\'. Much of this early work
that we recognize that stimulus when this \i- showed that cells which lost their normal in-
sual representation is reactivated by the same put as a consequence of injur)' could receive
scene at a later time. new connections (17). These early studies
Experiments on monkeys by Mishkin et ha\e contributed to more current views of the
al. (15) support this hypothesis and indicate CNS as a structure capable of dynamic
that these coded representations of visual changes in parallel pathways throughout dis-
stimuli are stored in higher-order sensor\' as- tributed sites.
sociation areas of the visual cortex. How does Continued research examining the re-
this representation get stored.' When we see a organization of neuronal circuits following
unique scene, this new set of visual stimuli is injur\' has shown that the CNS has amazing
coded by parallel neural circuits in the visual capacities for reorganization following injur)',
cortex, coding for size, color, texture, and and that this reorganization of neuronal cir-
shape of the stimulus. These circuits are in cuits has fianctional consequences (17). How
such places as Brodmann's areas, 18, 20, 21, much can the CNS reorganize following in-
and i7 of higher visual cortex. These parallel jur).- Can postinjur)- reorganizational pro-
pathways converge on a single set of inferior cesses be manipulated to facilitate CNS reor-
temporal cortex neurons, and they, in turn, ganization? These and other questions have
stimulate a reverberating circuit that includes critical importance for basic scientists and cli-
neurons in a cortico-limbo-thalamo-cortico nicians involved in the rehabilitation of brain-

pathway including neurons in the amygdala,
( injured patients.
hippocampus, thalamus, returning to the vi- Injur)' to the CNS can aftect neuronal
sual cortex). This circuit ser\es as a sponta- function through direct damage to the neu-
neous rehearsal mechanism that sen'es to rons themselves. In addition, disruption of
strengthen the connections that were part of neuronal ftinction can occur as the result of
the first activation of the circuit. indirect eft'ects of injuries that impair cerebral
When the neurons are later reactivated, blood flow, control of the cerebrospinal fluid,
the pathway can be considered the stored rep- or cerebral metabolism. As shown in Figure
resentation of that scene. This visual memoni' 4.5, whether the trauma occurs through a di-
will also interact with other memories that rect or indirect mechanism, the eft'ect on neu-
w ere laid down at the same time, such as sen- ronal function can include (a) interrupting
^ ^ Orthograde transneuronal
degeneration
9
^ Retrograde degeneration
Retrograde transneural
Secondary degeneration
Initial
disruption disruption
Figure 4.5. Schematic showing the secondary neuronal disruption that occurs as part of the cascade of events fol-
lowing neural injury. A, Normal neuronal function. B, Interruption of axonal projections from injured area. C, Sec-
ondary neuronal disruption.
axonal projections from areas injured (Fig. of input from the injured area, the symptoms
4. 55); (b) denervation of the population of caused by the disruption of input will be re-
neurons innervated by the injured neurons duced.

(Fig. 4.5C); and (c) removing some neurons
entirely (Fig. 4.5 C) (17).
SYNAPTIC EFFECTIVENESS
Aside from the loss of neurons damaged
at the site of injury, the consequences of syn- Neurons directly affected by the lesion
aptic loss from these neurons produces a cas- will show loss of synaptic effectiveness. Craik
cading degeneration along neuronal path-
(19) notes that edema at the site of neuronal
ways, increasing the extent of neuronal injur)' may lead to a compression of axons and
disruption with time (17).
physiological blocking of neuronal conduc-
tion. Reduction of the edema would then re-
Cellular Responses to Injury store a portion of the fijnctional loss.
This process is shown in Figure 4.6.
The following sections review some of
the events occurring within the nervous sys-
tem following injury. These events may con- DENERVATION SUPERSENSITIVITY
tribute to and limit recovery of ftinction.
Denervation supersensitivity can oc-
NEURONAL SHOCK cur when neurons show a loss of input from
another brain region. In this case, the post-
One of the first events following ner- synaptic membrane of a neuron becomes hy-
vous system injury is neuronal shock or di- peracdve to a released transmitter substance.
aschisis (18, 19). This includes the short-term For example, Parkinson's disease causes a loss
loss of fijnction in neuronal pathways at a of dopamine producing neurons in the sub-
distance from the lesion itself Such a situation stantia nigra of the basal ganglia. In response
could arise with the loss of the normal neu- to this disease-induced denerxation, their
ronal activation of the intact areas. If the in- postsynaptic target neurons in the striatum
tact areas recover to some extent from the loss beccjme hypersensitive to the dopamine that
Chapter Four I'hvsu)i,ck;v ov Moior Lharning and Recovery of Funci ion 95
is released by the remaining substantia nigra how this contributes to more global aspects
neurons, ty^jyiotx^^v) uW^y l^~i>cpr. \^ t'/V^cM'"- of plasticifv within the nervous system. For
example, how modifiable are the sensor\'-mo-
SILENT SYNAPSES/-^ <)P^&5jk^ /^ tor maps of our brain?
Research on the de\elopment of the vi-
Recruitment of previously silent synap- sual system has shown that the visual cortex is
ses also occurs during recover}' of function. highlv modifiable by experience during cer-
This suggests that structural synapses are pres- tain critical periods shortly after birth. Is this
ent in many of the brain that may not
areas modifiabilit)' also possible in other sensory
normally be flmctional due to competition and motor systems, and is it possible to
within neuronal pathways. However, experi- change these systems in the adult as well as
ential factors or lesions may lead to their being early in the developmental process? The an-
unmasked when they are released from these swers to these questions are YES!
previous effects. In Chapter 3, we talked about the pri-
mary' somatosensory' cortex areas 1,2, 3a, and
REGENERATIVE AND REACTIVE 3b, each having a separate sensor}' map of the
SYNAPTOGENESIS / <^CCKJî^'j body. Research (20) has shown that these
maps of the somatosensory cortex van.' from
Regenerative synaptogenesis occurs
when injured axons begin sprouting. Reac-
tive synaptogenesis, calleci c ollateral sprout -
ing, may occur when neighboring normal ax-
ons sprout to innersate synaptic sites that
were previously activated by the injured axon.
Examples of regenerative and reactive synap-
togenesis arc shown in Figure 4.7 (18, 19).
Global Aspects of Plasticity
With this understanding of some of the

responses of neurons to injur\-, we might ask
Figure 4.6. Diagram of recovery of synaptic effective- Figure 4.7. Examples of regenerative and reactive syn-
ness due to the resolution of edema, allowing nerve con- aptogenesis in related neurons following injury. (Adapted
duction to resume. (Adapted from Craik RL. Recovery from Held )M. Recovery of function after brain damage:
processes: maximizing function. In: Contemporary man- theoretical implications for therapeutic intervention. In:
agement of motor control problems. Proceedings of the Carr )H, Shepherd RB, Cordon F, et al., eds. Movement
IIStep Conference. Alexandria, VA: APIA, 1992:165- sciences; foundations for physical therapy in rehabilita-
173.) tion. Rockville, MD: Aspen Systems, 1987:155-177.)
individual to indi\idual according to past ex-

perience.
What have researchers learned about
how these maps change during recover)' of
fiinction? When the median ner\'e (which in-
nervates the cutaneous regions of the mon-
key's hand) of the monkey is severed, one
might expect that its corresponding parts of
Prior to differential
the somatosensory' cortex would become si- stimulation
lent, since there would be no input coming

into them. But when experiments were per-
formed to test the mapping of the cortex after
surgen,',it was found that neighboring maps
had expanded their receptive fields to cover

much of the denenated region. These repre-
sentations increased even further in the weeks
foOo\\'ing dener\'ation (21, 22).
Other research by Mortimer Mishkin
and his colleagues (23) has shown that so- After differential
stimulation
matosensory' cortex area II (SII) is also very
plastic. These researchers removed all of the Figure 4.8. Training causes an expansion of cortical
inputs for the hand representation coming representation (Adapted from Jenkins WM, Merzenich
into SII and noted that the area was initiall\' MM, Och MT, Allard T, Cuic-Robles E. Functional re-
unresponsive. However, within 2 months, the organization of primary somatosensory cortex in adult
owl monkeys after befiaviorally controlled tactile stim-
area was again responsi\'e and occupied b\' the
ulation. J Neurophysiol 1990;63:82-104.)
inputs from the foot. This reorganization in-
volved over half of the SII representation.
Additional studies (20) have shown that
the somatotopic maps in normal animals show or learning.* The mechanisms involved appear
extensive differences between individuals. But to be ver\' similar to those that we ha\e pre-
how do we kno\\' whether these differences viously discussed in relation to associative
are due to inherited genetic differences or to learning. In ftirther experiments with mon-
experience? To test for this, Merzenich and keys, Merzenich and his colleagues connected
coworkers (24) performed an experiment in r\\o fingers of the monkey together, so that
which monkeys were able to reach for food b\' these fingers would always be used together
using a strateg)' that involved use of their mid- in themonkey's actions (26). This means that
dle fingers only. After considerable experience the inputs from the two areas would always
with this task, the monkeys' cortical map be highly correlated in the cortex. This
showed an area for the middle fingers that was changed the mapping of area 3b in the so-
significantiy larger than normal. This reorga- matosenson- cortex, eliminating the sharp
nization in somatosensory' cortex resulting boundaries between the maps of these two
from training is shown in Figure 4.8. fingers. Thus, the normal sharp boundaries
It has also been shown that these between different parts of the body within our
changes occur at other levels of the ner\'ous sensor)' and motor maps may depend signifi-
system besides the cerebral cortex. The dorsal cantiy on the actix-it)' of these areas.
column which are the first synaptic
nuclei, What do these studies tell us? They sug-
juncture wthin the somatosenson' system, gest that we ha\e multiple path\\a)'s innervat-
also show reorganization after peripheral le- ing an\' given part of the sensor)' or motor
sions (25). cortex, with only the dominant pathway
What mechanisms contribute to the showing ftinctional activit)'. However, when
changes in receptive fields as a result of lesions a lesion occurs in one pathwax', the less ciom-
Chapter Four Physiology of Motor Le.\rning .\nd Recovery of Function 97
inant pathway may immediately show func- 3. CNS changes occur because of the
structural
tional connections. This leads us to the con- Interaction beKveen both genetic and experi-
ential factors.
clusion that cortical maps are vcr\' dynamic.
Even in adults there appears to be use-depen-
4. A key factor in is the concept of
experience
active competition, and this may be summed
dent competition among neurons for synaptic
up in the phrase "the squeaky wheel gets the
connections. So when one area becomes in-
oil," or in this case, it gets the new synaptic
active, a neighboring area can take over its
connections. This concept is applicable from
former targets and put them to fiinctional use. simple circuits to complex neural pathways.
These experiments also suggest that our 5. Research suggests that short-term and long-
senson- and motor maps in the corte.x are con- term memory may not be separate categories,
stantly changing in accordance with the but may be part of a single graded memory
amount to which they are activated by pe- function, involving the same synapses.
ripheral inputs. Since each one of us has been 6. Short-term changes reflect relatively tempo-
brought up in a different environment and has rary changes in synaptic effectiveness; struc-
tural changes are the hallmark of long-term
practiced ver\- different t>pes of motor skills,
memory.
the maps of each of our brains are unique and
7. Scientists believe that the circuits involved in
constandy changing as a result of these ex-
the storage of procedural and declarative
periences. learning are different, with procedural mem-
How can we apply this information to ory involving cerebellar circuitry and declar-
therapy.' First, it means that whenever a pa- ative memory involving temporal lobe cir-
tient experiences a neural lesion, the cortical cuitry.

maps show both (a) immediate reorganiza-
due to the unmasking of previously non-
tion,
fiinctional synaptic connections from neigh- References
boring areas, and ( b) a longer-term change,
1. Kupfermann I. Learning and Memory'. In:
where neighboring inputs to the areas take Kandel ER, Schwartz JH, Jessell TM, eds.
o\er the parts of the map that were previously Principles of neuroscience. 3rd ed. New York:
occupied by damaged or destroyed cells. ElseWer, 1991:997-1008.
Second, it tells us that experience is very 2. Kandel ER. Cellular mechanisms of learning
important in shaping cortical maps. Thus, if and the biological basis of individualin,'. In:
we leave patients without rehabilitation train- Kandel ER, Schwartz JH, Jessell TM, eds.
Principles of neuroscience. 3rd ed. New York:
ing for many weeks or months, their brains
Else\ier, 1991:1009-1031.
will show changes in organization, reflecting
3. Kandel ER, Schwarz JH. Molecular biolog>'
disuse, which be most detrimental to
will
of learning: modulation of transmitter re-
these patients. However, the good news is
lease. Science 1982;218:433^t43.
that training appears to make a difference no Shumway-Cook A, Horak FB. Rehabilitation
4.
matter when it is given, since the brain con- strategies for patients with vestibular deficits.
tinues to be plastic throughout our lives. Neurol Clin 1990;8:441-457.
5. Sweatt JD, Kandel ER. Persistent and tran-
scriptionally-dependent increase in protein
Summary phosphoniation in long-term facilitation of
Aplysia sensor\- neurons. Nature 1989;
The research on the neurophysiological
339:51-54.
basis for learning, memor\', and recovery' of
6. Kandel ER. Genes, ner\e cells, and the re-
function covered in this chapter suggests the membrance of things past. Neuropsychiatrx'
J
following important principles: 1989;1:103-125.
1. The brain is incredibly plastic, and has great 7. Bailey CH, Chen M. Morphological basis of
capacity to change; this includes not just the long-term habimation and sensitization in
immature brain but the mature adult brain. Aplysia. Science 1983;220:91-93.
2. The most important way in which the envi- 8. Abrams T\V, Kandel ER. Is conriguir\- detec-
ronment changes behavior in humans is tion in classical conditioning a system or a cel-
through learning. lular propcrtv? Learning in Aplysia suggests a
possible molecular site. Trends Neurosci peutic intenention. In: Carr JH, Shepherd,
1988;11:128-135. RB, Gordon F, et al., eds. Movement sci-
9. Penfield W. Functional localization in tem- ences: foundations for physical therapy in re-
poral and deep Sylvian areas. Res Publ Assoc habilitation. Rockville, MD: Aspen Systems,
Res Ner\- Ment Dis 1958;36:210-226. 1987:155-177.
10. Milner B. Amnesia following operation on 19. Craik RL. Recoven,' processes: maximizing
the temporal lobes. In: \Vhitt\' CWM, Contemporary management of
fianction. In:
Zang\\ill OL, eds. Amnesia. London: Butter- motor control problems. Proceedings of the
worths, 1966:109-133. II Step Conference. Alexandria, VA: APT A,
11. Bliss TVP, Lomo T. Long-lasting potentia- 1992:165-173.

tion of synaptic transmission in the dentate 20. Merzenich MM. Sources of intraspecies and
area of the anaesthetized rabbit following interspecies cortical map variabilit}' in mam-
stimulation of the pertbrant path. J Physiol mals: conclusions and hypotheses. In: Cohen
(Lond) 1973;232:331-356. MJ, Strumwassser F, eds. Comparative neu-
12. Morris RGM, Anderson E, Lynch GS, rology: modes of communication in the ner-
Baudn- M. Selective impairment of learning vous s\'stem. New York: John Wiley & Sons,
and blockage of long-term potentiation by an 1985:105-116.

N-methyl-D-aspartate receptor antagonist, 21. Merzenich MM, Kaas JH, Wall J, Nelson RJ,
APS. Nature 1986;319:774-776. Sur M, Felleman D. Topographic reorgani-
13. Gilbert PFC, Thach WT. Purkinje cell activity zation of somatosensor\' cortical areas 3B and
during motor learning. Brain Res 1977;70:1- 1 in adult monkeys following restricted de-
18. afferentation. Neuroscience 1983;8:33-55.
14. Melville-Jones G, Mandl G. Neurobionomics 22. Merzenich MM, Kaas JH, Wall JT, Sur M,
of adaptive plasticit\': integrating sensorimo- Nelson RJ, Felleman DJ. Progression of
tor fianction with environmental demands. change following median nen'e section in the
In: Desmedt JE, ed. Motor control mecha- cortical representation of the hand in areas 3b
nisms in health and disease. Adv Neurol and 1 in adult owl and squirrel monke\'s.
1983;39:1047-1071. Neuroscience 1983;10:639-665.
15. Mishkin MH, Malamut B, Bachevalier J. 23. Pons TP, GarraghtN' PE, Mishkin M. Lesion-
Memories and two neural systems. In:
habits: induced plasticity in the second somatosen-
McGaugh JL, Lynch G, Weinberger NM, sory' cortex of adult macaques. Proc Nad
eds. The neurobiology' of learning and mem- Acad Sci USA 1988;85:5279-5281.
ory. New York: Guilford Press, 1984:65-77. 24. Jenkins, WM, Merzenich MM, Och MT, Al-
16. Gordon J. Assumptions underlying physical lard T, Guic-Robles E. Functional reorgani-
therapv inter\'ention: theoretical and histori- zation of priman' somatosensorx' cortex in
cal perspectives. In: Carr JH, Shepherd RB, adult owl monkeys after behaviorally con-
Gordon F, et al., eds. Movement sciences: trolled tactile stimulation. J Neurophysiol
foundations for physical therapy in rehabili- 1990;63:82-104.
tation. Rock\ille, MD: Aspen Systems, 25. Wall PD, Egger MD. Formation of new con-
1987:1-30. nections in adult rat brains after partial deaf-
17. Steward O. Reorganization of neuronal con- ferentation. Nature 1971;232:542-545.
nections following CNS trauma: principles 26. Clark SA, .\llard T, Jenkins WM, Merzenich
and experimental paradigms. J Neurotrauma MM. Receptive fields in the body-surface
1989;6:99-151. map in adult cortex defined b\' temporally
18. Held JM. Recover)' of ftinction aft:er brain correlated inputs. Nature 1988;332:444-
damage: theoretical implications for thera- 445.
Chapter 5
A Conceptual Framework for

Clinical Practice
Introduction Abnormal Motor control

Conceptual Framework for Clinical Recovery of Function and the
Intervention Reacquisition of Skill
Clinical Decision-Making Process Task-Oriented Conceptual Framework for

L Hypothesis-Oriented Clinical Practice Clinical Intervention
K Models of Disablement Assessment
i WHO Model Performance-Based Functional
W Nagi Model Assessment
JF" Schenkman Model A General Taxonomy of Movement
Clinical Implications Tasks
Theories of Motor Control Limitations of Functional Tests
Parallel Development of Clinical Strategy Assessment
Practice and Scientific Theor\' Limitations
Neurological Rehabilitation: Reflex-Based Impairment Assessment
Neurofacilitation Approaches Integrating Hypothesis Testing into
Underlying Assumptions Assessment
Normal Motor Control Making the Transition from Assessment to
Abnormal Motor Control Treatment
Recovery of Function and the Interpreting Assessment Data
Reacquisition of Skill Setting Treatment Goals
Clinical Implications Long-Term Goals
Limitations Short-Term Goals
Changing Practices Clinical Implications —
Treatment
Systems-Based Task-Oriented Approach Retraining Strategies: Recovery Versus
Underlying Assumptions Compensation
Normal Motor Control Summary
Introduction propriate outcomes for evaluating the effects

of m\- treatment?
Clinicians responsible for retraining These questions reflect the critical need
movement in the patient \\ith neurological for a conceptual framework for clinical prac-
impairments are faced with an overwhelming tice. A conceptual framework is a logical
number of decisions. What is the most appro- structure that helps the clinician organize
priate way to assess my patient? How much clinical practices related to assessment and
time should be spent on documenting fimc- treatment into a cohesi\e and comprehensive
tional abilitv versus evaluating underlying plan. It provides the clinician with guidelines
problems leading to dysfianction? WTiat cri- for how to proceed through the clinical inter-
teria should I use in deciding what the priority \ ention process.
problems are? How do I establish goals that Clinical practices related to retraining
are realistic and meaningful? What should be the patient withmotor control problems are
treated? What is the best approach to treat- changing number of factors,
in response to a
ment and the most etfccti\e w ay to structure including new views on the physiological basis
my therapy sessions? WTiat are the most ap- of motor control. As new models of motor
99
control evolve, clinical practices are modified assumptions about the cause and nature
to reflect current concepts in how the brain of normal and abnormal movement are
controls movement. Thus, a conceptual derived.
framework for structuring clinical practice is
dynamic, changing in response to new scien- These assumptions guide the clinician in
tific theories about motor control.
making decisions about key elements to assess
The purpose of this chapter is threefold: and treat when retraining the patient with a
(«) to consider elements that contribute to a
movement disorder.
comprehensive conceptual fi-amework for The following sections describe each of
clinical practice; (b) to discuss the changing these important components in detail.
faceof clinical practice and its relationship to

underhing theories of motor control; and (c) Clinical Decision-Making Process
to describe a conceptual framework for re-
Mrs. Claire Stern has been referred for ther-
training the patient with mo\'ement disorders,
apy with a histor)' of recurrent falls. She is 72
which we call a task-oriented approach. A years old, and lives alone in an assisted li\ing
task-oriented approach is used in later chap- retirement center. She walks with a cane, and
ters as the framework for retraining posture, while she used to be fairly active, walking a half-
mobilitA', and upper extremity' control in the mile e\'er\' dav with her neighbor, since her last
patient with a neurological deficit ( 1 ). two falls, she is reluctant to leave her apartment.
She is becoming less and less active, is ha\ing
Conceptual Framework increasing difficuln,' in getting around her re-
tirement home, and referred for therapy. She
FOR Clinical is referred for balance
is
and mobility' retraining,

Intervention to reduce the likelihood of falling again.
Again, a conceptual framework for clin- Mr. George Johnson is a 68-year-old man
ical practice provides a structure for clinical who was diagnosed with Parkinson's disease ap-
intervention. It guides the clinician through proximateh' 15 years ago. He lives in his own
the inter\'ention process, unifying clinical home with his wife, who is in relatively good
practices related to assessment and treatment. health. He is spending more and more time sit-
ting, and his balance and walking have become
We propose that there are four key concepts
increasingly worse as has his abilirv' to assist in
or elements that contribute to a comprehen-
his own transfers. His wife is finding it increas-
sive conceptual framework for clinical prac-
ingly difficult to assist him during transfers.
tice. These include: Thev are referred for therapy to rs)- to improve
Mr. Johnson's mobilit\' skills, in particular, to
\. The clinical decision-making process,
improve his independence in transfer abilities.
which is a procedure for gathering in-
formation essential to developing a plan Sam Churchill is an 18-year-old with a re-

of care consistent with the problems cent histon.' of a motor vehicle accident in which
and needs of the patient; he suffered a closed head injur\-. Primar\' pa-
2. A hypothesis-oriented clinical prac- thology' was to the cerebellum. In addition, Sam
tice, which provides the means to sys- has significant cognitixe impairments, including
tematically test assumptions about the attention and memoiy problems. He is unable
to stand and walk independendy due to se\ere
nature and cause of motor control
ataxia, and is dependent in most of his activities
problems;
of daily living (ADL) due to dysmetria and dys-
3. A model of disablement, which im-
coordination. He spent 4 weeks in coma, but
poses an order on the eftects of disease
with the return of consciousness has been ad-
and enables the clinician to develop a mined to the unit to begin rehabilitation.
hierarchical list of problems towards
which treatment can be directed; and Sara is a 3-vear-old child who was bom
4. A theory of motor control from which with cerebral palsy, and has moderate spastic
Chapter Five CONCEPTUAL Framework FOR Clinical Practice 101
hemiplegia. She has been in an early inter\'en- fective treatment program requires that the
tion program was 4 months old. She
since she clinician identify' the patient's functional
has recently mo\ed into a new area, and is re- problems, and determine the underlying
ferred for a continuation of her therapy to im-
cause(s).
prove posture, mob)ilir\', and upper extremit\'
The process of identif\ing problems and
stalls.
tiicir underlying cause(s) is not always easy.
This diverse group of patients is D,'pical Most C-NS pathology affects multiple systems,
of those referred for retraining motor control resulting in a diverse set of impairments. This
problems aflecting their abilit)' to move and means that functional problems in the patient
earn- out activities of daily life. Can the same with a neurological deficit are often associated
approach used to assess motor control in an with many How does a ther-
possible causes.
elderly man with Paridnson's disease be ap- apist establish a link between impairment and
propriate for an 18-year-old head-injured pa- ftinctional disability? Which impairments arc
tient? Can the same approach to retrain- critical to loss of fiinction? Which impair-
ing posture and mobility,' problems in a 72- ments should be treated and in what order?
year-old elderly lallcr be used to habilitate What is the most efficacious approach to
mobility in a 3-year-oid child with cerebral treatment? Hypothesis-driven clinical practice
palsy? can assist the clinician in answering some of
As you will see, the answer to these these questions (4).
questions is yes. Despite the diversiri' of these
patients, the clinical ciecision- making process Hypothesis -Oriented Clinical
used to gather information and design an in- Practice
tenention program is similar for all patients.
While each patient's motor control problems What is a hypothesis and how do we use
and therapeutic solutions may be different, it in the clinic? A hypothesis can be defined
the process used to icientif)' problems and es- as a proposal to explain certain facts. In clin-
tablish a plan of care will be consistent across ical practice, it can be considered one possible
patients. explanation about the cause or causes of a pa-
Clinical decision making is the process tient's problem (4, 5). To a great extent, the
of gathering information essential to deyel- hypotheses generated reflect the theories a cli-
oping a plan of care consistent with the prob- nician has about the cause and nature of func-
lems and needs of the patient (2,3). The clin- tion and dysfiinction in patients w ith neuro-
ical decision-making process involves (a) logical disease (6). As noted in Chapter 1,
assessment of the patient, (b) analysis and in- there are many theories of motor control that
terpretation of the assessment data, (c) devel- present yar\'ing views on the nature and cause
opment of short- and long-term goals, (
d) de- of movement. As a result, there can be many
velopment of an appropriate treatment plan different hypotheses about the underlying
to achieve these goals, (e) carni'ing out the cause(s) of motor control problems in the pa-
treatment plan, and (/) reassessment of the tient with neurological dysfunction.
patient and treatment outcome. Clarifii'ing functional movement prob-
The purpose of clinical decision making lems requires the clinician to (rt) generate sev-
is to establish a scientifically sound and cost- eral alternative hypotheses about the potential
effective plan of care geared to the problems cause(s); (b) determine the crucial test(s)and
and needs of each individual patient. The first their expected outcomes, which would rule
step in establishing a plan of care is assess- out one or more of the hypotheses; (c) carr\'
ment. A good definition of assessment is the out the tests; and (//) continue the process of
systematic acquisition of information that is generating and testing hypotheses, refining
relevant and meaningful in providing the cli- one's understanding of the cause(s) of the
nician with a comprehensive picture of the pa- problem (5).
tient's abilities and problems. Planning an ef- The generation and testing of hypoth-
A WHO model
Pathology -
Impairments Disability —> Handicap
B Nagi model
Pathology - Impairments Functional —> Disability
limitations
C Schenkman model
Pathology
Example
> Force control —-^ Reach —>-ADLs

Tone Walk Self-care
Coordination Transfers Mobility
Sensory loss Lift Employment
Balance
etc
Figure 5.1.Models of disablement. Illustrated are three models of disablement. A, The WHO. B, The Nagi. C, The
Schenkman. The effects of a cerebral vascular accident (CVA) at the various levels are also described.
eses are an important part of clinical practice. chical system for categorizing patient prob-
However, there is a difference between hy- lems and can be used as a framework for
pothesis testing in a research laboraton' versus organizing and interpreting assessment data,
in a clinic. In the laborator\', it is often pos- and developing a comprehensive plan for
sible to set up a careftilly controlled experi- treatment (7). Three models are reviewed in
ment that will test the hypotheses. The out- this chapter.
come is a clean result, that is, a result that

accepts one hypothesis and rejects the alter- WORLD HEALTH ORG.\NIZATION
native hypothesis. In contrast, in the clinic, we MODEL
are often unable to get a clean result. Clinical
tests are often not sensitive and specific The International Classification of Im-
enough to clearly differentiate between two pairments, Disabilities and Handicaps is a
hypotheses. Rather, they indicate the likeli- model of disablement developed by the
hood for the origin of the problem. Despite World Healtii Organization (WHO) (8). The
the limitations of clinical tests, the generation, WHO model categorizes problems according
testing,and revision of alternative hypotheses to four levels of analysis: pathology', impair-
are important in the clinical decision-making ment, disability', and handicap. This model is
process. shown in Figure 5. 1 A.

The first level, the patholojjy le\'el of
Models of Disablement analysis, represents a description of the disease
or injun,' process at the organ level. The sec-
While the clinical decision making-pro- ond level, impairment, includes psychologi-
cess suggests how to proceed, it does not shed cal, physiological, or anatomical problems re-
light on what to assess. A different approach lated to structure or ftinction, such as
is needed to answer the questions: What shall decreased strength or range of motion
I assess.' Toward what goals should I direct ( ROM), or the presence of spastic hemiplegia.
my treatment.' In what order should problems The third le\'el, disabilit}', represents a distur-
be tackled.' A model of disablement suggests bance in task-oriented or functional behav-
a framework for structuring the effects of dis- iors, such as walking, climbing, transferring,
ease on the individual. It suggests a hierar- lifting, reaching, or maintaining a posture. Fi-
Chapter Five Conc;ei>tuai. Framkvvork for Ci,inic:ai. Practice 103
nally, the fourth level, handicap, is defined Schenkman fiirther divides impairments
with respect to the and family net\\()rk
societ)' into those that arc the direct effect of patho-
of the patient. Categories of handicap include physiology, those that result indirectly from
physical dependence and mobility', occupa- pathology, and those that are the composite
tion, social integration, and economic self- effects of both direct and indirect impair-
sufficiency. The degree of handicap is not usu- ments. important to differentiate between
It is
ally established by one professional, but rather direct, or primar\', impairments and indirect,
through the comprehensive assessment of the or secondary', impairments. Secondar\' im-
patient by a team of professionals. pairments develop as a result of the primary
impairments, not the patholog\' itself For ex-
NAGI MODEL ample, in the patient with UMN disease, mus-
culoskeletal contractures can dexelop second-
The Nagi model, shown in Figure 5. IB,
ary to weakness and immobilit\' (primary
also contains four levels of dysfunction (9-
impairments). However, secondan,' impair-
11). The first two levels, patholoffy and im-
ments can often be prevented with appropri-
pairment, are consistent with the WHO ter-
ate treatment.
minology'. The remaining two levels are con-
ceptually similar to the WHO model, but the CLINICAL IMPLICATIONS
terminology is difterent. In the Nagi model,
How do models of disablement assist
the next level of dysfijnction following im-
the clinician in formulating a clinical plan for
pairment is functional limitation (comparable
intervention? Figure 5. ID illustrates how
to the disability level in the WHO model).
these three models would potentially describe
Functional limitations describe a patient's
the effects of a cerebral vascular accident
problems with reference to fijnctional tasks.
(CVA) at the various levels. Clinicians are pri-
At the top of the disablement hierarchy is the
marily involved in identitv'ing and document-
disability level of dysfunction, which reflects
ing the effects of pathology at both the im-
the inability of the individuals to carrv' out
pairment and disabilit}' levels (10, 12).
their roles in society. This is roughly equiva-
During assessment, clinicians identify' and
lent to the handicapped level in the WHO document limitations in the patient's func-
model. Many clinicians prefer the Nagi model
tional capacity, for example, the ability to
because of the growing pressure in societ)' to
walk, transfer, reach for, and manipulate ob-
discontinue the use of the term "handi-
jects. In addition, clinicians determine and
capped" (10).
document the sensory, motor, and cognitive
SCHENKMAN MODEL impairments that constrain fianctional abili-

ties. These impairments can be the direct re-
Margaret Schenkman, a physical thera- sult of a neurological lesion, for example,
pist, has also suggested a model of disable- weakness, or the indirect effect of another im-
ment to be used as the basis for a multisystem pairment, such as contractures in the weak
evaluation anci treatment of individuals v\'ith and immobile patient.
neurological impairments (12). Her model,
shown in Figure 5.1 C, is composed of three
Theories of Motor Control
levels: pathophysiology, impairments, and The fourth element that contributes to
disabilities. Similar to the previous models, a comprehensive conceptual framework for
impairments refer to abnormalities within clinical practice is a theory of motor control.
organs and systems which constrain a
specific Theories of motor control have led to the de-
patient's ability' to fimction normally, for ex- velopment of clinical practices, which then ap-
ample, spasticity', weakness, or loss of joint ply assumptions from these theories to im-
mobility. Disability refers to functional re- proving the control of movement. Thus, the
strictions, for example, problems with gait, approach a clinician chooses when assessing
bed mobility, or transfers. and treating a patient with movement disor-
104 Section I THEORETICAL FR.\ME\VORK
ders is based in part on both implicit and ex- developed, resulting in a dramatic change in
plicit assumptions associated with an under- clinical treatment of the patient with neuro-
lying theon' of motor control (1, 13-15). logical impairments (1, 13). For the most
part, these approaches still dominate the way
PARALLEL DEVELOPMENT OF clinicians assess and treat the patient with
CLINICAL PRACTICE AND neurological deficits.
SCIENTIFIC THEORY Neurofacilitation approaches include
the Bobath Approach, developed by Karl and
Much has been written recendy about Berta Bobath (16-18), the Rood Approach,
the influence of changing sciendfic theories developed Margaret Rood (19-20),
by
on the treatment of padents with movement Brunnstrom's approach, developed by Signe
disorders. Several excellent articles discuss in Brunnstrom (21), Proprioceptive Neuromus-
detail the parallel development between cular FaciUtation (PNF), developed by Kabat
scientific theor\' and clinical practice (1, 13- and Knott and expanded by Voss (22), and
15). Sensor\' Integration Therapy, developed by
Neuroscience researchers identify the Jean Ayres (23-25). These approaches were
scientific basis formo\'ement and movement based largely on assumptions drawn fi-om
disorders, but up to the clinician to de-
it is both the reflex and hierarchical theories of
velop the applications of this research (13). motor control (1, 13, 15).
Thus, scientific theor\' pro\ides a framework Prior to the development of the neuro-
that allows the integration of practical ideas facilitation approaches, therapy for the patient
into a coherent treatment philosophy. As we with neurological dysfianction was largely di-
mentioned in Chapter 1, a theory' is not right rected at changing function at the level of the
or wrong in an absolute sense, but judged to muscle itself This has been referred to as a
be more or less useful in solving the problems muscle re-education approach to treatment (1,
presented by patients with movement dys- 13). While the muscle re-education approach
flinction (1, 13). was effective in treating movement disorders
Just as scientific assumptions about the resulting fi-om polio, it had less impact on al-
important elements that control movement tering movement patterns in patients with up-
are changing, so too, clinical practices related per motor neuron lesions. Thus, the neuro-
to the assessment and treatment of the patient facilitation techniques were developed in
with a neurological deficit are changing. New response to clinicians' dissatisfaction with
assumptions regarding the nature and cause previous modes of treatment, and a desire
of movement are replacing old assumptions. to develop approaches were more
that
Clinical practice evolves in parallel with sci- effective in solving the movement problems
entific theoPi', as clinicians assimilate changes of the patient with neurological dysfunction
in scientific theory and apply them to practice. (13).
Let's explore the evolution of clinical practice Clinicians working with patients with
in light of changing theories of motor control UMN began to direct clinical efforts
lesions
in more detail. towards modifying the CNS itself (13). Neu-
approaches focused on retrain-
rofacilitation
Neurological ing motor control through techniques de-

signed to facilitate and/or inhibit different
Rehabilitation: Reflex-
movement patterns. Facilitation refers to
Based Neurofacilitation treatment techniques that increase the pa-
Approaches tient's abilit)' to move in ways judged to be
appropriate by the clinician. Inhibitor\' tech-
In the late 1950s and early 1960s, the niques decrease the patient's use of move-
so-called neurofacilitation approaches were ment patterns considered abnormal.
Chapter Five Conceptlal Framework FOR CuxiCAL PR.\cncE 105
UnderhTng Assumptions ior\- to the lesion (13*. Thus, it is predicted

that in the child with L'MN lesions the pro-
NOIL\L\L MOTOR CONTROL cess of increasing corticalization is disrupted,
and motor control is dominated by
as a result,
Neurofacilitarion approaches are largely
primitive patterns of movement organized at
associated with both the reflex and hierarchi-
lower levels of the CNS. In addition, in the
cal theories of motor control (1, 13, 15).
adult with acquired L'MN lesions, damage to
Thus, clinical practices have been developed
based on assumptions regarding the nature
higher levels of the CNS probably results in a
release of lower centers from higher center
and cause of normal motor control, abnormal
control. Likewise, primitive and pathological
motor and the recoverv of function
control,
behaviors organized at these levels re-emerge
(1, 13; see also Chapter 1 of this text).
to dominate, pre\enting normal patterns of
For example, it is assumed that reflexes
movement from occurring ( 1, 14, 15 1.
are the basis for motor control. This approach
suggest that normal movement probably re-
sults from a chaining of reflexes that are or-
RECO\TRY OF FL'NCTION .\XD
ganized hierarchically within the CNS. Thus,
RE.\CQLTSlTION OF SKILL
control of movement is top-down. Normal
mo\'ement requires that the highest level of
A central assumption concerning the re-
the CNS, the cortex, be in control of both
covers of hinction in the patient with a L'MN
lesion is that recovery of normal motor con-
intermediate (brainstem) and lower ^ spinal
trol cannot occur unless higher centers of the
cord) levels of the CNS. This means that the
process of normal development, sometimes
CNS once again regain control over lower
centers.According to this approach, recovery
called corticalization, is characterized by the
of hinction in a sense recapitulates develop-
emergence of beha\iors organized at sequen-
ment, with higher centers gradually regaining
tially higher and higher levels in the CNS. A
their dominance over lower centers of the
great emphasis is placed on the understanding
CNS.
thatincoming senson." information stimulates,
and thus drives, a normal movement pattern
Two key assumptions are that a ( > fiinc-
once ab-
tional skills will automatically rettun
(D-
normal movement patterns are inhibited and
.\BNOR.\L\L MOTOR CONTROL normal mo\ ement patterns facilitated; and b) (
repetition of these normal movement patterns

Explanations regarding the physiologi- will automatically transfer to fiinctional tasks.
cal basis for abnormal motor control from a
reflex and hierarchical f>ersp>ective largely sug- Clinical Implications
gest that a disruption of normal reflex mech-
anisms underlies abnormal movement con- \\'hat aresome of the clinical implica-
trol. It is assumed that lesions at the highest tions of these assumptions? First, assessment
cortical levels of the CNS cause release of ab- of motor control should focus on idenrif\ing
normal reflexes organized at lower levels the presence or absence of normal and ab-
within the CNS. The release of these lower normal reflexes controlling movement. Also,
level reflexes constrains the patient's abilirv" to treatment should be directed at modif\ing the
move normally, .\nother prevalent assump- reflexes that control movement. The impor-
tion is that abnormal or atvpical patterns of tance of sensor.- input for stimulating normal
movement seen in the patient with L'MN le- motor output suggests a treatment focus of
sions are the direct result of the lesion itself, modifving the CNS through sensorv^ stimu-
as opp>osed to considering some behaviors as lation 1 1, 13). A hierarchical iheon- suggests
developing either secondary- to the lesion or thatone goal of therapy is to regain indep>en-
in response to the lesion, that is, compensa- dent control of movement bv higher centers
106 Section I THEORETICAL FRy\MEWORK
of the CNS. Thus, treatment is geared to- intervention, which is based on a systems the-
wards helping the patient regain normal patory' of motor control. As we mentioned in
terns of movement as a way of facilitating Chapter 1, a task-orientedapproach to re-
fiinctional reco\er\'. training is a term used to describe a newer
neurological rehabilitation approach that is
Limitations evolving in parallel with new theories of mo-

tor control ( 1 ). Others have referred to these
More recently, questions have been new clinical methods as a motor control ap-
raised about the assumptions related to neu- proach (13). In the past, we have referred to
rofacilitation models (13-15). Dissatisfaction this new clinical approach as a systems approach
with neurotacihtation approaches is reflected (14-15, 26). However, it has recentiy been
in a growing number of questions regarding suggested that separate names be given
their underlying assumptions, including: Can to each to distinguish between clinical treat-
inhibition of abnormal reflexes alone prociuce ment approaches and their theoretical bases
more normal patterns of movement? Will this (1).
carry over to improved fiinction? Are the atyp-
ical movement patterns seen in patients with Underlying Assumptions
neurological impairments the result of the ab-
normal CNS, or compensatorv' to the prob- NORMAL CONTROL OF MO\"EMENT
lem? Is it appropriate to train a patient to use
a particular pattern of movement when the Some underlying assumptions that
hallmark of normal fiinction is variabilitv' of guide a task-oriented approach to retraining
movement strategies? follow. First, normal movement emerges as an
interaction among many different systems,
Changing Practices each contributing different aspects of control.
In addition, movement is organized around a
The neurofacilitation approaches still behavioral goal; thus, multiple systems are or-
dominate the way clinicians assess and treat ganized according to the inherent require-
the patient with UMN lesions. However, just ments of the task being performed. These as-
as scientific theor\' about the nature and cause sumptions suggest that when retraining
of movement has changed in the past 30 mo\ement control, it is essential to work on
years, so too,many of the neurofacilitation identifiable fiinctional tasks, rather than on
approaches have changed their approach to movement patterns for movement's sake
practice. Currently, within the neurofacilita- alone.
tion approaches, there is a greater emphasis Another key assumption in this ap-
on and less em-
explicitly training fianction, proach is the recognition that organization of
phasis on inhibiting reflexes and retraining the various elements contributing to move-
normal patterns of movement. In addition, ment is determined bv various aspects of
also
there is more consideration of motor learning the environment. This means that strategies
principles when developing treatment plans. for moving and sensing emerge from an in-
The boundaries between approaches are less teraction of the individual with the environ-
distinct as each approach integrates new con- ment to accomplish a functional task. Thus,
cepts related to motor control into its theo- both fiinctional goals and environmental con-
retical base. straints play an essential role in determining
movement.
Systems -Based Task- Finally, the role of sensation in normal
Oriented Approach movement is not limited to a stimulus-re-
sponse reflex mode. Instead, sensation is hy-

One of the newer approaches to retrain- pothesized to contribute to predictive and
ing is the task-oriented approach to clinical adaptive control of movement as well.
Chapter Five Conceptual Framework HOR Cunical Practice 107
ABNORMAL MOTOR CONTROL ented approach intoa complete conceptual

framework which includes these other ele-
From a systems perspective, abnormal ments.
motor control results from impairments Using the clinical decision-making pro-
within one or more of the systems controlling cess, we can identih- the steps to follow during
movement. In addition,movements observed the course of clinical intcnention, including
in the patient with a UMX lesion represent
assessment, identification of problems and
behavior that emerges from the best mix of goals, and the establishment of a treatment
the systems remaining to participate. This plan to achieve those goals. Theof r\vo le\els
means that what is obsened is not just the analysis from Schenkman's model help us to
result of the lesion itself, but the eflbrts of the identifi' problems that are commonly assessed
remaining systems to compensate for the loss
and treated by physical and occupational ther-
and still be fijnctional. However, the compen- apists: impairments and functional disability-.
sators' strategies developed by patients are not In addition, we add a third level of analysis
always optimal. Thus, a goal in treatment
that focuses on identitx'ing the strategies pa-
may be to improve the efficiency of compen- tients use to perform functional tasks despite
saton,' strategies used to perform fianctional their impairments. This represents an inter-
tasks.
mediate of analysis, betvveen fiinctional
le\el
and underlying impairments.
disability'
RECO\'ERY OF FUNCTION AND THE Finalh', we draw on a systems theor\' of
REACQUISITION OF SKILL motor control to generate hypotheses about
the potential causes of fiinctional movement
A systems perspective suggests that pa-
disorders and their treatment. We will con-
tients learn by acti\ely attempting to sohe the
tinue to refer to this framework as a task-ori-
problems inherent to a frmctional task, rather
than repetitively practicing normal patterns of
ented approach. The remaining section of this
movement. Adaptation to changes in the en- book discusses in more detail the essential el-
vironmental context is a critical part of recov- ements of assessment and treatment based on
er}' of function. In this context, patients are
a task-oriented approach. In later chapters, we
helped to learn a variety' of ways to solve the
will show the specific application of this ap-
task goal rather than a single muscle activation

proach to retraining posture, mobility', and
pattern.
uppper cxtremip,' fiinction in the patient with
neurological dysfianction.
Task- Oriented
Assessment
Conceptual Framework
FOR Clinical We begin with assessment, the first step
Intervention in the clinical decision-making process. A

task-oriented assessment evaluates motor be-
In the beginning of this chapter, we dis- havior at three levels: (a) objective measure-
cussed the importance of a comprehensive ment of fiinctional skills, (b) a description of
conceptual framework for guiding clinical the strategies used to accomplish fiinctional
practice. We suggested there were four key and (c) quantification of the underlying
skills,
elements in a comprehensive conceptual sensor>', motor, and cognitive impairments

framework including: the clinical decision- that constrain performance (27-29).
making process, hypothesis-oriented practice, Since there is no single test that allows
models of disablement, and a theop,' of motor one to collect information at all levels, clini-
control. We have just discussed the assump- cians are required to assemble a batter\' of
tions underlying a task-oriented approach to tests, enabling them to document problems
retraining, based on a systems theori' of motor at all three levels of analysis. This concept is
control. We now incorporate our task-ori- illustrated in Figure 5.2, which examines the
Levels of Assessment
Functional Assessment -> Strategy Assessment ->- Impairment Assessment
Examples of Tests and Measurements

Gait Tests Gait Strategies Strength tests
3 MIn Walk Tests Movement ROM tests
Dynamic Gait Index Sensory Coordination tests
Tinetti Mobility Test Adaptive Muscle tone tests
Reflex tests
Sensory tests
Cognitive tests
Balance Tests Balance Strategies

Functional Reach Test Movement
Tinetti Balance Test Sensory
Ataxia Test Battery Adaptive
Sit-to-stand strategies
ADL Tests
Katz Index
Movement
Sensory
FIM
Adaptive
Barthel Index
lADL Tests
OARS
Lawton
SIADL
Figure 5.2. A conceptualization of the relationship between the three levels of testing within a task-oriented model,
and the types of tests a clinician could choose from within each of these levels.
three levels of testing within a task-oriented the Katz Index (30); Functional Indepen-
model, and the t>'pes of tests a clinician could dence Measure (FIM) (31); and the Barthel
choose from within each of these levels. The Index (32).
figure is not intended to present a compre- lADL scales assess activities in which the
hensive list of alland measurements
tests person interacts with the environment, in-
available within each level, but presents the cluding: telephone usage, traveling, shop-
concept itself ping, preparing meals, housework, and fi-
nances. An example is the Scale for
PERFORMANCE-BASED FUNCTIONAL Instrumental Activities of Daily Living

ASSESSMENT (L\DL)(33).
Some assessment scales limit their focus
Assessment tools that focus on measur- to specific tasks such as balance, mobilit)', or
ing the first level of performance, fijnctional upper extremity' control. Examples of these
abilities, are called performance-based func- t\'pes of assessment tools include: the Tinetti
tional measures. These tests allow the clini- Test of Balance and Mobilit)' (34), the Ataxia
cian to document a patient's level of indepen- Test Batter}' (35), or the Erhardt Test of Ma-
dence in carrying out daily life activities and nipulator}' Skills (36). These tests have been
are an important part of justifying ongoing developed to provide clinicians with a clearer
therapy to the patient, the patient's family, picture of the patient's functional skills related
and third-party insurers. to a limited set of tasks the clinician will be
There are a number of different ap- directly involved in retraining. These task-
proaches to assessing functional behavior. For specific tests will be covered in later chapters,
example, standardized assessment tools have which discuss retraining posture, mobility,
been developed to test Activities of Daily Liv- and upper extremity ftinctions.
ing (ADL), or Instrumental Activities of Daily A number of assessment tools have been
Living (lADL). ADL scales test the patients' developed to assess fimctional limitations and
ability to care for themselves including: bath- underlying impairments in specific patient
ing, dressing, toileting, feeding, mobility, and populations. The most prevalent of this t}'pe
continence. Examples of ADL scales include of tool relates to assessing function following
Chapter Five Concefiual Framework for Clinical Praciic;e 109
stroke. Examples of these include the Motor tient with a neurological disorder? The clas-
Assessment Scale for Stroke Patients (37), the sification of fijnctional movement tasks into
Fugl-Meyer Test (38), or the Motor Assess- distinct categories provides an inherent order
ment Hemiplegia by Signc Bninnstrom
in or structure to tasks that imolve the per-
(39). Several scales have been developed to former, the task, and the environment.
evaluate the severin,' of symptoms associated Gentile's classification of movement
with Parkinson's disease, including the Uni- tasks, shown Table 5.1, represents a hier-
in
fied Rating Scale for Parkinsonism (40) and archy of tasks that could form the basis for an
the Schwab Classification of Parkinson Pro- assessment profile, as well as a progression for
gression (41 ). retraining motor control in the patient with a
neurological disorder. Tasks that have mini-
A General Taxonomy of Movement Tasks mal variation and relatively fixed environmen-
tal features are consiciered simple closed tasks.
Ann Gentile, a motor control scientist Closed tasks require fixed and habitual pat-
from Columbia Universitv' in New York, has terns of movement, and therefore ha\e fairly
proposed a comprehensive approach to cate- limited information processing and atten-
gorizing functional movement tasks (corre- tional demands. In contrast, open tasks vary
sponding to a le\el 1 analysis) based on the gready from trial to trial, ha\e changing en-
goals of the task and the environmental con- vironmental features, and as a result, have
text in which the action takes place (29, 42). large information processing and attentional
She points out that different tasks have inher- demands. Movements used to perform open
end\' ciifferent requirements with respect to tasks are constantly changing, adapting to
the en\ironment and thus make different de- changing task and environmental demands.
mands on senson', motor, and cognitive pro- One limitation of Gentile's classification
cesses. These requirements can be used to scheme of movement tasks is that, while it
classify tasks into a hierarchy according to the represents an interesting theoretical fi-ame-
demands of the task. work for assessing and retraining motor con-
How does a classification of functional trol, a formal application of diis framework to
movement tasks help in the assessment and retraining the patient with movement disor-
treatment of movement disorders in the pa- ders has not yet been proposed.
Table 5.1 . Gentile's Taxonomy of Movement Tasks'
Body Staliilitv Body Transport

Environmental Context No Manipulation Manipulation No Manipulation Manipulation
Stationary Closed Closed Closed Closed

No intertrial Body stability Body stability plus Body transport Body transport
variability manipulation plus manipulation
Stationary Variable Variable Variable Variable

Intertrial Motionless Motionless Motionless Motionless
variability Body stability Body stability plus Body stability Body stability plus
manipulation manipulation
Motion Consistent Consistent Consistent Consistent

No intertrial Motion Motion Motion Motion
variability Body stability Body stability plus Body transport Body transport plus
manipulation manipulation
Open Open Open Open

Intertrial Body stability Body stability plus Body transport Body transport plus
variability manipulation manipulation
'From Gentile A. Skill acquisition: action, movement, and neuromotor processes. In: Carr J, Shepherd R, Cordon ), et al., eds. Movement
science: foundations for physical therapy in rehabilitation. Rockville, MD: Aspen Systems, 1987:1 5. 1
110 Section I THEORETIC\L FR.\ME\VORK
Limitations of Functional Tests factors. The first relates to the demands of the
task and the person's desire for particular
There are a number of limitations in- standards of achievement. The second relates
herent in functional performance-based test- to the capacities, both mental and physical,
ing. Performance-based measures will not that a person brings to the task. The third is
necessarily pro\'ide information as to win the the strategies that the person uses to meet the
patient is dependent in performing functional demands of the task, while the fourth is the
skills. As a result, flinctional tests will not al- abilin,' to choose the most efficient strateg}' for
low the therapist to test hypotheses about the a given task.
cause of motor dysfiinction. Therefore, per- Note that two of the four factors relate
formance-based fiinctional tests will not tell to strategies, emphasizing their importance in
the clinician what to treat, since treatment determining our level of performance. Thus,
strategies are often directed at underhing sen- the strategies we demands of the
use relate the
sorimotor impairments constraining function. task to our capacity- to perform the task. If we
Performance-based measures assess per- choose poor strategies, and the task is diffi-
formance quantitatively rather than qualita- cult, we may reach the limits of our capacities
tively. That is, they e\'aluate the degree to well before we ha\e met the demands of the
which a patient can earn' out a task, but not task. In contrast, inefficient strategies may still
how they perform the task. To understand be effective in carrying out simple, less de-
how a patient is performing a task, we need manding tasks. As capacit}' to perform a task
to focus on a strategy" level of analysis. x\n- declines either because of age or disease, we
other limitation of functional performance- may be unable to meet the demands of a task,
based tests is that they examine performance unless we use alternative strategies to main-
in one instant in time, under a fairly limited tain performance.
set of circumstances. Results from a fiinc- For example, as a young adult you rise
tional- based assessment do not always predict quickly out of a chair without the need to use
performance in less than ideal situations. For your arms. You rely on the abilit\' to generate
example, because a patient can walk safely and momentum using movements of your trunk
independendy with a cane in the clinic does to rise from the sitting position. As you age,
not necessarily mean the patient can (or will) strength may slow ly decline without affecting
walk safely and independendy in a cluttered, your abilit\' to use this strategy' for getting up.
poorly lit home environment. But at some threshold, the loss of strength no
longer allows you to get up using your once
STRATEGY ASSESSMENT effective momentum strategy'. Instead, you
begin to use your arms to get up, thereby
The second level of assessment of motor maintaining the functional ability- to rise from
control examines the strategies used to ac- a chair, albeit with a new strateg)'.
complish frinctional tasks. The term strategy Thus, in the individual with a neurolog-
is not limited to the evaluation of the move- ical deficit,maintaining functional perfor-
ment pattern used to accomplish a task, but mance depends on the capacit\- of the individ-
includes how the person organizes sensor)' ual tomeet the demands of the task in a
and perceptual information necessari' to per- particular environment. WTien impairments
forming a task and how this changes under limit the capacity- to use well-learned strate-
various conditions. gies, the patient must learn new ways to ac-
WTiy is it important for clinicians to ex- complish frinctional tasks despite these limi-
amine the strategies a patient uses when per-
forming a fiinctional task.' One answer is that
the strategies used to perform a task largely Limitations
determine the of performance. Accord-
level
ing to Welford (43), a psychologist from En- Clinicians are hampered in their ability'
gland, performance depends on four different to assess sensor)', motor, and cognitive strat-
Chapter Five CoNCEm.-.\L Fr.\aievvork for Clinical Practice 1 1
egies used to perform daily tasks because as- ative signs associated with UMN disease
sessment tools to evaluate these strategies are would be weakness, or senson- loss.
just being developed. There is only limited in-

formation defining sensorv', motor, and cog- INTEGR.\TING HYPOTHESIS TESTING
nitive strategies in neurologically intact sub- INTO ASSESSMENT
jects. In addition, we know ven," little about
how compensator)- strategies de\elop as a re- Earlier, we described the importance of
sult of neurological impairments. hypothesis testing in clarilying the cause(s) of
Researchers ha\e begun to quantity' fimctional movement problems. We sug-
movement strategies used in functional tasks gested it required the clinician to generate
such as gait, stance postural control, and other and test several alternative h\potheses about
mobilit)- skills such as moving from sit to the potential cause(s), and continue this pro-
stand, supine to prone, and supine to stance. cess until a clear understanding of the cause(s)
Clinical tools to assess movement strategies of the problem emerge (5).
have grown out of these analyses. .\n example For example, a patient with hemiplegia
is the use of obser%ational gait analysis to de- is referred for balance retraining because of
fine the movement strategies used during am- recurrent falls. During the course of your eval-
bulation. uation, vou observe that when standing, the

patient tgrids_tojall^ rimaril\' ia the-±iacka:ard •yr^/«
direction. Your knowledge of normal postural
IMPAIRMENT ASSESSMENT
control suggests the importance of the ankle
Finally, the third level of assessment fo- muscles during the recoverx' of stance balance.
cuses on identihing the impairmentsthat po- You generate three hypotheses that could ex-
tentiallv constrain hinctional movement skills. plain why the patient is falling backwards: («)
This requires an evaluation of the sensor>', weak anterior tibialis muscle, {,b) shortened
motor, and cognitive systems contributing to gastrocnemius, (c) a problem coordinating

movement control. Assessment of the motor the anterior tibialis muscle within a postural
system includes an evaluation of both the response svnergv. \Miat clinical tests can be
neuromuscular and musculoskeletal systems. used to distinguish among these hvpotheses?
Since perception is essential to action, assess- Strength testing indicates the patient is weak
ment of motor control requires the assess- but able to volimtaril\' generate force, thus
ment of sensoPi- and perceptual abilities in the weakening support for the first h\pothesis.
control of movement. Since task-specific Range of motion tests suggest normal passive
movement is performed within the context ot range of motion at the ankle, w eakening sup-
intent and motivation, cognitive aspects of port for the second h\pothesis. In response to
motor control including mental status, atten- the Nudge Test a brief displacement in the
(
tion, motivation, and emotional consider- backwards direction), the patient does not
ations must be assessed. dorsiflex the foot of the hemiplegic leg. The
Impairments that affect motor control inabilit\- to dorsiflex the foot, even though the
can be either direct or indirect effects of the capacit\- to generate force voluntarily is pres-
neural lesion (12). In addition, as first de- ent, suggests support for the third hypothesis.
scribed by Hughlings Jackson, upper motor If it were available, surface electromyography
neuron disease ( UMN
can result in both pos-
l could be used to investigate fiirther whether
itive and negative signs or impairments (44). the anterior tibialis is activated as part ot a
Positive signs refer to the emergence of be- postural svnergv' responding to backwards in-
haviors that are not normally present and con- stabilitv'.
strain motor fiinction. Examples of positive How much confidence can we have that
signs include the presence of increased muscle our clinical tests have given us a clean result,
tone, or involuntar\- movements such as trem- that is a result that clearly supports one hy-
ors. Negative signs refer to the absence of bc- pothesis and rejects the others.' A clean result
ha\iors normally present. An example of neg- depends on clinical tests that are valid ways to
differentiate between underlying problems. INTERPRETING ASSESSMENT DATA

Sometimes this is not the case. For example,
Interpreting the data collected during
in the case presented above, passive range of
motion tests may not be a valid way of pre-
the assessment process is no easy task. A num-
dicting the active range of a muscle during
ber of important issues arise when analyzing
assessment information. For example, by what
dynamic activities. In addition, manual mus-
criteria do we determine normalcy) Most of-
cle testing may not be way to test
a valid
ten, assessment is carried out to distinguish
strength in the patient with upper motor neu-
normal motor behavior from abnormal
ron disease.
behavior, and to determine the most appro-
Despite the limitations of clinical tests,
priate approach to retraining motor dys-
the generation, testing, and revision of alter-
control and regaining functional indepen-
native hypotheses is an important part of the
clinical decision-making process. Hypothesis
dence. This requires that we have some
criteria for determining what "normal"
generation assists the clinician in determining
means.
the relationship between functional limita-
Clinicians are hampered in their abil-
tions and underlying impairments. We treat
ity to discriminate normal from abnormal
those impairments that relate directly to func-
because there are no standards by which to
tional limitations and reach within the scope
judge normal movement fianction. Determin-
of treatments available to us (4).
ing normal performance is often based on our
In summary, a task-oriented approach
visual observations and assumption that if
to assessment is directed at answering the fol-
the person is using a typical strategy for
lowing questions:
moving, he/she is normal. Alternatively, the
1 To what degree can the patient perform patient using an atypical strategy is con-
fiinctional tasks.' sidered abnormal and in need of therapy.
2. What strategies does the patient use However, an important aspect of as-
to perform the tasks, and can he/she sessing motor control is determining whether
adapt strategies to changing task con- the patient is using an optimal strategy
ditions.'
(albeit atypical) given the constellation
3. What are the sensory, motor, and cog- of sensory, motor, and cognitive problems
nitive impairments that constrain how involved.
the patient performs the task, and can Once all three levels of assessment are
these impairments be changed through completed, the clinician can translate these as-
intervention.' sessment data into a list of patient problems

4. Is the patient performing optimally categorized according to functional disabihty,
given the current set of impairments, or problems associated with task-specific strate-
gies, and underlying sensory, motor, and cog-
can therapy improve the strategies be-
ing used to accomplish functional tasks nitive impairments. From a comprehensive
list, the therapist and patient identify the most
despite the impairments?
difficult problems, which will become the fo-
cus for intervention strategies. Thus, a

Making the Transition from initial
list of short- and long-term treatment goals

Assessment to Treatment
are established and a specific treatment plan is
formulated for each of the problems identi-

The next three steps in the clinical
fied.
decision-making process, analysis and inter-
pretation of the assessment data, develop-
SETTING TREATMENT GOALS
ment of short- and long-term goals, and de-
velopment of an appropriate treatment plan, Establishing a reasonable and rational
establish the link between assessment and plan of treatment requires setting appropriate
treatment. short- and long-term goals that are consistent
Chapter Five Conceptual Framework for Clinical PRAcrnch 1 1
with the patients' needs and desires, and lowing goals derived from the three levels of
within their capacity- t(i attain. assessment:
1. Resolve or prevent impairments;

Long-Term Goals
2. Develop effective task-specific strate-
and
gies;
Generally, long-term goals define the
3. Retrain fiinctional goal-oriented tasks.
patient's expected level of performance at the
end of the treatment process. Long-term A critical aspect of retraining functional
goals are often expressed in terms of fianc-
skills is helping the patient learn to adapt task-
tional outcomes, such as (a) amount of in-
specific strategies to changing environmental
dependence, {b) super\ision, or level of assis-
contexts.
tance required to carr\- out a task, or (c) in
These goals are not approached sequen-
relationship to the equipment or environmen-
tially, that is in a set order, but rather in par-
tal adaption needed to perform the task. An
allel. Thus, a clinician may utilize techniques
example of a long-term goal is: the patient will
designed to focus on one or more of the
be able to walk 350 ft using an ankle foot or-
aforementioned goals within the same therapy
thosis with quad cane in 3 minutes with no
session. For example, when retraining mobil-
loss of balance; or, the patient will need min-
ity in a patientwho has had a stroke, the cli-
imal super\ision in all dressing activities.
nician may have the patient work on (a)
strengthening exercises to remediate weak-
Short-Term Goals ness (impairment), (b) improving weightbear-
ing on the involved leg, to produce a more
Short-term goals are goals that are ex-
symmetrical gait pattern (strateg)-), (c) prac-
pected to be achieved in a reasonably short
ticing level walking (fijnctional task) and
period of time, for example, one month.
walking on slighdy uneven surfaces or around
Short-term goals are often defined with re-
obstacles (adaptation).
spect to expected changes at the impairment
level. For example, the patient will gain 15° RETRAINING STRATEGIES:
of knee flexion, or the patient will increase RECO\TRY VS. COMPENSATION
quadriceps strength as indicated by an in-
creased number of standing squats from four A question that frequently arises during
to eight. Alternatively, short-term goals may the course of rehabilitating the patient with a
be derived from long-term goals, which are UMN lesion is how much emphasis should be
broken down into interim steps. For example, placed on promoting recover)' of normal
the patient will walk 10 feet with minimum strategies versus teaching compensatory strat-
assistance. Thus, treatment strategies geared egies for performing a task.' Recover)' of nor-
to attaining short-term goals can focus on res- mal strategies for fijnction is defined as the
olution of impairments and/or achie\ing in- returning capability of the individual to per-
terim steps of functional tasks. form a task using mechanisms previously
used. CompensatoPi' strategies are atypical ap-
Clinical Implications —Treatment proaches to meeting the sensor\' and motor
requirements of the task using alternative
The remaining steps in the clinical prob- mechanisms not t}'pically used.
lem-solving process involve establishing a When to facilitate normal strategies ver-
comprehensive plan of care, carrying it out, sus teach compensator)' strategies is not easy
and evaluating its effectiveness in achie\ing to determine and will var)' from patient to pa-
the short- and long-term goals. tient. Often, the guideline used to determine
A task-oriented approach to establishing when compensator)' strategies should be
a comprehensive plan of care includes treat- taught is time. That is, in the acute patient,
ment strategies designed to achieve the fol- emphasis is on recover)' of normal ftmction.
while in the chronic patient, the emphasis ysisand interpretation of the assessment data,
shifts to maximizing function through com- (c)development of short- and long-term goals,
pensatory strategies. id) development of an appropriate treatment
plan to achieve these goals, carrying out
We have found it helpful in the decision-
(e)
the treatment plan, and (f) reassessment of the

making process to consider the nature of the
patient and assessment of treatment outcome.
impairments themselves. Compensatory strat-
3. During the course of clinical intervention, the
egies will be needed in the case of permanent,
clinician will be required to generate multiple
unchanging impairments, regardless of hypotheses, proposing possible explanations
whether the patient is acute or chronic. An regarding the problem and its cause(s), and
example would be teaching a patient with a must investigate these hypotheses through ob-
permanent loss of vestibular fianction to rely servation, tests, and measurement.
on alternative vision and somatosensor)' cues 4. A model of disablement provides a hierarchi-
for maintaining balance during functional cal for categorizing patient problems
system
impairments are tem-
tasks. Alternatively, if
that can be used as a framework for organiz-
ing and interpreting assessment data.
porary and changeable (either through natu-
5. Clinical practices evolve in parallel with sci-
ral recovery or in response to therapy), the
entific theory, as clinicians assimilate changes
emphasis would be on remediating impair-
in scientific theory and apply them to practice.
ments and recovery of normal strategies for
Neurofacilitation approaches to treatment
action. were developed in parallel with the reflex and
A
problem arises when it is not known hierarchical theories of motor control. New
whether impairments v\all resolve. For exam- approaches to treatment are being developed
ple, in the acute CVA patient with flaccidity, in response to changing theories of motor con-
it is ofiien not possible to predict whether the trol.
patient will remain flaccid or regain control 6. A task-oriented approach to clinical interven-
over affected extremities. In this case, the cli- tion is based on a systems theory of motor
approach is the as-
control. Crucial to this
nician may revert to a time-based decision-
sumption that movement emerges as an inter-
making process, working towards recovery of
action among many different systems that are
normal strategies in the acute patient, and
organized around a behavioral goal and var-
switching to a compensatory focus in the
ious aspects of the environment.
chronic patient. 7. A task-oriented assessment evaluates behavior
We will be discussing treatment strate- at three levels including (a) objective mea-
gies in greater depth in later chapters focusing surement of functional skills; (fa) a description
on retraining posture, mobility, and upper ex- of the strategies used to accomplish functional
tremit)' function. skills; and (c) quantification of the underlying
sensory, motor, and cognitive impairments
that constrain performance.
Summary 8. A task-oriented approach to treatment focuses
on (a) resolving or preventing impairments, (fa)
1 A comprehensive conceptual framework tor
developing effective task-specific strategies,
clinical practice is built upon four key ele-
and (c) retraining functional goal-oriented
ments: (a) the clinical decision-making pro-
tasks.
cess that establishes the steps for intervention;
9. A critical aspect of retraining functional skills
(b) hypothesis-oriented practice, which
is helping the patient learn to adapt task-spe-
provides a process for testing assumptions re-
cific strategies to changing environmental
garding the nature and cause of motor control
contexts.
problems; model of disablement that im-
(c) a
poses a hierarchical order on the effects of dis-
ease on the individual; and id) a theory of mo-
References
tor control that suggests essential elements to
assess and treat. 1. Horak F. Assumptions underlying motor
2. The clinical decision-making process in- control for neurologic rehabilitation. In:
volves: (a) assessment of the patient, (b) anal- Contemporary management of motor con-
Chapter Fi\Tc Concepti.'ai. Framework for Cunical Practice 115
trol pfobkms. Proceedings of tfac II Siq? 14. Woolbcott M, Shumw-ay-Cook A. Changes

Conference. Alcxandm, VA: APTA, in posture cootrol across the hfc spian: a sys-
1991:11-27. tems appnxach. Phys Thcr 1990;70:7W-
O'SuUivan S. Clinical decision making: plan- 807.
ning cffectrve treatments. In: O'SuIlivan S, 15. Horak F, Shumway-Cook .\. Clinical impli-
Schmitz T, eds. Ph>-sical rehabilitation: as- cationsof postural control research. In: P
sessment and treatment. 2nd ed. Philadephia: DutKan, ed. Balance. Alexandria, VA; .\PTA,
FADa\is, 1988:1-7. 1990:105-111.
Wolf S. Summation: identification of prind- 16. Bobath B, Bohath K. Motor dcvêiopmcnt in
pics undeit\'ing clinical decisions. In: Wolf S, of cerebral pals>-. Loodon:
difietent t>pes
ed. Clinical decision making in ph>'scai ther- Heinemann, 1976.
ap>-. Philadelphia: FA DaxTS, 1985:379-384. 17. Bobatfa K, Bobatfa B. The nctlrodc^'clop-
Rotfastein J, Echtemach JL. H\pothesis-ori- mcntal treatment. In: Sciunon D, ed. Man-
ented algorithm tor clinicians: a method for agement of the motor disorders of cerebral
and tncatment planning.
e\-aluation Phn pals\". CHn Dev Med No 90. London: Hd-
Thcr 1986;66:1388-1394. nemann, 1984:6-18.

Plan JR. Strong infereiKC. Science 1964; 18. Ma\-sion M. The Bobatfa concept: e\-olution
146:347-352. and H, Hirschfcid
application. In: Fotssbcrg
Sheperd K. Thcorj': criteria, importance, and H, eds. Movement disorders in children. Med
impact. In: Contempotarv' maiugemcnt of Sport Sd Basel: Karger, 1992.
motor control problems. Proceedings of the 19. Stockmyer S. An interpretation of the ap-
n STEP Conference. Alexandria, \Ai .\PT.\, proach of Rood to the treatment of neuro-
1991:5-10. muscular dysfiinction. .-Km I Ph>-s Med 1967;
Campbell S. Framework for the measurement 46:950-955.
of neurologic impairment and disabilii>'. In: 20. Minor NL\. Proprioceptix'e neuromuscular &-
Contemporary- management of motor con- dlitation and the approach of Rood. In: Con-
trol problems. Proceedings of the 11 Step temporaiy management of motor control
Conference. .\lcxandria, VA: APTA. probtcms. Proceedings fiom the II Step Con-
1991:143-153. fference. Alexandiia, VA: .\PTA, 1992:137-
International classification of impairment, 139.

disabilities and handicaps: a manual of dassi- 21 Bninnstrom S. Movement tberap\- in
fication relating to the consequences of dis- hemiplegia. New York; Harper & Row;
ease. Gene\a, Switzerland: World Health Or- 1970.
ganization, 1980. 22. Voss D, lonta M, .Myers B. Propriocepti\TC
Nagj SZ. Some conceptual issues in disability neuromuscular ^dlitation: patterns and tech-
and rehabilitation. In: Sussman MD, ed. So- niques. 3rd ed. New York; tiaipcr Ro»- &
ciolog\' and rehabilitation. Washington, E>C: 1985.
Am Sociological Assoc, 1965:100-113. 23. .\\Tes J. Sensory integradoo and learning db-
10. Jette AM. Diagnosis and classification b>- orders. Los Angeks: Western Ps>\JK)kîcaI
ph\-sical therapists: a special communication Ser^^ces; 1972.
Ph\^ Ther 1989;69:967-969. 24. MontgomcT)- P. Neurode\-elopniental treat-
Gucdone A.\. Phracal therapy diagnosis and ment and sensory intcgrati\'e theory. Con-
the relationship between impairments and temporary management of motor control
Ther 1991;" 1:499-504.
fimction. Ph\^ problems. Proceedings from II Step. .Mex-
12. Schenkman M, Buder RB. .\ model for mul- andiia VA: .\PT.\, 1991.
tis\-stem e\-aluation, interpretation, and treat- 25. Fisher .\, Bund>- .\C. Scnsor\- integration the-
ment of indiMduals with neurologic dysfimc- or\. In: Forssberg H, Hirschfcld H, eds.
tion. Ph\-s Ther 1989;69:538-547. Movement disorders in children. Basel: Kar-
13. Gordon J. Assumptions underl\ing ph\'sical ger. 1992:16-20.
therapy intervention: theoretical and histori- 26. Shumway-Cook -\, Horak FB. Balance reha-
cal f>erspectives. In: Carr J, Shepherd R, Gor- bilitation in the neurological p>aticnt. Seattle:
don I, et al., eds. Movement sdence: foun- N"ER.\, 1992.

dations for ph\-sical therapy rehabilitation. 27. Shumway-Cook .\. Retraining balance and
RDck\ille, MD: .^spcn S«tems, 1987. mobilit\': the inteeration of research into cHn-
116 Section 1 THEORETICAL FRAMEWORK
ical practice. Presentation at the APTA An- tion: theon,', assessment, treatment. Laurel,
nual Meeting, Cincinnati; 1993. MD: Ramsco Publishing Co., 1982.
28. WooUacott M, Shumway-Cook A. Clinical 37. Carr Shepherd R. Motor relearning pro-
J,
and research methodolog\' for the study of gramme for stroke. Rockville, MD: Aspen
posture and balance. In: Sudarsky L, ed. Gait Systems, 1985.
disorders of aging: mechanisms, falls and 38. Fugl-Myer AR, Jaasko L, Lcymanl,etal. The
therapy. In press. post-stroke hemiplegic patient: a method for
29. Gentile, A. The nature of skill acquisition: evaluation of physical performance. Scand J
therapeutic implications for children with Rehabil Med 1975;7:13-31.

movement disorders. In: Forssberg H, 39. Brunnstrom S. Motor testing procedures in
Hirschfeld H, eds. Movement disorders in hemiplegia: based on sequential recover\'
children. Basel: Karger, 1992:31-40. stages. Phys Ther 1966;46:357-375.
30. Katz S, Downs TD, Cash HR, Grotz RC. 40. Hoehn MM, Yahr MD. Parkinsonism: onset,
Progress in development of the index of progression and mortalirv'. Neurology' 1967;
ADL. Gerontologist 1970:20-30. 17:433^50.
31. Keidi RA, Granger CV, Hamilton BB, Sher- 41. Schwab RS. Progression and prognosis in
The fianctional independence mea-
\vin FS. Parkinson's disease. J Nerv Ment Dis 1960;
new tool for rehabilitation. In: Eisent-
sure: a 130:556-572.
berg MG, Grzesiak RC, eds. Advances in 42. Gentile A. Skill acquisition: action move-
clinical rehabilitation, vol 1. New York: ment, and neuromotor processes. In: Carr J,
Springer Verlag, 1987:6-18. Shepherd R, Gordon J, et al., eds. Movement

32. Mahoney RI, Barthel DW. Functional eval- science: foundations for physical therapy in
uation: the Barthel Index. Maryland State rehabilitation. Rockville, MD: Aspen Sys-
Medical Journal 1965;14:61-65. tems, 1987.
33. Lawton MP. The fianctional assessment of el- 43. Welford AT. Motor skills and aging. In: Mor-
derly people. J Am GeriatrSoc 1971;19:465- timer J, Pirozzolo FJ, Maletta G, eds. The
481. aging motor system. New York: Praeger,
34. Tinetti ME. Performance oriented assess- 1982:152-187.
ment of mobility problems in elderly patients. 44. Foerster, O. The motor cortex in man in the
J Am Geriatr Soc 1986;34:119-126. light of Hughlings Jackson's doctrines. In:
35. Fregly AR, Graybeil A. An ataxia test battery Payton OD, Hirt S, Newman R, eds. Scien-
not requiring rails. Aerospace Medicine tific bases for neurophysiologic approaches to
1968:277-282. therapeutic exercise. Philadelphia: FA Davis,
36. Erhardt RP. Developmental hand dysfianc- 1977:13-18.
Section II
POSTURE/BALANCE
Chapter 6
Control of Posture and

Balance
[ Introduaion Senses Contributing to Posture Control

Defining the Task of Postural Control Visual Inputs
Detlning Systems for Postural Control Somatosensor\ Inputs
Postural Requirennents Var\ with Functional Vestibular Inputs
Task During Quiet Stance
Sensorv' Strategies
StarKe Postural Control Senscy
Strategies During Perturbed
Motor Mechanisms for Postural Control Stance
Motor Control of Quiet Stance Adapting Senses for Postural Control
Alignment Adaptation to Rotational Support
Muscle Tone Surface Perturbations
Postural Tone Adapting Senses When Learning a
Motor Strategies During Perturbed New Task
Stance Sensorimotor Adaptation
Ankle Strategy- Anticipatory Postural Control
Hip Strateg\ Seated Postural Control
Stepping Strategy Summary
Adapting Motor Mechanisms
Sensory Mechanisms Related to Posture
Introduction nition of (xjsture and balance, nor agreement

on the neuiral mechanisms underlying the
Picture yourself getting out of the car at control of these functions.
our flight. You
the airport, in a hurry to catch y Cher the last several decades, research
pick up your suitcase and run towards the ter- into pHDSture and balance control and their dis-
minal building. On the way, you misjudge the orders has shifted and broadened. The very
height of the curb, trip, but reco\ering, go definitions of fxjsture and balance have
into the terminal and check your bags. You changed, as has our understanding of the un-
get on the mo\ing walkways that take you to derlying neural mechanisms. In rehabilitation
your gate, mo\"ing quickly to avoid running science, there are at least t\\o difrerent con-
into other jjeople. Finally, you board the ceptual theories to describe the neural control
plane and sink gratefully into your scat. of posture and balance: the reflex/hierarchical
The many tasks in\ oh ed in getting from theory and the systems theory 1-3 ). 1
y our car to your scat on the plane place hea\y A reflex/hierarchical theory suggests
demands on the systems that control posture that posture and balance result from hierar-
and balance. In examining some of these tasks chically organized reflex responses triggered
you can see that posture and balance invohe by independent sensory systems. According
not just the ability to recover from instability to this theory , during development there is a
but also the ability to anticipate and move in progressive shift from the dominance of prim-
ways that will help you a\oid instability. to higher levels of postural
itive spinal reflexes
While few clinicians would argue the reactions, mature cortical responses
until
imponance of posture and balance to inde- dominate. This theory of balance control will
pendence in acti\ities such as sitting, stand- t>e presented in more detail in the next chap-
ing, and walking, there is no universal defi- ter.
119
120 Section II POSTURE/BALANCE
This chapter discusses normal posture orientation of the body. In the process of es-
and balance control from a systems perspec- tablishing a vertical orientation, we use mul-
tive. Sitting and standing postural control are tiple sensor\' references, including graxit)' (the
described as well. Posture related to mobUit}', vestibular system), the support surface (so-
however, is covered in the next section of the matosensor\- system), and the relationship of
book. our body to objects in our environment (vi-
As noted in Chapter 1, the systems ap- sual system).

proach suggests that action emerges from an Postural stability is defined as the abil-
interaction of the indi\'idual with the task and ity' to maintain the position of the body, and
the environment (Fig. 6.1). The systems ap- specifically, the center of body mass (COM),
proach implies that the abiht)' to control our within specific boundaries of space, referred
body's position in space emerges from a com- to as stabilit^' limits. Stability limits are
plex interaction of musculoskeletal and neural boundaries of an area of space in which the
systems, collectively referred to as the postural body can maintain its position without chang-
control system. ing the base of support. Stabiht)' limits are not
fixed boundaries, but change according to the
Defining the Task of Postural task, the indixidual's biomechanics, and vari-
Control ous aspects of the environment. The term sta-

bility is used in this text interchangeably with
To understand postural behavior in the

balance or equilibrium. Stability' involves es-
individual, we must understand the task of tablishing an equilibrium between destabiliz-
postural control, and examine the effect of the ing and stabilizing forces (5).
environment on the task of posture. Stability' and orientation represent two

The task of postural control in\ol\es distinct goals of the postural control system
controlling the body's position in space for (6, 7). Some tasks place importance on main-
the dual piuposes of stability- and orientation. taining an appropriate orientation at the ex-
Postural orientation is defined as the ability pense of stabilit)'. The successful blocking of
a goal in soccer, or catching a flyball in base-
to maintain an appropriate relationship be-
tween the body segments, and between the ball, requires that the player always remain
body and the environment for a task (4). For oriented with respect to the ball, sometimes
most flinctional tasks, we maintain a vertical failing to the ground in an eftbrt to block a
goal, or to catch a ball. Thus, while postural
control is a requirement that most tasks have
in common, stabiliU' and orientation de-
mands change with each task (8).
Defining Systems for Postural

Control
Postural control for stabilit)' and orien-

tation requires (a) the integration of sensor\'
information to assess the position and motion
of the body in space, and (b) the abilit\' to
generate forces for controlling body position.
Thus, postural control requires a complex in-
teraction of musculoskeletal and neural sys-
Figure 6.1 . Postural actions emerge from an interaction
of the individual, the task with its inherent postural de-
tems, as shown in Figure 6.2.
mands, and the environmental constraints on postural Musculoskeletal components include
actions. such things as joint range of motion, spinal
Chapter Six Control of Posture and B.\lance 121
Thus, in a systems approach, postural

control results from a complex interaction
among many bodily systems that work coop-
eratively to control the body's position in
space. specific organization of postural
The
systems determined both by the functional
is
task and the environment in which it is being

performed.
Postural Requirements Vary with

Functional Task
The to control our body's posi-

abilit}'
tion in space fundamental to everything we

is
do! All tasks ha\e postural requirements. That

is, ever}' task has an orientation component
Figure 6.2. Conceptual model representing systems
contributing to postural control.
and component. However, the sta-
a stability
bilityand orientation requirements will vary
with the task and the environment.
flexibility, muscle properties, and biomechan- The task of sitting in a chair and reading
ical relationships among linked body seg- has an orientation requirement of keeping the
ments. head and gaze stable and fixed on the reading
Neural components essential to postural material. The arms and hands maintain an ap-
control encompass (n) motor processes, in- propriate task-specific orientation that allows
cluding neuromuscular response synergies; the book to be held in the appropriate posi-
(b) sensor)' processes, including the visual, tion in relationship to the head and eyes. The
vestibular, and somatosenson,' systems; (c) stability' requirements of this task are lenient.
sensory strategies that organi/.e these multiple Since the contact of the body with the chair
inputs; (li) internal representations important back and seat provides a fairly large base of
for the mapping of sensation to action; and support, the primar\' requirement is control-
(f) higher level processes essential for adaptive ling the unsupported mass of the head with
and anticipatory aspects of postural control. respect to the mass of the trunk.
In this book we refer to higher level In contrast, the task of standing and
neural processes as cognitive influences on reading a book has roughly the same orien-
postural control. It is very important to un- tation requirements with respect to the head,
derstand, however, that the term cognitive as eyes, arms, and book, but die stability re-
it isused here does not mean conscious con- quirements are considerably more stringent.
trol. Higher le\el cognitive aspects of postural This task requires that the center of mass be
control are the basis for adaptive and antici- kept within a much smaller base of support
patory aspects of postural control. Adaptive defined by the two feet.
postural control involves modif\'ing senson' Finally, a person standing on a moving

and motor systems in response to changing bus has to constantly regain stabilitv' which is
task and environmental demands. Anticipa- threatened bv the constant motion of the bus.
tory aspects of postural control prctune sen- The task of stability is more rigorous, reflect-
son,' and motor systems for postural demands ing the changing and unpredictable nature of
based on previous experience and learning. the task. In this case, the task demands vary
Other aspects of cognition that affect postural from moment to moment, requiring constant
control include such processes as attention, adaptation of the postural system.
motivation, and intent (6). Thus, you can see that while these tasks
demand postural control, the specific orien-

tation and stability requirements vat)' accord- ACTIVE LEARNING MODULE
ing to the task and the environment. Because
You can determine this for your-
of this, the sensory and motor strategies used
self. Try standing up with your feet
to accomphsh postural control must adapt to
shoulder distance apart. First notice:
varying task and environmental demands (6). are you standing perfectly still, or do you move
very slightly? In which direction do you feel your-
Stance Postural Control self swaying most? Try leaning forward and back-
wards as far as you can without taking a step. Does

How do the sensory and motor systems your body move the same way as when you only
work together to control stance? The task of lean forward or backward a little? What muscles
stance postural control has stringent stabilit)' do you feel working to keep you balanced when
demands, requiring that the center of body you sway a little? What muscles work when you
mass (COM) be kept within stabilit)' limits, sway further? What happens when you lean so far
forward that your center of mass moves outside the
defined principally by the length of the feet
base of support of your feet?
and the distance between them (5).
Stance postural control is usually asso-
ciated with the maintenance of a vertical ori-
entation, though this is not an invariant As you have already discoxered, no one
requirement of the task. That is, one could stands absolutelystill; instead, the body sways
maintain a standing position but be bent over, in small amounts, mostiy in the forward and
looking atsomething on the ground, or al- backward direction. This is why researchers
ternatively, stand with the head extended, have concentrated on understanding how
looking at a bird. In both instances, one can normal adults maintain stability in the sagittal
var)' the configuration of body parts to ac- plane.

complish these two standing tasks, but the Now we can explore the underlying
stabilit}' requirements do not vary. If the cen- control mechanisms in depth, beginning with
ter of body mass is not kept within the support the motor mechanisms underlying postural
base of the feet, a fall will occur, unless the motor mecha-
control. In our discussion of
base of support is changed by taking a step. nisms important to postural control, we first
Over the past decade, sensory and mo- consider the role of muscle tone and postural
tor strategies for controlling stance posture tone in controlling small oscillations of the
have been widely studied. What do we mean body during quiet stance. Then we re\'iew
by strategies for postural control? A strategy motor strategies and underlying muscle syn-
is a plan for action, an approach to organizing ergies that help us to recover stability when
individual elements within a system into a col- our balance is threatened.
lective structure. Postural motor strategies
refer to the organization of movements ap-
propriate for controlling the body's position
Motor Mechanisms for Postural
in space.Sensory strategies organize sensory Control

information from visual, somatosensory, and
vestibular systems for postural control. Fi- Postural control requires the genera-
nally, sensorimotor strategies reflect the and coordination of forces that
tion, scaling,
rules for coordinating sensor)' and motor as- produce movements effective in controlling
pects of postural control (6). the body's position in space. How does the
Research in stance postural control has nervous system organize the motor system to
focused primarily on examining strategies ensure postural control during quiet stance?
for controlling forward and backward sway. How does the organization change when sta-
Why? bility is threatened?

Chapter Six Control of Posture and Bai.\nc:h 123
iVlOTOR CONTROL OF QUIET background muscle tone, which exists nor-

ST.'VNCK mally in all muscles because of neural contri-
butions, and (f) postural tone, the activation
What arc the beha\i()ral characteristics of antigravity muscles during quiet stance.
of quiet stance, and \\ hat is it that allows us Let's look at these factors (9-12).
to remain upright during quiet stance or sit-

ting? Quiet stance is characterized by small Alignment
amounts of spontaneous postural swa)'. A
number of factors contribute to our stabilin,' How does alignment contribute to pos-
in this situation. First, bociy alignment can tural stabilinr In a perfectly aligned posture,
minimize the effect of gravitational forces, shown in Figure 6.3, the vertical line of grav-
which tend to pull us off center. Second, mus- it)' fills in the midline between (a) the mas-
cle tone keeps the body from collapsing in re- toid process; ( b) a point just in front of the
sponse to the pull of gravir,'. Three main fac- shoulder joints, (c) the hip joints (or just be-
tors contribute to our background muscle hind), (rf) a point just in front of the center
tone during quiet stance: (a) the intrinsic of the knee joints, and (f) a point just in front
stiffness of the muscles themselves, (b) the of the ankle joints (10). The ideal alignment
Abdominals(±)
Tensor fasciae
latae(-f)
- Tibialis
anterior(+)
j^^ -Soleus (+)
Figure 6.3. Diagrammatic and Bl the ideal alignment in stance, requiring minimal muscular effort
illuslration of (A
and C, the muscles that are tonically active during the control of quiet stance. (Adapted
to sustain the vertical position,
from Kendell FP, McCreary EK. Muscles: testing and function. 3rd ed. Baltimore: Williams & Wilkins, 1983:280.)
TECHNOLOGY BOX 1
ELECTROMYOGRAPHY is a technique used for measuring the activity of muscles through elec-
trodes placed on the surface of the skin, over the muscle to be recorded, or in the muscle itself.
The output from the electrode (the electromyogram or EMC) describes the output to the
signal
muscular system from the motor neuron pool. It provides the clinician with information about (a)
movement, (6) the timing and relative intensity
the identity of the muscles that are active during a
of muscle contraction, and (c) whether antagonistic or synergistic muscle activity is occurring.
Surface electrodes are most often used; however the ability of these electrodes to differentiate
between the activity of neighboring muscles is not very effective.
The amplitude of the EMC signal is often interpreted as a rough measure of tension gen-
erated in the muscle. However, caution must be used when interpreting EMC amplitude mea-
surements. There are many variables that can affect the amplitude of EMC signals, including how
rapidly the muscle is changing length, resistance associated with cutaneous tissue and subcuta-
neous fat, and location of the electrode. Thus, generally, it is not accurate to compare absolute
amplitudes of EMC activity of a muscle across subjects, or within the same subject across different
days. Researchers who utilize EMC amplitude data to compare temporal and spatial patterns of
muscle activity across subjects or within a subject on different days generally convert absolute
amplitude measures to relative measures. For example, one can determine the ratio between the
response amplitude and the amplitude of a maximum voluntary contraction of that muscle. Or,
the ratio between agonist and antagonist muscles at a joint can be determined. Likewise, the ratio
of synergistic muscles can be found. One can then examine how this ratio changes as a function
of changing task or environmental conditions (61 62). ,
TECHNOLOGY BOX 2
KINEMATIC ANALYSIS is the description of the characteristics of an object's movement, in-
cluding linear and angular displacements, velocities, and accelerations. Displacement data are
usually gathered from the measurement of the position of markers placed over anatomic land-
marks and reported relative to either an anatomic coordinate system, that is, relative joint angle,
or to an external spatial reference system.
There are various ways to measure the kinematics of body movement. Coniometers, or
electrical potentiometers,can be attached to measure a joint angle where a change in joint angle
produces a proportional change in voltage.
Accelerometers are usually force transducers that measure the reaction forces associated
with acceleration of a body segment. The mass of the body is accelerated against a force trans-
ducer, producing a signal voltage proportional to the acceleration. Finally, imaging measurement
techniques, including cinematography, video, or optoelectric systems, can be used to measure
body movement. Optoelectric systems require the subject to wear special infrared lights or re-
flectivemarkers on each anatomic landmark, which are recorded by one or more cameras. The
location of the light, or marker, is expressed in terms of x and y coordinates in a two-dimensional
system, or x, y and z coordinates in a three-dimensional system. Output from these systems is
expressed as changes in segment displacements, joint angles, velocities, or accelerations, and the
data can be used to create a reconstruction of the body's movement in space (61, 62).
Chapter Six Control of Posture and BaIj\nce 125
TECHNOLOGY BOX 3
KINETIC ANALYSIS refers to the analysis of the forces that cause movement, including both
internal and external forces. Internal forces come from muscle activity, ligaments, or from
friction in the muscles and joints; external forces come from the ground or external loads. Kinetic
analysis gives us insight into the forces contributing to movement. Force-measuring devices or
force transducers are used to measure force, with output signals that are proportional to the
applied force.
Force plates measure ground reaction forces, which are the forces under the area of the
foot, from which center of pressure data are calculated. The term center of gravity (CG) of
the body is not the same as the center of pressure (CP). The CG of the body is the net location of
the center of mass in the vertical direction. CP is the location of the vertical ground reaction force
on the forceplate and is equal and opposite to all the downward acting forces (61, 62).
in stance allows the body to be maintained in There are also neural contributions to
equilibrium with the least expenditure of in- muscle tone or stiffness, associated with the
ternal energ\'. activation of the stretch reflex, which resists
Before we continue reviewing the re- lengthening of the muscle. C'hanges in muscle
search concerning the control of posture and length are sensed by the muscle spindles. This
movement, be sure to review the information afferent information is sent to the motor neu-
contained in the adjacent boxes, which discuss rons,which alter their firing to achieve the
techniques for movement analysis at different needed force to change the muscle length to
levels of control, including electromyography, the desired value. In this way, the stretch re-
kinematics, and kinetics. flex loop acts continuously to keep the muscle
length at a set value. For a more detailed re-
Muscle Tone view of the role of the muscle spindle, review

Chapter 3.
What is muscle tone, and how does it The role of the stretch reflex as a con-
help us to keep our balance.' Muscle tone re- tributor to normal muscle tone is fairly clear.
fers to the force with which a muscle resists The role of stretch reflexes in controlling up-
being lengthened, that is, its stiffness (10). right stance posture, however, is not. Accord-
Muscle tone is often tested clinically by pas- ing to one theor\', stretch reflexes play a feed-
sively extending and flexing a relaxed patient's back role during the maintenance of posture.
limbs and feeling the resistance offered by Thus, this theor\' suggests that, as we sway
the Both non-neural and neural
muscles. back and forth while standing, the ankle mus-
mechanisms contribute to muscle tone or cles are stretched, activating the stretch reflex.
stiffness. in a reflex shortening of the mus-
This results
A certain le\el of muscle tone is present cle,and subsequent control of forward and
in a normal, conscious, and relaxed person. backward swa\'.
However, in the relaxed state no electrical ac- While some authors suggest that die
tivity is recorded in normal human skeletal stretch reflex is critical for maintaining pos-
muscle using EMGs. This has led researchers ture, others have questioned the role of the
to argue that non-neural contributions to stretch reflex in the control of quiet stance.
muscle tone are the result of small amounts Reports that the gain of the stretch reflex is
of free calcium in the muscle whichfiber, quite low during stance has led some re-
cause a low level of continuous recycling of searchers to question its relevance to control-
cross-bridges (13). ling sway (14).
126 Section II POSTURE/BAL.\NCE
Postural Tone major mechanism in supporting the body

against gravit)'. In particular, many clinicians
We have explained the mechanisms con- have suggested that postural tone in the trunk
tributing to the generation of tone in individ- segment is the key element for control of nor-
ual muscles, when a person is in a relaxed mal postural stabilit\- in the erect position (12,
state. This background level of acti\it\' 18, 19). How consistent is this assumption
changes in certain anti-gra\it\' postural mus- with EMG studies that have examined the
cles when we stand upright, thus counteract- muscles active in quiet stance?
ing the force of gra\it\-. This increased level Researchers ha\e found that many mus-
of activit\' in anti-gravit)' muscles is referred to cles in the body are tonically acti\'e during
as postural tone. WTiat are the factors that quiet stance (10). Some of these muscles are
contribute to postural tone.- shown in Figure 6.3C, and include {a) the
A number of factors influence postural soleus and gastrocnemius, since the line of
tone. E\idence from experiments showing gravity- falls slighth- in front of the knee and
that lesions of the dorsal (senson') roots of the ankle; (b) the tibialis anterior, when the body
spinal cord reduced postural tone, indicates sways in the backward direction; (
c) the glu-
that postural tone is influenced by inputs teus medius and tensor fasciae latae but not
coming from the somatosensor*' system
in the gluteus maximus; (d) the iliopsoas, which
(15). In addition, it has long been known that prevents hyperextension of the hips, but not
activation of cutaneous inputs on the soles of the hamstrings and quadriceps; and (e) the
the feet causes a placing reaction, which re- thoracic erector spinae in the trunk (along
sults in an automatic extension of the foot with intermittent activation of the abdomin-
toward the support surface, thus increasing als),because the line of gravin' falls in front
postural tone in extensor muscles. Somato- of the spinal column.
senson,' inputs from the neck acti\"ated by These studies suggest that muscles
changes in head orientation can also influence throughout the body, not just those limited
the distribution of postural tone in the trunk to the trunk, are tonically active to maintain
and limbs (15). These have been referred to the body in a narrowly confined vertical po-
as the tonic neck reflexes, and are discussed sition during quiet stance. Once the center of
further in the next chapter on postural devel- mass moves outside the narrow range defined
opment (9). by the ideal alijjnmeiit, more muscular eftbrt
Inputs from the \isual and vestibular is required to recover a stable position. In this
systems also influence postural tone. Vestib- situation, compensaton,' postural strategies
ular inputs, activated by a change in head ori- are used to return the center of gra\iD,' to a
entation, alter the distribution of postural stable position within the base of support.
tone in the neck and limbs, and have been
referred to as the vestibulocollic and vestibu- iMOTOR STRATEGIES DURING
lospinal reflexes ( 15-17). PERTURBED STANCE
Often, these reflex contributions to pos-
ture control are highly emphasized in the clin- Many research labs, including Lewis
ical literature. However, it is important to re- Nashner's lab from the United States and the
member that there are many influences on labs of Dichgans, Dietz, and Allum in Europe,
postural control in a normal, intact, function- have studied the organization of movement
ing individual (16). It is possible that in the strategies used to recover stability" in response
neurologically impaired indixidual, who has to brief displacements of the supporting sur-
lost var\ing amounts of nonreflex influences, face, using a variet\-of moving platforms such
reflex pathwa\-s take a more commanding role as the one shown (20-22). In
in Figure 6.4
in the control of posture. addition, characteristic patterns of muscle ac-
In the clinical literature, much emphasis tivit}', called muscle synergies, which are as-
is placed on the concept of postural tone as a sociated with postural movement strategies.
Chapter Six Control of Posiurh and Baianc k 127
dcrlie movement strategics for balance. Re-

sults from postural control research in ncu-
rologicaily intact young adults suggest the
ncr\'ous system combines independent,
though related muscles, into units called
muscle synergies. A synergy is defined as the
functional coupling of groups of muscles such
that they are constrained to act together as a
unit; this simplifies the control demands on
the CNS. It is important to keep in mind that
while muscle synergies are important, they

are only one of many motor mechanisms
that affect outputs for postural control
(23-25).
Wliat are some of the muscle synergies
underlying movement strategies critical for
stance postural contrcil.* How do scientists
know whether these neuromuscular responses
are due to neural programs (that is, synergies)
or are the result of independent stretch
if they
Figure 6.4. Moving platform posturography used to
study postural control. (Adapted from Woollacott MH, of the individual muscles at mechanically cou-
Shumway-Cook A, Nashner LM. Aging and posture con- pled joints?
trol: changes in sensory organization and muscular co-
ordination. Int I Aging Hum Dev I986;22:332.)
Ankle Stratee;\'
The ankle strategy' and its related muscle

svnerg\' were among the first patterns for con-
iiave been described (23-25). These move-
trolling upright sway to be identified. The an-
ment patterns are referred to as the ankle, hip,
and suspensor\'/or stepping strategies, and
kle strategy,' restores the COM
to a position of
stabilit}- through body movement centered
are illustrated in Figure 6.5.
primarily about the ankle joints. Figure 6.6A
These postural movement strategies are
shows the typical synergistic muscle activity
used inboth a feedback and feedforward (an-
and body movements associated with correc-
ticipator}')manner to maintain equilibrium in
tions for loss of balance in the forward direc-
a number of circumstances. Here are some ex-
tion. In this case, motion of the platform in
amples of such situations:
the backward direction causes the subject to
sway forward. Muscle acti\ ity begins at about
1. In response to external disturbances to 90 to 100 msec after perturbation onset in the
equilibrium, such as when the support
gastrocnemius, followed by activation of the
surface moves;
hamstrings 20 to 30 msec later, and finally b\'
2. To prevent a disturbance to the system,
the activation of the paraspinal muscles (23).
for example, prior to a voluntar}- move- Actixation of the gastrocnemius pro-
ment that is potentially destabilizing;
duces a plantarflexion torque that slows, then
3. During gait and in response to unex- reverses, the body's forward motion. Activa-
pected disruptions to the gait cycle; and tion of the hamstrings and paraspinal muscles
4. During \olitional center of mass move- maintains the hip and knees in an extended
ments in stance. position. Without the synergistic activation of
the hamstrings and paraspinal muscles, the in-
Nashner and his colleagues (23-25) direct effect of the gastrocnemius ankle
have explored the muscle patterns that un- torque on proximal body segments would re-
Figure 6.5. Three postural movement strategies used by normal adults for controlling upright sway. (From Shumway-
Cook A, Horak F. Vestibular rehabilitation: an exercise approach to managing symptoms of vestibular dysfunction.
Seminars in Hearing 1989,10:199.)
suit in forward motion of die trunk mass rel- provides stretch to the gastrocnemius muscle
ative to the lower extremities. and dorsiflexion of the ankle, but these inputs
Figure 6.6B shows the synergistic mus- are not associated with movements at the me-
cle activit)' and body motions used when rees- chanically coupled knee and hip. The neuro-
tablishing response to backwards
stabilit}' in muscular response that occurs in response to
instabilit)'. Muscle activity' begins in the distal toes-up platform rotation includes activation
muscle, the anterior tibialis, followed by ac- of muscles knee, and hip joints,
at the ankle,
tivation of the quadriceps and abdominal despite the fact that motion has occurred only
muscles. at the ankle joint. Evidence from these exper-
How do scientists know that the ankle, iments supports the hypothesis of a neurally
knee, and hip muscles are part of a neuro- programmed muscle synerg\' (20, 23, 24), in-
muscular synerg}', instead of being activated cluding knee and hip muscles on the same side
in response to stretch of each individual joint.' of the bodv as the stretched ankle muscle.
Some of the first experiments in postural con- Since these responses are destabilizing,
trol (23, 24) provide some evidence for syn- in order to regain balance, muscles on the op-
ergistic organization of muscles. posite side of the body are activated. These
In these early experiments the platform responses have been hypothesized to be acti-
was rotated in a toes-iip or toes-down direction. vated in response to visual and vestibular in-
In a toes-up rotation, the platform motion puts (21) and are sometimes referred to as M3
Chapter Six Control oh Postuiu-; and Baianc:k 129
ANKLE STRATEGY ment strateg}' (25). This strategy controls

motion of the COM
by producing large and
rapid motion at the hip joints with antiphase
rotations of the ankles ( refer back to Fig. 6.5).
Figure 6.7 A shows the topical synergis-
tic muscle activit}' associated with a hip strat-
eg^•. Motion of the platform in the backward
direction again causes the subject to sway for-
ward. As shown in Figure 6.7/1, the muscles
that typically respond to fonvard sway when
a subject is standing on a narrow beam are
different fi-om the muscles that become acti-
\ated in response to forward sway while
standing on a flat surface. Muscle activity be-
gins at about 90 to 100 msec after perturba-

tion onset in the abdominal muscles, followed
by activation of the quadriceps (25). Figure
6.7 B shows the muscle pattern and body mo-
tions associated with the hip strategy' correct-
ing for backward swa\'.
HIP STRATEGY
Figure 6.5. Muscle synergy and body motions associ-

ated with ankle strategy lor controlling A, forward sway
and B, backward sway. (From Horak F, Nashner L. Cen-
tral programming of postural movements: adaptation to

altered support surface configurations. | Neurophysio!
1986;55:1372.)
responses, as opposed to an Ml response, that

is, a monosynaptic stretch reflex, and the
longer latency stretch responses, which ha\e
been called M2 responses (22).
The ankle movement strateg%' described
earlier appears to be used most commonly in
situations in which the perturbation to equi-

librium is small and the support surface is
firm. Use of the ankle strategy requires intact
range of motion and strength in the ankles.
What happens if the perturbation to balance
is large, or if we are in a situation where we
are unable to generate force using ankle joint

muscles?
Figure 6.7. Muscle synergy and body motions associ-
ated with the hip strategy for controlling A, forward sway
Hip Strategy'
and B, backward sway. (From Horak F, Nashner L. Cen-
tral programming of postural movements: adaptation to

Scientists have identified another strat- altered support surface configurations. J Neurophysio!
egy' for controlling body sway, the hip move- 1986;55:1372.)
Horak and Nashner suggest that the hip when standing on a narrow beam ( Fig. 6. SB)
strateg)' is used to restore equilibrium in re- (27).
sponse to larger, faster perturbations, or when This information is interesting, but is it
the support surface is compliant, or smaller true thatwe modify the amplitude of postural
than the feet, for example, when standing on responses only when they are inappropriate to
a beam (25). the task? In fact, no. Recent research has
shown that we are constantiy modulating the
Stepping Strategy' amplitudes of our postural responses, even
when they are appropriate. For example,
When a postural perturbation is strong Woollacott and colleagues examined the re-
enough to displace the COM outside the base sponses of adults to repeated translational
of support of the feet, a step or hop (the step- platform movements, and found that with re-
ping strateg)') is used to bring the support peated exposure to the movements, the sub-
base back into alignment under the COM (rejects swayed less and showed smaller ampli-
fer back to Fig 6.5) (6,26). tude postural responses (28). Thus, with
While the aforementioned strategies repeated exposure to a given postural task,
and their associated muscular synergies are subjects refine their response characteristics to
presented as discrete entities, researchers have optimize response efficiency.
shown that most neurologically intact individ- How
do we modify our postural strat-
uals use various mixtures of these strategies egies to accommodate multiple task goals.'
when controlling forward and backward swa\' For example, if we are tning to stand on a
in the standing position (25). moving bus while carrying a cup of coffee, do
we use a different strategy' from when we are
ADAPTING MOTOR STRATEGIES trying to read a book? To answer this ques-
tion, researchers asked adults to stand on a
Studies have shown that normal subjects moveable platform while either keeping their
can shift relatively quickly from one postural arms at a ftxed angle, as if they were reading
movement strategy- to another (25). For ex-
ample, when asked on the narrow
to stand
beam, most subjects from an ankle to
shifted
a hip strategy' within fi\'e to 15 trials, and
when returned to a normal support surface,
they shifted back to an ankle strategy, within
six trials. During the transition from one strat-
egy to the next, subjects used complex move-

ment strategies that were combinations of the
pure strategies.
Scientists theorize that the CNS ma\-
represent the different movement strategies
with respect to the boundaries in space in
which they can be safely used. That is, the
CNS appears to map the relationship between Ankle Strategy
body mo\ements in motor
space and the
Figure 6.8. Changes in boundaries lor motor strategies
strategies used to control those movements
used to control sway change as a function of the support
(27). These conceptual boundaries are shown surface. Mapping the relationship between body move-
in Figure 6.8. Boundaries ma\' be dvnamic, ments in space and the motor strategies used to control
shifting in response to the demands of the task those movements while standing A, on a firm flat surface
vs. B, crosswise on a narrow beam. (From Horak FB.
and environment. For example, boundaries
Effects of neurological disorders on postural movement
for using hip, ankle, and stepping strategies
strategies in the elderly. In: Vellas B, Toupet M, Ruben-
when standing on a firm, flat surface (refer to stein L, Albarede )L, Christen Y, eds. Falls, balance and
Fig. 6.8^1) may be different from those used gait disorders in the elderly. Paris: Elsevier, 1992:147.)
Chapter Six CONTROL OF PosTURE ANP B.\lanc:e 131
abook, or keeping their finger at a fixed point an essential part of postural control. We know
they were tning to keep a glass
in space, as if the CNS must activate synergistic muscles at
of water from spilling (29, 30). They found mechanically related joints to ensure that
that people continued to use the ankle forces generated at one joint for balance con-
strategy' during both but changed the
tasks, trol do not produce instabilit\' elsewhere in
coupling of the arm to the trunk in order the body. We belie\e the CNS internally rep-
to perform the additional upper extremit\' resents the bod\'s position in space with ref-
task. erence to behavioral strategies that are effec-
To learn more about postural sway tive in controlling that movement; however,
strategies in other directions, MacPherson it is not clear whether these behavioral strat-
performed experiments in which she per- egies are internally represented as muscle syn-
turbed cats in 16 different directions, around ergies, movement strategies, or force strate-
a 360° continuum (31 ). Despite the tact that gies.
cats were perturbed in 16 different directions,

they responded w ith force vectors in only two Sensor\' Mechanisms Related to
directions. In addition, while some muscles Posture
appeared to be fi.mctionally coupled into syn-
ergies, others appeared to be controlled in-
Effective postural control requires more
than the abilit)' to generate and apply forces
dependently and used to fine-tune the syner-
for controUing the body's position in space.
gies.
How does this work w ith cats relate to In order to know when and how to apply re-
human postural control experiments? Until storing forces, the CNS must have an accurate
recently, human postural research stressed the picture of where die body is in space, and
importance of a limited number of muscle whether it is stationary' or in motion. How
synergies that are the basis for postural con- does the CNS accomplish this?
trol. The work with cats suggests that some

muscles within the synerg\' may be tightly
SENSES CONTRIBUTING TO
coupled, but other muscle activity may POSTURE CONTROL
be highly modifiable. Thus, the CNS may
The CNS must organize information
combine muscles more ways than was
in
from sensor\' receptors throughout the body
originally thought. However, the way in
before it can determine the body's position in
which forces are applied may be \er\' space. Normally, peripheral inputs from vi-
limited. This would change the emphasis in
sual, somatosensory' (proprioceptive, cutane-
postural control fi"om a limited number ot
ous, and joint receptors), and vestibular sys-
muscle synergies, to a limited number of the body's
tems are available to detect
force strategies.
position and movement in space with respect
There is some support for this hypoth-
to gravit\' and the environment. Each sense
esis in humans from postural experiments ex-
provides the CNS with specific information
amining muscle responses used to control about position and motion of the body; thus,
sway in various directions in young adults
each sense provides a different /)•«>«£• of ref-
(32). The experimenters found stereot\'pical
erenee for postural control (33, 34).
muscle response synergies when sway w as for- What information does each of the
ward or backwards, but the responses were senses provide for postural control? Is one
much more variable in other directions.
sense more important than others? Does the
As perturbation direction was changed,
CNS use all three senses all the time? If not,
activation of the recorded muscles varied
how does the CNS decide w hich sense to use?
continuously as a fimction of perturbatitin
direction. \'isual Inputs
In summan,', we know that the ability to
generate and apply forces in a coordinated \'isual inputs report information regard-
wav to control the body's position in space is ing the position and motion of the head
with respect to surrounding objects. Visual are standing on a surface that is moving rela-
inputs provide a reference for verticality, since tive to you, for example, a boat, or on a sur-
many things that surround us, like windows face that is not horizontal, like a ramp, then
and doors, are aligned vertically. In addition, not appropriate to establish a vertical ori-
it is
the visual system reports motion of the head, entation with reference to the surface. In
since as your head moves forward, surround- these situations, inputs reporting the body's
ing objects move in the opposite direction. position with respect to the surface become
Visual inputs include both peripheral visual less helpfijl in establishing a vertical orienta-
information, as well as foveal information, tion.

though there is some evidence to suggest that
Vestibular Inputs
a peripheral (or a large visual field) stimulus is
more important for controlling posture (35). Information from the vestibular system
Visual inputs are an important source of is also a powerful source of orientation infor-
information for postural control, but are they mation. The vestibular system provides the
No, since most of us can
absolutely necessary.* CNS with information about the position and
keep our balance when we close our eyes, or movement of the head with respect to gravity
are in a dark room. In addition, visual inputs and inertial forces, providing a jjravito-iner-
are not always an accurate source of orienta- tial frame of reference for postural control.
tion information about self-motion. If you are The vestibular system has two types of
sitting in your car at a stop light and the car receptors that sense different aspects of head
next to you moves, what do you do? You position and motion. The semicircular canals
quickly put your foot on the brake. In this (SCC) sense angular acceleration of the head.
situation, visual inputs signal motion^ which The SCC are particularly sensitive to fast head
the brain initially interprets as self-motion; in movements such as those occurring during
other words, my car is The brain
rollinjj. gait or during imbalance, e.g., slips, trips, or
therefore sends out signals to the motor neu- stumbles (7).
rons of the leg and foot, so you step on the The otoliths signal linear position and
brake and stop the motion. acceleration. Since gravity is detected in rela-
Thus, visual information may be misin- tion to our Unear position or movement in
terpreted by the brain. The visual system has space, the otoliths are an important source of
difficulty distinguishing between object mo- information about head position with respect
tion, referred to as exocentric motion, and to gravit)'. The otoliths mostiy respond to
self-motion, referred to as egocentric motion. slow head movements, such as those that oc-
cur during postural sway. Thus, the vestibular
Somatosensory Inputs
system reports position and motion of the
The somatosensory system provides the head, and is important in distinguishing be-
CNS with position and motion information tween exocentric and egocentric motion (7).
about the body's position in space with ref- It is also interesting to note that vestib-
erence to supporting surfaces. In addition, so- ular signals alone cannot provide the CNS
matosensory' inputs throughout the body re- with of how the body is moving
a true picture
port information about the relationship of in space. For example, the CNS cannot distin-
body segments to one another. Somatosen- guish between a simple head nod (movement
sory receptors include joint and muscle pro- of the head relative to a stable trunk) and a
prioceptors, cutaneous, and pressure recep- forward bend (movement of the head in con-
tors. junction with a moving trunk) using vestib-
Under normal circumstances, when ular inputs alone (7).
standing on a firm, somatosensory
flat surface, How does the CNS organize this sen-
receptors provide information about the po- sor)' information for postural control.'' Pos-
sition and movement of your body with re- tural demands during quiet stance, often re-
spect to a horizontal surface. However, if you ferred to as static balance control, are different
Chapter Six Control of Postl-re and Balance 133
from those during perturbations to stance or adigm in which subjects stood in a room that
during locomotion, which require more dy- had a fixed floor, but with waUs and a ceiling
namic forms of control. Therefore, it is likely that could be moved fonvard or backward,
that information is organized diffcrcndy for creating the illusion of sway in the opposite
these tasks. direction ( 38 ). The moving room can be used
to create slow oscillations, simulating visual
SENSORY STR.\TEGIES DURING cues during quiet stance sway, or an abrupt
QUIET ST.\NCE fjerturbarion to the visual field, simulating an
unexpected loss of balance.
Somatosenson.' inputs from of all parts If ven. small continuous room oscilla-
the bodv contribute to balance control during tions are used, neurologically intact adults be-
quiet stance. Studies by the French scientist gin to sway with the room's oscillations, thus
Roll and his colleagues used mini\ibrators to showing that visual inputs have an imponant
excite eye, neck, and ankle muscles 36 and i ), influence on p>ostural control of adults during
explored the contributions of proprioceptive quiet stance i 38 ).
inputs from these muscles to posture control Other studies have given adults slow,
during quiet stance. They found that \ibra- continuous platform oscillations (simulating
tion to the eve muscles of a standing subject quiet stance) vs. fast, transient platform per-
with eves closed produced bod\' swa\', with turbations (creating loss of stabilitv). The re-
swav direction depending on the muscle \i- sults from these studies indicate that visual,
brated. Body sway also was produced b\- vi- vestibular, and somatosensorv' inputs all influ-
bration to the sternocleidomastoid muscles of ence balance control in normal adults during
the neck or the soleus muscles of the leg. slow oscillations similar to quiet stance. In
When these muscles were \ibrated simulta- contrast, somatosensorv- inputs apf>ear to
neously, the effects were additive, with no dominate postural control in resp>onse to tran-
clear domination of one proprioceptive influ- sient surface perturbations ( 42 ).
ence over another. This suggests that proprio- UTiat can we conclude from all of these
ception from all parts of the body pla\s an studies.- Thev suggest that all three senses
important role in the maintenance of quiet contribute to postural control during quiet
stance body posture. stance.
Early studies examining the effect of vi-
sion on quiet stance examined the amplitude SENSORY STR\TEGIES DURING
of swav with eyes open vs. eyes closed, and PERTURBED ST.\NCE
found that there was a significant increase in
swav normal subjects with eyes closed.
in How do visual, vestibular, and somato-
Thus, was concluded that vision actively
it sensorv inputs contribute to jxjstural control
contributes to balance control during quiet during recoverv- from a transient pemu-bation
stance. The ratio of bodv' sway during eyes to balance? Let's look at some of the research
open and closed conditions has been referred examining this question.
to as the Romberg quotient [37). Moving rooms, as we just described,
Do we use visual cues in a different man- have also been used to examine the contri-
ner depending on whether we are standing bution of visual inputs to recoverv- from tran-
quiedv' or responding to an unexpected threat sient perturbations. When abrupt room
to balancer The answer appears to be yes. Sev- movements made, young children (1-
are
eral researchers have studied sensirivirv' to vear-olds comp)ensate for this illusorv" loss of
)
continuous vs. transient visual motion cues in balance with motor responses designed to re-
people of different ages ( 38—H i. store the vertical p)osition. However, since
The first experiments of this tvpc were there is no actual body sway, only the illusion
performed by David Lee and his colleagues of sway, motor responses hav e a destabilizing
from Edinburgh, Scodand, using a novel par- effect, causing the infants to stagger or fall in
the direcrion of the room movement (38, 43). the gastrocnemius muscle, this response is de-
This indicates that vision may be a dominant stabilizing, pulling the body backward. Al-
input in compensating for transient pertur- lum, a researcher ft-om Switzerland, has
bations in infants iirst learning to stand. shown that the subsequent compensator)' re-
Interestingly, older children and adults sponse in the tibialis anterior muscle, used to
t\picall\- do not show^ large sway responses to restore balance, is activated by the visual and
these movements, indicating that in adults, vestibular systems when the eyes are open.
vision does not appear to play an important When the eyes are closed, it is primarily (80%)
role in compensating for transient perturba- activated bv the vestibular semicircular canals
tions. (21).
Muscle response latencies to visual cues These studies, examining postural con-
signaling s\\a\- are quite slow, on the order of trol in response to transient horizontal per-
200 msec, in contrast to the somatosensory turbations to stance, suggest that neurologi-
responses that are activated in response to cally intact adults tend to rely on
support surface translations (80 to 100 msec) somatosenson,' inputs, in contrast to young
(24, 44). Because somatosensorv' responses to children, who may rely more heavily on visual
support surface translations appear to be inputs.
much faster than those triggered by vision, re- Regardless of the task, no one sense by
searchers have suggested that the nervous sys- itself can provide the CNS with accurate in-
tem preferentially relies on somatosenson' information regarding the position and motion
puts for controlling body sway when of the body in space in all circumstances. The
imbalance is caused by rapid displacements of abilitv' of the nenous system to adapt its use
the supporting surface. of sensorv' information under changing task
What is the relative contribution of the and environmental conditions is discussed in
vestibular system to postural responses to sup- the next section.
port surface perturbations? Experiments by
ADAPTING SENSES FOR POSTURAL
Dietz and his colleagues indicate that the con-
tribution of the vestibular system is much
CONTROL
smaller than that of somatosenson,' inputs We live in a constandy changing envi-
(44). In these experiments, the onset latency ronment. Adapting how we use the senses for
and amplitude of muscle responses were com- postural control is a critical aspect of main-
pared for two different tvpes of perturbations taining stabiHtv' in a wide variety of environ-
of stance: (a) the support surface was moved ments, and has been studied by several re-
forward or backward, stimulating somatosen- searchers.

sory' inputs; and (b) a forward or backward One approach to investigating how the
displacement of a load 2 kg attached to the
( ) CNS adapts multiple senson' inputs for pos-
head was given, stimulating the vestibular sys- tural control was developed by Nashner and
tem the response was absent in patients with
( coworkers. This approach uses a moving plat-
vestibular deficits). For comparable accelera- form with a moving visual surround (20, 45).
tions, muscle responses to vestibular signals A simplified version of Nashner's protocol was
were about 10 times smaller than the soma- developed by Shumway-Cook and Horak
tosensorv' responses induced by the displace- (46) to examine the role of sensorv' interac-
ment of the feet. This suggests that vestibular tion in balance.
inputs play only a minor role in recoverv' of In Nashner's protocol, body sway is
postural control when the support surface is measured while the subject stands quiedy un-
displaced horizontally. der six different conditions that alter the avail-
However, under certain conditions, ves- abilitv' and accuracy of visual and somatosen-
tibular and visual inputs are important in con- sorv inputs for postural orientation. In
trolling responses to transient perturbations. conditions 1-3, the subject stands on a nor-
For example, when the support surface is ro- mal surface with eyes open ( 1 ), eyes closed (2
tated toes-upward, stretching and activating or the visual surround movine; with bodv swav
Chapter Six Control of PosruRh and B.\lance 135
(3). Conditions 4—6 are identical to 1-3 ex- things about how the CNS organizes and
cept that the suppon surface now rotates with adapts sensor}- information for postural con-
body swav as well. These conditions are trol. It suppons the concept of hierarchical
shown in Figure 6.9. Dift'erences in the weighting of sensor}- inputs for posture based
amount of body sway in the different condi- on their relative accuracy in reporting the
tions are used to determine a subject's ability- body's position and movements in space.
to adapt senson- information for postural con- In environments where a sense is not
trol. providing optimal or accurate information re-
Many studies have examined the perfor- garding the bod}'s position, the wcijfbt given
mance of normal subjects \\ hen sensor}' inputs to that sense as a source of orientation is re-
for postural control are varied (45, 47, 48). duced, while the weight of other more accu-
Generally, these studies ha\e shown that rate sen,ses is increased. Because of the redun-
adults and children over the age of 7 easily dancy of senses available for orientation and
maintain balance under all six conditions. the abilit}- of the CNS to modi!}- the relative
A\erage differences in body sway across importance of any one sense for postural con-
the six sensor}- conditions within a large group trol, individuals are able to maintain stabilitv'
of neurologically intact adults are shown in in a variet} of environments.

Figure 6.10. Adults s\\a\- the least in the con- In summar}-, postural control includes
ditions where support surface orientation in- organizing multiple sensor}- inputs into sen-
puts are accurately reporting the body's po- son- strategies for orientation. This process
sition in space relative to the surface regardless appears to involve the hierarchical ordering of
of the availabilitA' and accuracy of visual inputs sensor}- frames of reference, thereb}- ensuring
(Conditions 1, 2, and 3). WTien support sur- that themost appropriate sense is selected for
face information is no longer available as an the environment and the task. Sensor}- strat-
accurate source of orientation information, egies, that is, the relative weight given to a
adults begin to sway more. The greatest sense, van- as a fiinction of age, task, and en-
amount of swav is seen in conditions 5 and 6, vironment. It appears that under normal con-
in which onlv one accurate set of inputs, the ditions, the nervous system may wfijfbt the
vestibular inputs, is a\ailable to mediate pos- importance of somatosensor}- information for
tural control (48). postural control more hea\il\- than vision/
This research suggests a number of vestibular inputs.
Q ^ Q
C i i
Sensory condition
Sensory information
Vest Vest Vest Vest Vest Vest

Accurate Vision Somato Somato Somato
Somato
Inaccurate None None Vision Somato Somato Vision

Somato
Figure 6.9. The six sensory conditions used to experimentally test how people adapt the senses to changing sensory
conditions during the maintenance of stance posture. (Adapted from Horak F, Shumway-Cook A, Black FO. Are
vestibular deficits responsible for developmental disorders in children. Insights into Otolaryngology 1988;3:2.)
100 1 the response amplitude over a series of ap-

FALL Normals proximately 10 trials. It has thus been hy-
pothesized that when subjects receive inac-
curate sensory information from one sense (in
this case, ankle joint inputs), they are able to
compare that information to the other avail-
able sensory systems and then readjust the
weighting of their sensory inputs driving pos-
tural responses, to shift to the remaining, ac-
5-50H curate inputs.
5
Adapting Senses When Learning a New Task
Thus far, we have talked about re-

weighting sensory information in environ-
ments when it is not appropriate to use a par-
ticular sense for postural control. Similar
reweighting of the senses appears to occur
during the process of learning new motor
skills.Lee and Lishman (38) found an in-
creased weighting of visual inputs when adults
were just learning a task. As the task became
more automatic, there appeared to be a de-
crease in the relative importance of visual in-
12 3 4
Sensory conditions
5 6 puts for postural control and an increased
weighting given to somatosensory inputs.
It has been suggested that adults recov-
Figure 6.10. Body sway in the six sensory conditions
used to test sensory adaptation during stance postural
ering from a neurological lesion also rely pre-
control. (Adapted from Woollacott MH, Shumway-Cook dominandy on vision during the early part of
A, Nashner L. Aging and posture control: changes in sen- the recovery process. As motor skills, includ-
sory organization and muscular coordination. Int ) Aging ing postural control, are regained, patients
Hum Dev 1986;23:108.)
become less reliant on vision, and are more
able to use somatosensory inputs (51).
Adaptatdon to Rotational Support Surface

Perturbations Sensorimotor Adaptation
Researchers have performed other t)'pes Up to this point in the chapter, we have
of experiments to explore postural adaptation. presented sensory and motor aspects of pos-
Rotational platform movements have been tural control separately, but postural control
used to study the adaptation of postural re- is truly a sensorimotor task, requiring the co-
sponses to different conditions (20, 49, 50). ordination of sensory information with motor
For example, toes-down rotational platform aspects of postural control. How we move in-
movements cause stretch to the tibialis ante- fluences how we sense, and in turn, how we
rior muscles, activating the T-Q-A synergy, sense affects how we move.
but when the synergy is first activated in this Researchers have found an important
situation, it is inappropriate and serves to pull difference in how the senses are used depend-
the subject more forward in the direction of ing on the type of movement strategy being
the platform rotation. Studies indicate that used to restore stability. Effective use of the
subjects adapt the responses by attenuating ankle strategy appears to depend on intact
Chapter Six Control OF Posture and Baijvnc;e 137
sensation from somatosensory' inputs that re- the bodv. They found that the sequence of
port the body's position in space relative to postural muscles activated, and thus the man-
the surface ( 52 ). In contrast, vestibular inputs ner of preparing for the movement, was spe-
are critical for executing the hip strateg}' (7). cific to the task.
Thus, apparently, there is a relative change in After it was discovered that postural re-
weighting a particular sense, depending on sponses involved in feedback control of pos-
how we mo\e. These experiments emphasize ture were organized into distinct synergies
the importance of adaptation in the postural (23), an important question was raised: Are
system. To maintain orientation and stability' the synergies used in feedback postural con-
in a wide range of tasks and environments, we trol the same synergies that are used in antic-
are constantly called upon to modif\' how we ipator}' posture control.' To answer this ques-
sense and move. This capacity' to atiapt is a tion, C'ordo and Nashner (54) performed
critical aspect of normal postural control and experiments in which they asked standing
is heavily dependent upon experience and subjects to forcefully push or pull on a handle,
learning. in a reaction-time task. They found that the
same postural response synergies used in
Anticipatory Postviral Control stanciing balance control were activated in an
anticipator}' fashion before the arm move-
Did you ever pick up a box expecting it ments. For example, when a person is asked
to be heavy and find it to be light.' The fact to pull on a handle, first the gastrocnemius,
that you lifted the box higher than you ex- hamstrings, and trunk extensors are activated,
pected shows that your CNS preprogrammed and then the prime mover, the biceps of the
force based on anticipation of what the task arm
required. Based on previous experience with One feature of postural adjustments as-
lifi:ing other boxes of similar and different sociated with movement is their adaptabiUty
shapes and weights, the CNS forms a repre- to the conditions of the task. In the afore-
sentation of what sensor}' and motor actions mentioned experiment (54), when the sub-
are needed to accomplish this task. It pretunes jects leaned forward against a horizontal bar
these systems for the task. Our mistakes are at chest height, the leg postural adjustments
evidence that the CNS uses anticipator^' pro- were reduced or disappeared. Thus, there is
cesses in controlling action. an immediate preselection of the postural
In the 1960s, scientists in Russia first muscles as a fiinction of their abilit}' to con-
began to explore the way we use posture in tribute appropriate support.
an anticipator*' manner to steady the execu- Though we usually think of anticipator}'
tion of our skilled movements. In a paper adjustments in terms of activating postural
published in 1967 (53), Belen'kii, Gurfinkel, muscles in advance of a skilled movement, we
and Paltsev noted that when a standing adult also use anticipation in scaling the amplitude
is asked to raise the arm, both postural (leg of postural adjustments depending on the size
and trunk) and prime mover (arm) muscles or amplitude of the perturbation we expect.
were activated. They observed that the pos- Horak et al. (55) examined the influ-
tural muscle activation patterns could be di- ence of prior experience and central set on the
vided into two parts. The first part was a pre- characteristics of postural adjustments by giv-
parators' phase, in which postural muscles ing subjects platform perturbations under the
were activated more than 50 msec in advance following conditions: (a) serial vs. random
of the prime mover muscles, to compensate conditions, (b) expected vs. unexpected con-
in advance for the destabilizing effects of the ditions, and (c) practiced vs. unpracticed con-
movement. The second part was a compen- ditions. They found that expectation played a
sator}' phase, in which the postural muscles large factor in modulating the amplitude of
were again activated after the prime movers, postural responses. For example, subjects
in a feedback manner, to additionally stabilize overresponded when they expected a larger
perturbation than they received, and under- by Massion and his colleagues to look at this ques-
responded when they expected a smaller one. tion in more detail (57). They trained animals to
perform a leg-lifting task that required the animal
Practice also caused a reduction in pos-
to simultaneously activate postural muscles in the
tural response magnitude and in the ampli-
other three legs when they lifted the prime mover
tude of antagonist muscle responses. How-
leg. They found that they could also directly stim-
ever, central set did not aft'ect EMG onset
ulate the motor cortex or the red nucleus in the
latencies. The authors noted that when dif-
area of the forelimb flexors and produce the leg-
ferent perturbations were presented in ran- lifting movement. When they did this, the move-
dom order, all scaling disappeared. Evidendy, ment was always accompanied by a postural ad-
scaling of postural responses is based on our justment in the other limbs, initiated in a
anticipation of what is needed in a given sit- feedforward manner. They hypothesized that the
uation. postural adjustments are organized at the bulbo-
It is important to realize that anticipa- spinal level, and that the pyramidal tract activates
tory postural adjustments are not isolated to these pathways as it sends descending commands
to the prime mover. Massion suggests that, while
tasks we perform while standing.
the basic mechanisms for postural adjustments
could be organized at this level, they appear to be
modulated by several other parts of the nervous
ACTIVE LEARNING MODULE system, including the cerebellum.
For example, right now, you can

test this with a partner. First, take a
heavy book inone hand, and ask your Seated Postural Control
partner to lift it off that hand. What did the hand
The maintenance of postural control in
Was it steady? Or did it move
holding the book do?
the seated position has not been studied to
upward as the book was lifted off? Now, put the
book back on your hand, and lift it off with your the extent of stance postural control. How-
other hand. What happened now? Was it steady? ever, many scientists believe that concepts im-
What you may have noticed is that you are able portant for stance postural control will be
to use anticipatory postural adjustments when you shown to be equally \'alid for understanding
are lifting the own hand, so that
book out of your the control of seated posture.
your hand does not involuntarily move upward, A recent study was performed to com-
while you cannot use these adjustments when pare the posttiral responses elicited by plat-
someone else is lifting the same book from your form translations vs. rotations of subjects
hand.
seated with the legs extended forward (58).
Scientists from France and Switzerland, Hugon,
The authors noted that forward platform
Massion, and Wiesendanger (56), first made this
discovery which they measured

experiments in
movements, causing the body to sway back-
in
the EMGs of the biceps of both the left and right ward, elicited well-organized, consistent re-
arms during a modification of the task just men- sponses in the quadriceps, abdominal and
tioned. In this case, either the subject or the ex- neck flexor muscles at 63 ±12 msec, 74±21
perimenter lifted a 1 kg weight from the subject's msec, and 77± 10 msec, respectively. Similar
forearm (Fig. 6.1 1). They found that in the active responses were elicited by legs-up rotations.
unloading of the arm by the subject, there was pre- However, in response to backward platform
paratory biceps muscle inhibition to keep the arm perturbations, causing forward sway, smaller
from moving upward when it was unloaded. The and more variable responses were elicited in
anticipatory reduction in the biceps EMC of the
the trunk and neck extensor muscles. These
arm holding the load is time-locked with the onset
differences reflect the asymmetr\' of the sta-
of the activation of the biceps of the lifting arm.
This reduction was not observed bility' limits during sitting.
in the passive un-
loading condition. The authors suggest that the postural
How are these anticipatory postural adjust- control system sets a threshold for activation
ments associated with movements centrally orga- of postural responses according to an internal
nized? Animal experiments have been performed representation of the body, including the re-
Chapter Six Control of Posture and Baij\nc :e 1 39
"Active" unloading
"Passive" unloading
Biceps L.
Biceps P I
..yUW^Y^y-W
R Potentiometer R.
Potentiometer
Load R Load R.
200 g
Figure 6.11. Experiments examining anticipatory postural activity associated with lifting a weight from a subject's
arm. (Adapted from Hugon M, Massion Wiesendanger M. Anticipatory postural changes induced by active unloading
),
and comparison with passive unloading in man, Pflugers Arch 1982;393:292-296.)
lationship bet\veen the center of gravit\- and that increased reach distance and decreased
the support surface. Since the rotational and support were associated with earlier, larger
translational perturbations caused ver*' differ- postural adjustments. It has also been shown
ent head movements, but vePi' similar muscle that leg muscles are consistently active during
response patterns, the authors conclude that anticipaton' postural adjustments in advance
somatosensor}' inputs from the backward ro- of voluntarx' reaching while sitting (60).
tation of the pelvis trigger the postural re-

sponse svnergies in sitting.
Summary
E.xperiments have also been performed
to examine the characteristics of anticipator\- 1 . The task of postural control involves con-
postural adjustments used in reaching for an trolling the body's position in space for (a)
object while sitting (59). Researchers found stability, defined as controlling the center of
body mass within the base of support, and 10. The maintenance of postural control in the
(b) orientation, defined as theabilit>'to main- seated position has not been studied in
tain an appropriate relationship bet\\'een the depth. However, many scientists believe that
body segments, and bet\s'een the body and concepts important for stance postural con-
the environment for a task. trol will be shown to be equally valid for the
2. A number of factors contribute to postural control of seated posture.
control during quiet stance (so-called static
balance), including (a) body alignment,
REFERENCES
which minimizes the effect of gravitational
forces, ib) muscle tone, and (c) postural tone, 1. Shumway-Cook A. Equilibrium deficits in
which keeps the body from collapsing in re- children. In: Woollacott M, Shumway-Cook
sponse to the pull of gravity. A, eds. Development of posture and gait
3. When quiet stance is perturbed, the recovery across the life span. Columbia, SC: University
of stability requires movement strategies that of Soudi Carolina Press, 1989:229-252.
are effective in controlling the center of mass 2. Woollacott M, Shumway-Cook A. Changes
relative to the base of support. in posture control across the life span: a sys-
4. Movement patterns used to recover stance tems approach. Phys Ther 1990;70:799-
balance from sagittal plane instability are re- 807.
ferred to as ankle, hip, and suspensory/or 3. Horak F, Shumway-Cook A. Clinical impli-
stepping strategies. Normal subjects can shift cations of postural control research. In: Dun-
relatively quickly from one postural move- can P, ed. Balance: proceedings of the APTA
ment strategy to another. Forum. .Alexandria, VA: .\PTA, 1990:105-
5. The CNS activates synergistic muscles at me- 111.
chanically related joints, possibly to ensure 4. Shum«'ay-Cook A, Horak F. Balance reha-
that forces generated at one joint for balance bilitation in the neurologic patient: course
control do not produce instability elsewhere syUabus. Seattie: NERA, 1992.
in the body. 5. McCollum G, Lcen T. The form and explo-
6. Inputs from visual, somatosensory (proprio- ration of mechanical stabilin." limits in erect
ceptive, cutaneous, and joint receptors), and stance. Journal of Motor Beha\ior 1989;
vestibular systems are important sources of 21:225-238.
information about the body's position and 6. Shumway-Cook A, Horak F. \'esDbular re-
movement in space with respect to gravity habilitation: an exercise approach to manag-
and the environment. Each sense provides ing s\Tnptoms of \estibular dysfionction. Sem-
the CNS with a different kind of information inars in Hearing 1989;10:196.
about position and motion of the body; thus, 7. Horak F, Shupert C. The role of the vestib-
each sense provides a different frame of ref- ular system in postural control. In: Herdman
erence for postural control. S, ed. \'estibular rehabilitation. New York:
7. In adults, all three senses contribute to pos- FA Da\is, 1994:22-16.
tural control during quiet stance; in contrast, 8. Shumway-Cook A, McCollum G. Assess-
in response to transient perturbations, adults ment and treatment of balance disorders in
tend to rely on somatosensory inputs, while the neurologic patient. In: Montgomer\' T,
young children rely more heavily on visual Connolly B, eds. Motor control and physical
inputs. therapy: theoretical fi-ame\%ork and practical
8. Because of the redundancy of senses avail- applications. Chattanooga, TN: Chattanooga
able for orientation and the ability of the Corp. 1990:123-138.
CNS to modify the importance of any one 9. Roberts TDM. Neurophysiolog\' of postural
sense for postural control, individuals are mechanisms. London: Buttenvorths, 1979.
able to maintain stability in a variety of en- 10. Basmajian fV, DeLuca C. Muscles alive. 5th
vironments. ed. Baltimore: Williams 8c Wilkins, 1985.
9. Postural adjustments are also activated be- 11. Kendall FP, McCrear\- EK. Muscles: testing
fore voluntary movements minimize po-
to and function. 3rd ed. Baltimore: Williams &
tential disturbances to balance that the Wilkins, 1983.
movement may cause. This is called antici- 12. Schenkman M, Butier RB. "Automatic Pos-
patory postural control. tural Tone" in posture, movement, and fiinc-
Chapter Six Control of Posture and Balance 141
tion. Forum on physical therapy issues related of the APTA Forum. Alexandria, VA: APTA,
to cerebrovascular accident. Alexandria, VA: 1989:5-12.
APTA, 1992:16-21. 17 . Horak F, Shupert C^, Mirka A. Components
13. Hoyle G. Muscles and their neural control. of postural dyscontrol in the elderly: a review.
NY: John Wiley & Sons, 1983. Neurobiol Aging 1989;10:727-745.
14. Gurfinkel VS, Lipshits MI, Popov KE. Is the 28. Woollacott M, Roscblad B, Hofsten von C.
stretch reflex the main mechanism in the sys- Relation between muscle response onset and
tem of regulation of the vertical posture of body segmental movements during postural
man? Biophysics 1974;19:761-766. perturbations in humans. Exp Brain Res
15. Chez C. Posture. In: Kandcl ER, Schwartz 1988;72:593-604.
JH, Jessell TM, eds. Principles of neural sci- 29. Sveistrup H, Massion J, Moore S, Hu MH,
ence. 3rd ed. NY: Elsevier, 1991:596-607. Woollacott MH. Are there differences in pos-
1 6. Anderson ME, Binder MD. Spinal and supra- tural support strategies for simple balance
spinal control of movement and posture. In: tasks vs. tasks requiring precise hand stabili-
Patton HD, Fuchs AF, Hille B, Scher AM, zation? Ncuroscience Abstracts 1991;
Steiner R. Textbook of physiology, vol. 1: 17:1388.
Excitable cells and neurophysiology. Phila- 30. Moore S, Sveistrup H, Massion M, Hu M-H,
delphia: WB Saunders, 1989:563-581. Woollacott MH. Postural control strategies
17. Massion J, Woollacott M. Normal balance for simultaneous control tasks. In: Woolla-
and postural control. In: Bronstein AM, cott M, Horak F, eds. Posture and gait: con-
Brandt T, Woollacott M. Clinical aspects of trolmechanisms. Eugene, OR: Univ. of Or-
balance and gait disorders. London: Edward egon Press, 1992:218-221.
Arnold. In press. 31. MacPherson J. The neural organization of
18. Bobath B. Adult hemiplegia: evaluation and postural control —do muscle synergies exist?
treatment. London: Heinemann, 1978. In: Amblard B, Berthoz A, Clarac F, eds. Pos-
19. Davies PM. Steps to follow. New York: ture and gait: development, adaptation and
Springer- Verlag, 1985. modulation. Amsterdam: Elsevier, 1988:
20. Nashner L. Adapting reflexes controlling the 381-390.
human posture. Exp Brain Res 1976;26:59- 32. Moore SP, Rushmer DS, Windus SL, Nash-
72. ner LM. Human automatic postural re-
21. Allum JHJ, Pfaltz CR. Visual and vestibular sponses: responses to horizontal perturba-
contributions to pitch sway stabilization in tions of stance in multiple directions. Exp
the ankle muscles of normals and patients Brain Res 1988;73:648-658.
with bilateral peripheral vestibular deficits. 33. Hirschfeld H. On the integration of posture,
Exp Brain Res 1985;58:82-94. locomotion and voluntary movement in hu-
22. Diener HC, Dichgans J, Bruzek W, Selinka mans: normal and impaired development.
H. Stabilization of human posture during in- Dissertation. Karolinska Institute, Stock-
duced oscillations of the body. Exp Brain Res holm, Sweden, 1992.
1982;45:126-132. 34. Gurfinkel VS, Levick Yu S. Perceptual and au-
23. Nashner LM. Fixed patterns of rapid postural tomatic aspects of the postural body scheme.
responses among leg muscles during stance. In: Paillard J, ed. Brain and space. NY: Ox-
Exp Brain Res 1977;30:13-24. ford Science Publishers, 1991.
24. Nashner L, Woollacott M. The organization 35. Paillard J. Cognitive versus sensorimotor en-
of rapid postural adjustments of standing hu- coding of spatial information. In: Ellen P,
mans: an experimental-conceptual model. In: Thinus-Blanc C, eds. Cognitive processes and
Talbott RE, Humphrey DR, eds. Posture and spatial orientation in animal and man. Dor-
movement. NY: Raven Press, 1979:243-257. drecht: Martinus Nijhoff Publishers B\\
25. Horak F, Nashner L. Central programming 1987:43-77.
of postural movements: adaptation to altered 36. Roll, JP, Roll R. From eye to foot: a proprio-
support surface configurations. J Neurophy- ceptive chain involved in postural control. In:
siol 1986;55:1369-1381. Amblard B, Berthoz A, Clarac F, eds. Posture
26. Nashner LM. Sensory, neuromuscular, and and gait: development, adaptation and mod-
biomechanical contributions to human bal- ulation. Amsterdam: Elsevier, 1988:155-
ance. In: Duncan P, ed. Balance: Proceedings 164.
37. Romberg MH. Manual of nervous diseases of 50. Hansen PD, WooUacott MH, Debu B. Pos-
man. London: Sydenham Society', 1853:395- changing task conditions.
tural responses to
401. Exp Brain Res 1988;73:627-636.
38. Lee DN, Lishman K Visual proprioceptive 51. Mulder T, Berndt H, Pauwels J, Nienhuis B.
control of stance. Journal of Human Mo\'e- Sensorimotor adaptability and in the elderly
ment Studies 1975;1:87-95. G, Homberg V, eds.
disabled. In: Stelmach
39. Butterworth G, Hicks L. Visual propriocep- Sensori-motor impairment in the elderly.
tionand postural stability in infancy: a devel- Dordrecht: Kluwer, 1993.
opmental study. Perception 1977;6: 255- 52. Horak F, Diener H, Nashner L. Postural
262. strategies associated \\'ith somatosensor\' and
40. Butterworth G, Pope M. Origine et fonction vestibular loss. Exp Brain Res 1991;82:167-
de la proprioception \isuelle chez le enfant. 177.
In: de Schonen S, ed. Le developpementdans 53. Belen'kii VY, Gurfinkel VS, Paltsev YI. Ele-
la premiere annee. Paris: Presses Universi- ments of control of voluntar\' movements.
de France, 1983:107-128.
taires Biofizika 1967;12:135-141.
41. Brandt T, VVenzel D, Dichgans J. Die En- 54. Cordo P, Nashner L. Properties of postural
nvicklung der visuellen Stabilisation des adjustments associated with rapid arm move-
aufrechten standes bein kind: Ein refezeichen ments. J Neurophysiol 1982;47:287-302.
in der kinderneurologie (Visual stabilization 55. Horak F, Diener HC, Nashner LM. Influence
of free stance in infants: a sign of maturity'). ot central set on human postural responses. J
Arch Psychiat Ner\-enkr 1976;223:1-13. Neurophysiol 1989;62:841-853.
42. Diener HC, Dichgans J, Guschlbauer B, 56. Hugon M, Massion J, Wiesendanger M. An-
Bacher M. Role of visual and static vestibular ticipator)' postural changes induced by active
influences on dynamic posture control. Hu- unloading and comparison with passive un-
man Neurobiolog\' 1986;5:105-113. loading in man. Pflugers Arch 1982;
43. Lee DN, Aronson E. Visual proprioceptive 393:292-296.
control of standing in human infants. Percep- 57. Massion J. Role of motor cortex in postural
tual Psychophysiolog>' 1974;15:529-532. adjustments associated with movement. In:
44. Dietz M, Trippel M, Horstmann GA. Signif- Asanuma H, Wilson VJ, eds. Integration in
icance of proprioceptive and vestibulospinal the ner\'ous system. Tokyo-New York: Igaku-
'
reflexes in the control of stance and gait. In: Shoin, 1979:239-260.

Patia AE, ed. Adaptability of human gait. 58. Forssberg H, Hirschfeld H. Postural adjust-
Elsevier: Amsterdam, 1991:37-52. ments in sitting humans following external
45. Nashner LM. Adaptation of human move- perturbations: muscle activity and kinematics.
ment to altered environments. Trends in Exp Brain Res 1994;97:515-527.
Neuroscience 1982;358-61. 59. Moore S, Brunt D, Nesbitt ML, Juarez T.
46. Shumway-Cook A, Horak F. Assessing the Investigation of evidence for anticipatory
influence of sensory' interaction on balance. postural adjustments in seated subjects who
Phys Ther 1986;66:1548-1550. performed a reaching task. Phys Ther 1992;
47. WooUacott MH, Shumway-Cook A, Nashner 72:335-343.
L. Aging and posture control: changes in sen- 60. Shepherd RB, Crosbie J, Squires T. The con-
sor\' organization and muscular coordination. tribution of the ipsilateral leg to postural ad-
Int J Aging Hum Dev 1986;23:97-114. justments during fast voluntary reaching in
48. Peterka RJ, Black FO. Age related changes in sitting. Abstract of International Society for
human posture control: sensor\' organization Biomechanics, 14th Congress 1993, Paris.
tests. I Vest Res 1990;1:73-85. 61 Gronley JK, Pern' J. Gait analysis techniques:
49. Keshner E, Allum J. Plasticity' in pitch sway Rancho Los Arnigos Hospital gait laborator}'.
stabilization: Normal habituation and com- Phys Ther 1984;64:1831-1837.
pensation for peripheral vestibular deficits. In: 62 Winter DA. Biomechanics and motor control
files W, Brandt T, eds. Disorders of posture of human movement. New Y'ork: John Wiley
and gait. New York: Elsevier, 1986:289-314. & Sons, 1990.
Chapter 7
Development of Postural
Control
E Introduction Emergence of Independent Sitting
Postural Controland Development Motor Coordination
Motor Milestones and Emerging Postural Sensory Contributions
Control Relating Reflex to Systems Theory
Theories of Developing Postural Control Transition to Independent Stance
Reflex/Hierarchical Theory Motor Coordination
Postural Reflexes in Human Role of Strength
Development Development of Muscle Synergies
Attitudinal Reflexes Sensory Contributions
Righting Reactions Development of Adaptive Capability
Balance and Protective Reactions Relating Reflex to Systems Theory
Role of Reflexes in Development Refinement of Stance Control
New Models of Development Motor Coordination
Assessment Based on Newer Models Quiet Stance
Development of Postural Control: A Systems Compensatory Postural Control
Perspective Sensory Contributions
Emerging Head Control Development of Sensory Adaptation
Motor Coordination Development of Anticipatory Postural
Sensory Contributions Actions
Relating Reflex to Systems Theory Summary
Introduction the development of postural control and how

it contributes to the emergence of stability
During the early years of life, the child and mobilit)' skills. Later chapters consider
develops an incredible repertoire of skills, in- the implications of this research when assess-
cluding crawling, independent walking and ing postural control.

running, climbing, eye-hand coordination,
and the manipulation of objects in a variety of Postural Control and
wavs. The emergence of all of these skills re- Development
quires the development of postural activirv' to
support the priman,' mo\ement. Let's first look at some of the e\idence
To understand the emergence of mo- showing that postural control is a critical part
bility' and manipulatory- skills in children, ther- of motor development. Research on early de-
apists need to understand the postural sub- velopment has shown that the simultaneous
strate for these skills. Similarly, understanding development of the postural, locomotor, and
the best therapeutic approach for a child with manipulative systems is essential to the emer-
difficulties in walking or reaching skills re- gence and refinement of skills in all these ar-
quires the knowledge of any limitations in eas. In the neonate, when the chaotic move-
their postural abilities. Understanding the ba- ments of the head that regularly disturb the
sis for postural control, then, is the first step infant's seated balance are stabilized, move-
in determining the best therapeutic approach ments and behaviors normally seen in more
for improving related skills. mature infants emerge (1). For example, as
This chapter discusses the research on shown in Figure 7.1, the newborn may begin
143
Figure 7.1. Stabilizing the head in a neonate can produce dramatic changes in behavior. A, Uncontrolled movements
of the head produce a Moro response. B, External support to the child's head and trunk result in more mature behaviors
including attending to people and objects, and even reaching. (Adapted from Amiel-Tison C, Grenier A. Neurological
evaluation of the human infant. New York: Masson, 1980:81.)
12-13 months 14-18 months
Figure 7.2. Motor milestones that emerge with the development of postural control. (Adapted from Shumway-Cook
A, Woollacott M. Theoretical issues in assessing postural control. In: Wilhelm ed. Physical therapy assessment in
I,
early infancy. NY: Churchill Livingstone, 1993:163.)

Chapter Seven Development of Postural Control 145
to attend to the examiner, reach for objects, to a more complex alternating arm pattern as
anci maintain his arms at his sides, with the the skill of crawling is perfected. When the
fingers open, suggesting inhibition of the child first begins to creep, there is a return to
grasp and Moro reflexes. the svmmetrical arm pattern. Eventually, as

These results support the concept tiiat creeping becomes perfected, the emergence
an immature postural system is a hmiting fac- of an alternating arm pattern occurs.
tor or a constraint on the emergence of other Thus, as children progress to each new
behaviors such as coordinated arm and hand stage in the development of a skill, they may
movements, as well as the inhibition of re- appear to regress to an earlier form of the be-
flexes. It has also been suggested that delayed havior as new, more mature and adaptive, ver-
or abnormal development of the postural sys- sions of these skills emerge.
tem may also constrain a child's ability to de- Most of the traditional assessment scales
velop independence in mobilit\- and manipu- created to evaluate the emergence of motor
lator)' skills. behaviors use developmental norms estab-
lished by McGraw (3) and Gesell. Using these
Motor Milestones and Emerging scales, the therapist evaluates the performance
Postural Control of the infant or child on functional skills that
require postural control. These skills include
The development of postural control sitting, standing, walking unsupported,
has been traditionally associated with a pre- reaching forward, and moving from sitting to
dictable sequence of motor behaviors referred stanciing position. Evaluations follow normal
to as motor milestones. Some of the major mo- development and are used to identifv' children
tor milestones in development are shown in at risk for developmental problems.
Figure 7.2. They include crawling, sitting,

creeping, pull-to-stand, independent stance, Theories of Developing
and walking. The sequence and timing of the Postural Control
emergence of these motor milestones has
been well described by several developmental What is the basis for the development of
researchers. postural control underlying this predictable
In 1946, Arnold Gesell, a pediatrician, sequence of motor behaviors.' Several theories
described the emergence of general patterns of child development tn,' to relate neural
of behavior in the first few years of life. He structure and behavior in developing infants.
noted the general direction of behavioral de- Classic theories of child development place
velopment as moving from head to foot, and great importance on a reflex substrate for the
from proximal to distal within segments. emergence of mature human behavior pat-
Thus, he formulated the law of developmental terns. This means that in the normal child the
direction (2). emergence of posture and movement control

In addition, Gesell portrayed develop- is dependent on the appearance and subse-
ment as a spiralling hierarchy. He suggested quent integration of reflexes. Accorciing to
that the development of skilled behavior docs these theories, the appearance and ciisappear-
not follow a strict linear sequence, always ad- ance of these reflexes reflect the increasing
vancing, constantly improving with time and maturitv' of cortical structures that inhibit and
maturity'. Instead, Gesell believed that devel- integrate reflexes controlled at lower levels
opment is much more dynamic in nature and within the C'NS into more functional postural
seems to be characterized by alternating ad- and voluntarv' motor responses (refer to Fig.
vancement and regression in abilitv' to per- 1 .6 in Chapter 1 ). This classic theorv' has been
form skills. referred to as a reflex/hierarchy theon*' (4, 5).

Gesell gave the example of children .\lternatively, more recent theories of
learning to crawl and then creep. Initially, in motor control, such as the systems, ecologi-
learning to crawl, the child uses a primarily cal, and dynamic theories, have suggested that
symmetrical arm pattern, eventually s\\ itching posture control emerges from a complex in-
teraction of musculoskeletal and neural sys- Postural reflexes in animals \\ere classi-
tems collectively referred to as the postural fied by Magnus as local static reactions, seg-
control system. The organization of elements mental static reactions, general static reac-
within the postural control system is deter- tions, and righting reactions. Local static
mined both by the and the en\'ironment.
task reactions stiffen the animal's limb for support
Systems theor\' does not deny the existence of of body weight against gravit}'. Segmental
reflexes, but considers them as only one of static reactions invoke more than one body
many influences on the control of posture and segment, and include the flexor withdrawal
movement. reflex, and the crossed extensor reflex. Gen-
Let's briefly review the reflexes that have eral static reactions, called attitudinal re-
been associated with the emergence of pos- flexes, in\'olve changes in position of the whole
tural control. body in response to changes in head position.
Finally, Magnus described a series of five
Reflex-Hierarchical Theory of righting reactions, which allowed the animal

Postural Control to assume or resume a species specific orien-
tation of the body with respect to its environ-
Postural reflexes were studied in the ment.
early part of this century by investigators such
as Magnus (6), DeKleijn (7), Rademaker (8), POSTURAL REFLEXES IN HUMAN
and Schaltenbrand (9). In this early work, re- DEVELOPMENT
searchers selectively lesioned different parts of
the CNS and examined an animal's capacity' Examination of reflexes has become an
to orient. Magnus and associates took the an- essential partof the study of motor develop-
imal down to what they referred to as the zero ment. Many researchers have tried to docu-
condition^ a condition in which no postural ment accurately the time frame for appearance
reflex activity could be elicited. Subsequent and disappearance of these reflexes in normal
animals underwent selective lesions, leaving children, with wideh' \ar\dng results. There is
systematically greater and greater amounts of little agreement on the presence and time
the CNS intact. In this way, Magnus identi- course of these reflexes, or on the significance
fied individually and collectively all the re- of these reflexes to normal and abnormal de-
flexes that worked cooperatively to maintain velopment (10).
postural orientation in various t)'pes of ani- Figure 7.3 summarizes the results from
mals. a number of studies examining the presence
Age (Months)
Source
PRE- NEW 1 2 3 4 5 6 7 8 9 10
MAT B
Magnus and De Kleijn o o o o C' r;,
Schaltenbrand m^mm'^
Landau 1 1 1 K:
Gesell Ames ^^^^^^^^^^^HM •la _
Prechtl and Beintema

Milani-Comparetti
W'" B ^:
?
Bobath
lllingworth
Peiper and Isbert

^K
^^ ^
LI
|
i
r ^^ o
a o
Figure 7.3. A summary of various studies that examined the presence and time-course of the asymmetric tonic neck
reflex in normal development. O= reflex not present. (Adapted from Capute A), Accardo P|, Vining EPC, et al. Primitive
reflex profile. Baltimore; University Park Press, 1978:36.)
Chapter Seven Dfatlopment OF PosTUR-M. Control 147
STNR
Figure 7.4. The attiiudinal reflexes. A, The ATNR reflex produces extension in the face arm, and flexion in the skull
arm when the head is turned. B, The STNR reflex results in extension in the upper extremities and flexion in the lower
extremities when the head is extended. C, The tonic labyrinthine reflex produces an increase in extensor tone when
the body is supine, and flexion when prone. Also shown is the time-course for these reflexes. (Adapted from Barnes
MR, Crutchfield CA, Heriza CB. The neurophysiological basis of patient treatment. Morgantown, VA:Stokesville W
Publishing, 1978:222.)
and time course of the asymmetric tonic neclc and (f) the tonic labyrinthine reflex (TLR)
reflex in normal development. This chart (Fig. 7.4C){11).
shows obvious disagreement o\er whether the
reflex is present in infancy, and regarding the Righting Reactions
time course for its appearance and disappear-
According to a reflex-hierarchical
model, the interaction of five righting reac-
Attitudinal Reflexes tions produces orientation of the head in
space, and orientation of the body in relation-
According to the reflex dieor\' of pos- ship to the head and ground. Righting reac-
tural control, tonic attitudinal reflexes pro- tions are considered automatic reactions that
duce persisting changes in body posture, enable a person to assume the normal stand-
which result from a change in head position. ing position and maintain stability when
These reflexes are not obligator)' in normal changing positions (12).
children, but have been reported in children The three righting reactions that orient
with various types of neural pathology. These the head in space include (a) the optical
reflexes include {a) the asymmetric tonic righting reaction (Fig. 7. 5 A), which con-
neck reflex ( ATNR) ( Fig.7AA), (
b) the sym- tributes to the reflex orientation of the head
metric tonic neck reflex (STRR) (Fig. 7AB), using visual inputs; (b) the labyrinthine
Optical righting (ORR) Labyrinthine righting (LRR) Body-on-head righting (BOH)
AW B
1 2 3 4 5 6 7 8 9 10 11 12 months 12 3 4 5 years (persists)
ORR
LRR
BOH
Figure 7.5. The righting reactions that orient the head. A, The optical righting reaction orients the head to vision. B,
The labyrinthine righting reaction orients the head in response to vestibular signals. C, The body-on-head righting
reaction uses tactile and neck proprioceptive information to orient the head. Also shown is the time-course for these
reflexes. (Adapted from Barnes MR, Crutcht'ield CA, Heriza CB. The neurophysiological basis of patient treatment.
Morgantown, W VA;Stokesville Publishing, 1978:222.)
B 1 2 3 4 5 6 7 8 9 10 11 12 14 16 18 2) 21 22 23 31 months
Landau
Figure 7.6. Shown is the Landau reaction and its time-course during development, which combines the effects of all
three head-righting reactions. (Adapted from Barnes MR, Crutchfield CA, Heriza CB. The neurophysiological basis of
patient treatment. Morgantown, W VAiStokesville Publishing, 1978:222.)
righting reaction (Fig. 7. SB), which orients Balance and Protective Reactions
the head to an upright vertical position in re-
sponse to vestibular signals (9, 13, 14); and According to reflex-hierarchical theory,
(c) the body-on-head righting reaction balance control emerges in association with a
(Fig. 7.5C), which orients the head in re- sequentially organized series of equilibrium
sponse to proprioceptive and tactile signals reacticins. Balance reactions are often sepa-
from the body in contact with a supporting rated into three categories. The tilting reac-
surface. The Landau reaction, shown in Fig- tions, shown in Figure 7.8 A-C, arc used for
ure 7.6, combines the effects of all three head- controlling the center of gravit\' in response
righting rcacrions (9, 15). to a tilting surface. Postiu-al fixation reac-
Two reflexes interact to keep the body tions, shown in Figure 7.9^-C, are used to
oriented with respect to the head and the sur- recover from forces applied to the other parts
face. The neck-on-body righting reaction,

of the body (17). Parachute or protective
shown in Figure 7.7^1, orients the body in re- responses protect the body from injury dur-
sponse to cervical atferents, which report ing a tall and are shown in Figure 7 AQA-C
changes in the position of the head and neck. (12).
Two forms of this reflex have been reported: Table 7.1 summarizes the postural reflex
an immature form, resulting in log rolling, mechanism purported to underlie the emer-
which is present at birth, and a mature form gence of postural and balance control in chil-
producing segmental rotation of the body dren.
(16).The body-on-body righting reaction, Many investigators have suggested that
shown in Figure 7.7B, keeps the body ori- emerging balance reactions are necessary pre-
ented v\ith respect to the ground, regardless cursors to the acquisition of associated devel-
of the position of the head. opmental milestones; however, perfection of
Neck-on-body righting reaction (NOB) Body-on-body righting reaction (BOB)
B 1 2 3 4 5 6 7 8 9 10 11 12 months 1 2 3 4 5 years
NOB
BOB
Figure 7.7. The righting reactions of the body. Shown are the mature form of A, the neck-on-body (NOB) righting
reaction and B, the body-on-body (BOB), and their time-course for emergence. (Adapted from Barnes MR, Crutchfield
CA, Heriza CB. The neurophysiological basis of patient treatment. Morgantown, W VA: Stokesville Publishing,
1978:222.)
Figure 7.8. The tilting reactions. Tilting responses are purported to emerge first in A, prone, then supine (not shown),
then B, sitting, then emerge in all fours (not shown) and finally C, standing. Also shown is the time-course for these
reflexes. (Adapted from Barnes MR, Crutchfield CA, Heriza CB. The neurophysiological basis of patient treatment.
Morgantown, W VA:Stokesville Publishing, 1978:222.)
the tilting reaction does not occur until the dination since movement of the head (and
child has moved onto the next developmental eyes) brings thehand within view (21, 22).
milestone (18-20). However, another study showed no relation-
ship between reaching behavior and the pres-
ROLE OF REFLEXES IN ence or absence of this reflex in a 2- to 4-
DEVELOPMENT month-old group of infants (23). Various

researchers have intimated that theasymmet-
What is the role of reflexes inmotor de- ric tonic neck reflex contributes to move-
velopment.> Scientists do not know for sure; ments in adults since there is facihtation of
as a result, the role of reflexes in motor control extension in the extremities when the head is
is controversial. Many theorists believe that rotated (24-27).
reflexesform the substrate for normal motor The neck-on-body and body-on-body
control. For example, it has been suggested righting reactions are reported to be the basis
that the asymmetric tonic neck reflex is part of for rolling in infants. An immature form of
the developmental process of eye-hand coor- rolling at 4 months of age is purported to be
Chapter Seven Denxlopment of Postur.\l Control 151
8 10 12 21 months Persists
B 1 2 3 4 5 6 7 9 11
Prone
Supine
Sitting --
All fours
Stance —
Figure 7.9. The postural fixation reactions. Fixation reactions stabilize the body in response to destabilizing forces
applied to the body from anywhere but the supporting surface, and emerge in parallel to the tilting reactions. Shown
are reactions in A, prone, B, sitting, and C, stance. Also shown Is the time-course for these reflexes. (Adapted from
Barnes MR, Crutchfield CA, Herlza CB. The neurophysiological basis of patient treatment. Morgantown, W VA:
Stokesville Publishing, 1978:222.)
predictive of CNS pathoiog\', including ce- ated theories of motor development. These
rebral and developmental delay
palsv (28) newer theories are consistent in suggesting
(29). The role of these reflexes in more ma- that development involves much more than
ture rolling patterns has recently been ques- the maturation of reflexes within the CNS.
tioned (30). Development is a complex process, with new
Clearlv, there is considerable uncer- beha\iors and skills emerging from an inter-
tainty' about the contribution of reflex testing action of the child (and its maturing ner\'OUS
in clarifving the basis for normal and abnor- and musculoskeletal system) with the envi-
mal development in children. ronment.
With this framework, the emergence of
New Models of Development postural control is likewise ascribed to com-
plex interactions between neural and muscu-
Manv of the newer theories of motor loskeletal systems. These include (please refer
control presented in Chapter 1 have associ- back to Fig. 6.2);
Forward Sideways Backward
B 1 2 34567 89 10 11 12 13 14 1 5 months Persists
Sideways
Backward
Parachute
legs
Staggering
Figure 7.10. The protective reactions. These reactions protect the body from injury resulting from a fall, and develop
A, in the forward direction, then B, sideways, and C, backwards. Also shown is the time-course for these reflexes.
first
(Adapted from Barnes MR, Crutchfield CA, Heriza CB. The neurophysiological basis of patient treatment. Morgantown,
W VA:Stokesville Publishing, 1978:222.)
1. Changes in the musculoskeletal system, Apparently, an important part of inter-

including development of muscle preting senses and coordinating actions for
strength and changes in relative mass of postural control is the presence of an internal
the different body segments; representation or body schema providing a
2. Development or construction of the co- postural frame of reference. It has been hy-
ordinative structures or neuromuscular pothesized that this postural frame of refer-
response synergies used in maintaining ence is used as a comparison for incoming
balance; sensory inputs, as an essential part of inter-
3. Development of individual sensory sys- preting self-motion, and to calibrate motor
tems including somatosensory, visual, actions (32).
or vestibular systems; Development of sensory and motor as-
4. Development of sensory strategies for pects of postural control has been hypothe-
organizing these multiple inputs; sized to involve the capacit)' to build up ap-
5. Development of internal representa- propriate internal representations related to
tions important in the mapping of per- posture, which reflect the rules for organizing
ception to action; sensory inputs and coordinating them with
6. Development of adaptive and anticipa- motor actions. For example, as the child gains
tory mechanisms that allow children to experience moving in a gravity environment,
modify the way they sense and move for sensor^'-motor maps would develop. These
postural control (31). maps would relate actions to incoming sen-
Chapter Seven Dev'elopment of Postural Control 153
Table 7.1 Reflex Model of Postural Development

Birth 12 3 4 5 6 9 10 11 12 15 20 24 3 4 5 6 Persists
//
//
months years
Optical righting reaction

ORR
Labryrinthine righting reaction
j-,|. Body-on-head righting reaction
Immature mature form
NOB — ^ NOB Neck-on-body righting reaction
BOB — BOB Body-on-body righting reaction
Prone tilting and postural fixation reaction
Supine tilting and postural fixation reaction
Sitting tilting and postural fixation reaction
Stance tilting fixation
Protective reaction-upper extremities
Staggering
sor)' inputs from vision, somatosensory', and important to understand which components
vestibular systems. In this way, rules for mov- are rate-limiting at each developmental stage,
ing would develop and be reflected in altered or conversely, which ones push the system to
synaptic relationships. Thus, researchers ar- a new level of function when they have ma-
gue, the path from sensation to inotor actions tured. According to newer models of devel-
proceeds via an internal representational opment, finding the connection between crit-
structure or body schema (32, 33). ical postural components and development
ultimately guides the clinician in determining

ASSESSMENT BASED ON NEWER which systems should be assessed, and how
MODELS the contribution of these systems changes at
various developmental stages. It also allows
According to these newer theories, as-
the clinician to determine appropriate inter-
sessment of early motor development includes
ventions specific to the system that is dys-
the evaluation of both emerging behavioral
fimctional.
motor milestones and the supporting systems
for postural control. In addition, evaluation
Development of Postural
must occur within the context of different
tasks and environments. The child's capacity' Control: A Systems
to anticipate and adapt to a changing envi- PERSPECTrVTE
ronment, as evidenced by variabilit^' of per-
formance, is also included in an analysis of de- Since Gesell's original studies in 1946
velopment. The abilitv' to adapt how we sense describing the cephalo-caudal nature of de-
and how we move is a critical part of normal velopment, many researchers have found ex-
development. As a result, it is as crucial to as- ceptions to some of his general developmental
sess as the acquisition of stereotypical motor rules. For example, recent studies have found
milestones. that infants show control of the legs in kicking

Since different systems affecting pos- and supported walking behaviors well before
tural control develop at different rates, it is they can control their head and trunk in space
154 Section II POSTURE/B.\L.\NCE
(34, 35). However, of balance and

in the area down, noting any antigravitv' responses. New-
postural control, it does appear as if de\"el- borns and infants up to 8 to 10 weeks did not
opment follows a cephalo- caudal sequence. respond either to head downwards or up-
wards tilts. However, by 8 to 10 weeks, with
Emerging Head Control the onset of spontaneous head control, infants
MOTOR COORDINATION showed clear EMG patterns in response to the

tilting surface, and this response became con-
Heinz Prechd, a researcher and physi- sistent at about the third month of age.
cian from the Netherlands (36), used ultra- This research suggests that the emer-
sound techniques to study the spontaneous gence of coordinated postural responses in
postural beha\ior of infants during prenatal neck muscles, underlying both spontaneous
development. He observed spontaneous pos- head control and responses to perturbations,
tural changes and described several different occurs at about 2 months of age. However, it
motor patterns responsible for these changes. does not give us specific information about
Positional changes occurred as often as 20 the abilitv' of individual sensorv' systems to
times per hour in the first half of pregnancy, drive postural responses in the neck.
but decreased in later pregnane}-, perhaps due
to space restriction. SENSORY CONTRIBUTIONS
Prechd (36) also attempted to test re-
sponses to perturbations, and noted that he Babies as young as 60 hours old are able
was unable to activate vestibular reflexes in to orient themselves toward a source of visual
utero. He reported that the vestibulo-ocular stimulation, and can foUovv a moving object
reflexand the Moro response were absent pre- by correctiy orienting the head (37, 38).
natally but were present at birth, and sug- These orientation movements appear to be
gested that these reflexes \\ ere inhibited until part of a global form of postural control in-
the umbilical cord was broken, thus prevent- volving the head and entire body.
ing the fetus from moving even,' time the WTien do visuaUy controlled postural re-
mother turned. sponses become available to the infant? To ex-
Prechd also examined spontaneous head amine visual contributions to spontaneous
control in neonates and noted that infants had control of head movements, Jouen and col-
verv' poor postural or antigravitv" control at leagues 39 performed a study with preterm
( )
birth. He hv-pothesized that this could be due infants (32 to 34 weeks of gestation), exam-
either to lack of muscle strength ( a musculo- ining head alignment both with and without
skeletal constraint) or alternatively to lackof visual feedback (goggles were worn). They
maturitv' of the motor processes controUing kept the infant's head initially in a midline po-
posture of the head and neck at this age (mo- sition, then released it and measured the re-
tor coordination constraint). To test this, he sulting movements of the head. They found
examined spontaneous head movements us- that without vision, there was a significant
ing both electromyographic (EMG) record- tendencv- to turn the head to the right, but
ings and video recordings to determine if co- with vision, the neonate oriented to midline.
ordinated muscle activitv was present. He Thus, from at least 32 to 34 weeks of gesta-
found no organized patterns of muscle activ- tion, infants show a simple tvpe of head pos-
ity', which appeared to counteract the force of tural control that uses vision to keep the head
gravitv- on anv' consistent basis. This finding at midline.
suggests that the lack of head control in new- A second study examined the capabilitv'
boms is not solelv- the result of a lack of of neonates to make responses to visual stim-
strength, but also results from a lack of or- uli giving the illusion of a postural pciturba-
ganized muscle activitv'. tion (39, 40). Infants were placed in a room
To examine infants' responses to per- in w hich a pattern of stripes moved either for-
turbations of balance, he placed infants on a ward or backward. Postural responses were
rocking table that could be tipped up or measured with a pressure -sensitive pillow be-
Chapter Seven Dextlopment OF PoSTL'R.\L CONTROL 155
hind the infant's head. The neonates made sensor\-motor mapping is occurring in these
postural adjustments of the head in response sensor\- systems as well.
to the optical flow; for example, when the vi- According to a reflex model, the Landau
sual patterns moved backwards, the infants reflex, which requires the integration of all
appeared to percei\e forward swa>" of the three righting reactions, docs not emerge un-
head, because they moved the head back- til 4 to 6 months. This finding is consistent
wards, as if to compensate. with Jouen's findings, which suggest that

Research has also examined the early de- mapping between vision and vestibular sys-
velopment of sensorv contributions to anti- tems for postural action is present at 2 1/2 to
graxitv responses in infants. In these experi- 5 months of age. Thus, both theories are con-
ments, infants of 2.5 or 5 months were placed sistent in suggesting that mapping of individ-
in a chair that could be tilted to the right or ual senses to action may precede the mapping
left 25°. During some trials, a red wool ball of multiple senses to action. This tvpc of scn-
was placed in the visual field, to catch the in- sor\-to-sensory and sensorv'-to-motor map-
fant's attention (41, 42 I. The infants showed ping may represent the beginning of internal
an antigravitv- response (keeping the head neural representations necessan- for coordi-
from tailing to the side to w hich the bab\- w as nated postural abilities.
tilted which improved with developmental

I,
level, with the older infants dropping the head Emergence of Independent Sitting
less than the younger infants. Interestingly,
when the wool ball was placed in the \isual .\s infants begin to sit independently,
field, both age groups tilted the head less, and thus develop trunk control, they must
with the eflect being strongest in the younger learn to master the control of both sponta-
group. The authors conclude that these re- neous background sway of the head and crunk
sults show a significant effect of vision on the and to respond to perturbations of balance.
vestibular antigravitv' response in the infant This requires the coordination of sensorv^-mo-
and improvement in this response with
a clear tor information relating two body segments
age. However, in this paradigm it is difficult together in the control of posture. To accom-
to determine if the improvement is due to en- plish this, they need to extend the rules they
hanced neck muscle strength, somatosen- learned for sensorv -motor relationships for
sorv/motor processing in neck muscles, or head postural control to the new set of mus-
vestibular/motor processing. cles controlling the trunk. It is possible that
once these rules have been established for the
RELATING REFLEX TO SYSTEMS neck muscles, they could be readily extended
THEORY to the control of the trunk muscles.
How consistent are svstems and reflex MOTOR COORDINATION

theories in describing the development of
head control? Reflex-hierarchical theor\- sug- With the emergence of independent sit-
gests that visual-motor coordination appears ting, infants develop the abilit)- to control
at approximately 2 months of age and is the spoiitatieous sway sufficiently to remain up-
resultof mattiration of the optical righting re- right. This occurs at appro.ximately 6 to 7
action. Systems theorv' suggests that certain months of age (43).
basic visual-postural mapping is present at The abilitv- to respond to postural per-
birth and with experience in moving, the child turbations with organized postural adjust-
develops more refined rules for mapping vi- ments appears to develop simultaneously.
sual information to action. How do the muscles that coordinate sway re-
Reflex theon,- suggests that since body- sponses develop in the neck and trunk.- Both
righting reactions acting on the head and lab- and longitudinal studies have
cross-sectional
vrinthine-righting reactions also emerge be- been used to explore the development of
tween birth and 2 months, this tvpe of muscle coordination underlving neck and
156 Section II POSTURE/B.\L.\NCE
trunk control in infants 2 to 8 months of age variable postural muscle response ssnergies
(33, 44). EMGs were used to record muscles than those causing forward swav. This mav
in the neck and trunk in infants either seated be caused by the larger base of postural
in an infant seat or sitting independently on a support in the forward direction in seated
moveable platform, shown in Figure 7.11yl. infants.
Motion of the platform forward or backward
caused a disturbance of the infant's head and SENSORY CONTRIBUTIONS
trunk posture, requiring a subsequent com-
pensator)- adjustment to regain balance. Other research has examined the capa-
Two-month-olds did not show consis- bilit>' of infants sitting unsupported to make
tent, directionally appropriate, responses to responses to \isual stimuli, gi\ing the illusion
the platform perturbations. By 3 to 4 months, of a postural perturbation the mo\ing room
t
infants showed direcnonalh' specific responses paradigm) (43, 45, 46 Infants with vaning
1.
in the neck muscles 40 to 60% of the time. By amounts of sitting experience were studied,
5 months, as infants were beginning to sit in- including infants with to 3 months' expe-
dependently, coordinated postural acti\it\' in rience, 4 to 6 months' experience, and 7 to
the trunk muscles in response to platform mo- 12 months' experience. In the to 3 month
rion was occurring approximately 40% of the group, a complete loss of balance was often
time. By 8 months of age, muscles in the neck recorded in response to the \isual stimulation,
and trunk \sere coordinated into effective pat- even though the infant could maintain bal-
terns for controlling for\\ard and backward ance when sitting quietiy. .\fter the first 3
sway in the seated position. months of experience sitting, the response
A recent study using similar support sur- amplitude declined. This implies that newly
face perturbations to balance ( 33 has
) also in- sitting infants rely hea\ily on \isual inputs to
dicated that platform movements causing maintain d\namic posture, and decrease this
backward s^vay give much stronger and less dependence, rehing more on somatosensory
Figure 7.11. Moving platform posturography used to study p)OStural response patterns in infants in response to a
moving surface in A, sitting, and B, standing.
Chapter Se\en Development of Postural Control 1 57
inputs, with experience in independent sit- occurs at approximately 6 to 8 months of age,

ting. coincident to the emergence of independent
In addition, VVooliacott and co\vori«.ers sitting. These results are quite similar to find-
found that talcing away visual stimuli did not ings from studies using a reflex-hierarchical
cause a disruption in muscle activation pat- approach. In those studies, orientation of the
terns in response to a moving platform. They body reportedly emerges at about 6 months
concluded that somatosen.sor\' and vestibular of age with the emergence of the mature
systems are capable of eliciting postural ac- neck-on-body and body-on-body righting re-
tions during seated perturbations in isolation actions.

from vision in infants first learning to sit While the neck-on-body and bodyon-
(44). body righting reactions have traditionally
What is the primar\' sensor\' system con- been used to describe the emergence of roll-
trolling responses to postural perturbations in ing patterns, we have chosen to describe their
seated infants.' To address this question, ex- actions as Magnus did, as they affect body ori-
periments were performed in which head orientation to the head/neck (neck-on-body)
entation was systematically varied, in an eftbrt and supporting surface (body-on-body).
to change the relationshipbetween inputs re- Thus, there appears to be agreement between
lated to head motion (vestibular and visual), the two theories concerning the emergence of
and proprioceptixc inputs from the trunk trunk control, but a difference in the under-
(33). Coordinated muscle acuvit\' stabilizing lying explanation for these emerging behav-
the trunk did not change regardless of how iors.
the head was oriented. This suggests that in

the seated position, postural responses to per- Transition to Independent
turbations are largeh' controlled by somato- Stance
sensor\' inputs at the hip joints, not by vestib-
ular or visual stimulation. These results are During the process of learning to stand
similar to those found in adults for standing independently, infants must learn to (a) bal-
perturbations. ance within significantly reduced stabilitv' lim-
These studies suggest that coordinated its compared to those used during sitting, and
postural activit\- in the neck and trunk devel- (
b) control many additional degrees of fi-ee-
ops gradually at about the same time the in- dom, as they add the coordination of the leg
fant is developing independent head control and thigh segments to those of the trunk and
and the abilit\' to sit independendy. First, in- head.
fants appear to map relationships between
sensoPi' inputs and the neck muscles for pos- MOTOR COORDINATION
tural control; thisis later extended to include
the trunk musculature with the onset of in- The following sections examine the
dependent sitting. These studies do not tell emergence of this control during both quiet
us whether it is nervous system maturation or stance and in response to perturbations of bal-
experience that allows neck and trunk muscle ance.
responses to emerge, since maturation anci
the refinement of synergies through experi- Role of Strength
ence are both gradual, and they seem to occur
synchronously. Several researchers have suggested that
emer-
a primarv' rate-limiting factor for the
RELATING REFLEX TO SYSTEMS gence of independent v\alking is the devel-

THEORY opment of sufficient muscle strength to sup-
port the body during static balance and
The research we just reviewed suggests walking (47). Can leg muscle strength be
that the child's abilin,' to orient the trunk with tested in the infant to determine if this is the
respect to the head and the support surface ca.se?
Researchers have shown that by 6 shown in Figure 7.115, infants stood with
months of age infants are producing forces vaning degrees of support on the moving
well beyond their own body weight (48). platform while EMGs were used to record
These experiments suggest that the abilit\' to muscle activity' in the leg and trunk in re-
support weight against die force of gravitN' in sponse to loss of balance.
the standing position occurs well before the Figure 7.12 shows EMG responses from
emergence of independent stance, and so is one child during the emergence of coordi-
probably not the major constraint to emerg- nated muscle acti\ini' in the leg and trunk
ing stance postural control in infants. muscles in response to a fall in the backward
direction. Infants tested at 2 to 6 months of
age, before the onset of pull-to-stand behav-
Development of Muscle S\'nergies
ior, did not show coordinated muscle re-
sponse organization in response to threats to

How do postural response synergies balance (Fig.7.12yl). During early pull-to-
compensating for perturbations to balance stand behavior (7 to 9 months), the infants
begin to emerge in the newly standing infant? began to show directionally appropriate re-
Longitudinal studies have explored the emer- sponses in their ankle muscles (Fig. 7.125).
gence of postural response synergies in infants As pull-to-stand skills improved, muscles in
from ages 2 to 18 months, during the tran- the thigh segment were added and a consis-
sition to independent stance (49-52). i\s tent distal-to-proximal sequence began to
Pre-pull-to-stand Early pull-to-stand
Late pull-to-stand Independent stance
Gr 1
— J'^^*Â^^«v
Q D
Figure 7.1 2. EMG responses from one child during the emergence ol coordinated muscle activity in the leg and trunk
muscles in response to platform perturbations in A, pre-pull-to-stand, B, early pull-to-stand, C, late pull-to-stand, and
D, independent stance. Abbreviations: C, gastrocnemius; H. hamstrings; T, tibialis anterior; Q, quadriceps muscles.
(Adapted from Sveistrup H, Woollacott MH. Systems contributing to the emergence and maturation of stability in
postnatal development. In: Savelsbergh CJP, ed. The development of coordination in infancy. Amsterdam: Elsevier,
1993:331.)
Chapter Seven Dex'elopmbnî of Postural Control 159
emerge (Fig. 7.12C-D) late pull-to-stand and room were examined in infants and children
independent stance (9 to 11 months), trunk of varving ages and abilities and compared to
muscles were consistendy activated, resulting those of young adults 54 ( ). Figure 7.13 shows
in a complete synergv'. an example of an infant positioned in a mov-
To determine if experience is important ing room. The sway was recorded
child's
in the development of postural response char- through a one-way mirror with a video cam-
acteristics in infants learning to stand, postural era mounted outside the room, and muscle
responses were compared in two groups of in- responses were recorded from the legs and
fants in the pull-to-stand stage of balance de- hips. Infants who were unable to stand inde-
velopment (53). One group of intants was pendendy were supported by their parents
given extensive experience with platform per- about the hip.
turbations, receiving 300 perturbations over Children as young as 5 months of age
3 days. The control group of infants did not swayed in response to room mo\ements; sway
receive this training. amplitudes increased in the pull-to-stand
Infants ho had extensive experience on
\\ stage, peaking in the independent walkers,
the platform were more likely to activate pos- and dropped to low levels of sway in experi-
tural muscle responses, and these responses enced walkers (54). Sway responses were as-
were better organized. However, onset laten- sociated with clear patterns of muscle re-
cies of postural responses did not change. sponses that pulled the child in the direction
These results suggest that experience has the of the visual stimulus.
capabilin,- of influencing the strength of con- These experiments suggest that the \i-
nections between the sensor\' and motor sual system will ehcit organized postural re-
pathwavs controlling balance, thus increasing sponses in standing infants at an earlier time
the probability' of producing postural re- than the somatosensor\' system, and that the
sponses. However, the lack of a training effect somatosensorv' system develops postural syn-
on muscle response latency suggests that neu- ergies separately in association with somato-
ral maturation may be a rate-limiting factor in sensort' inputs signalling sway.
latency reduction with development. It is
probable that the myelination of ner\ous sys- DE\"ELOPMENT OF ADAPTIVT.

tem pathways responsible for reducing laten- CAPABILITY'
cies of postural responses during development
is not affected by training. To determine if higher level adaptive
processes are available to the infant during
SENSORY CONTRIBUTIONS pull-to-stand behavior, independent stance,
and early walking, the abilirv' of the infants to
Once an infant learns how to organize attenuate postural responses to the visual flow
synergistic muscles for controlling stance in created by the mo\ing room was monitored
association with one sense, will this automat- (54). None of the infants in any of these be-
ically transfer to other senses reporting sway.- havioral categories was able to adapt inappro-
This may not always be the case. It appears priate postural responses to low levels, over a
that \ision maps to muscles controlling stance period of five trials. The researchers con-
posture at 5 to 6 months, prior to somatosen- cluded that higher level adaptive processes re-
sor\- system mapping, and long before the in- lated to postural control ha\e not yet matured
fant has much experience in the standing po- by the emergence of independent walking.
sition (54). This suggests that the infant has
to rediscover the synergies when somatosen- RELATING REFLEX TO SYSTEMS
son,' inputs are mapped for stance postural THEORY
control.
EMG responses and sway patterns in re- Differences in focus between reflex-hi-
sponse to visual flow created by a mo\ing erarchical and systems models make it diflicult
Figure 7.13. Diagram showing the moving room paradigm used to examine the development of visual contributions
to postural control. (From Sveistrup H, Woollacott MH. Systems contributing to the emergence and maturation of
stability in postnatal development. In: Savelsbergh CJP, ed. The development of coordination in infancy. Amsterdam:
Elsevier, 1993:324.)
to relate findings examining the emergence of gest that, for the most part, experience within
independent stance. Reflex-hierarchical the- a specific posture is important for sensory in-
ory distinguishes the righting reactions un- formation signalling the body's position in
derlying orientation from the tilting and pos- space to be mapped to muscular actions,
tural fixation reactions essential to the which control the body's position in space.
emergence of balance, suggesting different
neural mechanisms are involved in these two Refinement of Stance Control
fiinctions. Studies of tilting and postural
fixation reactions have not examined the As children mature, postural adjust-
importance of individual sensory systems ments are refined. The emergence of adult
to these reactions, nor their capability for levels of control occurs at different times for
adaptation. different aspects of postural control. The fol-
Systems-based research suggests that lowing sections review the literature on the
the time-course for emerging stability behav- refinement of stance postural control.
iors is different in each of the sensory systems.
Visual inputs relating the body's position in
MOTOR COORDINATION
space map to muscular actions controlling the
body's position earlier than do inputs from Quiet Stance
the somatosensory system. It is not known yet
how early vestibular inputs map to stance pos- Ho\\' does the control of spontaneous
tural actions. sway during quiet stance change as children
Results from systems-based studies sug- develop.' Are children inherendy more stable
Chapter Seven DhVELOPMENT of POSTL'RAL CONTROL 161
than adults? Children are shorter and there- using a moveable platform to examine
fore closer to the ground. Does their height changes in postural control (58-61). Re-
make balancing an easier task? Anyone who search has shown that compcnsator\' postural
has watched a tearless young child ski down a responses of young children (15 months of
steep slope with relative ease, falling and age) are more variable and slower than those
bouncing back up might assume that their of adults (58). These slower muscle responses
task is easier. They don't have as far to fall! It and the more rapid rates of sway acceleration
turns out that while children are shorter than obser\ed in young children cause sway am-
adults, they are proportioned ditVerently. plitudes that are bigger and often more oscil-
Children are top-heavy. The relative size of the lat(5ry than those of older children and adults.
head, in comparison to lower extremities, Even children of 1 1/2 to three years of
places the center of mass at about TI2 in the age generally produce well-organized muscle
child, compared to L5-S1 in the adult. Be- responses to postural perturbations while
cause of their shorter height, and the differ- standing (59). However, the amplitudes of
ence in the location of their center of mass, these responses are larger, and the latencies
children sway at a faster rate than adults. and durations of these responses are longer
Thus, the task of static balance is slightly more than those of adults. Other studies have also
difficult since the body is moving at a faster found a longer duration of postural responses
rate during imbalance (55). in voung children and have additionally noted
A number of studies have examined the activation of monosynaptic stretch re-

changes in spontaneous sway with develop- flexes in young children in response to plat-
ment (56, 57). One study examining children form perturbations. These responses disap-
from 2 to 14 years of age showed that the pear as the children mature (60, 61 ).
amplitude of sway decreased with age. There Surprisingly, postural responses in chil-
was considerable variabilit)' in sway amplitude dren 4 to 6 xears of age are, in general, slower
in the young children. This variance system- and more variable than those found in the 15-
atically lowered with age and with the chil- month- to 3-year-olds, 7- to-10-\'ear-olds, or
dren's improved balance. Effects of eye clo- adults, suggesting an apparent regression in
sure were represented by the Romberg the postural response organization. Figure
quotient (eyes-closed sway expressed as a per- 7.14 compares EMG responses in the four age
centage of eyes-open sway), giving an indi- groups.
cation of the contributions of vision to bal- In these studies, by 7 to 10 years of age,
ance during quiet stance. Ver\' low Romberg postural responses were basically like those of
quotients were recorded for the youngest the adult. There were no significant differences
children who completed the task (4-year- in onset latency, variabilit)', or temporal co-
olds) with values less than 100%. This indi- ordination bersveen muscles within the leg syn-
cates that these children were swaying more erg\' between this age group and adults 59 ). (
with eyes open than with eyes clo.sed (56). VVhv are postural actions so much more
Spontaneous sway in children reaches adult variable in the 4- to 6-year-old child? It may
levels by 9 to 12 years of age for eyes-open be significant that the variability' in response
conditions and at 12 to 15 years of age for parameters of 4- to 6-year-old children occurs
eyes-closed conditions. Sway velocity also de- during a period of disproportionate growlJi
creased with age, reaching adult levels at 12 with respect to critical changes in body form
to 15 years of age (57). (59). It has been suggested that discontinu-
ous changes seen in the development of many
Compensatory Postural Control skills including postural control, may be the
result of critical dimension changes in the

Refinement of compensaton,' balance body of the growing child (62). The system
adjustments in children 15 months to 10 years would remain in a state of stability until di-
of age has been studied by several researchers mensional changes reached a point where pre-
\_yV^x-A_^
h\» ^
~' —^ -^.^^
\c:X3
^ûu
Figure 7.14. A comparison of muscle activation patterns in leg and trunk muscles in response to forward platform
perturbations causing backward sway in four age groups of normal subjects. Three successive responses to platform
perturbations are shown for each child. Platform perturbation started at the onset of the electromyogram recording.
Abbreviations: 7", tibialis anterior; Q, quadriceps; G, gastrocnemius; H, hamstring muscles. (From Shumway-Cook A,
Woollacott M. The growth of from a developmental perspective. Motor Behav 985;1 7:1 36.)
stability: postural control | 1
viousmotor programs were no longer highly of different segments of the body, in response
At that point, the system would
effective. to platform perturbations in both children
undergo a period of transition marked by in- and adults (63), has shown tliat the kinemat-
stability and variability, and then a new pla- ics of passive body movements caused bv plat-
teau of stabiiit\'. form translations are \'er\' similar in the 4- to
Recent work analyzing the movements 6-year-old, 7- to 9-year-old, and adult. Thus,
C;haptcr Sc\cn DKVhLOPMEKT OF POSTURAL Con IROl. 163
it is more probable that changes in response trol involves determining the accuracy of in-
latencies and \ariabilit\- seen in 4- to 6-year- coming sensorv' inputs for orientation pur-
olds represent developmental changes in tiic poses, and selecting the most appropriate
nervous system itself. sense for orientation, given the context. This
process entails changing the relati\e weight-
SENSORY CONTRIBUTIONS ing of sensor\' inputs for postural control, de-
pending on their accuracy for orientation (64,
Visual inputs affect balance control in a
65). How does the CNS learn to interpret in-
number of ways. To determine these effects,
formation from vision, vestibular, and soma-
one can stimulate balance responses with vi-
tosenson' receptors and relate it to postural
sual inputs. Alternatively, one can remove vi-
actions?
sion and see if there are any deficits in balance
We have already described evidence
function.
from moving room experiments suggesting
Removing visual inputs with opaque
that the visual system plays a predominant
goggles during horizontal platform move-
role in the development of postural actions.
ments has a surprising effect on the organi-
That is, visual inputs reporting the body's po-
zation of postural responses in children ages
sition in space appear to map to muscular ac-
2 to 7 years (44). Previous studies had found
tions earlier than other sensor)' systems. In
that adults wearing opaque goggles showed
\'oung children, the invariant use of visual in-
no significant differences in the organization
puts for postural control can sometimes mask
or timing of muscle responses. In contrast, in
the capability' of other senses to activate pos-
2- to 3-year-olds, postural responses were
tural actions. Results from the experiments in
more likely to be activated with shorter onset
which children balanced without \ isual inputs
latencies. In the 4- to 6-year-olds, muscle re-
suggest that in certain age groups, postural
sponse patterns were again more likely to be
actions activated by other sensoPi' inputs can
activated, but the timing of the responses was
be better organized than those associated
more variable.
with vision!
What is the significance of more consis-
Moving platform posturography in con-
tently organized and faster postural responses
junction with a moving visual surround has
when vision is removed? It implies that visual
also been used to examine the development
cues are not required to activate postural re-
of intersensorv- integration for postural con-
sponses in children as young as 2 years of age.
trol. The platform protocols used to study the
In fact, removal of visual cues may actualh'
organization and selection of senses for pos-
increase the sensitivity- of the postural system
tural control are described in detail in the pre-
to the remaining proprioceptive ancf \cstibu-
\ious chapter.
lar These findings support the concept
cues.
The development of sensor\' adaptation
that vision may be the dominant sense for
in children ages 2 to 9 was studied using a
postural control in the 2- to 3-year-old age
modification of this protocol (59). Four- to
group. When vision is removed, a shift occurs
6-year-olds swayed more than older children
from the use of longer latency visual input
and adults, e\en when all three sensor)' inputs
with eves open to shorter latency propriocep-
were present (condition 1 ). With eyes closed
tive inputs with eyes ck)scd (44).
(condition 2), their stabilit)' decreased fur-
ther, but they did not
DEVELOPMENT OF SENSORY fall.
Reducing the accuracy of somatosen-

ADAPTATION
sor\- information for postural control by ro-
Postural control is characterized b\' the tating the platform surface (condition 3) fur-
abilit\' to adapt how we use sensor>' informa- ther reduced the stability' of 4- to 6-year-olds,
tion about the position and movement of the and half of them lost balance. When children
body in space to changing task and environ- 4 to 6 vears of age had lo maintain balance
mental conditions. The process of organizing using primarily vestibular information alone
and acfapting sen.sor\' inputs for postural con- for postural control, all but one fell. In con-
none of the older children 7 to 9 years

trast, Development of Anticipatory
of age lost balance. Figure 7.15 compares
Postural Actions
body sway in children of various ages and
adults in these four sensory conditions (59). Skilled movement has both postural and
These results suggest that children un- voluntary components; the postural compo-
der 7 years are unable to balance efficiendy nent establishes a stabihzing framework that
when both somatosensory and visual cues are supports the second component, that of the
removed, leaving only vestibular cues to con- primar)' movement (66). Without this sup-
trol stability. In addition, children under 7 porting postural framework, skilled action de-
show a reduced abiUty to adapt senses for pos- teriorates, as seen in patients with a variety of
tural control appropriately when one or more motor problems.
of these senses are inaccurately reporting body The development of reaching in infants
orientation information. shows changes that parallel postural devel-
100%
FALL
r^
.§ S 60-
4-6
dl.
7-10 A '
4-6
m^ li
7-10 A 4-6 7-10 A 4-6 7-10 A
Figure 7.15. A comparison of body sway in 4- to 6-year-olds, 7- to 10-year-olds, and adults in the four sensory
conditions. A, Eyes open, firm support surface. B, Eyes closed, firm support surface. C, Eyes open, sway-referenced
surface. D, Eyes closed, sway-referenced surface. (Adapted from Shumway-Cook A, Woollacott M. The growth of
stability: postural control from a developmental perspective. ) Motor Behav 1985;17:141 .)
Chapter Seven Dkvelopment of Postural Control 165
Table 7.2 Systems Model of Postural Development

9 10 11 12 15 20 24 3 4 5 6 7
months
Sense and motor systems operational
No rules for posture
Coordinated muscle action at neck for posture
Visual system maps to neck muscles
Somatosensory system maps to neck muscles
Vestibular system maps to neck muscles
Multisensory mapping to neck muscles for head control
"Rules" for sensing and moving for posture extend to trunk
Visual system maps to leg muscles
Somatosensory system maps to leg muscles
Stance "ankle" synergy present, mapped to senses
Stepping synergy energes
Visual dominance
beginning to decline ^
Somatosensory
system dominance
— >
"Adult"-like
postural control
opment. Later sections of this book detail the

development of manipulator)' fianction. Summary
Infants as young as 9 months show ac-
ti\ ation of the postural muscles of the trunk
The development of postural control is an es-
sential aspect of the development of skilled
in advance of most but not all reaching move-
actions, like locomotion and manipulation.
ments (67). By the time infants are able to sit
Consistent with Gesell's developmental prin-
independently, and are showing relatively ma-
ciples, postural development appears to be
ture reaching movements, they are also show-
characterized by a cephalo-caudal progres-
ing advance activation of postural muscles to sion of control.
stabilize voluntarv' movements in the seated The emergence of postural control can be
position. characterized by the development of rules
Children as young as 12 to 15 months that relate sensory inputs reporting the body's
are able to activate postural muscles in ad- position with respect to the environment, to
vance of arm movements while standing 58 ( ). motor actions which control the body's posi-
tion.
By 4 to 6 years, anticipator\' postural adjust-
a. Control begins in the head segment. The
ments preceding arm movements while stand-
first sense that is mapped to head control
ing are essentially mature (68, 69).
appears to be vision.
Table 7.2 summarizes the emergence of
b. As infants begin to sit independently, they
postural control from a systems perspective.
learn to coordinate sensory-motor infor-
By comparing Tables 7.1 and 7.2, you can see mation relating the head and trunk seg-
the similarities and differences bet\veen this ments, extending the sensorimotor rules
model and the reflex-hierarchical model in de- for head postural control to trunk mus-
scribing the emergence of posture control in cles.
neurologically intact children. c. The mapping of individual senses to ac-

tion may precede the mapping of multiple 3. McGraw MB. From reflex to muscular con-
senses to action, thus creating internal assumption of an erect posture and
trol in the
neural representations necessary for co- ambulation in the human infant. Child Dev
ordinated postural abilities. 1932;3:291.
4. Anticipatory, or proactive postural control, 4. Woollacott M, Shumway-Cook A. Changes
which provides a supportive framework for in postural control across the lifespan — a sys-
skilled movements, develops in parallel with tems perspective. Phys Ther 1990;70:799-
reacf/ve postural control. 807.
5. Adaptive capabilities that allow the child to 5. Horak F, Shumway-Cook A. In: Duncan P,
modify sensory and motor strategies to changed. Balance: Proceedings of the APTA Fo-
ing task and environmental conditions devel- rum. .Alexandria, VA: APTA, 1990: 1 05-1 1 1
ops later. Experience in using sensory and mo- 6. Magnus R. Some results of studies in the
tor strategies for posture may play a role in the physiology' of posture. Lancet 1926;2:531-
development of adaptive capacities. 588.
6. The development of postural control is best 7. DeKleijn A. Experimental physiologii' of the
characterized as the continuous development labyrinth. J Lar\-ngol Otol 1923;38:646-
of multiple sensory and motor systems, which 663.
manifests behaviorally in a discontinuous 8. Rademaker GGJ. De Beteekenis der Roode
step-like progressionmotor milestones.
of Kernen en van the overige Mesencephalon
New strategies for sensing and moving can be voor Spiertonus, Lichaamshouding en Laby-
associated with seeming regression in behav- rinthaire Reflexen. Leiden:Eduarol Ijdo,
ior as children incorporate new strategies into 1924.
their repertoire for postural control. 9. Schaltenbrand G. The development of hu-
7. Not all systems contributing to the emergence man motilit\' and motor disturbances. Arch
of postural control develop at the same rate. Neurol Psychiatr 1928;20:720.
Rate-limiting components limit the pace at 10. Claverie P, Alexandre F, Nichol J, Bonnet F,
which an independent behavior emerges. Cahuzac M. L'activite tonique reflexe du
Thus, the emergence of postural control must nourisson. Pediatric 1973;28:661-679.
await the development of the slowest critical 11. Milani-Comparetti A, Gidoni EA. Pattern
component. analysis of motor development and its disor-
8. Much debate has occurred in recent years ders. Dev Med Child Neurol I967;9:625-
over the relative merits of the reflex-hierar- 630.
chical vs. systems models in explaining pos- 12. Barnes MR, Crutchfield CA, Heriza CB. The
tural development. In many respects, the two neurophysiological basis of patient treatment.
models are consistent. Their differences in- Vol II: Reflexes in motor development. Mor-
clude (a) the reflex-hierarchical model views gantown, WV: Stokesville Publishing, 1978.
balance control from a reactive perspective, 13. Peiper A. Cerebral function in infancy and
while the systems model stresses the impor- childhhod. N\': Consultants Bureau, 1963.
tance or proactive, reactive, and adaptive as- 14. Ornitz E. Normal and pathological matura-
pects of the system, and (b) the reflex-hierar- tion of vestibular fiinction in the human
chical model tends to weight the role of CNS child. In: Romand R, ed. Development of au-
maturation more heavily than experience, ditor)' and vestibular systems. NY: Academic
while the systems model does not emphasize Press, 1983:479-536.
the role of one over the other. 15. Cupps C, Plescia MG, Houser C. The Lan-
dau reaction: a clinical and electromyographic
analysis. Dev Med Child Neurol 1976;18:41-
'
53.
REFERENCES
16. Paine RS. The evolution of infantile postural
1. Amiel-Tison C, Grenier A. Evaluation neu- reflexes in the presence of chronic brain syn-
rologique du nouveau-ne et du nourrisson. dromes. Dev Med Child Neurol 1964;
Neurological evaluation of the human infant. 6:345-361.
New York: Masson, 1980:81. 17. Martin JP. The basal ganglia and posture.
2. Gesell A. The ontogenesis of infant behavior. Philadelphia: JB Lippincott, 1967.
In: Carmichael L, ed. Manual of child psy- 18. Bobath B, Bobath K. Motor development in
chologN-. NY: John Wiley & Sons, 1946:335- different t\pesof cerebral palsy. London:
373. Heinemann, 1976.
19. Capute AJ, VVachtel RC, Palmer FB, Shapiro mans: normal and impaired development.
BK, Accardo PJ. A prospective study of three Dissertation. Stockholm: Nobel Institute for
postural reactions. Dcv Med Cîiild Neurol Neurophvsiolog\', Karolinska Institute, 1992.
1982;24:314-320. 34. Thelen E, Ulrich, BD, Jensen JL. The devel-
20. Haley S. Sequential analyses of postural re- opmental origins of locomotion. In: Wool-
actions in nonhandicapped infants. Phys Ther lacott MH, Shumway-Cook A, eds. Devel-
1986;66:531-536. opment of posture and gait across the life
21 . Gesell A. Behavior patterns of fetal-infant and span. Columbia, SC: Universit\'of South Ca-
child. In: Hooker D, Kare C, eds. Genetics rolina Press, 1989:25^7.
and inheritance of neuropsychiatric patterns. 35. Forssberg H. Ontogeny of human locomotor
Res Publ Assoc Res Ner\- Ment Dis 1954; control. I: Infant stepping, supported Icko-
33:114-126. motion, and transition to independent loco-
22. Cor\'ell Henderson A. Role of the asym-
J, motion. E.\p Brain Res 1985;57:480^93.
hand visualiza-
metrical tonic neck reflex in 36. Prcchtl HFR, Prenatal motor develoment.
tion in normal infants. Am J Occup Ther In: Wade MC, Whiting HTA, eds. Motor de-
1979; 33:255-260. veloment in children: aspects of coordination
23. Larson MA, Lee SL, Vasque DE. Compari- and control. Dordrecht: Martinus NighofF,
son of ATNR presence and developmental 1986:53-64.
actixities in 2^
month old infants. Alexan- 37. BuUinger A. Cognitive elaboration of senso-
dria, VA: APTA; Conference Proceedings, rimotor behaWour. In: Butterworth G, ed.
June, 1990. Infancy and epistemology: an evaluation of
24. Fukuda T. Studies on human dynamic pos- Piaget's theon-. London: The Harvester
tures from the viewpoint of postural reflexes. Press, 1981:173-199.
Acta Otolar\ngol Suppl 1 96 1 1 6 1 1 -52
{ ) ; : 38. Bullinger A, Jouen F. Sensibilite du champ de
25. Hellebrandt FA, Schode M, Cams ML. detection peripherique qux variations postur-
Methods of evoking the tonic neck reflexes in aleschez le bcbe. Archives de Psychologic
normal human subjects. Am J Phys Med 1983;51:41-48.
1962;41:90-139. 39. Jouen F, Lepecq JC, Gapenne O. Early visual
26. Hirt S. The tonic neck reflex mechanism in vestibular relations in newborns. Child Dev,
the normal human adult. Am J Phys Med in press.
1967;46:56-65. 40. Gapenne O, Jouen F. Effect of visual inputs

27. Tokizane T, Murao M, Ogata T, Kordo T. on head's spontaneous oscillations in new-
Electromvographic studies on tonic neck, borns. Child Dev, in press.
lumbar and labyrinthine reflexes in normal 41. Jouen F. Visual-vestibular interactions in in-
persons. Jap J Physiol 1951;2:130-146. fancy. Infant Behavior and Development

28. Campbell SK, Wilhelm IJ. Development 1984;7:135-145.
from birth to 3 vears of age of 15 children at 42. Jouen F. Early \isual-vestibular interactions
high nenous system dysfiinc
risk for central and postural development. In: Bloch H,
tion. Phys Ther 1985;65:463^69. Bertenthal BI, eds. Sensor\ -motor organiza-
29. Molnar GE. Analysis of motor disorder in re- tions and development in infana* and early
tarded infants and young children. American childhood. Dordrecht: Kluwer, 1990:199-
Journal of Mental Deficiency- 1978; 83:213- 215.
222. 43. Butterworth G, Cicchetti D. Visual calibra-
30. Van Sant AF. Life-span development in function of posture in normal and motor retarded
tional tasks. Phys Ther 1990;70:788-798. Down's svndrome infants. Perception 1978;
31. Woollacott M, Shumway-Cook A, Williams 7:513-525.
H. The development of posture and balance 44. Woollacott M, Debu B, Mowan M. Neuro-
control. In: Woollacott MH, Shumway-Cook muscular control of posture in the infant and
A, eds. Development of posture and gait child: is vision dominant? Journal of Motor
across the life span. Columbia, SC: Universit)' Behavior 1987;19:167-186.
of South Carolina Press, 1989:77-96. 45. Butterworth G, Hicks L. X'isual propriocep-
32. Gurfinkel VS, Levik YS. Sensory complexes tion and postural stability in infancy. Percep-
and sensorimotor integration. Fiziolog\'a tion;6:255-262.
Cheloveka 1978;5:399-414. 46. Buttenvorth G, Pope M. Origine et fonction
33. Hirschfeld H. On the integration of posture, de la proprioception visuelle chez I'enfant. In:
locomotion and voluntary movements in hu- de Schonen S, cd. Le developpemcnt dans la
premiere annee. Paris: Presses Universitaires tion to altered support and visual condi-
de France, 1983:107-128. tions during stance. J Neurosci 1982;2:545-
47. Thelen E, Fisher DM. Newborn stepping: an 552.
explanation for a "disappearing reflex." De- 59. Shumway-Cook A, Woollacott M. The
velopmental Psychology', 1982;18:760-775. grovnh of stabilit)'; postural control from a
48. Roncesvalles NC, Jensen J. The expression of developmental perspective. Journal of Motor
weightbearing ability in infants between four Behavior 1985;17:131-147.
and seven months of age. Sport and Exercise 60. Berger W, Quintern J, Dietz V. Stance and
Psycholog>' 1993; 15:568. developmental
gait perturbations in children:
49. Woollacott M, Sveistrup H. The develop- aspects of compensatory mechanisms. Elec-
ment of sensorimotor integration underlying troencephalogr Clin Neurophysiol 1985;
posture control in infants during the transi- 61:385-395.
tion to independent stance. In: Swinnen SP, 61. Hass G, Diener HC, Bacher M, Dichgans J.
Heuer H, Massion J, Casaer P, eds. Interlimb Development of postural control in children:
coordination: neural, dynamical and cogni- short-, medium-, and long latency EMG re-
tive constraints. San Diego, CA: Academic sponses of leg muscles after perturbation of
Press, 1993. stance. Exp Brain Res 1986;64:127-132.
50. Sveistrup H, Woollacott MH, Shumway- 62. Kugler PN, Kelso JAS, Turvey MT. On the
Cook A, McCollum G. A longitudinal study control and coordination of naturally devel-
on the transition to independent stance in oping systems. In: Kelso JAS, Clark JE, eds.
children. Neuroscience Abstracts 1990; The development of movement control and
16:893. coordination. NY:John Wiley & Sons,
5 1 Woollacott MH, Sveistrup H. Changes in the 1982:5-78.
sequencing and timing of muscle response 63. Woollacott M, Roseblad B, Hofsten von C.
coordination associated with developmental Relation between muscle response onset and
Human Move-
transitions in balance abilities. body segmental movements during postural
ment Science 1992;11:23-36. perturbations in humans. Exp Brain Res
52. Sveistrup H, Woollacott MH. Systems con- 1988;72:593-604.
tributing to the emergence and maturation of 64. Horak F, Shupert C. The role of the vestib-
stability in postural development. In: Savels- ular system in postural control. In: Herdman
bergh GJP, ed. Advances in psychology" the S, ed. Vestibular rehabilitation. NY: FA Da-
development of coordination in infancy'. Am- vis, in press.
sterdam: Elsevier, 1993:319-336. 65. Berthoz A, Pozzo T. Intermittent head sta-
53. Sveistrup H, Woollacott M. Can practice bilization during postural and locomotor},'
modif\' the developing automatic postural re- tasks in humans. In: Posture and gait: devel-
sponse? Child Dev, in press. opment, adaptation and modulation. Am-
54. Foster E, Sveistrup H, Woollacott MH. Tran- blard B, Berthoz A, Clarac F, eds. Amster-
sitions in visual proprioception: a cross-sec- dam: Elsevier, 1988;189-198.
tional developmental study of the etfect of vi- 66. Gahery Y, Massion J. Coordination between
sual flow on postural control. Journal of posture and movement. Trends Neurosci
Motor Behavior, in press. 1981;4:199-202.
55. Zeller W. Konstitution und Entwicklung. 67. Hofsten von C, Woollacott M. Anticipator}'
Gottingen:Verlag firr Psychologic, 1964. postural adjustments during infant reaching.
56. Hayes KC, Riach CL. Preparatory' postural Neuroscience Abstracts 1989;I5:1199.
adjustments and postural sway in young chil- 68. Nashner L, Shumway-Cook A, Marin O.
dren. In: Woollacott MH, Shumway-Cook Stance posture control in selected groups of
A, eds. Development of posture and gait children with cerebral palsy: deficits in sen-
across the life span. Columbia, SC: University son' organization and muscular coordination.
of South Carolina Press, 1989:97-127. E.xp Brain Res 1983;49:393^09.
57. Taguchi K, Tada C. Change of body sway 69. Woollacott M, Shumway-Cook A. The de-
with growth of children. In: Amblard B, Ber- velopment of the postural and voluntary mo-
thoz A, Clarac F, eds. Posture and gait: de- tor control svstem in Down's syndrome chil-
velopment, adaptation and modulation. Am- dren. In: Wade M, ed. Motor skill acquisition
sterdam: Elsevier, 1988:59-65. of the mentally handicapped: issues in re-
58. Forssberg H, Nashner L. Ontogenetic devel- search and training. Amsterdam: Elsevier,
opment of postural control in man; adapta- 1986:45-71.
Chapter 8
Aging and Postural Control
1 Introduction Sensory Systems

1 Models of Aging Changes in Individual Sensory Systems
1 Primary and Secondary Factors and Aging Somatosensory
< Heterogeneity of Aging Vision
Behavioral Indicators of Instability Vestibular
Systems Analysis of Postural Control Multisensory Deficit
• Musculoskeletal System Adapting Senses for Postural Control
1 Neuromuscular System Anticipatory Postural Abilities
' Changes in Quiet Stance Cognitive Issues and Posture Control
Changes in Motor Strategies During Balance Retraining
Perturbed Stance Summary
Adapting Movements to Changing Tasks
and Environments
Introduction in many older adults, scientists do not agree

on how and why we age (1-6). This has led
VVhy is it that Mr. Jones at the age of 90 to a number of models of aging (7-9). Two
is able to run marathons, while Mr. Smith at models are shown in Figure 8.1. The first
the age of 68 is in a nursing home, confined model (Fig. 8.1^1) describes the process of ag-
to a wheelchair, and unable to walk to the ing as a linear decline in neuron function
bathroom without assistance? Clearly, the an- across all levels of the central nervous system
swer to this question is complex. Many factors (CNS). It predicts that, as the number of neu-
outcomes with respect to health and
affect rons declines in a specific part of the CNS,
These factors contribute to the tre-
mobilit)'. \arious disease states become evident (8). Al-
mendous differences in abilities found among ternatively, a second model of aging (Fig.
older adults. 8.15) suggests that the CNS continues to
This chapter does not describe all as-
function at a relatively high level until death,
pects of aging. Rather, the focus is on age- unless there is a catastrophe or disease that
related changes occurring in systems critical affects a specific part of the CNS. Thus pa-
to postural control. We re\iew the research thology within individual parts of the CNS
examining age-related changes in systems may result in a rapid decline in a specific neu-
whose dysfunction may contribute to insta- ral fiinction (8).
bilin,' among older adults and recent studies These models lead to ver\' different con-
that look at the effects of training on improv- clusions regarding the inevitability of func-
ing balance function in these systems. Some tional decline with aging. The first model of-
introductory comments about research ex- fers a rather pessimistic view of aging, since it
amining changes in older adults are important suggests that neuronal loss is andinevitable,
to keep in mind. thus functional loss is an invariant part of
growing old. This r\pe of reasoning can lead
Models of Aging
to self- limiting perceptions on the part of
Though many studies have examined older indi\iduals regarding what they can do
the process of aging and have shown a decline (10). These self-limiting perceptions can be
in a number of sensor\' and motor processes inadvertendy reinforced by the medical pro-
169
Primary and Secondary Factors

Normal aging
and Aging
Many scientists beheve that factors con-

tributing to aging can be considered either
primar\' or secondan* (9). Priman' factors,
such as genetics, contribute to the inevitable
decline of neuronal fijnction in a system. An
example of a genetic predisposition to a con-
Normal aging
dition would be the situation where a person
Different catastrophes carries the genes for degeneration of auditor)'

"
and disease neurons and suffers hearing loss in old age.
^
^ Parkinson's Genetic predisposition can interact with en-
^
, X threshold vironmental factors. For example, a person
who comes from a family with a tendency' to-
Age
B wards hearing loss and who works in a noisy
Two models of aging. A, The model
environment may e.xperience accelerated
Figure 8.1. first
suggests that aging is associated with an inevitable de- hearing loss due to a combination of genetic
cline in neuronal function in all systems. B, Alternatively, and environmental influences. Primani' factors
the second model suggests that neuronal function re- do not necessarily lead to a generalized de-
mains optimal with aging unless specific catastrophes or
cline, but rather to a loss of fiinction within
Wool-
disease affect specific parts of the system. (From
specific systems (9).
lacott M. Aging, posture and movement prepa-
control,
ration. In; Woollacott MH, Shumway-Cook A, eds. De- Research is beginning to suggest that
velopment of posture and gait across the lifespan. secondan' factors haxe a profound effect on
Columbia, SC: University of South Carolina Press, aging (10). Secondan', or experiential, factors
1989:156.)
are more or less under our control. Some of
these include nutrition, exercise, insults, and
fessional, who may hold a limited view re- pathologies that affect our mind and body.
garding what older adults can accomplish. For Environmental factors such as air pollution
example, when assessing an older adult, a and carcinogens our drinking water also
in
therapist may perceive that the patient's fall into this categon*, though you may not
strength is ôod, considering the patient's age. agree that these factors are under your
As a result, a strength grade of 3 out of 5, control!
which would ne\êr be accepted in a 30-year- Scientists have shown that proper nutri-
old, is often accepted in a 70-year-old as nor- tion results in prolonged and healthier Lives
mal. (11). Further, animal studies have shown that
In contrast, the alternative model of ag- span (12,
dietan,' restriction increases the life
more optimistic view (8). In this

ing leads to a 13). In addition, exercise programs have been
model, one expects optimal function from the shown to impro\'e cardioxascular health, con-
CNS unless unexpected patholog\' occurs and trol obesit\', and increase physical and mental
if optimal experiential factors are present. Ex- fiinction. The resultant gains in aerobic
periential factors involve leading a healthy and power, muscle strength, and flexibilit}' can im-
active life. In this case, when a therapist eval- prove biological age by 10 to 20 years. This
uates an older person, it is anticipated that can result in delaying the age of dependency
function will be optimal. If a decline is de- and increasing the quality' of the remaining
tected in any area of the nervous system, this years of life (13, 14). This knowledge that
perspecti\'e will allow the therapist to work on how we age is largely determined b)' how we
rehabilitation aimed at returning
strategies live leads to an emphasis on preventative
fiinction toward that of a normal \'oung health care measures (15). It also has impli-
adult. cations for rehabilitation. Therapists work to
Chapter Eight Aging .\ND PosTURAL CONTROL 171
assist older patients who have experienced pa- og\-, that is, had no disorders of the muscu-
tholog\' to return to optimal life-sr\'les. loskeletal, neurological, or cardiovascular sys-
Thus, the factors that determine the tems, or any previous histon' of falls (16).
health and mobility- of Mr. Jones \s. Mr. Interestingly, this study found no significant
Smith are a combination of primniy njjiiijj t'ic- differences between their younger and older
tors, primarily genetics, over which they have adult groups when comparing four parame-
limited control; and sccoudmy ajjhiji, primar- ters measuring the variabilit)' of gait. They
ilv experiential factors, over which the\' ha\e thus concluded that an increase in variabilitv'
considerable control. in the gait cycle among older adults was not
It appears that aging, whether it is pri- normal, but always due to some pathology.
mar\' or secondar\-, may not necessarily be These t>'pes of results suggest that there
characterized bv an overall decline in all fiinc- is much heterogeneity' among older adults.
tions. Rather, decline may be limited to spe- This amazing variabilit}' reminds us that it is
cific neural structures and functions. This is important not to assume that declining phys-
consistent with a major theme in this book, ical capabilities occur in all older adults.
that fiinctionand dysfimction are not gener-
alized, but emerge through the interaction of Behavioral Indicators of
the capacities of the individual carr\'ing out Instability
tasks within specific en\ironmental contexts.
Statistics on injuries and accidents in the
Heterogeneity of Aging older adult indicate that falls are the seventh
leading cause of death in people over 75 years
Certain studies show no change in func- of age ( 18 ). WTiat are the factors that contrib-
tion of the neural subsystems controlling pos- ute to these losses of balance.' Many early
ture and locomotion with age (16), while oth- studies on balance loss in the elderly expected
ers show a severe decline in fiinction in the to isolate a single cause of falls for a given
older adult (17). How can there be such a older adult, such as vertigo, senson,' neurop-
discrepancy in studies reporting ajie-relnted athy, or postural hypotension. In contrast,
changes in systems for posture and gait? This more current research indicates that falls in
may be due to tundamental differences in the the elderly have multiple contributing factors,
definition that researchers use in classifxing an including intrinsic physiological and muscu-
indi\idual as elderly. loskeletal factors and extrinsic environmental
For example, some researchers have factors (19 to 21).
classified the elderly adult as anyone over 60 To examine these factors, Lipsitz and
vears of age. WTien no exclusionan,' criteria are his colleagues followed a group of commu-
used in the study of older adults, results can nit}-dwelling older adults over 70 years of
be ven,- different from when researchers use age, for 1 year, and identified all falls tiiat oc-
restrictive criteria for including subjects for curred (21). They found that a number of fac-
studv. For example, a study on the effects of tors were associated with an increased risk of
aging on walking abilit\' selected agroup of falling, including reduced physical activirt', re-
71 subjects ranging in age from 60 to 99 duced proximal muscle strength, and reduced
years, using no exclusion criteria for possible stabilitN' while standing. Other significant fac-
pathology (17). These researchers noted that tors included arthritis of the knees, stroke, im-
the mean walking velocities for their older pairment of gait, hypotension, and the use of
adults were slower than any other studies had ps\chotropic drugs. The conclusions of this
previously reported. studv were that most falls in older adults in-
In contrast, another study examined \(>1\ and that many of
e multiple risk factors,
walking in healthy older adults. In this stud\-, these factors may be remediated. Thus, it was
1,187 individuals of 65 years and over were suggested that the clinician who is working
screened to find 32 who were tree of pathol- with an older adult should determine both in-
trinsic and extrinsic factors associated with a many older adults, including Buchner's and
particular fall and reduce or correct as many Wolfson's labs from the U.S., and Anni-
of these as possible. ansson's lab from Scandinavia (25-27).
The study of intrinsic factors leading to Lower extremit)' muscle strength can be re-
falls has included examining the role of bal- duced by as much as 40% between the ages of
ance control. Several researchers, including 30 and 80 years (26). This condition is more
Tinetti, fi-om the U.S., Berg, from Canada, severe in older nursing home residents with a
and Mathias and colleagues, from England, histor\' of falls (27). In these subjects, the
have measured functional skills related to bal- mean knee and ankle muscle strength were re-
ance in order to identify people at high risk duced two- and fourfold, respectively, com-
for falls (19, 22-24). Functional skills include pared with non-fallers.
sitting, standing and walking unsupported, Researchers have shown that the
standing and reaching forward, performing a association between strength and physical
360° turn, and moving from sit to stand po- fijnction is large, with over 20% of the vari-
sition. ance in fianctional status explained by relative
A more recent approach to understand- strength (25). However, the amount of
ing balance function in the elderly examines strength needed for physical function is de-
specific variables relating to normal postural pendent on the task. For example, it has
control and determines the extent to which been suggested that the r\'pical healthy 80-
deterioration in their fianction contributes to year-old woman is very near, if not at, the
loss of stabilit}^ and mobilit\' in the elderly.
In the remaining sections of this chapter

we examine the intrinsic factors related to bal-
ance problems in the older adult from a sys-
tems perspective. We discuss changes in the
motor system, the sensor)' systems, higher-
level adaptive systems, as well as the use of
anticipator)' postural responses before making
a voluntary' movement. Studies on the ability
of older adults to integrate balance adjust-
ments into the step cycle are covered in the
mobility section of this book.
Systems Analysis of
Postural Control
In previous chapters, we defined pos-
tural control as the ability to control the
body's position in space for the purpose of
stabilit)' and orientation, and discussed the
many systems that contribute to postural con-
trol (refer to Fig. 6.2). What have researchers
learned about how changes in these systems
contribute to an increased likelihood for falls
in the elderly?
Figure 8.2. Changes in spinal flexibility can lead to a

Musculoskeletal System stooped or flexed posture in many elderly people.
(Adapted from Lewis C, Bottomley ). Musculoskeletal
Several researchers ha\'e reported changes with age. In: Lewis C, ed. Aging: health care's
changes in the musculoskeletal system in challenge. 2nd ed. Philadelphia: FA Davis, 1990:146.)
Chapter Eight Aging and Postl'ral Control 173
threshold value for quadriceps strength nec- Similarly, a study examining spontaneous
essar>' 28 ). When strength
to rise from a chair ( sway in older adults with and without a his-
falls below the threshold needed for a task, ton,' of falls found a significant increase in
functional disabiiit\- occurs. sway in even healthy older adults compared to
Decreased range of motion (29) and young adults, with the greatest amount of
lossof spinal flexibilit>- in many older adults sway found in older of
people with a histon'
can lead to a characteristic flexed or stooped recent However, not all studies
falls (33).
posture (Fig. 8.2 ) ( 1 ). This can be associated have been consistent in showing increased
with other changes in postural alignment, in- postural swav among healthy elderly adults
cluding a shift in the vertical displacement of (30-37).
the center of body mass backwards towards .\nother study by Fernie and colleagues,
the heels (30). Other conditions, such as ar- examined both sway amplitude and velocity'
thritis, can lead to decreased range of motion in a population of institutionalized elderly
in many joints throughout the body. In ad- and determined that sway velocin,' (but not
dition, pain may limit the functional range of amplitude) was significantly greater for those
motion of a particular joint (30). who fell one or more dmes in a year than for
those who had not fallen (38).
Neuromuscular System On the w hole, these studies suggest that

older adults tend to sway more than young
adults during quiet stance. A possible conclu-
The neuromuscular system contributes
sion from these studies is that increased sway
to postural control through the coordination
is of declining balance control as
indicative
of forces eftective in controlling the body's
people age. This is based on the assumption
position in space.
that sway is a good indicator of balance con-
trol. It isimportant to realize that measures
CHANGES IN QUIET ST.\NCE of sway are not always good indicators of pos-
tural dyscontrol. There are several t\'pes of pa-
Traditional methods for assessing bal- tients with severe neurological disorders, such
ance function in the older adult have u.sed as Parkinson's disease, vestibular disorders, or
global indicators of balance control, such as peripheral neuropathy, who have normal sway
determination of spontaneous sway during in quiet stance (39). Therefore, caution must
quiet stance (31). One of the earliest studies be used when interpreting results from studies
examined the extent to which subjects in age that use spontaneous sway measures as indi-
groups from 6 years through 80 \ears swa\ed cators of balance control.
during quiet stance. Subjects at both ends of
the age spectrum (ages 6 to 14, and ages 50 CH.\NGES IN MOTOR STRATEGIES
to 80) had greater difficult}- in minimizing DURING PERTURBED STANCE
spontaneous sway during quiet stance than
the other age groups tested (31). This study Is the older adult capable of activating
tested a great variet\' of older adults, and did muscle response synergies with appropriate
not tr\' to limit subjects in the older groups timing, force, and muscle response organiza-
to those who were free of patholog\'. tion when balance is threatened? Most re-
More recent studies have measured search addresses this question by using a mov-
spontaneous sway in different age groups using platform to provide an external threat to
ing stabilometr\-, or static force plates. One balance. The organization of muscle re-
study examined 500 adults, aged 40 to 80 sponses used to compensate for the induced
years, who were free of patholog\', and found swav is examined. This approach was de-
that postural sway increased with each decade scribed in detail in the chapter on normal pos-
oflife. Thus, the greatest amount of sponta- tural control.
neous sway was seen in the 80-year-olds (32 ). Remember that w hen the balance of a
YOUNG ADULT
NORMAL TIMNG
„olNGl
^^^^ ^ *GING2
Figure 8.3. Changes in temporal structure of muscle resfDonse synergies in the elderly. A, Normally coordinated
muscle response pattern in a young adult, compared to B, a pattern of temporal delay, and C, temporal reversal.
(Reprinted with permission from Woollacott MH, Shumway-Cook A, Nashner LM. Aging and posture control: changes
in sensory organization and muscular coordination. Int ) Aging Hum Dev 1 986;23:335.)
young adult is disturbed by support surface between the two groups in certain response
movements, he or siie typically regains stabil- characteristics. The older adults showed sig-
ityby using an ankle movement strategy' in nificantly slower onset latencies in the ankle
which sway is focused at the ankle joint, and dorsiflexors in response to anterior platform
muscle responses are activated first in the movements, causing backward sway (29, 35).
stretched ankle muscle, and then radiate up- In addition, in some older adults, the muscle
ward to the muscles of the thigh and hip (refer response organization was disrupted, with
back to Fig. 6.5 ). How do the postural muscle pro.ximal muscles being activated before distal
response characteristics of healthy older adults muscles. This response organization has also
compare to those of younger adults? Wool- been seen in patients with central nerxous sys-
lacott, Shumway-Cook and Nashner com- tem dysfiinction (40). Figure 8.3 shows ex-
pared the muscle response characteristics of amples of muscle responses to anterior plat-
older adults (« = 12, aged 61 to 78 years) form moN'ements causing posterior sway in a
and younger adults (aged 19 to 38 years), and young adult, temporal delays in an elderly
found that the response organization was adult, and temporal dyscoordination in an-
generally similar between the older and other elderly adult.

younger groups, with responses being acti- The older adult group also tended to
vated first in the stretched ankle muscle and coactivate the antagonist muscles along with
radiating upward to the muscles of the thigh the agonist muscles at a given joint signifi-
(35). cantly more often than the younger adults.
However, there were also differences Thus, many of the elderly people studied
Chapter Eight Agino .\nd Postural Control 175
tended to stiffen the joints more than young adults, changes in the motor s\-stems affecting
adults when compensating for sway pertur- postural control can contribute significandy
bations. to an inability to maintain balance. Some of
these changes include i a ) impaired range of
.\D.\PTING MO\EMENTS TO motion and flexibility-, iî weakness, (c) im-
CH.\NGING TASKS .\ND paired organization among mus- s\Tiergistic
ENMRONMENTS cles acti\ated in response to instability-, and

(d) linutations in the abilit\- to adapt move-
Several labs, including those of Horak ments for balance in resf>onsc to changing
and Woollacott, have found that many older task and en\ironmental demands.
adults generally used a strategy' invohing hip
movements rather than ankle mo\ements sig- Senson,- Systems
nificandymore often than young adults 30, 1
41 Hip movements are t\pically used by

1. How do changes in the sensor\- systems
young adults when balancing on a shon sup- important for (Xjsture and balance control
pon surface, which doesn't allow them to use contribute to declining stability- as f>eople age?
ankle torque in comf>ensating for sway. The following sections re\iew changes within
It has been h\pothesized that this shift indi\idual senson- systems and then examine
towards use of a hip strategy for balance con- how these changes affect stabilit\ in quiet
trol in older adults ma\ be related to patho- stance, as well as our abilin.- to recover from
logical conditions such as ankle muscle weak- loss of balance.
ness or loss of peripheral senson. tunction ( 30,
41 1. With this shift toward a preferential use CH.\NGES IX IXDI\TDU.\L SENSORY
of the hip strategy-, older adults may alter the SYSTEMS
boundaries for discrete movement strategies
within internally mapped stabilit\' limits. This Somatosensory
concept shown in Figure 8.4 30
is ( ).
Horak has suggested that in older adults Studies ha\e shown that cutaneous \i-
some falls, particularh those associated with braton.- sensation thresholds at the knee are
slipping, may be the result of using a hip strat- increased in the elderly 1 70 to 90 years i com-
egy- in conditions w here the surface cannot re- pared to young adults 1 42 i. In this research,
sist the sheer forces of the feet, w hich are as- the authors reported an inabilin,' to record \i-
sociated with the use of this strategy', for brator\- responses from the ankle because
example, when on ice (30). man\ of the older subjects were not able to
In summary we see that for manv older f>erceive sensation there. Sensor\" neuropa-
Figure 8.4. A diagrammatic repfcsentation o»' the relationship hypothesized to occur between movements o* the
center of gravitv and the strategies used by a normal subject and older adults with some pathology. (From Horak F,
Shupert C, Mirka A. Components oi postural dyscontrol in the elderly: a review. Neurobiol Aging 1989;10:745.)
thies and diseases such as cervical spondylosis Multisensory Deficit

affect the transmission of sensory information
important for balance control. Multisensory deficit is a term used by
Brandt (46) to describe the loss of more than
Vision one sense important for balance and mobility
fijnctions. In many older people with multi-
Studies on the visual system show simi-
sensory deficits, the ability to compensate for
lar declines in fimction. Because of multiple
loss of one sense with alternative senses is not
changes within the structure of the eye itself,
possible because of numerous impairments in
less light is transmitted to the retina. In ad-
all the sensory systems important for postural
dition, there is of visual con-
typically a loss
control (46).
trast sensidvity,which causes problems in
contour and depth perception (43, 44). This
informadon is critical to postural fiinction.
ADAPTING SENSES FOR POSTURAL
Loss of visual acuity can result from cataracts,
CONTROL
macular degeneradon, and loss of peripheral
In addition to showing declines in func-
vision due to ischemic retinal or brain disease.
tion within specific sensory systems, research
Vestibular
from many labs, including those of Wolfson,
Horak, Stelmach, WooUacott, and Brandt,
The vestibular system also shows a re- has indicated that some older adults have
duction in fianction, with a loss of 40% of the more difficulty than younger adults in main-
vestibular hair and nerve cells by 70 years of taining steadiness under conditions where
age (45). In young adults, even fairly severe sensory information for postural control is se-
vestibular problems often do not affect bal- verely reduced (30, 32, 34-37, 47, 48).
ance control significantly because of the avail- To understand the contribution of vi-
ability of other senses providing orientation sion to the control of sway during quiet stance
information to the CNS. Imbalance can bein older adults, researchers examined sway un-
come apparent in environments where sen- der altered visual conditions (30, 32, 34-37,
sory cues for balance are reduced or inaccu- 47, 48). When young people close their eyes,
rate. For example, when subjects with they show a slight increase in body sway, and
vestibular loss were asked to balance under this is also true for healthy older adults (37).
conditions with reduced or conflicting so- However, research is contradictory in this
matosensory and visual inputs, they showed area, since many researchers have found that
excessive sway or loss of balance (30). healthy older adults do not tend to sway more
Dizziness, an additional consequence of with vision removed than do voung adults
some types of vestibular dysfiinction, can also (36,37).
contribute to instability among older adults. In addition, when their eyes are open,
Dizziness is a term used to describe the illu- healthy older adults are often as steady as
sion of movement. It can encompass feelings young adults when standing on foam, a con-
of unsteadiness and imbalance, as well as feel- dition that reduces the effectiveness of so-
ings of faintness or the sense of being light- matosensory inputs reporting body sway
headed. Dizziness can be a symptom of a va- (37).However, when healthy older adults are
riety of diseases, including those of the inner asked to stand with their eyes closed on a
ear. Partial loss of vestibular function can lead foam surface, thus using vestibular inputs
to complaints of dizziness, which can be a sig- alone for controlling posture, sway signifi-
nificant factor contributing to imbalance in cantly increases compared to young adults

the elderly. Degenerative processes within the (37).
otoliths of the vestibular system can produce Several studies have examined the ability
positional vertigo and imbalance during walk- of healthy older adults to adapt senses to
ing. changing conditions during quiet stance us-
Chapter Eight Aging and Postural Control 177
ing posturograpliy testing (30, 34-36). These voung people when there is a recluction in the
studies found that healthy active older adults availabilir\' or accuracy of a single sense for
did not show significant differences from postural control. However, in contrast to
young adults in amoimt of body sway (Fig. young adults, reducing the availability' of two
8.5) except in conditions where both ankle senses appears to have a significant effect on
joint inputs and visual inputs were distorted postural steadiness in even apparently healthy
or absent (conditions 5 and 6). older adults.
When both and somatosenson,'
visual Are the changes summarized above the
inputs for postural control were reduced result of an inevitable decline in ner\'ous sys-
(conditions 5 and 6), half of the older adults tem function, or are they the result of bor-
lost balance on the first trial for these condi- derline pathology in specific subsystems con-
tions and needed the aid of an assistant. How- tributing to postural fiinction?
ever, most of the older adults were able to To determine if evidence of borderline
maintain balance on the second trial within pathology existed in subjects v\ho participated
these two conditions. Thus, they were able to in a postural study and who considered them-
adapt senses for postural control, but only selves fit, active older adults, researchers gave
with practice in the conciition (35). each subject a neurological exam, and then
These results suggest that healthy older correlated the existence of borderline pathol-
adults do not sway significantly more than og\' with performance on the balance tasks
(41 ). Although all the older adults considered
themselves to be healthy, a neurologist par-
100
FALL
D
Younger
U
Older
ticipating in the study
ment, such as
found neural impair-
diminished deep tendon re-
Adults Adults flexes, mild peripheral nerve deficits, distal
weakness in tibialis anterior, and gastrocne-
mius and abnormal nystagmus in many adults

in the population. Loss of balance in two sub-
jects accounted for 58% of total losses of bal-
ance (41).
These subjects had no history of neu-
.1
OT
50 rological impairment, but the neurologist di-
agnosed them as having borderline pathology
of central ner\'ous system origin. These results
again suggest the importance of pathologies
within specific subsystems as contributing to
imbalance in the older adult, rather than a
generalized decline in performance.
Other researchers have also studied the
^nri rl
adaptation of sensory information during
quiet stance in older adults (30). One group
of older adults was active and healthy and had
no previous history of falls (labeled asymp-
tomatic). The second group was symptomatic
for falHng. Figure 8.6 illustrates some of the
results of their study, showing that over 20%
Figure 8.5. A comparison of body sway in the six sen- of the elderly (both symptomatic and asymp-
sory conditions in young versus a group of active healthy
when visual information
tomatic) lost balance
elderly. (Adapted from Woollacott MH, Shumway-Cook
was inaccurate for balance (Condition 3)
A, Nashner LM. Aging and posture control: changes in
sensory organization and muscular coordination. Int |
compared to none of the subjects ages 20 to
Aging Hum Dev 1986;23:340.) 39. Forty percent of the asymptomatic elderly
suggest a slower ability' to adapt postural con-

trol in this population.
A propensitA' for falls in the first trial of
a new condition is a recurring finding in many
different studies examining postural control
in older adults (30, 34-36). Perhaps this
means that a slowing occurs, rather than a to-
Figure 8.6. A comparison of number of falls in the six
tal lack of adaptability', in manv elderlv people.
sensory conditions in young, elderly non-fallers, and el-
derly fallers. [Open box equals 20-39 years; black box

A propensit}' to fall in new or novel situations
equals more than 70 years asymptomatic; shaded box

could also be the result of impaired anticipa-
equals more than 70 years symptomatic.) (From Horak tory mechanisms. Anticipaton,' processes re-
F, Shupert C, Mirka A. Components of postural dyscon- lated to postural control enable the selection
trol in the elderly: a review. Neurobiol Aging 1989: of appropriate sensor\' and motor strategies
10:732.)
needed for a particular task or environment.
lost balance in condition 6 when both visual Anticipatory Postural

and somatosensory information were inaccu- Abilities
rately reporting body sway. By contrast, less
than 10% of the normal young adults fell in Postural adjustments are often used in a
this condition. The symptomatic elderly had body before
proactive manner, to stabilize the
a larger percentage of falls in any condition making a voluntar\' movement. Adults in their
which was sway-referenced, that is, with mis- 70s and 80s may begin to have more difficult)'
leading somatosensory cues (conditions 4, 5, maneu\'ering in the world because the\' have
and 6). lost some of their abilit)' to integrate balance
This led researchers to conclude that the adjustments into ongoing voluntary' move-
abilitv' to select and weight alternadve orien- ments such as lifting or carr\'ing objects.
tation references adaptively is a crucial factor Thus, it is important to study the effects of
contributing to postural dyscontrol in many age on the abilit)' to use postural responses
older adults. This is especially true for those proactiveh' within the context of x'oluntary
who are symptomatic for balance problems
'
movements. It is in these dynamic conditions,
(30,48). including walking, lifting, and carr\'ing ob-
Why are there differences among rejects, that most falls occur.
searchers reporting on the capability' of older One of the first researchers to study age-
adults to maintain steadiness under altered related changes in anticipaton,' postural ad-
sensory conditions? These differences may justments was Man'kovskii, from Russia (49).
simply be related to the variet\' of subjects He compared the characteristics of anticipa-
studied. A neurological exam of older adults tor)' mover (vol-
postural responses and prime
without obvious signs of patholog)', may untar)') responses foryoung (ages 19 to 29),
bring out subtle signs of neural deficits con- medium old (ages 60 to 69), and ven,' old
tributing to balance dysfiinction. (ages 90 to 99) adults who were asked to do
Another approach to studying adapta- the simple task of flexing one leg at the knee
tion of sensor)' systems involves the use of ro- (prime mover response) while using the other
tational movements of a platform. These ex- leg for support (postural response), both at a
periments were described in more detail in comfortable and at a fast speed. Both the me-
earlier chapters. Results from platform rota- dium old adults and ven,- old adults showed a
tion studies with older acquits found that 50% slowing in both the postural (contralateral
of the healthy older subjects lost balance on rectus femoris) and prime mover (ipsilateral
the first trial. However, all but one of the biceps femoris) muscle response latencies, for
subjects were able to maintain balance on the movements at a comfortable speed, but
subsequent trials (35). This finding could this slowing did not result in an increased
Chapter Eight AGING AND Postural Control 179
probability t)f losing balance. However, at the larger for the voluntary' muscles than the pos-
fast speeds, for both medium and vcr\' old tural muscles, there may be a slowing in both
adults, (a) the correlation between the pos svstems in the older adult ( 50 ).
rural and prime mover muscles decreased and This studv also pointed out a number of
(/;) there was a decrease in the time period other interesting differences between this
between the onset of postural and prime population of elderly and young adults. Mus-
mover muscles. In the ver\' old, postural and cle response latencies were much more vari-
prime mover muscles were activated almost able in the elderly group than in the young
simultaneously. This inability to activate pos- adults. In addition, the organization of mus-
tural muscles tar enough before the prime cle synergists was disrupted in the elderly as
mover caused a loss of balance on many trials compared with the young adults.
(49). In a similar study, researchers found that

we mentioned that
In the last chapter, older adults showed more variabilit)' in the or-
in thenormal young adult, the same postural ganization of their postural adjustments than
response synergies that are activated during young adults. The majoritv' of older subjects
stance balance control are activated in an an- showed a change in the ordering of postural
manner
ticipator}' before making a voluntar\' response activation, tonic co-contraction of
movement while standing. Thus, when a agonist, and antagonist postural muscles and/
voung adult is asked to pull on a handle, first or activation of postural muscles following ac-
the gastrocnemius is activated, followed by tivation of prime mover muscles (51). This
the hamstrings, trunk extensor, and then the was associated wth longer reaction times and
prime mover muscle, the biceps of the arm. smaller center of pressure shifts for the older
A slowing in onset latency or a disrup- adults in the movement tasks.
tion of the sequence of activation of these These studies suggest that many older
postural synergies could affect the ability of adults have problems making anticipator)'
an older adult to make such movements as postural adjustments quickly and efficiently.
lifting objects.Experiments were performed This inability to stabilize the body in associ-
by the labs of Woollacott, in the U.S., and ation with voluntar\- movement tasks such as
Frank, in Canada, to explore age -related lifting or carr\'ing may be a major contributor
changes in the abilit\' of older adults to acti- to tails in many elderly people.
vate postural muscle response synergies in an

anticipatoPi' manner (50, 51). In one study, Cognitive Issues and
standing young (mean age 26 years) and older Posture Control
(mean age 71 years) adults pushed or pulled
on a handle that was adjusted to shoulder Mrs. Beaulieu, who is 80 years old, nor-
level, in response to a visual stimulus. Results mally has no problems with falls. She is walk-
of the study showed that the onset latencies ing down a busy sidewalk in the cit>', talking
of the postural muscles were significantly to a friend, while carr\'ing a fragile piece of
longer in the older adults than in the younger crystal she just bought at the department
adults when they were activated in a complex store. Suddenly, a dog runs in front of her,
reaction time task. There were large age-re- bumping into her. Will she be able to balance
lated increases in onset times for voluntan,' in this situation as well as she does when she
muscles. According to a systems perspective, is walking down a quiet street by herself?
this slowing in voluntan' reaction time in the Mrs. BeauUeu's friend, Mr. Cham-
older adult could be caused either by the need pagne, has within the 6 months recovered
last
for advanced stabilization by the already de- from a series of serious falls. These falls have
laved and weaker postural muscles or to slow- led to a loss of confidence and fear of falling,
ing in the voluntary' control system itself which has resulted in a reduction in his overall
Since the absolute differences in onset times activity' level and an unwillingness to leave the
between the young and the older adults were safety of his own home. Can fear of falling
significantly affect fiow we percei\'e and move similar to the young adults and others show-
in relation to balance control? Determining ing significant impairments. He concluded
the answer to these and other questions re- that dual-task designs were much more sen-
lated to the complex role of cognitive issues sitive measures of subtle processing deficits
in postural control may be a key to under- across different age groups.

standing loss of balance in some older adults. Although many studies have explored
As we mentioned in the first part of this the differences in postural performance be-
chapter, the capacitv' of an individual, the de- t\\een fallen and non-fallers, \cr\ few have
mands of a task, and the strategies the person explored the effect of fear of falling on the
uses to accomplish a task are important factors control of balance (53). There is now exper-
that contribute to the of a person to
abilit\' imental evidence that anxiet)' and fear of fall-
flinction in different environments. As indi- ing affect the performance of older adults on
viduals get older, their capacities to perform tests of balance control ( 10, 53). As a result,
certain tasks such as balance control may be older adults probably modulate strategies for
reduced compared to their abilities at age 20, postural control based on their perception of
but they will still be able to function in normal the level of postural threat. Thus, those older
situations where they can focus on the task. adults who have a great deal of anxiet)' about
Howe\'er, when they are faced \\'ith situations falling related to poor perceptions regarding
in wfiich they are required to perform multi- their level of balance skills will move in ways
ple tasks at once, such as the one just de- that reflect these perceptions. More work is
scribed, they may not have the capacit)' to per- needed to fully understand the relationship
form both tasks. between fear of falling and postural control.
Researchers are beginning to explore
the question of how our attentional capacities
affect our balance abilities in different envi-
Balance Retraining
ronments. Theo Mulder, a researcher from
the Netherlands, used a rather humorous Our review of pre\'ious research has
method for exploring these changes in the el- shown that there is a significant loss of balance
derly (52). He asked both young and older flinction in many older adults, and that there
adults to walk at their preferred speed down are specific decreases in function of the differ-
a walkway, either under normal conditions, ent neural and musculoskeletal systems con-
while making mental calculations, while wear- tributing to postural control. Can these losses
ing scuba diving flippers, or while doing both of balance flinction be reversed with training?
calculations and wearing the flippers. He In recent years, many research labs have be-
noted that the older adults had significantiy gun to design and test different training pro-
more problems than the young adults in per- grams with the specific goal of balance im-
forming the concurrent tasks, and walked provement. Training programs have included
much slower. In fact, he noted that the data such diverse components as aerobic exercise,
of the oldest subjects in this experiment re- strength, and balance training.
sembled data fi-om amputees who were just One t\'pe of balance training program
starting their rehabilitation. It was as if in both has focused on general aerobic exercise as a
groups the brain had to deal with a break- wav of improving stabilit}'. In one study, the
down in their normal control strategies and exercise program included stretching, walk-
the system became more \Talnerable. ing, reaction time maneuvers, and
static and
Although the single tasks were impaired active balance exercises performed for 1 hour,
somewhat in the older adults, the dual tasks three times a week, for 16 weeks (54). The
were most significantiy impaired. He also no- study did not show significant differences be-
ticed that the variabilin,' in the older adults tween the exercise and control groups of el-
was great, with some sho\\ing performance derlv women when measured on one- and
Chapter Eight Aging AND POSTURAL Control 181
two-legged balance tests with eyes open and characteristics to platform perturbations, in-
eyes closed. It is possible that the study did cluding significantly less coactivation of an-
not find significant impnnements in the ex- tagonist muscles after training than before
ercise group because it didn't focus on train- training when compared with the control
ing a specific subsystem related to balance group. These experiments suggest that a sen-
control, and thus the effects on any single sys- sor\' program in balance control may
training
tem were too small to be significant. result in significant improvements in balance
A second t>'pe of training program em- under altered sensor)' conditions, and this im-
phasized muscle strength training to impro\'c provement may transfer to other balance
balance. One study focused .specifically on tasks.
strengthening the leg muscles, and had con-

siderably greater success than general exercise
Summary
programs (55). This study used high-resist-
ance weight training of the quadriceps, ham- 1 Two models of aging include (a) the concept
strings, and adductor muscle groups in fi-ail that aging involves a linear decline in neuron
residents of nursing homes. The authors function across all levels of the central ner-
noted highly significant and clinically mean- vous system (CNS); and (b) the concept that
ingful gains in muscle strength in all subjects. during aging, the CNS continues to function
well until death, unless there is a catastrophe
In addition, there was a decrease in walking
or disease that affects a specific part of the
dme, and two subjects no longer used canes
CNS.
to walk at the end of the study.
2. Many scientists believe that factors contrib-
A study from our own laborator\' (56,
uting to aging canbe considered either pri-
57) used a balance training protocol that fo- mary or secondary. Primary factors, such as
cused on the use of different senson.' inputs genetics, contribute to the inevitable decline
and the integration of these inputs under con- of neuronal function in a system. Secondary
ditions in which sensor)' inputs were reduced factors are experiential and include nutrition,
or altered. Subjects ranged in age ft-om 65 to exercise, insults, and pathologies.
87 years. Differences in the amount of sway 3. Researchers In all areas find much heteroge-
of the subject from the beginning to the end neity among older adults, suggesting that as-
sumptions about declining physical capabili-
of the training period were determined. Sig-
ties cannot be generalized to all older adults.
nificant improvements were found in the
4. Falls are the seventh leading cause of death in
training group between the first and the last
people over 75 years of age. Falls In the el-
day of training in five of the eight training
derly have multiple contributing factors in-
conditions. cluding intrinsic physiological and musculo-
Although the subjects improved signif- skeletal factors and extrinsic environmental
icantly in the training paradigm itself, it was factors. Understanding the role of declining
necessars' to determine if this training could postural and balance abilities is a critical con-
transfer to other balance tasks. Therefore, the cern In helping to prevent falls among older
trained and control groups of subjects were adults.
also tested up to 4 weeks after the end of 5. Many factors can contribute to declining bal-
training on t\\o other balance tasks. We found ance control In older adults who are symp-
tomatic for Imbalance and falls. Researchers
the training group lost balance significandy
have documented impairments in all of the
less often than the control group did. In ad-
systems contributing to balance control; how-
dition, the training group performed signifi-
ever, there is no one predictable pattern that
cantly better on the t\vo additional tests of Is characteristic of all elderly fallers.
balance, including standing on one leg with 6. On a positive note, there are many older
eyes open and eyes closed. Finally, increased adults who have balance function that is
stability' in the training group \\as accompa- equivalent to young people, suggesting that
nied by specific changes in muscle response balance decline is not necessarily an inevita-
ble result of aging. We suggest that experien- Rats: 1. Physical, metabolic and longevitv'
tial factors such as good nutrition and exercise characteristics. J Gerontol 1985;40:657-
can aid in the maintenance of good balance 670.
and decrease the likelihood for falls as people 1 3 McCarter RJ, Kelly NG. Cellular basis of ag-
age. ing in skeletal muscle. In: Coe RM, Peny
HM, eds. Aging, musculoskeletal disorders
and care of the frail elderly. N\': Springer,
REFERENCES
1993:45-60.
1. Leuis C, Bottomley J. Musculoskeletal 14. Shephard RJ. Benefits of exercise in the el-
changes widi age. In: Lewis C, ed. Aging: derly. In: Coe RM, Perry HM, eds. Aging,
health care's challenge. 2nd ed. Philadelphia: musculoskeletal disorders and care of the frail
FA Dans, 1990:145-146. elderly. N\': Springer, 1993:228-242.

2. Duncan P\V, Chandler J, Studenski S, 15. Tinetti M. Performance-oriented assessment
Hughes M, Presort B. How do physiological of mobility' problems in elderly patients. J Am
components of balance aftect mobility in el- GeriatrSoc 1986;34:119-126.
derly men? Arch Phys Med Rehabil 1993; 16. Gabell A, Nayak USE. The effect of age on
74:1343-1349. variability in gait. J Gerontol 1984;39:662-
3. Tinetti ME, Ginter SF. Identif\ing mobiUt\- 666.
dysflinctions in elderly patients. JAi\L\ 1988; 17. Imms FJ, Edholm OG. Studies of gait and
259:1190-1193. mobiliDi' in the elderly. Age Ageing 1981;
4. Kosnik W, Winslow L, Kline D, Rasinski BC, 10:147-156.
Sekuler R. Visual changes in daily life 18. Ochs AL, Newbeny- J, Lenhardt ML, Harkins
throughout adulthood. J Gerontol Psych Sci SW. Neural and vestibular aging associated
1988;43:63-70. with falls. In: Birren JE, Schaie KVV, eds.
5. Sloane P, Baloh RW, Honrubia V. The ves- Handbook of psychology- of aging. NY: Van
tibular system in the elderly. Am J Otolar- Nostrand & Reinholdt,^1985:378-399.
yngol 1989;1:422^29. 19. Tinetti ME, VViUiams TF, Mayewski R. Fall
6. Aniansson A, Grimby F, Gedberg A. Muscle risk index for elderly patients based on num-
function in old age. Scan J Rehab Med 1978; bers of chronic disabilities. x\m J Med 1986;
6(Suppl):43-19. 80:429^34.
7. Davies P. Aging and Alzheimer's disease: new- 20. Campbell AJ, Borrie MJ, Spears GF. Risk fac-
light on old problems. Paper presented at tors for falls in a communit%-- based prospec-
Neuroscience Society- Annual Meeting, New
'
tive study of people 70 years and older. J Ger-
Orleans, LA, 1987. ontol 1989;44:MII2-MII7.
8. Woollacott, M. Aging, posture control, and 21. Lipsitz LA, Jonsson PV, Kellev MM, Koest-
movement preparation. In: Woollacott MH, ner JS. Causes and correlates of recurrent falls
Shumway-Cook A, eds. Development of pos- in ambulator\- frail elderlv. J Gerontol 199 1;

ture and gait across the lifespan. Columbia, 46:M114-M122.
SC: Universit)' of South Carolina Press, 22. Mathias S, Nayak USL, Isaacs B. Balance in
1989:155-175. elderly patients: the "get-up and go" test.
9. Birren JE, Cunningham W. Research on the Arch Phys Med Rehabil 1986;67:387-389.
psychology,' of aging: principles, concepts and 23. Berg K, Wood-Dauphinee S, Williams J, Gay-
theor\'. In: Birren JE, Schaie KW, Hand-
eds. ton D. Measuring balance in the elderly: pre-
book of the psycholog}' of aging. 2nd ed. NY: liminar\- development of an instrument. Phys-
Van Nostrand & Reinholdt, 1985:3-34. iodierapy Canada 1989;41:304-308.
10. Tinetti M, Richman D, Powell L. Falls effi- 24. Speechlev M, Tinetti M. Assessment of risk
cacy as a measure of fear of falling. J Gerontol and prevention of falls among elderly persons:
1990;45:239-243. role of the physiotherapist. Physiotherapy
1 1 Lee Manson J, Hennekens C, Paffenbarger
I, Canada 1990;2:75-79.
R. Body weight and mortalit\': a 27 vear fol- 25. Buchner DM, deLateur BJ. The importance
low up of middle aged men. JAAL\ 1993; of skeletal muscle strenglJi to physical func-
270:2623-2628. tion in older adults. Annals of Behavioral
12. Yu BP, Masossro EJ,MciMahan CA. Nutri- Medicine 1991;13:12-21.
tional influences on aging of Fischer 344 26. i\nniansson A, Hedberg M, Henning G, et
Chapter Eight Aging .\nd Postl'r.\i. Control 183
al.Muscle morpholog\', enzymatic acti\it>' 39. Horak FB. Effects of neurological disorders
and muscle strength in elderly men: a follow on disorders on postural movement strategics
up study. Muscle Nenc 1986;9:585-591. in the elderly. In: Vellas B, Toupet M, Rub-
27. Whipple RH, Wolfton LI, Amerman PM. enstein L, .\lbaredc JL, Christen Y, eds. Falls,
The relationship of knee and ankle weakness balance and gait disorders in the elderly.
to falls in nursing home residents: an isoki- Paris: Elsevier, 1992:137-152.
netic study. J Am Geriatr Soc 1987;35:13- 40. Nashner L, Shumway-Cook A, Marin O.

20. Stance posture control in selected groups of
28. Young A. Exercise physiolog\' in geriatric children \vith cerebral palsy: deficits in sen-
practice. Acta Scand 1986;711(Suppl):227- sory organization and muscular coordination.
232. Exp Brain Res 1983;49:393-409.
29. Studenski S, Duncan PW, Chandler J. Pos- 41. Manchester D, VVoollacon M, Zederbauer-
tural responses and eftector factors in persons Hvlton N, Marin O. Visual, vestibular and so-
with unexplained falls: and mcthodo-
results matosensor\' contributions to balance control
logic issues. J ,\m Geriatr Soc 1991;39:229- in the older adult. J Gerontol 1989;
234. 44:M118-M127.
30. Horak F, Shupert C, Mirka A. Components 42. WTianger A, Wang HS. Clinical correlates of
of posturaJ dyscontrol in the elderly: a renew. the vibraton' sense in elderly psychiatric pa-
Neurobiol Aging 1989;10:727-745, tients. I Gerontol 1974;29:39-45.
31. Sheldon JH. The eftect of age on the control 43. Pins DG. The etfects of aging on selected vi-
of sway. Gerontolog)' Clinics 1963;5:129- sual functions: dark adaptation, \isual acuit\',
138. stereopsis,and brightness contrast. In: Sek-

32. Toupet M, Gagey PM, Heuschen S. Vestib- ular R, Kline D, Dismukes K, eds. Modem
ular patients and aging subjects lose use of aging research: aging and human \isual fiinc-
\isual input and expend more energ\- in static tion. NT: .^lan £
lIss, 1982: 131-160.
postural control. In: Vellas B, Toupet M, 44. Pastalan L.\, Mantz RK, Merrill J. The sim-
Rubenstein L, .-Mbarede JL, Christen Y, eds. ulation of age-related sensor\' losses: a new
Falls, balance and gait disorders in the elderly. approach to the study of environmental bar-
Paris:Else\ier, 1992:183-198. riers. In: Preiser WFE, ed. Environment
33. Shumway-Cook A, Baldwin M, Kerns K, design research, vol 1. Stroudsberg, PA:
WooUacon M. The etfects of cognitive de- Dowden, Hutchinson & Ross, 1973:383-
mands on postural control in elderly fallers 390.
and non fallers. Society' for Neuroscience Ab- 45. Rosenhall U, Rubin W. Degenerative
stracts, 1993;2:257. changes in the human vestibular sensorv' ep-
34. Wolfton L, WTiipple R, .\merman P, Kaplan Acta Otolaryngol 1975;79:67-81.

ithelia.
J, Kleinberg A. Gait and balance in the el- 46. Brandt T, Daroft'RB. The multisensorv- phys-
derly. Clin Geriatr Med 1985;1:649-659. iologicaland pathological vertigo s\ndromes.
35 VVooUacott MH, Shumway-Cook A, Nashner .WNeurol 1979:7:195-197.
LM. Aging and posture control: changes in 47. Brandt T, Paulus W, Straube A. Vision and
sensors' organization and muscular coordi- posture. In: Blex W, Brandt T, eds. Disorders
nation. Int J Aging Human Dev 1986;23:97- of posture and gait. Paris: Elsevier, 1986:157-
114. 176.
36. Peterka RJ, Black FO. Age-related changes in 48. Horak FB, Mirka A, Shupert CL. The role of
human posture control: sensor\' organization peripheral vestibular disorders in postural
tests. Journal of Vestibular Research 1990; dyscontrol in the elderiy. In: Woollacon MH,
1:73-85. Shumway-Cook A, eds. The development of
37. Teasdale N, Stelmach GE, Breunig A. Pos- posture and gait across the lifespan. Colum-
tural sway characteristics of the elderly under bia, SC: Universitv of South Carolina Press,
normal and altered visual and support surface 1989:253-279.
conditions. J Gerontol 1991;46:B238-B244. 49. Man'kovskii NB, Mints AY, Lyscnvoik VP.
38. Fernie GR, Gn,fe CI, Holliday PJ, Llewellyn Regulation of the preparaton,' period for
A. The relationship of postural sway in stand- complex voluntarv' movement in old and ex-
of falls in geriatric subjects.
ing: the incidence treme old age. Human Physiologv' (Moscow)
Age Ageing 1982;11:11-16. 1980;6:46-50.
50. Inglin B, WooUacott MH. Anticipaton' pos- 54. Lichtenstein MJ, Shields SL, Shiavi RG,
tural adjustments associated with reaction Burger C. Exercise and balance in aged
time arm movements: a comparison between women: a pilot controlled clinical trial. Arch
young and old. J Gerontol 1988;43:M105- Phys Med Rehabil 1989;70:138-143.
M113. 55. Fiatarone MA, Marks EC, Ryan ND, Mere-
5 1 Frank JS, Patla AE, Brown JE. Characteristics dith CN, Lipsitz LA, Evans WJ. High-inten-
of postural control accompanying voluntary sity strength training in nonagenarians: ef-
arm movement in the elderly. Society for fectson skeletal muscle. JAMA 1990;
Neuroscience Abstracts 1987;! 3:335. 263:3029-3034.
52. Mulder T, Berndt H, Pauwels J, Nienhuis B. 56. Hu M, WooUacott M. Multisensor)' training
Sensorimotor adaptability in the elderly and of standing balance in older adults. I. postural
disabled. In: Stelmach G, Homberg V, eds. stabilit\' and one-leg stance balance. J Ger-
Sensorimotor impairment in the elderly. Dor- ontol 1994;49:M52-M61.

drecht: Kluwer. 1993:413^26. 57. Hu M, WooUacott M. Multisensorv training
53. Maki B, HoUiday PJ, Topper .\K. Fear of fall- of standing balance in older adults. II. kinetic
ing and postural performance in the elderly. and electromx'Ographic postural responses. J
J Gerontol 1991;46:M123-M131. Gerontol 1994;49:M62-M71.
Chapter 9
Abnormal Postural Control

Introduction Loss of Anticipatory Postural Control
Postural Dyscontrol: A Systems Perspective Summarizing Motor Problems by Diagnosis
Musculoskeletal Impairments Cerebral Vascular Accident— Spastic
Neuromuscular Impairments Hemiplegia
Weakness Parkinson's Patients
Abnormalities of Muscle Tone Cerebellar Disorders
Dyscoordination within Motor Strategies Sensory Disorders
Alignment Misrepresentation of Stability Limits
Movement Strategies lnabilit>' to Adapt Senses
Timing Problems Sensorimotor Adaptation

Scaling Problems Summary
m Motor Adaptation Problems
Introduction during performance of a task. In the rehabil-

itation emironment, emphasis is often placed
The recoven.- of functional indepen- on positive systems, such as abnormalities of

dence following a neuroiogical insult is a muscle tonus, at the expense of negative
complex process requiring the reacquisition s\mptoms, like loss of strength, when at-
of many skills. Since controlling the body's tempting to understand performance deficits
position in space is an essential part of regain- in the neiu-ological patient (2, 3). In addition,
ing functional independence, restoring pos- manv secondary' effects of CNS lesions also
tural control is a critical part of rehabilitation. contribute to the postural beha\-ior seen in pa-
Postural control ensures task-specific stability' tients.These secondarv- problems are not the
and orientation for functional skills. direct resultof the CNS lesion, but rather de-
In the therapeutic emironment, the abil- velop as a result of the original problem. For
it\- to retrain postural control requires an un- example, the patient with gastrocnemius spas-
derstanding of the physiological basis for nor- due to an upper motor neuron lesion
ticir\'
mal postural control, as well as an appreciation may de\elop secondan- tightness in the achilles
for the basis for instability- in the neurological tendon, limiting ankle range of morion. Lim-
patient. Yet, understanding the beha\iors re- ited range of motion at the ankle joint, which
lated to abnormal postural control seen in oiu" develops secondan- to the neurological lesion,
patients is complicated for several reasons. mav ultimatelv impair function as much as the
Hughlings Jackson described upper original impairment of spasticity (4).
motor neuron lesions as lesions of cortical and Interpreting patients' behaxiors related
subcortical structures, producing motor dys- to posture and movement is fiirther compli-
control because of the presence of abnormal cated because beha\iors (except in the most
behaviors, so-called positive symptoms, and acute cases) are not solely related to the out-
the loss of normal behaviors, negative symp- come of the CNS lesion, but most often re-
toms 1 ). Positive symptoms might include
( flect the CNS's best attempt to compensate
the presence of exaggerated reflexes, hyper- for that lesion. Compensatory strategies are
kinetic, or associated movements. Negative alternative approaches to sensing and mo\ing

symptoms may involve inability to generate used to accomplish the goal of maintaining
force, or inappropriate selection of muscles the body's position in space (5).
185
An example of a compensator' motor exclusively on vision for controlling the

strateg}' for postural control might be that of body's position in space.
the CVA patient who stands with the icnee Thus, understanding posture and move-
hyperextended because of an inability' to gen- ment behaviors seen in the patient with an
erate enough force to keep the knee from col- upper motor neuron (UMN) lesion is a com-
lapsing while standing (Fig. 9.1). Standing plicated process. It involves sorting out be-
with the knee in hyperextension ensures that ha\'iors (both positive and negative symp-
the line of gravit\' falls in front of the knee toms) that are the direct result of the lesion,
joint, keeping the knee passively extended those that have developed consequent to the
when loaded and preventing knee collapse original lesion (secondan,' factors), and those
while standing. Therapeutic interventions de- that are compensator.' behaviors.
signed to keep the patient from hyperextend- This chapter reviews the senson,' and
ing the knee will not necessarily be effective motor basis for instabilit\' in the neurological
until the patient either develops sufficient patient from a svstems perspective. Not in-
strength to control the knee position, or de- cluded in this chapter is a discussion of pos-
velops an alternative strategy' for pre\enting tural dyscontrol from the perspective of neu-
knee collapse. rological diagnosis, that is, what is the basis
An example of a compensator\' sensor^' for instability' in the patient with cerebral vas-
strateg)' might be that of the patient with a cular accident, or traumatic brain injur)', or
loss of vestibular flinction who learns to relv cerebral palsy. Instead, the chapter uses a
problem-based approach to focus on how def-
icits in the sensor}' and motor systems impor-
tant to postural control can contribute to loss
of the ability' to control the body's position in
space.
Postural Dyscontrol: A Systems

Perspective
Neurological disorders represent a wide

variet\' of upper motor neuron diseases. Since
lesions can occur an^'where in the CNS, there
can be many causes of postural dyscontrol in
the patient with a neurological insult. In ad-
dition, the capacity' of the individual to com-
pensate for a neural lesion will also van'. Thus,
the patient with a neurological deficit \\'ill
show a wide range of abilities and disabilities

owing to the mixture of t)'pe and severity' of
deficits in the many component systems of
posture and movement control.

A systems perspective to postural dys-
control focuses on identif\'ing the constraints
or impairments in each of the systems essential
to controlling body posture. Impairments
are defined as limitations within the individual
Figure 9.1. Compensatory postural strategies develop
movement
which restrict sensory and strate-
to accommodate primary impairments such as weakness.
gies for postural control. Impairments can be
By hyperextending the knee and flexing the trunk, the
line of gravity falls in front of the knee joint, preventing musculoskeletal, neuromuscular, sensory,
collapse of the knee in a hemiparetic patient. perceptual, or cognitive (Fig. 9.2). Our cur-
Chapter Nine ABNORMAL PosTURAL Control 187
most often sccondar\' to the neurological le-

sion. Yet, musculoskeletalproblems can be a
major limitation to normal postural function
in the neurological patient. Atypical postures
and movements in sitting (Fig. 9.SA} and
standing (Fig. 9.3B, and C) often develop as
a result of restrictions in movement associated
with shortened mu.scles.
Musculoskeletal restriction can limit
mcnement strategies used in balance. Since an
ankle movement strateg\' for controlling up-
right posture requires intact range of motion
and strength in the ankle, loss of ankle range
or strength will limit the patient's ability to
use this movement for postural control. Ther-

apeutic interventions, such as the use of an
Figure 9.2. Constraints on postural control can be the
anklc-foot-orthosis externally constrain mo-
results of impairments in the musculoskeletal, neuro-
tion at the ankle; this may prevent the patient
muscular, sensory, perceptual, and/or cognitive systems.
who has adequate ankle range of motion from
using it eftectively in controlling body sway.
Musculoskeletal impairments constrain-
rent knowledge of the effects of impairments
ing the ability' to mcjve have been reported in
in certainsystems on posture control is
a wide variet\' of neurolcjgically impaired pa-
greater than that of impairments in other sys-
tients. Loss of spinal flexibilit\' ma\' be a major
tems. For example, we know more about the
limitation in capacity' to move in patients with
effects of musculoskeletal impairments on
Parkinson's disease (6). Changes in spinal
posture control than we do about many cog-
flexibilitv' in patients with this disease can also
nitive impairments. In addition, impairments
afreet the alignment of the center of mass
arise from interacting deficits; thus, the re-
moving it forward with respect to the base of
sulting effect on motor behavior may be com-
support, which can be seen in Figure 9.4.
plex.
Following stroke, paralysis and immo-
During the recoven,' of postural control
bilit)- lead to loss of range of motion and sub-
following a neurological lesion, the therapist
sequent contracture. Of particular concern is
must help the patient develop a broad range
loss of range of motion in the ankle joint due
of scnsor\' and mott)r strategies effective in
to contractures in the gastrocnemius and so-
meeting the postural demands of a task. A key
leus muscle groups (7).
to developing effective strategies for balance
Immobilization of a joint decreases the
is understanding musculoskeletal and neural
flexibilit)' of the connective tissue, and in-
constraints or impairments that affect the abil-
creases that tissue's resistance to stretch (8).
ity' to both sense and control the body's po-
Paralvsis and subsequent immobilization also
sition in space. In the following sections we
result in disuse atrophy, which afreets trophic
discuss the constraints on motor control re-
factors in the muscle itself This can result in
sulting from d\'sfunction in the different sys-
a reduction in sarcomere numbers, a relative
tems contributing to posture control.
increase in connective tissue, and a decreased
rate of protein synthesis (9).
Musculoskeletal (Children with cerebral palsy frequently
Impairments show restricted range of motion in many
joints, including the ankle, knee, and hip.
In the patient with upper motor neuron C'ontractures of the hip, knee, and ankle mus-
lesions, musculoskeletal disorders develop cles are frequent con,sequences of disordered
Pelvis tipped
^
backward
Shortened
hamstring
muscles
Figure 9.3. Atypical postures due to musculoskeletal impairments. A, Excessive posterior tilt of pelvis in sitting ac-
commodates shortened hamstrings. B, Shortening of the gastrocnemius muscle results in toe walk. C, Hip flexor tight-
ness can result in tilting of the pelvis and flexion of the knee. (Adapted from ReimersJ. Clinically based decision making
for surgery. In: Sussman M, ed. The diplegic child. Rosemont, IL; American Academy of Orthopedic Surgeons,
1992:155,156, 158.)
patterns of movement (10). Using a habitual lems can contribute to an inability to sustain
crouciied postural pattern during stance and an ideal alignment of body segments in the
subsequent shortening of
gait results in the upright position, requiring excessive force to
the hamstring muscles, ensuring the contin- counter the effects of gravity and sustain a ver-
ued use of a habitual crouched posture. tical posture.
Patients with vestibular abnormalities
can show restrictions in cervical range of mo- Neuromuscular
Often these patients minimize motion of
tion.
Impairments
the head in an effort to reduce complaints of
dizziness. This strategy leads to secondary Neuromuscular limitations encompass a
cervical dysfiinction, which can restrict the pa- diverse group of problems that represent a
tient's abilit)' to move
ways that are nec-
in major constraint on postural control in the
essary to overcome the primary vestibular dys- patient with neurological dysftmction.
function (4).
In summary, musculoskeletal problems, Weakness
while often not a primary result of a neuro-
major constraint to
logical lesion, present a Neural lesions affecting the abilit)' to
normal posture and movement control in generate forces, both voluntarily and within
many neurologicaliy impaired patients. Loss the context of a postural task, are a major lim-
of range of motion and flexibility' can limit the itation in many neurologicaliy impaired pa-
ways in which a patient can move for postural tients. Strength is defined as the ability to
control. In addition, musculoskeletal prob- generate sufficient tension in a muscle for the
Chapter Nine Abnormal Postural Control 189
(15-17). In addition, stroke patients have

been shown to have abnormal and reduced
firing rates of motor neurons (9). Instabilit)'
in the weak
patient results from an inability to
generate sufficient force to counter destabiliz-
ing forces, particularly the force of gravitv', in
the \ertical position.
Abnormalities of Muscle Tone
The presence of abnormalities of muscle

tone with upper motor neuron
in the patient
lesions is well known (18-21). However, the

exact contribution of abnormalities of muscle
tone to functional deficits in posture, loco-
motion, and movement control is not well
understood.
The term spasticif\' is used clinically to
cover a wide range of abnormal behaviors. It
is used to describe («) hyperactive stretch re-

flexes, {b) abnormal posturing of the limbs,
(
c) excessive coactivation of antagonist mus-
cles, (d) associated movements, {e) clonus,
and (/) stereotyped movement synergies (22).
Thus, one word (spasticity') is used to describe
many abnormal behaviors often seen in pa-
tients with a neurological disorder.
Figure 9.4. In a Parkinson patient, changes in spinal The range of muscle tone abnormalities
flexibility can also affect the alignment of the center of found within patients who have UMN
lesions
mass with respect to the base of support. (Adapted from
is broad (Fig. 9.5). At one end of the time
Schenkman M. Interrelationship of neurological and me-
chanical factors in balance control. In: Duncan P, ed.
spectrtmi is flaccidin,' or complete loss of mus-
Balance: proceedings of the APIA forum. Alexandria, cle tone. Moving up the tone continuum is
VA: APTA, 1990:37.) hypotonicit\', defined as a reduction in the
stiffness of a muscle to lengthening. H\po-
tonicity is described in many different kinds
purposes of posture and movement (11). of patients, including those with spinocere-
Strength results from both properties of the bellar lesions (21 ), and in many developmen-
muscle itself (musculoskeletal aspects of tally delayed children, such as children with
strength) and the appropriate recruitment of Down syndrome (23).
motor units and the timing of their activation On the upper end of the tone spectrum
(9, 12-14). Neural aspects of force produc- is hvpertonicitN' or spasticity'. Spasticity' is de-
tion reflect (a) the number of motor units refined as "a motor disorder characterized by a
cruited, (b) the r\pe of units recruited, and (c) velocit\-dependcnt increase in tonic stretch
the discharge frequency ( 12-14). reflexes (muscle tone) with exaggerated ten-
Weakness, or the inabilirv' to generate don jerks, resulting from hyperexcitabilit>' of
tension, major impairment of fiinction in
is a the stretch reflex, as onecomponent of the
many patients with upper motor neuron le- upper motor neuron syndrome" (24).
sions. Several authors have documented the Remember, in the chapter on normal
selective atrophy of t)'pc I (slow) and II (fast) postural control, we defined normal muscle
muscle fibers in patients with a UMN lesion tone as the muscle's resistance to being
190 Section II POSTURE/BAL.\NCE
Normal range
Flaccldity Hypotonia of Hypertonia Rigidity
muscle tone
Figure 9.5. Range ot tonus found in a patient with a neurological impairment. On one end of the tonus continuum
is flaccidity or low tone. At the other end of the continuum are problems related to hypertonicity including spasticity
and rigidit>'.
lengthened, its stijfness. Normal muscle stiff- are close to their threshold for excitation. In-
ness or tone is the result of both non-neural creased depolarization could arise because of
and neural components. The non-neural ( a ) increased tonic excitator\- input from seg-
components reflect the mechanical-elastic mental afterents or descending pathways such
characteristics of the muscle and connective as the lateral vestibulospinal tracts, and/or b) (
tissue that resist lengthening. The neural basis a tonic reduction in inhibitor)' synaptic input
motor unit
for stiffriess reflects the degree of from the inhibiton,' interneurons (26).
most importandy, stretch reflex-gen-
acti\it\% The second mechanism that could pro-
erated muscle activit}', which resists muscle duce an enhanced reflex response to stretch
lengthening. Several mechanisms have been ( hypertonicit}' ) is disorder within the stretch
suggested as the basis for spastic hypertonus reflex mechanism itself Disorders in the
in the patient with a neurological disorder. stretch reflex mechanism could be alterations
One mechanism for Increased muscle in the threshold and or the gain of the stretch
stiffness in spastic hypertonia may be changes reflex in spastic hypertonicit\' (26).
in the intrinsic properties of the muscle fibers Most studies examining alteration in
themselves. Researchers analyzing gait in chil- stretch reflex mechanisms with spasticity' have
dren with cerebral palsy have found that in- been consistent in showing changes in the set-
creased tension in the gastrocnemius muscle point, or angular threshold of the stretch re-
is not always associated with increased muscle flex. It has been shown that the threshold for
acti\it\' in that muscle. Based on these find- motoneuron recruitment in response to

ings, so-called spastic gait (equinus foot po- stretch is reduced in patients with spasticit^'.
sition at foot strike) may be partiy due to As a result, a smaller or slower stretch can re-
changes in intrinsic properties of the muscle flexly excite the motoneurons. Changes in the
rather than hyperexcitabilit}- of the stretch re- threshold of the stretch reflex are purported
flexmechanism (25). to result from an enhanced net descending ex-
The predominant hypothesis regarding citatoPi' drivefrom higher centers, especially
the neural mechanism underlying spastic hy- vestibulospinal and reticulospinal pathways. It
pertonia is abnormalities within the segmental remains unresolved whether this augmented
stretch reflex. Researchers have suggested t^vo drive is caused solely by increased excitatory'
possible mechanisms that produce an en- descending input from these pathways, or
hanced reflex response to muscle stretch fol- whether it reflects a reduced inflow from de-
lowing a UMN lesion (26). scending or regional inhibitor\- svstems (26,
The first mechanism is an increase in a- 27).
motoneuron excitabilits', which results in an Despite the change in threshold for ac-
increased response to stretch evoked input. tivating the reflex response to stretch, the gain
Increased a-motoneuron excitability' could of the stretch reflex appears to be normal in
occur because the motoneurons are continu- the spastic muscle. This means the force-
ously depolarized more than normal and so length relationship in the spastic muscle re-
Chapter Nine Abnormal Postural Contrdi. 191
mains the same as for a normal muscle. It used cus in regaining motor control may have lim-
to be believed tiiat spastic hypertonia was due ited impact on helping patients regain
to hyperactivit)' of the 7-efferent fibers (so- functional independence. This is because loss
called y spasticitN'), causing an increased sen- of functional independence is often the result
sitiviu'of the muscle spindle receptor to of many factors, which may be more limiting
stretch,and a subsequent change in the gain to the recover)' of motor control than the
of the reflex. However, this concept has lost presence of abnormal muscle tone. Some of
support, since there is no evidence to support those factors include problems within the co-
the concept of increased dynamic fusimotor ordination of synergistic muscles acti\ated in
activity' as the basis for spasticity' (26, 27). In response to instability,'.
summaPi', spastic hypertonicit)' in upper mo-

tor neuron lesions is probably the result of Dyscoordination within Motor
changes in the threshold of the system to Strategies
stretch, rather than the gain of the system.
Though we ha\e a greater understand- Neurological lesions also affect the abil-
ing of the neural mechanisms underlying spas- ity' to organize multiple muscles into coordi-
tic hvpertonicit\', there is still no agreement nated postural movement synergies.
on the role of spastic hvpertonicit)' (a positive
sign of UMN
disease) in the loss of fimctional ALIGNMENT
performance (a negative sign) (26, 27).
It has been suggested that spastic hy- Alignment of the body refers to the ar-
pertonicit}' limits a patient's abilit\' to move rangement of body segments to one another,
quickly, since acti\ation of the stretch reflex is as well as the position of the body with ref-
\elocit\' dependent. Excessive activation of erence to graviU' and the base of support (4).
the stretch reflex mechanism would sen'e to Alignment of body segments over the ba,se of
reflexly prevent the lengthening of the antag- support determines to a great extent the effort
onist muscle during shortening of the agonist. required to support the bod\' against gravit\'.
This has been referred to as antajjotiist re- In addition, alignment cietermines the con-
straint (18, 28) or spastic restraint (20). It stellation of movement strategies that will be
would be expected that evidence for antago- effective in controlling posture (4).
nist restraint would appear as coactivation of Changes in initial position or alignment
the agonist and antagonist muscle associated are often characteristic of the patient with a
with movement. UMN lesion. Abnormalities can reflect
A growing number of research studies changes in the alignment of one body part to
are finding evidence against this argument. another. Examples include the patient who
Instead, researchers are arguing that inade- sits or stands with the pelvis rotated back pos-
quate recaiitment of agonist motor neurons, teriorly, with excessive trunk kyphosis and the
not increased activit}' in the antagonist, is the head in a fonvard-flexed position (refer back
primar)' basis for ciisorders of motor control to Fig. 9.3), or the child with cerebral palsy,
following UMN lesions (29-36). Thus, other who uses a habitual crouched postural pattern
problems such as inabilit}' to recruit moto- during stance and gait ( 10).
neurons (weakness), abnormalities of recip- Abnormal alignment can also be ex-
rocal inhibition between agonist and antago- pressed as a change in the position of the body
nist, and dyssynergia may be more disabling with reference to gravit)' and the base of sup-
to motor control than simph' hvpertonicirv port. For example, asymmetric alignment in
(26). sitting anci standing is often characteristic of

This research has tremendous implica- patients with a unilateral neural lesion such as
tions for clinical practice. It suggests that cerebral vascular accident (9). These patients
treatment practices directed primarily at re- tend to stand with weight displaced towards
ducing spastic hypertonicit)' as the major fo- the noninvolved side. Other patients, most
notably, patients with cerebellar lesions, tend cluding delays in the onset of a postural re-
to stand with a wide base of support (21 ). sponse, and problems in the temporal coor-
Finally, many patients stand with the dination among muscle synergists.
center of mass displaced either forward to
backward. For example, it has been reported
that elderly patients with a fear of falling tend
to stand in a forward lean posture with the
center of mass displaced anteriorly (37). (For- Let's do another experiment. Get a
ward displacement of the center of mass is partner and have him/her stand facing
you. Hold your partner by the hips and
shown in Fig. 9.4.) However, there are other
gently push him/her in the backward direction.
types of patients who stand with the center of
Watch the feet, notice how quickly the toes come
mass displaced posteriorly (4).
up as he/she is pushed in the backward direction.
Changes in alignment can be viewed as Do both feet react at about the same time? For
both a musculoskeletal impairment or as a most people, the anterior tibialis in both legs con-
strategy compensating for other impairments. tracts quickly, bringing the toes up symmetrically
For example, in the elderly person, alignment, in both legs. Remember from the chapter on nor-
which is often characterized by a prominent mal postural control, the actual onset time for the
kyphosis and for\vard-flexed head position, following a perturbation is approximately
tibialis
represents a musculoskeletal impairment that 100 msec. What would you expect to see if the
constrains movements necessary' for posture anterior tibialis in one leg was slow in becoming
and balance (38). In contrast, the asymmetric active? Probably that foot would be slow to come
off the ground compared to the other foot when
alignment commonly seen in the hemiplegic
you shift the person in the backwards direction.
patient who and stands with weight
sits
shifted to the nonhemiplegic side, is often a

strateg)' that develops to compensate for Delays in the onset of postural motor
other impairments such as weakness (4). Un- activity during recovery of balance result in
derstanding these differences is important, delayed corrective responses, increased sway,
since achieving a symmetrically aligned posi- and, in many cases, subsequent loss of bal-
tion may not be a reasonable goal for the ance. To study the timing of muscle activation
hemiparetic patient until underlying impair- for postural control, researchers use a moving
ments have resolved sufficiently to ensure that platform to induce sway in a standing subject
the hemiparetic leg will not collapse under the and EMGs to record how quickly muscles re-
weight of the body. spond to sway. Using this approach with
stroke patients, researchers have found that
MOVEMENT STRATEGIES muscle onset latenciesv\'ere often ven' slow,
approximately 220 msec compared with 90 to

We divide coordination problems that 100 msec found in normal controls (39).
manifest within postural movement strategies Inabilit}' to respond quickly to loss of
into (a) disorders related to the timing of pos- balance was also found in a number of trau-
tural actions, and (b) disorders related to the matic brain-injured patients (40). Signifi-
scaling of postural actions. cantly delayed activation of postural muscles
occurred in TBI patients with focal cortical
Timing Problems contusions. Interestingly, patients with mild
to moderate traumatic cerebral concussions
In many patients with neurological dis- did not show a similar delay in onset latencies
orders, postural dyscontrol not entirely
is re- of postural muscles.
lated to the ability to generate force, but re- Significant delays in the onset of pos-
sults fi-om an inabilit\' to time the application tural activit)' were reported in de\elopmental
of forces effectively for recovering stabilit)'. A abnormalities including Down syncHrome (41
number of different timing problems related and some forms of cerebral palsy (42).
to postural control have been described, in- Other t\'pes of timing problems can af-
Chapter Nine Abnori\l\l Postur.'U. Control 193
feet the coordination of muscles responding It is important to remember that during

synergistically to recover balance. When there the normal control of movement, the CNS
is a disruption in the timing among the mus- makes use of muscular synergies as a way of
of body
cles actixated to control the center simplifxing the control of movement. As we
mass, movements become dyscoordinated described in the chapter on normal postural
and can hamper the restoration of equilib- control, a synerg}- is a group of muscles that
rium. Disruption of the timing and sequenc- are constrained to act together to achieve a
ing of muscles that work synergistically has An important feature of nor-
fiinctional task.
been referred to as dyssynergia. Dyssynergia mal postural svnergies, which distinguishes
is a general term used to describe a variety' of them from abnormal synergies, is their abilits'
problems related to timing or sequencing of to be modified. Normal synergies are not in-
muscles for action. variant, that is, immutable, but are assembled
In the rehabilitation literature, the term to accomplish a task, and are therefore flexible
synergy has often been used to describe ab- and adaptable to changing demands. In the
normal or disordered motor control 18, 19). ( ncurologically impaired patient, dyssynergia,
Abnormal synergies are stereot\'pical patterns or the absence of normal synergies of mo\'e-
of movement that cannot be changed or ment, constrains the recover)' of normal mo-
adapted to changes in task or environmental tor control, including postural control.
demands. A variet\' of abnormal synergies that Timing among postural muscles can be
impair normal movement have been de- disrupted in a number of different ways. Yet,
scribed in hemiplegic patients (9, 19). Figure all t\'pes of timing problems are classified as
9.6 illustrates an example of an abnormal dyssyncvffia. Whsx are some of these problems,
flexor synergy in the arm in supine (A)^ sitting and what t\pes of patients topically have these
(B) and standing ( C) { 19). The process of re- problems?
cover*' during stroke rehabilitation has been Dyssynergia has been reported in pa-
described as the dissolution of abnormal syn- tients with spastic hemiplegia, either due to
ergies of movement in favor of independent cerebral palsy (42), or cerebral \ascular acci-
or selective control (19). dent (39).
Figure 9.6. Abnormal synergies of movement in a patient with hemiplegia in A, supine, B, sitting, and C, standing.
(Adapted from Brunnstrom S. Movement therapy in hemiplegia: a neurophysiological approach. Hagerstown, MD:
Harper & Row, 1970:12, 13, 15.)
Using the moving platform technique stretched is the gastrocnemius. But during
described in earlier chapters, postural muscle forward sway, imposing a stretch on the hy-
patterns were studied in a group of cerebral peractive gastrocnemius^ the first muscles to
palsy children ages 7 to 12 (42). In spastic respond were the hamstrings! The gastroc-
hemiplegic children, muscle responses in the nemius muscle was slow to become active,
affected leg were abnormally sequenced pri- and the amplitude of the muscle acti\-it)' was
marily due to delayed acti\ation of the distal low compared to the uninvolved side. These
muscle. When the platform moved back- findings are consistent with those of other au-
wards, the child swayed forward, and the mus- thors who have noted that one major finding
cle activation pattern in the nonhemiplegic in neurologically impaired patients with spas-
leg wasfrom distal to proximal, with a 30 to tic hypertonia isto recruit and reg-
an inabilin,'
50 msec delay (Fig. 9.7). In contrast, in the of motor neurons.

ulate the firing frequency
hemiplegic leg, the first set of muscles to be- This same disruption was found in re-
come active in response to forward swav was sponse to backward sway, that is, instead of
the hamstrings, followed by activation of the the normal activation of tibialis anterior,
gastrocnemius. quadriceps, and abdominals found in the non-
This finding was surprising for a number involved leg, the cerebral palsy hemiplegic
of reasons. On clinical examination, this child children activated the quadriceps muscle first,
showed spasticity in the gastrocnemius mus- followed by the tibialis anterior. The biome-
cle. Signs included: increased stiffiiess in re- chanical effect of the disordered sequencing
sponse to passive stretch, clonus, equinus gait, was hyperextension of the knee and forward
and lack of dorsiflexion at the ankle in re- fle.xion of the trunk (Fig. 9.8). When seen
sponse to a backwards displacement. One clinically, this movement pattern is often as-
possible explanation for all these clinical find- cribed to hyperactivity' of the gastrocnemius,
ings was a priman' impairment of gastrocne- which pre\ents appropriate activation of the
mius spasticity. tibialis anterior (TA) because of antagonist re-
Given these clinical findings, one might straint. In the case of the children tested in
predict a h\'peracti\e response in the gastroc- this stud}', however, acti\'it\' in the gastroc-
nemius when the child stood on the platform nemius muscle was not the cause of this par-
and swayed in the forward direction, since ticular movement pattern.
during forward sway the first muscle to be ^-Vnalysis of postural patterns in adult
Nonhemiplegic leg
1^^ Hemiplegic icy
nciiiiîcyn., leg
200 400 600 800 200 400 600 800 msec
Figure 9.7. Abnormal sequencing of muscles in a hemiplegic child responding to a backward translation of a moving
platform. EMG records show an inappropriateactivation of muscles responding to forward sway, with proximal muscles
firing in advance of the so-called spastic distal muscles. (Abbreviations: Ham, hamstrings; Quad, quadriceps; Cast,
gastrocnemius; Tib, tibialis anterior.) The arrow signals platform movement onset. (Adapted from Nashner LM, Shum-
way-Cook A, Marin D, Stance posture control in select groups of children with cerebral palsy: deficits in sensory
organization and muscular coordination. Exp Brain Res 1983:49:401 .)
brain-injured adults with focal cortical con-

tusions (40). The biomechanical conse-
quences of delayed acti\'ation of proximal
muscles compared to the distal muscles in-
clude excessive motion at the knee and hip.

This is because the muscle timing pattern is
not efficient in controlling the indirect effects

of forces generated at the ankle on more prox-
imal joints.
Dyssynergia can also be characterized by
cocontraction of muscles on both anterior
and posterior aspects of die body. Researchers
have found that Parkinson's padents use a
complex movement strategy' when respond-
ing to threats to balance. (This is shown in
Fig. 9.9.) This activation of muscles on both

sides of the bodv results in a stiffening of the
bodv, and is a ver\' inefficient strategy' for the
recover)' of balance, since it is not direction-
ally specific (44).
These results are not consistent with the

classic \\'ork on Parkinson's
patients by Pur-
due Martin, who reported an absence of equi-
librium and righting reactions in Parkinson's
patients (45). The rigidit)' and loss of balance
found in patients during tilt tests imply that
Figure 9.8. The biomechanical consequences of a disequilibrium reactions were absent. Placing
ruption in the timing of muscles responding to backwards EMGs on the muscles of Parkinson's patients
sway includes back-kneeing and forward flexion of the has allowed researchers to see that Parkinson
trunk. The numbers represent the order in which the mus-
patients do indeed respond to disequilibrium,
cles are activated. (Adapted from Shumway-Cook A,
but the pattern of muscular activity' useti is
McCollum G. Assessment and treatment of balance def-
icits. In; Montgomery P, Connolly B, eds. Motor control
ineffective in recovering balance.
and physical therapy. Hixson, TN: Chattanooga Croup,
Scaling Problems
1991:130.)
Maintaining balance requires that forces

generated to control the body's position in
Stroke patients has also revealed disordered space be appropriately scaled to the degree of
patterns of muscle activity', including abnor- instability'. This means that a small perturba-
mal timing and sequencing of muscle activation to stabilit\' is met with an appropriately
tion, excessive cocontraction, and greater sized muscle response. Thus, force output
variabilit\' in the timing of responses among must be appropriate to the amplitude of in-
hemiplegic subjects (9). Disorders in initial Researchers are beginning to exam-
stability'.
standing posture also affected the organiza- ine the physiological mechanisms underlying
tion of postural strategies in some hemiplegic the scaling of postural responses in neurolog-
subjects (43). ically intact subjects. In adciition, researchers
Patients with dyssynergia sometimes are looking at the effects of lesions in the cere-
have abnormally long delays in the onset time bellum or basal ganglia on the abilit\' to scale
of proximal muscle synergists. This t\-pe of the amplitude of postural responses to differ-
dyssynergia has been reported in children ent sized perturbations to balance (46, 47).
with Down syndrome (41) and in traumatic Results from these studies have shown
196 Section II POSTUR£/BALANCE
that neurologically intact subjects use a com- to changes in support has been found in pa-
bination of feedforward, or anticipatory, and tients with Parkinson's disease (44). In this
feedback control mechanisms to scale forces study, normal controls and a group of Parkin-
needed for postural stability (46). Grading or son's patients were asked to maintain stance
scaling force output probably involves ante- balance in a variet)' of situations, including
rior portions of the cerebellum, since an in- standing on a flat surface, standing across a
ability to anticipate and scale forces appropri- narrow beam, and sitting on a stool with the
ate to changes in the size of a postural feet unsupported. Normal subjects can adapt
perturbation was found in patients with an- the muscles used for postural control in re-
terior cerebellar lesions (47). sponse to changing task demands (Fig. 9.9^1).
Postural responses that are too large are In contrast, Parkinson's patients were unable
called hypermetric, and are associated with to modify the complex movement strategy
excessive compensator^' body sway in the di- used in recovering balance while standing on
rection opposite the initial direction of insta- a flat surface to the beam or seated conditions,
bility. For example, patients with unilateral showing an inabilit^' to modif\' how they
cerebellar pathology aflfecting the anterior moved in response to changes in environmen-
lobe, can show hypermetric responses on the tal and task demands (Fig. 9.9B).
involved side of the body. This will often re-
sult in a fall in the direction opposite the af-
fected side due to excessive activity in the hy- Sitting
permetric extremity. By contrast, many

hemiparetic patients will fall in the direction
of weakness due to an inabilit}' to generate
sufficient force to counter the destabilizing
forces. Patients with hypermetric responses
may also show excessive oscillation of the cen-
ter of mass (4).
MOTOR ADAPTATION PROBLEMS

Normal postural control requires the Ankle strategy Hip strategy Trunk strategy
ability to adapt responses to changing tasks

and environmental demands. This flexibility'
requires the availabiht}' of multiple movement
strategies and the ability to select the appro-
priate strategy for the task and environment.
The inabilit)' to adapt movements to changing
task demands is a characteristic of many pa-
tients with neurological disorders.
become fixed in stereot\'pical
Patients
movement, showing a loss of
patterns of
movement flexibilit)' and adaptability. The B Complex Complex
strategy strategy
fixed movement synergies seen in the patient
with hemiparesis are an example of impair-
Figure 9.9. Normal and abnormal adaptation. A, Nor-
ments related to loss of flexibilit)' and adapt- mal adaptation of muscle activity in response to three
ability of movements. Infants with cerebral different postural tasks. B, By contrast, EMG patterns in
palsy who have trouble dissociating move- Parkinson's patients revealed a complex strategy of mus-
cle activity that did not adapt to changes in task de-
ments of their legs are constrained to kick
mands. (Adapted from Horak FB, Nashner LM, Nutt JG.
symmetrically because of these obligatory
Postural instability in Parkinson's disease: motor coor-
movement patterns in the legs (48). dination and sensory organization. Neurology Report
Inabilin,' to aciapt movement strategies 1988;! 2:55.)
Loss of Anticipatory Postural In this next section, we ts)' to summa-

rize some of the research on motor problems
Control
related to postural dyscontrol by diagnosis.
We have seen that an inability- to adapt Several warnings must be stated prior to be-
how we move in response to changing task ginning this section. Remember that even pa-
and environmental conditions can be a source tients with the same diagnosis can be ver)' dif-
of instability' in many neurologically impaired ferent. Thus, no two stroke patients look alike
patients. Another source of postural dyscon- because of the difference in t)'pe, location,
trol is the loss of anticipaton,- processes that and extent of neural lesion. Other factors such
activate postural adjustments in advance ot as age, premorbid status, and degree of com-
potentially destabilizing voluntan,' move- pensation, also have a profound impact on be-
ments. Anticipator}' postural acti\'it}' is heavily havior seen. Nonetheless, we provide a sum-
dependent on previous experience and learn- mar}' here of the kinds of problems one is
ing. likely to see in various t}'pes of neurologically
Inabilin,' to activate postural muscles in impaired patients, based on current postural

anticipation of voluntary' arm movements has control research. This information is also
been described in many neurologically im- summarized in Table 9.1.

paired patients, including stroke patients
(49), children with cerebral palsy (42), chil- CEREBRAL VASCULAR ACCIDENT-
dren with Down syndrome (41), and Parkin- SPASTIC HEMIPLEGL\
son's patients (50).
Postural control research has reported

Summarizing Motor Problems by the multiple kinds of motor problems in
Diagnosis stroke patients with hemiplegia. A number of
articles have re\'iewed the senson,' and motor
Until now, our discussions of postural who has had a
impairments in the patient
dyscontrol in the neurologically impaired pa- Weakness is frequently
stroke (9, 43, 51, 52).
tient have focused on presenting a wide \'ari-
a primar\' impairment. Abnormal muscle tone
et\' of motor problems leading to problems in
is common, ranging from complete flaccidity
stability' and orientation. You can see that the
to spastic h\pertonicit>'. Postural responses
range of problems is great, and this reflects the
are often delayed. In addition, dyssynergia, or
complexity' of problems that affect the central
a breakdown in the s}'nergistic organization
ner\ous svstem. In some cases, the same t\'pe
of muscles, is widely reported. This can in-
of problem can be found in patients with ven,'
clude proximal muscles firing in advance of
different diagnoses. For example, delayed on-
distal muscles, or in some patients, quite late
set of postural responses can be found in adult
in relationship to distal muscles. Loss of an-
hemiplegic patients, in children with Do\\'n of postural muscles dur-
ticipator}' activation
syndrome, and in elderly people with periph- movements common,
ing voluntan' is also as
eral neuropathies.
is an inabilit}' to modif}' and adapt movements
On the other hand, some problems ap-
to changing task demands. Neuromuscular
pear to be unique to a diagnosis. For example,
problems often produce secondar}' musculo-
a particular t\'pe of complex movement strat-
skeletal problems including shortening of the
eg\' used in all task conditions has been found
gastrocnemius/soleus muscle groups and loss
in Parkinson's patients. In most cases, what
of ankle range of motion.
we don't know about postural dyscontrol in
patients far outweighs what wc do know! This
is because this area of research is only about PARKINSON'S DISEASE
20 years old. New information is rapidly be-
coming available as scientists expand the study Motor problems such as bradykinesia
of postural control to more groups of pa- and produce many disabling muscu-
rigidit}'
tients. loskeletal problems, including loss of flexibil-

Table 9.1. Motor Problems by Diagnosis

Hemiplegic Cerebellar
Adult CVA Pediatric CP Adult Pediatric Parkinson's
Force Problems
Weakness
Abnormal tone
Hypermetric response
Timing Problems
Delayed onset
Dysynergia
Impaired adaptation
Impaired anticipatory control
Musculoskeletal problems
iU' and joint range of motion (6, 50, 53, 54). symptoms. For example, lesions of the mid-
Motor problems do not appear to be the refine vermis and fastigial nuclei affect primarily
sult of muscle weakness (45). Interestingly, trunk and upper extreinities; they thus can
despite the fact that bradykinesia or slowed manifest as truncal tremor, wide- based ataxic
voluntary' movement is common in Parkin- gait, and dysarthric speech. In contrast, le-
son's padents, onset latencies of automatic sions to the anterior lobe (vermis, and leg ar-
postural responses are reported to be normal eas) produce movement disturbances in the
(44). EMG found that Parkin-
studies have legs,which result in poor performance on the
son's patients use a complex pattern of muscle heel-shin test, dyssvnergia, and abnormal gait
activity involving muscles on both sides of the (21).
body when responding to instabilit)'. This Much of the research on postural con-
coactivation results in a rigid body and an in- trol in cerebellar patients has been with pa-
ability to adequately recover stability'. In ad- tients who have anterior lobe cerebellar de-
dition, patients appear to be unable to modify generation. Thus, findings from these studies
movement patterns in response to changing may not necessarily be found in patients with
task demands. Finally, anticipator^' postural lateral hemisphere lesions or vestibulocere-
activity' is disrupted in many Parkinson's pa- bellar lesions. Onset latencies are reported to
tients (50, 54). be normal in adult cerebellar patients, though
delayed in cerebral palsy children with cere-
CEREBELLAR DISORDERS bellar ataxia. An inabilit)' to scale postural ac-
tivity leading to hypermetric postural re-
Signs and symptoms associated with dis- sponses has also been reported in cerebellar
orders of the cerebellum were first described patients (47).
in the 1920s and 1930s (55). Principal deficits
associated with cerebellar disorders include: Sensory Disorders
(a) hypotonia or decreased resistance of the
limb to stretch; {b) ataxia, which is described As we mentioned earlier, effective pos-
as a delay in initiation of movement, or errors more than the abilin,' to
tural control requires
in the range, force, or metrics of movement, generate and apply forces for controlling tiie
often referred to as dysmetria or dyssynergia; body's position in space. In order to know
and (f) action or intention tremor, partic- when to apply restoring forces, the CNS must
of movement (21).
ularly at the termination have an accurate picture of where the body is
Lesions of the cerebellum tend to produce in space,and whether it is stationary or in mo-

disorders ipsilateral to the lesion. tion. As a result, normal postural control
In addition, lesions to the various parts requires the organization of sensory infor-
of the cerebellum have distinctive signs and mation from visual, somatosensory', and ves-
tibular systems about the body's position and body's position in space appears to be the
movement with respect to the environment, presence of an internal representation or body
and the coordination of sensor,' information schema, providing an accurate representation
with motor actions. or postural frame of reference. Figure 9.10
Disruptions of senson,' information pro- provides an example of this concept. Illus-
cessing may aftect postural control in a num- proposed stability

trated are tlie limits for the
ber of ways (4, 56). First, sensory problems task of independent stance on a firm, flat sur-
may prevent the development of accurate in- face in a neurologically intact adult with nor-
ternal models of the body for postural con- mal postural control (Fig. 9.10v4) (57). Fig-
trol. This can aftect a patient's ability to ac- ure 9.10B, however, depicts modified stability
curately determine the orientation of the limits for a hemiplegic patient who requires a
body with respect to and the environ-
gravity cane for support due to unilateral weakness.
ment. Second, disruption of central sensory Stabilit\' limits now exclude the left leg, which
mechanisms may affect a patient's ability' to cannot support the body due to weakness, but
adapt sensor\' inputs to changes in task and include the cane, which ser\'es as an addition
environmental demands. Third, sensory prob- to the base of support ( 5 )
lems can disrupt motor learning, affecting a It has been suggested that an accurate
patient's abilitv' to adapt to change. Finally, representation, or model, of stability limits is
loss of sensory information can impair the essential to the recover^' of postural control.
ability to andcipate instability, and thus cause This allows the development of new sensory
a compensatory modification in the strategies and motor strategies while the patient remains
a patient uses to sense instability' and move. within his/her new stability' limits, regardless
of the impairments resulting from the neu-
Misrepresentation of Stability rological lesion (5). Thus, the process of re-
Limits covering postural control after a lesion in-
cludes development of accurate new

the
An important part of interpredng senses representations of the body's capability as it
and coordinaung acdons that control the relates to postural control. Usually, the indi-
Figure 9.10. Conceptual model of stability limits
for stance postural control. A, Normal stability lim-

its in a neurologically intact adult, compared to B,
modified stability limits in a left hemiplegic patient,
excludes the weak leg, but includes the cane,
which is now part of the patient's base of support.
(Adapted from Shumway-Cook A, McCollum G.
Assessment and treatment of balance deficits. In:
Montgomery P, Motor control and

Connolly B, eds.
physical therapy. Hixson, IN: Chattanooga Croup,

1991:129.)
vidual's model of stability limits are consistent many types of cerebral vascular accidents,
with actual stability limits. In many patients, leaving the hemiplegic patient with hemisen-
however, percei\'ed stability limits may be in- soty losses that profoundly afreet posture and
consistent with actual stability limits, which movement control (60). In addition, many
have changed as a result of sensoty and motor such patients have disorders within the visual
limitations following a neurological lesion. system, including impaired ocular motility, vi-
A discrepancy between actual and inter- sual field defects,and impaired convergence
nal limits of stability can result in instability leading to flisional problems (40). Finally,
and potential falls (5). In the drawing in Fig- many patients with central neurological dis-
ure 9.1 OB, the patient's actual stability limits orders have associated problems in peripheral
exclude the hemiparetic leg, which is incapa- or central vestibular structures (61). Trau-
ble of generating sufficient force to control matic injuty to the head can result in several
the body in the upright position. If the pa- types of injur)' to the vestibular system that
tient's internal model of stability limits in- can complicate the recover^' of postural con-
cludes the affected leg as part of the base of trol (61).
support, the patient will have a tendency to In many patients, despite intact periph-
fall to that side, when the center of mass shifi:s eral sensation, lesions in a wide \'ariety of
to that side. central ner\'ous system structures can aiFect
On the other hand, inaccurate represen- the ability to adapt senses for postural
tations of the body with respect to postural control.
control can limit the patient's ability to use Sensoty adaptation problems can man-
new skills for postural control (5). For ex- ifest as an inflexible weighting of sensoty in-
ample, if the hemiplegic patient's internal formation for orientation, and/or an inability
model of stability limits doesn't change dur- to maintain balance in any environments
ing the course of recovety to reflect new abil- where sensoty information is inaccurately re-
ities to control the left leg for purposes of supporting self-motion. The inability to adapt
port, the patient may continue to stand and weighting of senses for orientation in different
walk asN'mmetrically. environments is somewhat analogous to the

Many patients with neurological disor- inflexibility in the use of movement strategies
ders fail to develop accurate models of their seen in many neurologically impaired pa-
body related to the dynamics of moving and tients.
sensing for postural control (5, 58, 59). In- Researchers examining the effect of
accurate internal models result in patterns of neurological injuty on patients' ability to
moving and sensing that seem inconsistent adapt sensoty information for postural control
with the patient's apparent abilities. This as- have primarily focused on the use of comput-
pect of disordered postural control is just be- erized force platforms in conjunction with
ginning to be explored, and much research is mo\'ing visual surrounds, first developed by
needed in this area. Nashner and colleagues (62-66). This ap-
proach, described in detail in the chapters on
Inability to Adapt Senses normal postural control, tests the ability of
the patient to maintain stance balance under
In the neurologically impaired patient, situations where sensoty information is lost or
inability to adapt how the senses are used for made inaccurate for postural control. A clas-
postural control can result from patholog)' sification scheme for identifying different
v\ithin from
individual sensoty systems or problems related to organizing sensoty infor-
damage to central sensoty structures impor- mation for stance postural control has been
tant in organizing sensoty' information for proposed based on patterns of normal and ab-
postural control (4, 5). normal sway in six sensoty conditions used
The loss of somatosensoty, kinesthetic, during dynamic posturography testing (4).
and proprioceptive information is common in Patterns of sway associated with different cat-
egories of senson' organization problems are n Normal control subjects

summarized in Figure 9.11.
Vestibular loss subjects
What is the effect of loss of a sensors-
input on postural control? It depends! Some
important factors include («) the availabilit\-
of other senses to detect position of the body

in space, (b) the availabilit\' of accurate ori-
entation cues in the environment, and {c) the
abilit\' to correcdy interpret
information for orientation
and
(4).
select sensor)-
.. « d] J] d]
As shown in Figure 9.12, patients with
loss of vestibular information for postural
control may be stable under most conditions
as long as alternative senson- information
from vision or the somatosensor\' systems is
available for orientation. In situations where
vision and somatosensory- inputs are reduced, Figure 9.1 2. A comparison of body sway in the six sen-
sory conditions in neurologically intact adults vs. patients
leaving mainly vestibular inputs {the last two
with loss of vestibular function. Results show that insta-
conditions in Fig. 9.12) for postural control, bility in patients with loss of vestibular function occurs
the patient mav experience a sudden tall (62). only in conditions where vision and somatosensory In-
Functionally, patients with this t\-pe of puts are not available for postural control (conditions 5
and (Adapted from Horak F, Nashner LM, Diener HC.
postural dyscontrol might perform normally 6).
Postural strategies associated with somatosensory and

on most of balance as long as they are
tests
vestibular loss. Exp Brain Res 1990:418.)
performed in a well-lit environment and on a
firm, flat surface. However, performance on
balance tasks under ideal senson,- conditions bathroom at night and negotiating a carpeted
will not necessarily predict the patient's like- surface in the dark.
lihood for falls when getting up to go to the How does disruption of somatosensory
information affect postural control.' One
might expect that a patient with sudden loss
of somatosensory- information could maintain
stability as long as alternative information
from vision and vestibular senses were avail-
able. A group of researchers examined this
question bv applying pressure cuffs to the an-
N N N N kles of normal subjects and inflating them un-

Adults (7-60) N N
Children (1-7) N N A A A A til cutaneous sensation in the feet and ankles
Abnormal was These neurologically intact sub-

lost (63).
jects were able to maintain balance on all six
Visually dependent N N/A A N N/A A
of the sensory conditions (Fig. 9.13) since
Surface dependent N N N A A A they always had an alternative sense a\'ailable
N A A for orientation.
Vestibular loss N N N
Overrehance on vision for postural con-
Sensory selection N N A A A A
problems trol is referred to as a visual dependence pat-
tern for sensory organization. In this pattern,
Figure 9.1 . A classification scheme for identifying dif- sway is abnormally increased in any condition
ferent problems related to organizing sensory informa- reduced or inaccurate (condi-
where \ision is
tion for stance postural control based on patterns of nor-
tions 2, 3, 5, and 6 in Fig. 9.11). We saw this
mal and abnormal sway in six sensory conditions used
during dynamic posturography testing. (N = normal type of pattern in ver)' young normal children,
sway; A= abnormal sway.) as noted earlier in the chapter on normal de-
Normal control subjects for postural control in environments where

Somatosensory loss subjects one or more orientation cues inaccurately re-
port the body's position in space is referred to
as a sensory selection problem (64, 65). Patients
with a sensory selection problem are often
able to maintain balance in em'ironments
where sensor)' information for postural con-
trol is consistent; however, they are unable to
maintain when
J among
stabilit\' there is
the senses (64, 65). Patients with a

sensory selection problem do not
incongruity
necessarily
show of overreliance on any one
a pattern
sense, but rather appear to be unable to cor-
recdy select an accurate orientation reference;
Figure 9.13. Body sway In the six sensory conditions in

therefore, they are unstable in any environ-
normal subjects before use of pressure cuffs at the ankle ment in which a sensor)' orientation reference
and after subsequent temporary loss of cutaneous sen- is not accurate. This is shown in Figure 9.1 1,
sation with use of pressure cuffs. Loss of somatosensory
where abnormal sway is seen in conditions 3,
inputs did not affect the ability of these neurologically
4, 5, and 6.
intact subjects to maintain balance, due to the availabil-
ity and the capacity to adapt re-

of alternative senses
Sensory selection problems have been
maining senses to the changing demands. (Adapted from reported in stroke patients (60), traumatic
Horak F, Nashner LM, Diener HC. Postural strategies as- brain injur)' patients (40), and in children
sociated with somatosensory and vestibular loss. Exp
with developmental disorders, including ce-
Brain Res 1990:418.)
rebral palsy (42), Down syndrome (41),
learning disabilities (65) and the deaf (64).
velopment of postural control. A visual de-
pendence pattern has also been reported in Sensorimotor Adaptation
other t)'pes of neurologically impaired pa-
tients, including those with specific types of Sensor)' problems can affect the ways in
positional vertigo due to vestibular pathology which we move for postural control (56). As
(66). we mentioned earlier, certain movement
Alternatively, some patients may dem- strategies for controlling the body's position
onstrate an inflexible use of somatosensory in- in space depend on certain senses more than
puts for postural control, becoming unstable others. When the sense needed for controlling
in condidons where surface inputs do not al- that movement is not available, the abilit)' of
low patients to establish and maintain a ver- the individual to use that movement strateg)'
tical orientation (56). This t)'pe of pattern is for postural control is lost. For example, we
referred to as a surface-dependent pattern, and mentioned earlier that somatosensor)' inputs
is seen in patients who show excessive are very important when the ankle strateg)' is
amounts of body sway in conditions 4, 5, and used to compensate for support surface move-
6 (Fig. 9.11). Thus, when standing on a comments. Alternatively, visual and vestibular
pliant surface, like sand or thick carpet, or on senses appear to be more important when a
a tilted surface, like a ramp, or on a moving hip postural movement strateg)' is used to
surface, like a boat, the position of the ankle control balance in this situation.
joint and other somatosensory' and proprio- During the experiments in which pres-
ceptive information from the feet and legs sure cuffs were applied to neurologically in-
does not correlate well with the orientation of tact subjects, thereby reducing the availabilit)'
the rest of the body (56). An overreliance on of cutaneous inputs for orientation, subjects
somatosensor}' inputs for postural control in were able to maintain balance under the six
these environments will result in instability'. sensor)' conditions. However, in the absence
Inability' to appropriately select a sense of somatosensor)' inputs for orientation, sub-
jects tended to alter how they moved when dent Increase in tonic stretch reflexes (muscle
controlling balance. Instead of using an ankle tone) with exaggerated tendon jerks, resulting
strateg)' to control body sway, subjects tended from changes In the threshold of the stretch
reflex. The exact contribution of abnormalities
to increase the use of hip movements. This led
of muscle tone to functional deficits in posture
researchers to suggest that changes in the
are not well understood.
availability' of sensory inputs for orientation
7. Other neuromuscular factors contributing to
result in a change in how people move to con-
postural dyscontrol Include a wide range of
trol balance (56). abnormalities leading to an inability to orga-
Similarly, patients who have lost visual nize multiple muscles into coordinated pos-
and/or vestibular inputs for postural control tural movement synergies.
are ofiien unable to use a hip postural move- 8. Disruptions of sensory Information can affect
ment strateg)' and are constrained to move postural control In the following ways: (a) sen-
only at the ankles (56). sory problems can prevent the development
of accurate internal models of the body for
postural control, affecting a patient's ability to
Summary accurately determine the orientation of the
body with respect to gravity and the environ-
1 An enormous range of problems can contrib-
ment; (6) disruption of central sensory mech-
ute to postural dyscontrol In the neurologl-
anisms can affect a patient's ability to adapt
cally impaired patient. This Includes positive
sensory inputs to changes In and environ-
task
and negative which occur as a direct
signs,
mental demands; (c) sensory problems can
problems that occur In-
result of the lesion, or
disrupt motor learning, affecting a patient's
directly or compensatory to the lesion. As a
ability to adapt to change; (cO loss of sensory
result, understanding posture and movement
information can impair a patient's ability to
behaviors seen In such patients Is a compli-
anticipate instability, modifying the way he or
cated process. she senses and moves to prevent disruptions
2. A systems perspective to postural dyscontrol to postural control.
focuses on identifying the constraints or Im-
pairments each of the systems essential to
in
controlling body posture. Impairments are de-

References
fined as limitations within the Individual
which restrict sensory and movement strate- 1. Walshe F. Contribudons of John Hughlings
gies for postural control. Impairments can be Jackson to neurology: a brief introduction to
musculoskeletal, neuromuscular, sensory, his teachings. Arch Neurol 1961;5:119-131.
perceptual, or cognitive. 2. Gordon J. Assumptions underlying physical
3. In the patient with upper motor neuron le- therapy inter\'ention: theoretical and histori-
sions, musculoskeletal disorders develop most cal perspectives. In: Carr J, Shepherd R, Gor-
often secondary to the neurological lesion. don J, Gentile AM, Held J, cds. Movement
Yet, musculoskeletal problems can be a major science foundations for physical therapy in re-
limitation to normal postural function in the habilitation. Rockville: Aspen Publications,
neurologlcally Impaired patient. 1987:1-30.
4. Neuromuscular limitations encompass a di- 3. Katz R, Rymer Z. Spastic hypertonia: mech-
verse group of problems that represent a major anisms and measurement. Arch Phys Med
constraint on postural control In the patient Rehabil 1989;70:144-155.
with neurological dysfunction. 4. Shumway-Cook A, Horak F. Balance reha-
5. Weakness, or the inability to generate tension, bilitation in the neurologic patient: course
Is a major Impairment of function In many pa- syllabus. Seattle: NERA, 1992.

tients with upper motor neuron lesions. 5. Shumway-Cook A, McCoUum G. Assessment
6. Abnormalities of muscle tone are found in and treatment of balance deficits. In: Mont-
many patients with upper motor neuron le- gomery P, Connolly B, eds. Motor control
sions. The spectrum of muscle tone abnor- and physical therapy. Hixson, TN: Chatta-
malities is broad, ranging from flaccldlty In the nooga Group, 1991:123-137.
acute stroke patient, to rigidity In the Parkin- 6. Schenkman M. Interrelationships of neuro-
son patient. Spasticity Is defined as a motor logical and mechanical factors in balance con-
disorder characterized by a velocity-depen- trol. In: Duncan P, ed. Balance: proceedings
of the APTA Forum. Alexandria, VA: APTA, plegia: a neurophysiological approach. Hag-
1990:29-11. erstown, MD: Harper & Row, 1970.
7. Shiverick. D. Loss of gastrocnemius length in 20. Knutson E. Richards C. Different tii-pes of
hemiplegic patients. Neurology Report 1990; disturbed motor control in gait of hemiplegic
3:4-6. patients. Brain 1979;102:405-i30.
8. Woo SLV, Matthews JV, Akerson WH, et al. 21. Ghez C. The cerebellum. In: Kandel ER,
Connective tissue response to immobility'. Ar- Schwartz JH, Jessell TM, eds. Principles of
Rheum 1975;18:257-264.
thritis neural science. NY: Elsevier, 1991:627-
9. Duncan P, Badke MB. Determinants of ab- 646.
normal motor control. In: Duncan P, Badke 22. Horak FB. Assumptions underlying motor
MB, eds. Stroke rehabilitation: the recovery control for neurologic rehabilitation. In:
of motor control. Chicago: Year Book Med- Contemporary management of motor con-
ical Publishers, 1987:135-159. trol problems. Proceedings of the II Step
10. Perr\' J, Newsam C. Function of the ham- Conference. Alexandria, VA: APTA,
strings in cerebral palsy. In: Sussman M, ed. 1991:11-27.
The diplegic child. Rosemont, IL: American 23. Shea A. Motor attainments in Down's syn-
Academy of Orthopedic Surgeons, drome. In: Contemporan,' management of
1992:299-307. motor control problems. Proceedings of the
1 1 . Smidt GL, Rogers MW. Factors contributing II Step Conference. Alexandria, VA: APTA
to the regulation and clinical assessment of 1991:225-236.

muscular strength. Phys Ther 1982; 24. Lance JW. Symposium synopsis. In: Feldman
62:1283-1290. RG, Young RR, Koella WP, eds. Spasticity':
12. Buchner DM, DeLateur BJ. The importance disordered motor control. Chicago: Year
of skeletal muscle strength to physical fiinc- Book Medical Publishers, 1980:485.
tion in older adults. Annals of Behavioral 25. Berger W, Horstmann GA, Dietz VL. Ten-
Medicine 1991;13:1-12. sion development and muscle activation in
13. Amundsen LR. Isometric muscle strength the leg during gait in spastic hemiparesis: the
testing with ii.\ed-load cells. In: Amundsen independence of muscle hypertonia and ex-
LR, ed. Muscle strength testing instrumented aggerated stretch reflexes. J Neurol Neuro-
and non-instrumented systems. New York: surg Psychiatry 1984;47:1029-1033.
Churchill Li\ingstone, 1990:89-122. 26. Katz RT, Rymer WZ. Spastic hypertonia:
14. Rogers MM. Musculoskeletal considerations mechanisms and measurement. Arch Phys
in production and control of movement. In: Med Rehabil 1989;70:144-155.
Montgomery P, Connolly BH, eds. Motor 27. Burke D. Critical examination of the case for
control and physical therapy. Hixson, TN: or against ftisimotor involvement in disorders
Chattanooga Group, 1991:69-82. of muscle tone. In: Desmedt JE, ed. Motor
15. Edstrom L. Selective changes in the sizes of control mechanisms in health and disease.
red and white muscle fibers in upper motor NY: Raven Press, 1983:133-150.
lesions and parkinsonism. J Neurol Sci 1970; 28. Davies P. Steps to follow. New York:
11:537-550. Springer- Verlag, 1985.
16. Edstrom L, Grimby L, Hannerz J. Correla- 29. Gowland C, deBruin H, Basmajian J, Plews
tion between recruitment order of motor N, Burcea I. Agonist and antagonist activity
units and muscle atrophy patterns of upper during voluntary upper-limb movement in
motor neuron lesions: significance of spasti- patients with stroke. Phys Ther 1992;
city. Experientia 1973;29:560-561. 72:624-633.
17. Mayer RF, Young JL. The effects of hemiple- 30. Dietz V, Trippel M, Berger W. Reflex activity'
gia with spasticity. In: Feldman RG, Young and muscle tone during elbow movements in
RR, Koella WP, eds. Spasticity': disordered patients with spastic paresis. Ann Neurol
motor control. Chicago: Year Book Medical 1991;6:767-779.
Publishers,1980:133-146. 31. Lacquaniti F. Quantitative assessment of so-
18. Bobath B. Adult hemiplegia: evaluation and matic muscle tone. Funct Neurol 1990;
treatment. London: William Heinemann, 5:209-215.
1978. 32. Bohannon RW, Andrews AW. Correlation of
19. Brunnstron S. Movement therapy in hemi- knee extensor muscle torque and spasticiri'
\\'ith gait seed in patients with stroke. Arch In:Brandt T, Paulus lO, Bles W, et al., eds.
Phys JVIed Rehabil 1990;71:330-333. Disorders of posture and gait. Stuttgart:
33. Sahrmann SA, Norton BS. The relationship George Thieme Verlag, 1990:370-373.
of \oluntary movement to spasticity in tiic 48. Kamm K, Thelen E, Jensen, ]. A dynamical
upper motoneuron syndrome. Ann Neurol systems approach to motor development. In:
1977;2:460-465. Rothstein J, ed. Movement science. Alexan-
34. Tang A, Rymer WZ. Abnormal force-EMG dria,VA: APTA, 1991:11-23.

relations in paretic limbs of hemiparetic hu- 49. Horak FB, Anderson M, Esselman P, Lynch
man subjects. Neurol Neurosurg Psychiatry K. The effects of mo\'ement velocity, mass
1981;44:690-698. displaced and task certainty on associated
35. McLellan DL. Co-contraction and stretch re- postural adjustments made by normal and
flex in spasticity during treatment with baclo- J Neurol Neurosurg
hemiplegic individuals.
fen. Neurol Neurosurg Psychiatry' 1973; 1984;47:1020-1028.
Psychiatr>'
40:30-38. 50. Rogers MW. Control of posture and balance

36. Whitley DA, Sahrmann SA, Norton BJ. Pat- during voluntar\' movements in Parkinson's
terns of muscle activity' in the hemiplegic up- disease. In: Duncan P, ed. Balance: proceed-
per extremity. PhysTher 1982;62:641-651. ings of the APTA Forum. Alexandria, VA:
37. Maki B, HoUiday PJ, Topper AK. Fear of tail- APTA, 1990:79-86.
ing and postural performance in the elderly. 51. Badke MB, DiFabio RP. Balance deficits in
JGerontol 1991;46:M123-M131. patients with hemiplegia: considerations for

38. Lewis C, Phillippi L. Postural changes with assessment and treatment. In: Duncan P, ed.
age and soft tissue treatment. Phys Ther Fo- Balance: proceedings of the APTA Forum.
rum 1993;9:4-6. Alexandria,VA: APTA, 1990:73-78.
39. Badke M, Duncan P. Patterns of rapid motor 52. Duncan PW. Stroke: physical therapy assess-
responses during postural adjustments when ment and treatment. In: Contemporary man-
standing in healthy subjects and hemiplegic agement of motor control problems. Pro-
patients. Phys Ther 1983;63:13-20. ceedings of the II Step Conference.
40. Shumway-Cook A, Olmscheid R. A systems Alexandria, VA:APTA, 1991:209-217.
analysis of postural dyscontrol in traumati- 53. Cote L, Crutcher MD. The basal ganglia. In:
cally brain-injured patients. J Head Trauma Kandel ER, Schwartz JH, Jessell TM, eds.
Rehabil 1990;5:51-62. Principles of neural science. NY: Elsevier,
41. Shumway-Cook A, Woollacott M. Postural 1991:645-659.

control in the Down's syndrome child. Phys 54. Rogers MW. Motor control problems in Par-
Ther 1985;9:211-235. kinson's disease. In: Contemporary manage-

42. Nashner LM, Shumway-Cook A, Marin O. ment of motor control problems. Proceed-
Stance posture control in select groups of ings of the II Step Conference. Alexandria,
children with cerebral palsy: deficits in sen- VA: APTA, 1991:195-208.
sory organization and muscular coordination. 55. Holmes G. The cerebellum of man. Brain
Exp Brain Res 1983;49:393-^09. 1939;62:1-30.
43. Badke MB, DiFabio RP, Duncan PW. Lat- 56. Horak FB, Shupert CL. Role of die vestibular
erality of rapid motor responses in hemiplegic system in postural control. In: Herdman SJ,
subjects during postural adjustments in ed. Vestibular rehabilitation. Philadelphia:FA
standing. PhysTher 1983;63:13-20. Davis, 1994:22-16.

44. Horak FB, Nashner LM, Nutt JG. Postural 57. McColium G, Leen T. Form and exploration
instability in Parkinson's disease: motor co- of mechanical stability limits in erect stance.
ordination and sensory organization. Neu- J Motor Behav 1989;21:225-236.
rology Report 1988;12:54-55. 58. Shumway-Cook A, Horak, F. Vestibular re-
45. Martin JP. The basal ganglia and posture. habilitation: an exercise approach to manag-
London: Pitman, 1967. ing symptoms of vestibular dysfunction. Sem-
46. Horak FB, Diener HC, Nashner LM. Influ- inars in Hearing 1989;10:196-209.
ence of central set on human postural re- 59. Shumway-Cook A, Horak F. Rehabilitation
sponses. J Neurophysiol 1989;62:841-853. strategies for patients with vestibular deficits.
47. Horak FB. Comparison of cerebellar and ves- Neurol CHn 1990;8:441-457.
on scaling of postural responses.
tibular loss 60. DiFabio R, Badke MB. Relationship of sen-
sory organization to balance function in pa- and vestibular loss. Exp Brain Res 1990;
tients with hemiplegia. Phys Ther 1990; 82:167-177.
70:543-552. 64. Shumway-Cook A, Horak FB, Black FO.
61. Shumway-Cook A. Vestibular rehabilitation Critical examination of vestibular function in
in traumatic brain injury. In: Herdman S, ed. motor-impaired learning disabled children.
Vestibular rehabilitation. Philadelphia: FA Int I Fed Otorhinolaryngol 1988;14:21-30.
Davis, 1994:347-359. 65. Horak FB, Shumway-Cook A, Crowe T,

62. Black FO, Shupert C, Horak FB, Nashner Black FO. Vestibular fijnction and motor
LM. Abnormal postural control associated proficiency in children with hearing impair-
with peripheral vestibular disorders. In: Pom- ments and in learning disabled children with
peiano O, Allum J, eds. Vestibulo-spinal con- motor impairments. Dev Med Child Neurol
trol of posture and movement. Progress in 1988;30:64-79.
brain research. Amersterdam: Elsevier Sci- 66. Black FO, Nashner LM. Vestibulo-spinal
ence Publishers 1988;76:263-275. control differs in patients with reduced versus
63. Horak FB, Nashner LM, Diener HC. Pos- distorted vestibular hmction. Acta Otolarv'n-
tural strategies associated with somatosensory gol (Stockliolm) Suppl. 1984;406:110-114.
Chapter 10
Assessment and Treatment of

Patients with Postural
Disorders
Interpretation of Assessment
Introduction
feU^H
Assessment Treatment ^^S
Safety— First Concern Treating at Impairment Level %^
Functional Assessment Cognitive Impairments
Up and Co Test Musculoskeletal Impairments
Functional Reach Test Neuromuscular Impairments
Balance and Mobility Scale Strength
Functional Balance Scale Muscle Tone
Limitations of Functional Assessment Sensory Impairments
Strategy Assessment Perceptual Impairments
Motor Strategies Treating at Strategy level
Alignment in Sitting and Standing Alignment
Movement Strategies Movement Strategies
Sensory Strategies Developing a Coordinated Ankle
Systems Assessment; Identifying Strategy
Impairments Treatment of Timing Problems
Cognitive Systems Treatment of Scaling Problems
Level of Consciousness Developing a Coordinated Hip
Mental Status Strategy
Musculoskeletal System Developing a Coordinated Step
Range of Motion Strategy
Flexibility Sensory Strategies
Neuromuscular System Perceived Limits of Stability
Strength Treating at the Functional Task Level
Muscle Tone Summary
Coordination
Sensory Systems
Assessing Perceptions Relevant to
Postural Control
Stability Limits
Motion Perception ^^^^^
Introduction a theory of motor control. We referred to this

framework as a task-oriented approach. We
This chapter ciiscusses a task-oriented now combine this approach with our knowl-
approach to assessing and treating postural edge of normal and abnormal postural con-
disorders in the patient with neurological dys- trol, and show how it is applied to the clinical
ilinction. In Chapter 5, we introduced a con- management of postural disorders. It is im-

ceptual framework for clinical practice, which portant to remember that the development of
incorporateci four key elements: the clinical clinical methods based on a systems theory of
decision- making process, hypothesis-oriented motor control is just beginning. As systems-

clinical practice, a model of disablement, and based research provides us with an increased
207
understanding of normal and abnormal pos- jective documentation about change in fiinc-
tural control, new methods for assessing and tional status. There are a number of tests
treating postural disorders will emerge. available to measure functional skills related
to postural control. In addition to the fijnc-
Assessment tional assessment, it is good to gather infor-
mation on number and types of falls and near
A task-oriented approach assesses pos- falls, and to include this in a balance and falls
tural control on three levels: (a) the func- history (see Appendix A).
tional skills requiring posture control, (
b) the
sensor\' and motor strategies used to maintain GET UP AND GO TEST
posture in various contexts and tasks, and (c)
the underlying sensor)', motor, and cognitive The Get Up and Go test (
I ) was devel-
impairments that constrain posture control. oped as a quick screening tool for detecting
The information gained through assessment balance problems in elderly patients. The test
isused to develop a comprehensive list of requires that subjects standup from a chair,
problems, establish short- and long-term walk 3 meters, turn around, and return. Per-
goals, and formulate a plan of care for retrain- formance is scored according to the following
ing posture control. A
thorough assessment scale: 1 normal; 2 ver)' slightly abnormal; 3
must include a review of the patient's medical mildly abnormal; 4 moderately abnormal; 5
and social history, as well as a review of cur- severely abnormal. An increased risk for falls
rent symptoms and concerns. was found among older adults who scored 3
or higher on this test.
Safety —First Concern The Up and Go test modifies the orig-

inal test by adding a timing component to
During the course of evaluating postural performance (2). Neurologically intact adults
control, patients will be asked to perform a who are independent in balance and mobility'
number of tasks that will likely destabilize skills are able to perform the test in less than
them. Safet\' is of paramount importance. All 10 seconds. This test correlates well to fijnc-
patients should wear an ambulation belt dur- measured by the Barthel In-
tional capacity as
ing testing, and be closely guarded at all dex (3). Adults who took greater than 30 sec-
times. In determining what tasks and activities onds to complete the test were dependent in
cause loss of balance, the patient must be almost activities of daily living, and mobilit\'
lowed to experience instability'. However, the skills.
therapist should protect the patient at all
times to prevent a fall. FUNCTIONAL REACH TEST
Functional Assessment The Functional Reach Test (4) is an-

other single item test developed as a quick
A task-oriented approach to evaluating screen for balance problems in older adults.
postural control begins with a fijnctional as- As shown in Figure lO.Iyl, subjects stand
sessment to determine how well a patient can with feet shoulder distance apart, and with the
perform a variety' of skills that depend on pos- arm raised to 90° flexion. Without moving
tural control. A functional assessment can their feet, subjects reach as far forward as they
provide the clinician with information on the can while still maintaining their balance (Fig.
patient's level of performance compared to 10. IB). The distance reached is measured and
standards established with normal subjects. compared to age-related norms, shown in Ta-
Results can indicate the need for therapy, ble 1 0.1. The Functional Reach Test has es-
serve as a baseline level of performance, and tablished inter-rater reliabilit}', and is shown
when repeated at regular intervals, can pro- to be highly predictive of falls among older
vide both the therapist and patient with ob- adults (4).
Chapter Ten Assessment .\kd Treatment of Patients with Postural Disorders 209
Figure 10.1 The Functional Reach Test. A, Subjects begin by standing with feet shoulder distance apart, arm raised
to 90° flexion, and reach as far forward as they can while stillmaintaining their balance.
Table 10.1. Functional Reach Norms FUNCTIONAL BALANCE SCALE - &er^

Men Women
Norms (in inches) (in inches)
The Functional Balance Scale was de-
20-40 yrs 16.7 + 1.9 14.6 + 2.2 veloped by KatJiy Berg, a Canadian physical
41-69 14.9 + 2.2 13.8 + 2.2 therapist (7). This test uses 14 difterent items,
70-87 13.2 + 1.6 10.5 + 3.5 which are rated to 4. The test is shown in
Appendix A as part of a comprehensive bal-
'From Duncan PW, Weiner DK, Chandler ), Studenski S. Func-
new clinical measureof balance. Gerontol 1990:
tional reach: a
ance assessment form. The test is reported to
I
45:M195. have good test-retest and inter-rater reliabil-
it)'; however, to date, there are no norms pub-

PERFORMANCE ORIENTED ,
•
lished for this test.

"^'^tH*
MOBILITY' ASSESSMENT - '
Mar\' Tinetti, a physician researcher at LIMITATIONS OF FUNCTIONAL

Yale University', has published a test to screen
ASSESSMENT
for balance and mobilit\' skills in older adults
and to determine the likelihood for falls (5, As noted in Chapter 5, fianctional as-
6). Table 10.2 presents TLnetti's balance and sessments have a number of limitations. These
mobilin,- scale, which rates performance on a include the inabilit>' to («) assess a patient's
three-point scale. performance of tasks under changing en\iron-

210 Section II POSTURE /'B.\L.\NCE
Table 10.2. Balance and Mobility Assessment'
I. Balance Te>t< II. Gait Tests

Initial instructions: Subject is seated in a hard, armless Initial instructions: Subject stands with the examiner,
chair. The following maneuvers are tested. walks down hallway or across room, tlrst at usual
pace, then back at rapid, but safe pace (usual
1. Sitting balance
walking aids)
Leans or slides in chair =
Steady, safe = 1 1 0. Initiation of gait (immediately after told to "go")
Any hesitancy or multiple attempts to start =
2. Arises
No hesitancy = 1
Unable without help =
Able, uses arms to help = 1 1 1 . Step length and height
Able without using arms = 2 a. Right swing foot
Does not pass left stance foot with step =
3. Attempts to arise
Passes left stance foot = 1
Unable without help =
Right foot does not clear tloor completely
Able, requires >I attempt = 1
with step =
Able to rise, 1 attempt = 2
Right foot completely clears tloor = 1
4. Immediate standing balance (first 5 seconds) b. Left swing foot

Unsteady (staggers, moves feet, trunk sway) = Does not pass right stance foot with step =
Steads , but uses walker or other support = 1 Passes right stance foot = 1
Steady w ithout walker or other support = 2 Left foot does not clear floor completely with
step =
5. Standing balance
Left foot completely clears tloor = 1
Unsteady =
Steady but wide stance (medial heels > 4 inches 12. Step symmetry
and uses cane or other suf)port =
apart) 1 Right and left step length not equal (estimate) =
Narrow stance without support = 2 Right arxl left step appear equal = 1
6. Nudged (subject at maximum position with feet as 13. Step continuity

close together as possible, examiner pushes Stopping or discontinuity between steps =
lightly on subject's stemum with palm of hand 3 Steps appear continuous = 1
times)
14. Path (estimated in relation to tloor tiles, 12-inch
Begins to fall =
diameter: observe excursion of 1 foot over about
Staggers, grabs, catches self = 1
10 ft of the course)
Steady = 2
Marked deviation =
7. Eyes closed (at maximum position no. 6) Mild'moderate deviation or uses walking aid = 1
Unsteady = Straight without walking aid = 2

Stead> = 1
15. Tfunit
8. Turning 360 degrees Marked sway or uses walking aid =
Continuous steps = No sway, but flexion of knees or back pain or
Discontinuous steps = 1 spreads arms out while walking = 1
Unsteady steps (grabs, staggers) = 2 No sway, no flexion, no use of arms, and no use
if walking aid = 2
9. Sitting down
Unsafe (misjudged distance, falls into chair) = 1 6. Walking time
Uses arms or not a smooth motion = 1 Heel apart =
Safe,smooth motion = 2 Heels almost touching while walking = 1
Balance score: /1
Cait score: /1
Balance and gait score: /28
•"Fram Tineni, M. PenormanceKxiented assessment of mobility problems in elderly patients. JAGS 1 986:34:1 1 9-1 26.
Chapter Ten .ASSESSMENT .\XD Tre.\tment OF Patients wtth Postltl\l Disorders 211
mental contexts, {b) determine the quality of

movement used, and ( c\ idenal\" sf>ecific neu-
ronal or musculoskeletal subsystems within
the body responsible for a decline in perfor
mance.
Strategy- Assessment
The next level of assessment examines

the motor and sensor\' strategies used to con-
trol the body's position in space under a \a
riet>" of conditions.
MOTOR STRATEGIES
Assessment of motor strategies for pos
examines both the alignment o:
tural control
body segments during unperturbed sitting
and standing and the patient's abilit\" to gen-
erate multi-joint movements, or strategies,
which etFectiN ely control morion of the center
of mass relative to the base of suppon 8-1 1 ( 1.
Alignment in Sitting and Standing
The patient's alignment in sitting and

standing is obser\ ed. Is the patient vertical? Is Figure 1 0.2 The use of a static t'orceplate can be helpful
weight s\Tnmetrically distributed right to left, \s hen quantif>'ing static alignntient changes In standing.
and forward and backward? A plumb line in

conjunction with a grid can be used to quan-
tify- changes in alignment at the head, shoul- swav, and anticipaton.^ to a potentially desta-
ders, trunk, pehis, hips, knees, and ankles. In bilizing upper extremitN' movement 10 ( ).
addition, the width of the patient's base of Moxements used to control self-initi-
support up>on standing can be measured and ated bodv sway are obser\ed while the patient
recorded using a tapte to measure the distance volimtarily shifts the weight forward, then
t>etw"een the medial malleoli i or alternatively, backwards, then side to side. The patient is
the metatarsal heads k tested both in sitting and in standing. Figure

Altemati\e ways to quantih' placement 10.4 illustrates the range of movement pat-
of the center of mass in the standing position terns seen in a seated neurologically intact in-
include the use of static force plates to mea- di\idual as he/she shifts the trunk fiurher and
sure placement of the center of pressure Fig ( ftirther laterally. As weight is transferred to
10.2 ), or the use of two standard scales to de- one of the body, the trunk begins to
side
termine if there is weight discrepancy" bersxeen cime towards the unweighted side, resulting
the two sides iFig 10.31. in elongation of the w eightbearing side and
shortening of the trunk on the unweighted
Movement Strategies side (Fig. 10.4u4). .\s weight continues to be
shifted laterally, maintaining stabilitv- requires
Movement strategies are examined un- the patient to abduct the arm and leg in order
der three different task conditions: self-initi- to keep the trunk mass within the base of sup-
ated s\vay, in response to externally induced port (Fig. 10.45). Finally, when the center of
response to an external perturbation, or push

(10, 12, 13). Holding the patient about the
hips, the therapist displaces the patient for-
ward, backward, right, and then left. Figure
10.6yl illustrates the use of an ankle strateg}'
used to recover from a small backward dis-
placement.
A larger displacement by the therapist
usually results in a greater amount of hip and
trunk motion, that is, a hip strategv', as the
subject continues totr\' to keep the center of
mass within the base of support and not take

a step (Fig. 10.65). Finally, if the therapist
enough, and the cen-
displaces the subject far
ter of body mass moves outside the base of
support, the subject will take a step to avoid
a fall (Fig. 10.6C) (10).
The most common approach to evalu-
ating multi-joint dyscoordination within task-
specific movement strategies is through ob-
servation and subjective analysis. For example,
the clinician may note that during recovery of
stance balance the patient demonstrates ex-
cessive flexion of the knees, or asymmetric
movements in the lower extremities, or ex-
cessive fle.xion or rotation of the trunk. How-
e\er, the underlying nature of the dyscoordi-
nation, that is, specific timing and or
amplitude errors in synergistic muscles re-
Figure 10.3 Two standard scales can also be used to
sponding to instabiliu', cannot be determined
quantify static asymmetric standing alignment.
without using technical apparatus such as
electromyography (8).
mass of the trunk exceeds the base of support, Finally, movement strategies used to
the patient must protecti\ely extend the arm minimize instability' in anticipation of poten-
to prevent a fall (Fig. 10.4C). tially destabilizing movements can be assessed
Figure 10.5 illustrates two t\'pes of by asking a patient to lift a hea\y object as
movement strategies being used to control rapidly as possible. If the patient is standing,
self- initiated sway in standing. Two patients amount of backward sway of the whole
a small
have been asked to sway forward as far as they body should precede the lift, indicating the
can without taking a step. Patient A (Fig. presence of anticipator\' postural adjustments
10.5A) is swaying fon\'ard primarily about the in the legs. If the patient is sitting indepen-
ankles, using what has been referred to as an dently, one would expect to see backward
ankle strateg\' to control center of mass mo- sway in the trunk, if anticipator^' postural ad-
tion. In contrast. Patient B (Fig. 10.55) is justments are used. Fon\ard instabilit}' is
moving primarily the trunk and hips (a hip found in patients who do not make anticipa-
strateg)'), which minimizes fon\ard motion of tor)' adjustments (10).
the center of mass.
Movement strategies used to recover SENSORY STRATEGIES
from a perturbation are also assessed. Figure
10.6 illustrates one approach to assessing The Clinical Test for Sensor*' Interac-
movement patterns used to control swav in tion in Balance (CTSIB) is one method that
Chapter Ten ASSESSMENT AND TREATMENT of PATIENTS with PoSTUBAL DiSORJDERS 213
Figure 10.4 Controlling self-initiated trunk movements in sitting. A, Small movements produce adjustments at the
head and trunk. B, Larger movements require counterbalancing with the arms and legs. C, When the line of gravity
for the head and trunk exceeds the base of support, the arm reaches out to prevent a fall.
has been proposed for clinically assessing the

influence of sensor)' interaction on postural
stability' in the standing position ( 14, 15). The
technique uses 24" by 24" piece of medium-
a
densiU' Temper foam in conjunction with a

modified Japanese lantern. A large Japanese
lantern is cut down the back and attached to
a headband. Vertical stripes are placed inside
the lantern, and the top and bottom of the
lantern are covered with white paper (Fig.
10.7).
The method is based on concepts de-
veloped by Nashner (16), and requires the
subject to maintain standing balance for 30
seconds under six different sensory conditions
that either eliminate input or produce inac-
curate visual and surface orientation inputs.
These six conditions are shown in Figtire
10.8.
Patients are tested in the feet together
position, with hands placed on the hips. Us-
ing condition 1 as a baseline reference, the
therapist observes the patient for changes in
the amount and direction of sway o\'er the
Figure 10.5 Controlling selt-inituted bwa) in stance.
subsequent five condidons. If the patient is
Shown are two types of movement strategies being used
to control self-initiated sway in standing. A, the ankle, unable to stand for 30 seconds, a second trial
and B, the hip. is given (15).

Figure 10.6 Movement strategies used to recover from an external perturbation to balance. A, An ankle strategy is
used to recover from a small displacement at the hips. B, A larger displacement produces a hip strategy. C, Movement
of the COM outside the base of support requires a step to recovery stability.
Neurologically intact young adults are

able to maintain balance for30 seconds on all
six conditions with minimal amounts of body
sway. In conditions 5 and 6, normal adults

sway on the a\erage 40% more than in con-
dition 1 (16).
Results from a number of research stuci-
ies that have used either a moving platform or
the CTSIB suggest the following scoring cri-
teria ( 17-20). A single fall, regardless of the

condition, is not considered abnormal. How-
ever, two or more falls are indicative of diffi-
culties adapting senson,' information for pos-

tural control.
A
proposed model for interpreting re-
sults summarized in Figure 10.9. This
is
model is in the process of being validated. Pa-

tients who show increased amounts of sway
or lose balance on conditions 2, 3, and 6 are
thought to be visually dependent, that is,
highly dependent on vision for postural con-
Figure 10.7 A modified Japanese lantern is used to
change the accuracy of visual input for postural orien- trol. Patients who have problems on condi-
tation. tions 4, 5, and 6 are thought to be smface-

Chapter Ten ASSESSMENT AND TREATMENT OF Patients with Postural Disorders 215
VISUAL CONDITIONS
NORMAL BLINDFOLD DOME
naJa
îrr/mMA îC'n^-
•'iU(J
Figure 1 0.8 Six sensory conditions used to examine postural orientation under altered sensory contexts. The approach
tests the ability to adapt how senses are used to maintain orientation. (From Shumway-Cook A, Horak F. Assessing the
influence of sensory interaction on balance. Phys Ther 1986;66:1S49.)
dependent, that is, dependent primarily on altering the availability of incoming sensory
somatosensory information from the feet in information for postural orientation, addi-
contact with the surface, for postural control tional factors can affect performance in this
(9,10). condition. Standing on foam changes the dy-

However, it is important to remember namics of force production with respect to the
the following caution when interpreting re- surface, and this may be a significant factor
sults showing increased sway on a compliant affecting performance in this condidon. There
surface. While we suppose that the primary has been no research examining the dynamics
effect of standing on a foam surface relates to of standing on foam, thus clinicians should be
Q e Q Q © o
Patterns
Visually Dependent N N/A A N N/A A

Surface Dependent N N N A A A
Vestibular Loss N N N N A A
Sensory Selection N N A A A A
N= Body sway within normal limits
A= Body sway abnormal
Figure 10.9 A proposed model for interpreting the CTSIB test based on information gained through dynamic postu-
rography testing.
careful in interpreting results when using the Problems in arousal, attention, memory,
ities.
foam condition. and judgment can affect a patient's abilit)' to

Patients who sway more, or fall, on con- attend to and perform behaviors being as-
ditions 5 and 6 of the CTSIB, demonstrate a sessed (21). In addition, these factors can af-
vestibular loss pattern, suggesting an inability' fect a patient's abiUty to comply with a treat-
to select vestibular inputs for postural control ment regimen.
in the absence of usehil visual and somatosen- During the course of assessment, many
sory cues. Finally, patients who lose balance aspects of cognitive flmction are evaluated by
on conditions 3, 4, 5, and 6 are said to have the chnician. Some are tested formally; how-
a sensory seleetion problem. This is defined as ever, more often, cognitive status is judged
an inability to effectively adapt sensory infor- subjectively, based on observations of the pa-
mation for postural control (10). tient's behavior during the course of the as-
sessment process.
Systems Assessment: Identifying
Level of Consciousness
Impairments
The Rancho Los Amigos Scale (22) is
The next step in a task-oriented assess-
probably the most well-known approach to
ment involves evaluating the senson,', motor
quantif^'ing level of consciousness in the pa-
(neural and musculoskeletal), and cognitive
tient with neurological impairments. This
subsystems that underlie task-based perfor-
scale is shown in Table 10.3. Assessment of
mance. This allows the clinician to identify the
of consciousness, arousal, or state, is an
level
impairments constraining flinctional abilities.
of assessing motor control, since
essential part
motor behavior is ver)' dependent on arousal
COGNITIVE SYSTEMS level (23).
Understanding cognitive factors is an Mental Status

important part of the assessment process,
since these factors can preclude an accurate Mental status can be determined infor-
and valid assessment of a patient's motor abil- mally by determining the patient's orientation
Chapter Ten Assessment and Treatment of Patients with Postural Disorders 21
Table 10.3. Rancho Los Amigos Cognitive 7^ Determining the patient's motivation
Scale^
and goals are an important part of assessment
I. No response: unresponsive to any stimulus. as well. Remember from the research on mo-
II. Generalized response: limited, inconsistent, tor learning described in Chapter 2, that
nonpurposeful responses, often to pain only. learning is facilitated by working on tasks that
III. Localized response: purposeful responses; may
are perceived as important to the learner.
follow simple commands; may focus on presented
object.
What are the patient's goals? How strongly is
IV. Confused, agitated: heightened state of activity; the patient committed to work towards these
confusion, disorientation; aggressive behavior; goals? Do patients feel that the goals are
unable to do self-care; unaware of present events; within their capacity to achieve? The answers
agitation appears related to internal confusion.
to these questions can help a therapist struc-
V. Confused, inappropriate; nonagitated; appears
alert; responds to commands: distractable; does
ture a therapeutic program that is both rele-
not concentrate on task; agitated responses to vant and meaningful to the patient.
external stimuli; verbally inappropriate; does not
learn new information. MUSCULOSKELETAL SYSTEM
VI. Confused, appropriate: goal directed behavior,
needs cueing; can relearn old skills as activities of Assessment of the musculoskeletal sys-
daily living (ADLs); serious memory problems; tem includes evaluation of range of motion
some awareness of self and others.
and flexibility'. Alignment, which is often con-
VII. Automatic, appropriate: appears appropriate,
sidered an aspect of the musculoskeletal sys-
oriented; frequently robot-like in daily routine:
minimal or absent confusion; shallow recall;
tem, was covered earlier under motor strate-
increased awareness of self, interaction in gies for postural control. In addition,

environment; lacks insight into condition; strength, which has both non-neural and neu-
decreased judgment and problem solving; lacks ral aspects, is discussed as part of the neuro-
realistic planning for future.
muscular systems in the next section. This
VIM. Purposeful, appropriate: alert, oriented; recalls
and integrates past events; learns new activities
chapter does not discuss techniques for as-
and can continue without supervision; sessing the musculoskeletal system in depth;
independent in home and living skills; capable of instead, the reader is urged to consult other
driving; defects in stress tolerance, judgment, texts (26-29).
abstract reasoning persist; many function at
reduced levels in society.
Range of Motion
'Reprinted with permission for Rancho Los Amigos Medical Cen- Range of motion is evaluated using slow
ter, Downy, Calif, USA, Adult Brain Injury Service.
passive movements. Passive range of motion
(PROM) in a particular joint may var)' among
normal subjects according to age and sex.
to person, place, and time. more A formal
measurement of mental may be done
status
PROM can be measured quantitatively using
Mini-Mental State Exam equipment such as a goniometer, or can be
by using either the
described subjectively using a scale similar to
(24) or the Short Portable Mental Status
the one in Table 10.4 (30).
Questionnaire (25). The Mini-Mental State
Evaluation of range of motion can also
Exam is shown in the Appendix as part of the
include active range of motion, that is the
comprehensive postural assessment.
Other aspects of cognitive hanction that joint range achieved by the patient without
are subjectively evaluated include: attention,
assistance from the examiner. This is often
communication, and motivation. Attention is

smaller than PROM. Finally, many clinicians
test joint play during an assessment of mus-
often evaluated informally through obsen'a-
tion of the patient's ability' to selectively mon-
culoskeletal parameters of motor control.
itor task-relevant stimuli, while ignoring ir-

Flexibility
relevant Cominunication abilities,
stimuli.
including both receptive and expressive com- Flexibility' is sometimes described with
munication skills, are also noted (22). reference to loss of mobility' in a two-joint
Table 10.4. Range of Motion Scoring Scale"' tween impaired force generation and func-
= no movement (ankyloslsl tional outcomes in patients with CNS lesions,
1 = considerable decrease in movement (moderate providing justification for including strength

hypomobility) testing within a motor control assessment bat-
2 = slight decrease in movement (mild hypomobility) ten,' (32, 36).
3 = normal
Strength can be measured under three
4 = slight increase in movement (mild hypermobility)
5 = considerable increase in movement (moderate
conditions: isometrically, isotonically, or iso-
hypermobility) kinetically (32). However, in the clinic, the
6 — severe increase in movement (severe most common approach is to examine iso-
hypermobility) metric or isokinetic strength during a short-
ening contraction (32, 33). Manual muscle
"Adapted trom: Jensen GM. Musculoskeletal analysis: introduc-
tion. In: ScullyRM, Barnes MR, eds. Physical therapy. Philadel- testing is the most common clinical approach
phia: JB Lippincott, 1989:331. to testing strength. This test assesses a sub-
ject's abihty to move a body segment through
a range, against gravity, or against externally
muscle (31). Decreased flexibility in a two-
applied resistance (37). An ordinal scale is
joint muscle prevents the simultaneous com-
used to grade strength ft-om 0, no contrac-
pletion of complete range of motion in related
tion, to 5, fiiU movement against gravit)' and
joints. The most common example of de-
maximal resistance (37). A limitation of man-
creased flexibility' is a gastrocnemius contrac-
ual muscle testing is that it does not examine
ture, which limits ankle joint dorsiflexion
the abilit}' of a muscle to participate in a fianc-
when the knee is extended. Since the gastroc-
tional movement pattern (38).
nemius spans both the ankle and the knee
An alternative approach to quantifying
joints, fuUrange of ankle dorsiflexion may be
strength incorporates the use of hand-held
more difficult to achieve with the knee ex-
dynamometers, which provide an objective
tended than when the knee is flexed and the
indication of muscle group strength (35, 37).
gastrocnemius is not on stretch. Thus, a joint
Hand-held dynamometers measure the force
may be functionally limited in range of mo-
required to break the patient's position during
tion secondan,' to loss of muscular flexibiliti,'.
manual muscle testing (31, 35, 37). Finally,
Neuromuscular System muscle performance can be tested dynamically
through the use of instrumented isokinetic
Assessment of neuromuscular impair- systems. Isokinetic testing assesses power, or
ments includes measurement of strength, the abilit)' to generate force throughout the
muscle tone, and nonequilibrium forms of co- range of motion, at different speeds and over
ordination. several repetitions (22, 31, 39).
Strength Muscle Tone
The
ability to produce a volimtary con- Abnormal muscle tone ranging from
traction depends on both non-neural and h\potonicit\' to hypertonicity may limit a pa-
neural elements. Strength results from both tient's ability to recruit muscles necessary for
properties of the muscle itself (musculoskel- postural control. The extent to which abnor-
etal aspects of strength) and the appropriate mal muscle tone is a limitation in controlling
recruitment of motor units and the timing of movements is currentiy under considerable
their activation (22, 32-34). Although weak- debate in the rehabilitation literature. Muscle
ness is a predominant feature of upper motor tone is assessed clinically by describing a mus-
neuron lesions, the measurement of muscle cle's resistance to passive stretch. Subjective
strength in patients with brain lesions is still rating scales, such as the one shown in Table
questioned by many clinicians (35). There is 10.5, are often used to describe alterations in
some evidence to support the relationship be- muscle tone (40).
Chapter Ten Assessment AND Treatment OF Patients WITH Postural Disorders 219
Table 10.5. Modified Ashworth Scale for Assessment of the sensoPi' components begins
Gradin g Spasticity' with an evaluation of the individual senses im-
= No increase in muscle tone portant to postural control. Particular atten-
1 = Slight increase in muscle tone, manifested by a tion is paid to evaluating somatosensation
slight catch and release or by minimal resistance (muscle, joint, touch, pressure) in the lower
end of the range of motion when the
at the
extremities. A more in-depth discussion of
affected part(s) is moved in flexion or extension.
senson' testing, including commercially avail-
\ + = muscle tone, manifested by a
Slight increase in
catch, followed by minimal resistance able tests to document sensor)- impairment, is
throughout the remainder (less than half) of the found in the chapter on the clinical assessment
range of motion (ROM). of upper extremity' control.
2 = More marked increase in muscle tone through
Some of the tests used to evaluate so-
most of the ROM, but affected part(s) easily
matosensation include:
moved.
3 = Considerable increase in muscle tone, passive
movement difficult. 1. Light touch — using a cotton swab,
4 = Affected part(s) rigid in flexion or extension. lighdy touch the patient on the face,
arm, and legs. The patient should be
•Adapted from Bohannon RW, Smith MB. Interrater reliability of when and where the
able to identify
a modified Ashworth scale of muscle spasticity. Phys Ther 1 987;
67:206-207. stimulus is being applied.
2. Two-point discrimination alternately —
touch one or two points on the pa-
Coordination
tient's skin. Determine whether the pa-
tient can feel one or two points, and
Tests of coordination ha\e been divided
intononequilibrium and equilibrium tests
how much distance between the Uvo
points is necessary' for the patient to dis-
(41^3). Equilibrium tests of coordination
criminate two points of pressure.
generally reflect the coordination of muld-
joint movements for posture and gait. Coor- 3. Extinction test touch —
t^\'o corre-
sponding points on different sides of
dination testing related to postural control
the body simultaneously. Ask the pa-
was discussed in the section on assessing strat-
tient to indicate where die touch has
egies.
of coordination
occurred. The patient should perceive
Nonequilibrium tests
motor both touches.
are important to all aspects of control,
including posture, mobilit)', and upper ex-
4. —
Temperature touch the patient's skin
with small containers of hot \s. cold wa-
tremity' control. These tests are often used to
ter.
indicate specific patholog\' within the cere-
bellum (42 ). These tests can include: finger to
5. Pain — test patient's abilit}' to distin-
guish sharp fi-om dull, using a pin.
nose, rapid alternating inovements, past
pointing, heel to shin, finger opposition, tap-
6. Position sense —can be tested in several
hand wavs. Most often, die patient's big toe

ping (hand or foot), or drawing a circle (
is grasped on the sides and moved up

or foot). Performance is graded subjectively
or down. Without looking, the patient
using the following scale: 5 norinal, 4 minimal
is asked to report whether the toe is up
impairment, 3 moderate impairment, 2 severe
or do\\'n.
impairment, 1 cannot perform.
7. Movement sense —
move one limb pas-
and ask the patient to imitate the
sively,
SENSORY SYSTEMS
motion with the opposite limb.
Postural control requires the organiza- 8 . Stereognosis —place a series of common
tion of vision, somatosenson,', and vestibular objects in the patient's hand and ask the
inputs, which report information about tiie patient to identity' them.
position of the body relative to external ob- 9. Vibration —using a tuning fork on bony
jects, including the support base and gravit)'. prominences such as the malleoli, knee,
or wrist, the patient should be able to tive judgment regarding whether the patient
identify where the vibration has oc- is moving to ma.ximum stabilit)' limits in all
curred anci when it starts and stops. directions (8 — 10).
In addition, problems in the \'isual sys- Motion Perception
tem are noted, such as glaucoma, cataracts,
retinal degeneration, decreased visual acuit)', Motion perception is the conscious
diplopia, and peripheral visual field cuts. sense of whether the body is still or in motion.
Dizziness is a misperception of motion (either
self or environment) that results when sensor*'
ASSESSING PERCEPTIONS RELEVANT
inputs are inconsistent in reporting body mo-
TO POSTURAL CONTROL tion (9, 44, 45). The term dizzinessis used by
patients to describe a variety of sensations, in-
Two aspects of perception particularlv
cluding spinning (referred to as vertigo),
important to postural control are evaluated
rocking, tilting, unsteadiness, and lighthead-
(9,10).
edness.
Assessment begins with a carefiil histors-
StabiliU' Limits to determine the patient's perceptions of
whether dizziness is constant or provoked,
The patient's internal representation of and the situations or conditions that stimulate
stability limits in sitting and standing is eval- dizziness. The Vertigo Positions and Move-
uated. In particular, the consistency between ment Test (9) examines the intensit)' and du-
the patient's perceived vs. actual Umits of sta- ration of dizziness in response to mo\ement
bLlit\' is subjectively determined. The patient and or positional changes of the head while
is asked to sway voluntarily as far as possible sitting, standing, and walking. The patient is
in all directions without falling. This deter- asked to rate the intensity' of dizziness on a
mines the individual's limits of perceived sta- scale of to 10. In addition, duration of
bilit)'. Alternatively, the patient is asked to symptoms is timed and recorded, as are the
reach for an object held at the outer edge of presence of nystagmus and autonomic ner-
his/her stability' limits. The therapist obsen'es vous system symptoms including nausea,
the extent to which the patient is willing to sweating, and pallor. For a detailed descrip-
move the center of mass, and makes a subjec- tion of assessment and treatment of dizziness.
Table 10.6. Task-Oriented Assessment of Postural Control
Levels: Function Strategies Impairment
Tests/measurements; Get up/go Alignment R.O.M.

(examples) Functional reach Sitting Strength
Tinetti Standing MMT
Berg Movement Dynamometry
Ankle Tone
Hip Passive movement
Step Pendulum test
Sensory Reflex testing
CTSIB Coordination
Equltest Individual senses
Cognition
Mini Mental Test
Rancho Scale
Perception
Stability limits
Dizziness
the reader is referred to other sources (9, 44, functional problems include difficulty with trans-
fers (sit to stand, chair to chair), standing with a
45).
reduced base of support, and maintaining balance
In summary, a task-oriented approach
during dynamic activities such as stepping or turn-
to assessing postural control uses a variety of
ing.
tests, measurements, and observations to: (a)
An assessment of the patient's motor strategies
document fijnctional abilities related to pos-
indicates an asymmetric alignment, with weight
ture and balance control, (b) assess underlying side in both sitting and stand-
displaced to the left
sensory and motor strategies, and (c) deter- ing. In addition, movement strategies indicate pri-
mine the level of function of underlying sen- mary use of a hip strategy to control body sway,
sory,motor, and cognitive systems contrib- inability to use an ankle strategy in the hemiplegic
uting to postural control. This concept is leg, and difficulty taking a step with the nonin-
shown inTable 10.6. In addition, an example volved leg when the center of mass exceeds the
of a task-oriented assessment form for assess- base of support.
ing an adult patient with an UMN deficit is

An assessment of sensory strategies indicates
the patient is unable to maintain balance when
shown in the Appendix. This particular as-
any sensory information is reduced (falls on con-
sessment form is geared to the assessment of
ditions 2, 3, 4, 5, and 6 of the CTSIB).
an adult patient in a rehabilitation, outpatient,
The third level of assessment indicates the fol-
or home health program, rather than an acute lowing impairments: (a) decreased cognitive sta-
care patient. tus; specific problems with orientation to time and
place, attention, memory, and emotional lability.
Interpretation of In moderate problems with

addition, there are
receptive and expressive aphasia; (b) musculo-
Assessment skeletal impairments, including: 5° of ankle dor-
siflexion in the right leg; (c) neuromuscular im-
Following completion of the assess- pairments, including: reduced ability to generate
ment, the clinician must interpret the assess- force voluntarily (2±5 manual muscle testing in
ment, identify the problems, both at the level right lower extremity muscles), decreased ability
of fianction and impairments, and establish to recruit ankle muscles in the right leg for postural
the goals and plan of care. control, and moderate increase in muscle tone in
the right elbow flexor and ankle extensors; and {d}
sensory/perceptual problems, including: de-
creased sensory discrimination (somatosensation)
ACTIVE LEARNING MODULE in the right arm and leg, and right hemianopsia.
With this initial understanding of the patient's

Before moving on, take a moment
problems, the clinician can move ahead to estab-
and work on the following case study.
lishing goals and planning treatment. It is difficult
Your task is to create a problem list for
to gain an understanding of all of the patient's
Phoebe Mines, a 53-year-old patient with right-
problems in the one or two therapy sessions.
first
sided hemiplegia, referred for evaluation of bal-
Rather, understanding and insight continue to
ance 5 weeks following her stroke. (Refer to the
grow with each session over the course of treat-
evaluation form in Appendix A.) Based on your
ment. Before we establish goals and a plan of care
knowledge of normal and abnormal postural con-
for Ms. Mines, let's review a task-oriented ap-
trol and the type of problems likely to be found
proach to treating postural dyscontrol.
following a stroke, complete the evaluation. Once
completed, make a problems drawn from all
list of
three levels of your assessment. Use this problem
list to develop both short- and long-term goals for
treatment.
Treatment
What did you predict? We found that problems
drawn from the first level of assessment indicate
the patient appears to be having moderate func- The goals of a task-oriented approach to
tional balance problems, as indicated by a score retraining postural control include: to resolve
of 42/56 on the Functional Balance Scale. Specific or prevent impairments; to develop effective
task-specific strategies, to retrain flinctionai Table 10.7. Strategies for Working with the
Patient with Cognitive Impairments
tasks, and to adapt task-specific strategies so
that functionaltasks can be performed in 1 Reduce confusion — make sure the task goal is
changing environmental contexts. clear to the patient

2. Improve motivation — vi/ork on tasks that are
relevant and important to the patient
Treating at the Impairtnent Level 3. Encourage consistency of performance be —
consistent in your goals and reinforce only those
behaviors that are compatible with those goals
The goal of treatments aimed at the im-
4. Reduce confusion — use simple, clear, and concise
pairment le\el is to correct those impairments instructions
that can be changed, and to prevent the de- 5. Improve attention —accentuate perceptual cues
velopment of secondan' impairments. Allevi- that are essential to the task, and minimize the
ating underlying impairments enables the pa- number of irrelevant stimuli in the environment
tient to resume using previously developed

6. Improve problem-solving ability —begin viiith
relatively simple and gradually increase the

tasks,
strategies for postural control. When perma- complexity of the task-demands
nent impairments make resumption of previ- 7. Encourage declarative as well as procedural
ously used strategies impossible, new strate- learning —
have a patient verbally/and or mentally
gies wdll have to be de\eloped. rehearse sequences when performing a task
8. Seek a moderate level of arousal to optimize
learning — moderate the sensory stimulation in the
COGNITIVE IMPAIRMENTS environment; agitated patients require decreased

intensity of stimulation (soft voice, low lights, slow
touch) to reduce arousal levels; stuporous patients
Many patients with UMN lesions dem- require increased intensity of stimulation (use brisk,
onstrate significant cognitive impairments loud commands, fast movements, working in a
vertical position).
that affect the patient's ability' to participate
9. Provide increased levels of supervision, especially
in a retraining program. With this in
fiilly
during the early stages of retraining.
mind, Table 10.7 provides a few suggestions 1 0. Recognize that progress may be slower when
for modifying treatment strategies when working with patients who have cognitive
working with a patient who has cogniti\e impairments.
problems. Howe\'er, it is not within the scope
of this book to discuss in detail issues related

to retraining cogniti\'e impairments affecting NEUROMUSCULAR IMPAIRMENTS
motor control in the patient with neurological
dysfianction. Numerous neuromuscular limitations
leading to instability in the patient with a neu-
rological deficit are described in Chapter 9.
MUSCULOSKELETAL IMPAIRMENTS Since stabilit}' requires the abilit\' to generate
and coordinate forces necessary for moving
Musculoskeletal problems can be the center of mass, upper motor lesions pro-
treated using traditional physical therapy ducing limitations in strength, force control,
techniques, including modalities such as heat, and muscle tone will produce concomitant
ultrasound, massage, and biofeedback. Pas- limitations in stability'.
siverange of motion exercises are used to im-
prove joint mobility and muscle flexibilit}'. Strength
Manual therapies focus on regaining passive
range and joint play. Finally, plaster casts and The
abilit}' to produce a voluntan' con-
splints are used to passively increase range and traction depends on both the characteristics
flexibilit\' in the patient with neurological im- of the muscle itself, and on the appropriate
pairments. For an in-depth discussion of recruitment and timing of motor units. Tech-
treatment of this important area of musculo- niques to improve strength can focus on gen-
skeletal impairments, the reader is referred to erating force to move a body segment, or al-
other sources (26-29). ternativelv, the abilitA' to resist a movement.
Chapter Ten ASSESSMENT AND Treatment of Patients with Postural Disorders 223
Progressive resistive exercises are com- applied. For example, ice can facilitate muscle
monly used to increase strength vvithin indi- activity when applied quickly, as in a brief
vidual muscles. Isokinetic equipment can also sweep over a muscle. Alternatively, prolonged
be used to improve a patient's abilit\' to gen- icing is considered inhibitory, decreasing the
erate force throughout the range of motion, level of activation.
at different speeds of motion, and through re- Vibrators have also been used to either
peated efforts within individual and groups of facilitate or inhibit activity in a muscle. High-
muscles (22). Proprioceptive Neuromuscular frequency vibration tends to facilitate muscle
Facilitation techniques can be used to im- activit}', while low frequency inhibits muscle
prove the timing of force generation, as well activity levels (50, 51).
as the reciprocal interaction between agonist Techniques such as approximation,
and antagonist muscles (46). which activates joint receptors, have also been
Biofeedback and fiinctional electrical used to facilitate muscle activity in the patient
stimulation can also be used to assist patients with neurological impairments. Joint approx-
in regaining volitional control over isolated imation involves compressing a joint either
muscles and For example, stimulation
joints. manually (46), or through the application of
of the peroneal nen'e is commonly performed weights.Manual techniques that apply trac-
in hemiplegic patients to improve control tion to a joint are also used to facilitate muscle
over the anterior tibialis muscle during a vol- activity (46).
untary contraction. Quickstretch to a muscle facilitates ac-
A number of studies have shown that tivation of the muscle through the stretch re-
biofeedback is effective in helping the patient flex. In contrast, prolonged stretch (either
with a neurological impairment learn to ini- manually, or through the use of casts, splints,
tiate, sustain, and/or relax a voluntar}' muscle or orthoses) decreases activity levels.
contraction (47-49). There is some evidence Brisk touch or tapping also facilitates
that improved control over an isolated muscle muscle activity. In contrast, slow repetitive
has some carryover to gait. Thus, patients touching is considered inhibitory.
given therapy related to muscle control in- Altering a patient's position has also
creased gait velocitv', although this was not been suggested as a technique that can be
trained specifically (47). used to alter muscle tone and postural tone
(54). The underlying assumption, drawn
Muscle Tone from a reflex hierarchical theor\' of motor
control, is that placing patients in certain po-
Considerable effort has been directed at of muscle
sitions will alter die distribution
developing therapeutic techniques to alter (and postural) tone, primarily through the
muscle tone in the patient with neurological changes in reflex activity. For example, it has
impairments. One way to alter mus-
possible been suggested that placing a patient in the
cle tone is to change the background level of supine position will facilitate extensor tone,
activit}' in the motor neuron pool of the mus- while flexor tone is facilitated when the pa-
cle. As background level of activity in the mo- tient is prone, due to the presence of released
tor neuron pool increases so does the likeli- tonic labyrinthine reflexes in the patient with
hood that the muscle will respond to any UMN lesions. The use of a side-lying position
incoming stimulus, whether from the periph- is often suggested as an approach to inhibiting
ery or as part of a descending command. The the effects of the asymmetric tonic neck reflex
opposite is also true; as background levels of on muscle tone, facilitating bilateral symmet-
activity decrease, the muscle is less likely to ric activities (54).
fire. What techniques can be used to alter
background activity' of motor neuron pools.' SENSORY IMPAIRMENTS

Sensory stimulation techniques can be
used to facilitate or inhibit motor activity, de- Often, clinicians tend to view sensory
pending on the type of stimulus and how it is impairments such as loss of limb position
sense or somatosensor\' deficits leading to de- Treating at the Strategy Level

creased object recognition, as being perma-
nent, or not modifiable by treatment. How- The goal of retraining at the strategy
ever, a number of interesting studies suggest level involves helping or guiding patients to
that treatment can affect the patient's abilit)' recover, or develop, sensory and motor strat-
to process sensor\' stimuli. egies that are effective in meeting the postural
Based on some studies examining the demands of fiinctional tasks. To fiilly retrain
reorganization of somatosensory cortex in strategies, the clinician must understand the
primates (55), which were previously dis- inherent requirements of the task being per-
cussed in Chapter 4, a number of researchers formed.
have developed structured sensory reeduca- For example, both seated and stance
tion programs to improve the patient's ability postural control require that the center of
to discriminate and interpret sensory infor- gravity' of the body be within the base of sup-
mation (56-58). The goal of these interven- port. In the case of standing, the base of sup-
tions is to improve a patient's ability to detect port is limited to the feet, unless the patient
and process information in the environment is using an assistive device. In the case of
and thereby improve motor performance. seated postural control, the trunk mass must
Suggestions for retraining sensor^' discrimi- stay within the base of support defined by the
nation are presented in more detail in the bottom and feet. Thus, in order to regain the
chapter on retraining upper extremity con- ability to stand or sit independently, the pa-
trol. tient must develop movement strategies that
are successful in controlling the center of mass
PERCEPTUAL IMPAIRMENTS relative to the baseof support. These include
(a) strategies that move the center of mass
Treatment of dizziness varies, depend- relative to a stationary base of support, in
ing on the underlying cause. Vestibular Re- standing, for example, an ankle or hip strat-
habilitation is an exercise approach to treating egy, and (
b) strategies for changing the base
symptoms of dizziness and imbalance that re- of support when the center of mass moves be-
sult from patholog)' within the vestibular sys- yond it, for example, a stepping strategy in
tem. Since there are many potential causes of standing, or a protective reach in sitting.
dizziness, including metabolic disturbances,
side-effects of medication, cardiovascular ALIGNMENT
problems, such as orthostatic hypotension,
and patholog)' within peripheral or central The goal when retraining alignment is
vestibular structures, it is essential that the to help the patient develop an initial position
therapist know the underlying diagnosis prior that (a) is appropriate for the task, (b) is effi-
to beginning an exercise-based approach. cient with respect to gravity, that is, with min-
Vestibular Rehabilitation uses repeated imal muscle activity requirements for main-
exercises to habituate symptoms of dizziness. taining the position, and (c) maximizes
The patient is instructed to repeat the position stability, that is, places the vertical line of grav-
or movements that provoke dizziness five ity' well within the patient's stability limits;
times in a row, two to three times per this allows the greatest range of mo\'ements
day. Exercises are progressive in nature. for postural control. Many tasks utilize a sym-
The patient begins with fairly simple exer- metrical vertical position, but this may not al-
cises, such head movements in

as horizontal ways be a realistic goal for all patients (10).
the seated position, and progresses to A number of approaches can be used to
more difficult tasks, such as horizontal head help patients develop a symmetrically vertical
movements integrated into gait. This ap- posture. Commonly, verbal and manual cues
proach is discussed in more detail elsewhere are used by the clinician to assist a patient in
(9,44,45). finding and maintaining an appropriate ver-

tical posture. Patients practice with eyes open
and closed, learning to maintain a vertical po-

sition in the absence of visual cues.
Mirrors can also be used to pro\'ide pa-
tients with visual feedback about their posi-
The effect of a mirror can be
tion in space.
enhanced by having the patient wear a white
T-shirt with a vertical stripe down the center,
and asking him/her to tr\' to match the stripe
on the T-shirt to a vertical stripe on the mirror
(Fig. 10.10). The patient can use the mirror
and T-shirt approach while performing a va-
riety' of tasks, such as reaching for an object,
which require that the body be moved away

from the vertical line and then reestablish a
vertical position.
Another approach to retraining vertical
alignment is shown in Figure 10.11, and uses
flashlights attached to the patient's body in

conjunction with targets on the wall (10). In
this task, the patient is asked to bring the light
(or lights) in line with the target(s). Lights
Figure 10.11 Using a flashlight in conjunction with tar-
getson a wall to help a patient learn to control center of

mass movements.
can be turned on and off during the task so

that visual feedback is intermittent.
Another approach to retraining vertical
posture involves having patients stand (or sit)
with their back against the wall, which pro-
vides enhanced somatosensor)' feedback
about their position in space. This feedback
can be further increased by placing a yard stick
or small roll \ertically on the wall and having
the patient lean against it. Somatosenson,'
feedback can be made intermittent by having
the patient lean away from the wall, only oc-
casionally leaning back to get knowledge of
results (KR).
Kinetic or force feedback devices are of-
ten used to provide patients with information
Figure 10.10 Using a mirror when retraining align-

about postural alignment and weightbearing
ment; the patient is asked to line up the vertical stripe on status (60-64). Kinetic feedback can be pro-
his T-shirt with a vertical stripe on the mirror. \'ided with devices as simple as bathroom
scales (Fig. 10.3). Alternatively, kinetic feed- MOVEMENT STRATEGIES

back can be given through either load-limb
monitors (60) or forceplate biofeedback sys- The goal when retraining movement
tems (Fig. 10.2) (61). Other types of feed- strategies involves helping the patient develop
back devices include using a feedback cane to multijoint coordinated movements that are
improve patients' weightbearing status (63). effective in meeting the demands for posture
Clinicians routinely provide the un- and balance in sitting and in standing. We re-
steady patient with assistive devices, such as train strategies within the context of a task,
canes or walkers. What effect does providing since optimal fianction is characterized
an external support such as a cane have on by strategies that are efficient in accomphsh-
balance? As illustrated in Figure 10.12, an asing a task goal in a relevant environment
sistivedevice such as a cane increases the base (10).
of support. Since stabilit)' requires keeping the Retraining strategies involves both the
center of gravity within the base of support, of motor strategies and the devel-
recover)'
increasing the base of support makes the task opment of compensator^' strategies. As we
of stability' easier. Researchers have studied mentioned in Chapter term recovery
2, the
the effects of a cane on standing balance in through original
refers to achieving fiinction
patients with hemiparesis, using a forceplate processes, while compensation is defined as

to record changes in center of pressure under behavioral substitution, or the adoption of
various conditions of support. They found new strategies to complete a task.
that using a cane results in a significant shift: Patients are encouraged and guided to
in the position of the center of pressure to- develop strategies for both seated and stance
wards the cane side, and a decrease in both postural control, including the ability to move
anterior-posterior and medial-lateral postural the body in all directions to accomplish func-
sway. Thus, although using a cane will reduce tional tasks. We use as our example of strategy
postural sway, it increases the asymmetric retraining, the development of coordinated
alignment of patients towards the side hold- ankle, hip, and stepping strategies for stance
ing the cane (65). postural control, and show how these strate-
Figure 10.12 The effects of holding a cane while standing include widening the base of support and shifting the mean
position of the center of pressure laterally toward the cane side. (Adapted from Milezarek J), Kirby LM, Harrison ER,
MacLeod DA. Standard and four-footed canes: their effect on the standing balance of patients with hemiparesis. Arch
Phys Med Rehabil 1993;74:283.)
gies can be developed within the context of ance. Small perturbations can facilitate the use
self-initiated voluntary sway, in response to of an ankle strategy' for balance control, while
external perturbations, and during tasks re- larger perturbations encourage the use of a
quiring anticipatory postural adjustments. hip or step.
Remember, just because we limit our discus- Finally, patients are asked to carr)' out a
sion to activities that could be used to retrain varietA' of manipulation tasks, such as reach-
strategies for sagittal plane stance postural ing, lifting, and throwing, thus helping pa-
control, it does not mean that, in actualit)', tients to develop strategies for anticipatory
retraining postural control in the patient postural control. A hierarchy of tasks reflecting
should be limited to retraining these strate- increasing anticipator}' postural demands can
gies. be helpful when retraining patients in this im-
portant area. The magnitude of anticipator)'
postural activity is directly related to the po-
Developing a Coordinated Ankle Strateg)'
tential for instability inherent in a task. Poten-
tial instabilit\' degree
relates to speed, eftbrt,
Prior to retraining the use of an ankle of external support, and task complexity.
strategy for postural control, it is essential to Thus, asking a patient who is externally sup-
remember that this strategy' requires the pa-
tient to have adequate range of motion and
strength at the ankle (8, 10). In the face of
persisting impairments that preclude the use
of an anklestrategy', patients would be en-
couraged to develop the use of alternative

strategies, such as the hip or step, when con-
trolling body sway.
When retraining the use of an ankle
strategy during self-initiated sway, patients are
asked to practice swaying back and forth, and
side to side, within small ranges, keeping the
body and not bending at the hips or
straight
knees. Knowledge of results regarding how
far the center of mass is moving during self-
initiated sway can be facilitated using static

forceplate retraining systems (10). Flashlights
attached to the patient in conjuncdon with
targets on the wall can also be used to en-
courage patients to move from side to side
(refer back to Fig. 10.11).
Patients who are \'ery unsteady or ex-
tremely fearfiil of falling can practice move-
ment while in the parallel bars, or when stand-
ing close to a wall, or in a corner with a chair
or table in front of them (Fig. 10.13). Mod-
ifying the environment (either home or clinic)
in this manner allows a patient to continue
practicing movement strategies for balance
control safely and without the continual su-
per\ision of a therapist.
Use of perturbations applied at the hips
Figure 1 0.1 3 Placing a patient near a wall with a chair
or shoulders is an effective way to help pa-
in front of her increases safety when retraining standing
tients de\elop strategies for recovery of bal- balance in a fearful or unstable patient.
ported by the therapist to lift a light load

slowly, requires minimal anticipatory postural
activity. Conversely, an unsupported patient
who must lift a heavy load quickly, must util-
ize a substantial amount of anticipators' pos-

tural activity to remain stable.
Treatment of Timing Problems
How can a clinician help a patient re-

cover an ankle strategy in the face of coordi-
nation problems that affect the timing and
scaling of postural movement strategies.^
When a patient is unable to activate distal

muscles quickly enough to recover stability
during a postural task, the clinician can use a
variet)'of techniques to facilitate muscle ac-
tivation. These include icing, tapping, and vi-
bration to the distal muscles while the patient
is standing, immediately prior to, and during,
perturbations to standing balance, or self-ini-
tiated sway (10). This is shown in Figure

10.14.
Biofeedback and electrical stimulation Figure 10.14 The use on the anterior tibialis mus-
of ice
can also be used to improve the automatic re- cle just prior to a small backward displacement is used
cruitment and control of muscles during task- to facilitate its activation during recovery of balance.
specific movements strategies for posture (67)

and gait (62). For example, electrical stimu-
lation in conjunction with a foot switch can of tibialis activation was sufficient to trigger
be used to decrease onset latencies of postural stimulation of the quadriceps. This set-up was
responses (67). As shown in Figure 10.15, a used in conjunction with external perturba-
foot switch can be placed under the heel so tions to balance, and was successfitl in chang-
that increased weight on the switch triggers a ing the timing of quadriceps activation within
tetanic stimulation of the anterior tibialis the postural response synergy (66).
muscle. Electrical stimulation to recruit a There is no established research that
muscle within a postural movement strategy' provides guidelines to the clinician regarding
can be done during self-initiated sway or dur- the optimal frequency and duration of stim-
ing perturbed balance. ulation techniques during postural retraining.
Several clinicians have combined the use We ha\e found through trial and error that 5
of biofeedback and functional electrical stim- minutes of stimulation, twice daily, for 3 to 4
ulation (FES) during retraining motor con- weeks appears to be effective in altering tim-
trol, and found that the combined used of ing parameters. Howe\'er, fiirther research is
biofeedback and FES was superior to either in needed in this area.
isolation (65). One approach we have tried
successfially is to use EMG biofeedback on the Treatment of Scaling Problems
tibialis anterior muscle, and to link the bio-
feedback with a ftmctional electrical stimula- To produce effective movements of the
tor whose electrodes were placed on the center of body mass during postural control,
quadriceps muscle of the same leg (FES). The the level of muscle activation must be scaled,
two units were set up such diat a minimal level or graded, appropriate to the amplitude of
Chapter Ten ASSESSMENT AND Treatment of Patients with Postural Disorders 229
mass voluntarily to different targets displayed

on a screen. Targets are made progressively
smaller and are placed closed together, re-
quiring greater precision in force control.
Knowledge of results is given with respect to
mox'ements that overshoot the target, indicat-
ing an error in amplitude scaling.
another approach to treating
Finally,
scaling problems in patients with cerebellar
patholog)' producing ataxia, is to add weights
to the trunk or limbs (69, 70). Two rationales
are proposed to explain the potential benefits
of weighting. The first is that joint compres-
sion associated with weights would facilitate
coactivation of muscles around a joint,

thereby increasing stiffness. The other expla-
nation is mechanical; adding weights increases
the mass of the system. In this way, the in-
creased forces generated in the cerebellar pa-
tient match the increased mass of the system
(69). Researchers have found that adding
weights to cerebellar patients has inconsistent
effects. Some patients become more stable,
while others are destabilized by the weights
(69,70).
Developing a Coordinated Hip Strategy
Figure 10.15 Use of electrical stimulation of the distal A hip strategy can be facilitated by ask-
muscle in conjunction with a foot switch to facilitate ac- ing the patient to maintain balance without
tivation of the anterior tibialis muscle during stance bal-
taking a step and by using cHisplacements in
ance retraining.
larger ranges than those used for an ankle
strategy. Use of a hip strategy' can also be fa-
cilitated by motion at the ankle

restricting
body sway. Normal subjects use a combina-
joints either through the use of plaster casts
tion of feedforward and feedback control
(bivalved so they can be taken on and off) or
mechanisms to scale forces for postural con-
the use of ankle orthoses (10).
trol (68). To improve amplitude scaling of
Patients can be asked to maintain vari-
postural synergies, patients may practice re-
ous equilibrium positions that require the use
sponding to perturbations of various ampli-
of a hip strategy for stability. Possible exam-
tudes. Feedback regarding the appropriate-
ples include standing on a narrow beam,
ness of their response is provided by the
standing heel/toe, or adopting a single limb
clinician. Interestingly, it is easier for many
stance (10).
cerebellar patients, who consistently overre-
spond to small pushes, to appropriately scale
postural movements to large perturbations Developing a Coordinated Step Strateg}'
(10).
Static forceplate retraining systems can Stepping to avoid a fall recjuires tiie ca-
also be used effectively to retrain scaling prob- pacity' to maintain the body's weight on a sin-
lems. Patients are asked to move the center of gle limb momentarily, without collapse of
that limb. Stepping is normally used to pre- helping a patient develop the ability' to step
vent a fall when the center of mass has (or is for postural control, it is important to tell the
rapidly) moving outside the base of support. patient that the goal of the exercise is to take
Traditionally, stepping is taught within the a step to prevent a fall.
context of step initiation during gait retrain-

ing. Unexpected stepping is often viewed by SENSORY STRATEGIES
on the part of the pa-
the clinician as a failure
tient to maintain balance. However, learning The goal when retraining sensory strat-
to step when the center of mass exceeds the egies is to help the patient learn to eft'ecdvely
base of support is an essential part of postural coordinate sensory information to meet the
retraining. demands of postural control. This necessitates
Stepping can be manually by
facilitated correctly interpreting the position and move-
the clinician by shifting the patient's weight ments of the body in space. Treatment strat-
to one side and quickly bringing the center of egies generally require the patient to maintain
mass towards the unweighted leg (Fig. balance during progressively more difficult
10.16). The clinician can further assist the pa- static and dynamic movement tasks while the
tient with a step by manually lifting the foot clinician systematically varies the availability
and placing it during the maneuver. To en- and accuracy of one or more senses for ori-
sure a patient's safety, stepping can be done entation (9, 10, 44).
within the parallel bars, or near a wall. When Patients who show increased reliance on
vision for orientation are asked to perform a
variety of balance tasks when visual cues are
absent (eyes closed or blindfolded), or re-
duced (blinders or diminished lighting). Al-
ternatively, visual cues can be made inaccurate
for orientationthrough the use of glasses
smeared with petroleum jelly (shown in Fig.
10.17), or prism glasses. Decreasing a pa-

motion cues in their
tient's sensitivity to visual
environment can be done by asking the pa-

tient to maintain balance during exposure to
optokinetic stimuli, such as moving curtains
with stripes, moving large cardboard posters
with vertical lines, or even moving rooms ( 10,
71).
Patients who show increased reliance on
the surface for orientation are asked to per-
form tasks while sitting or standing on sur-
faces providing decreased somatosensory cues
for orientation, such as carpet or compliant
foam surfaces, or on moving surfaces, such as
a tilt board.
Finally, to enhance the patient's ability
to use remaining vestibular information for

postural stability, exercises are given that ask
the patient to balance while both visual and
somatosensory inputs for orientation are si-
Figure 10.16
multaneously reduced, such as standing on
Facilitating a stepping strategy by manu-
ally shifting the patient's center of mass laterally and compliant foam (Fig. 10.18) or an inclined
manually moving the patient's foot into a step. surface with eyes closed.
Chapter Ten Assessment and Treatment of Patients with PosTURAi Disorders 231
Figure 10.18 Facilitating the use of vestibular inputs for

postural control requires that the patient maintain bal-
ance when orientation cues from visual and somatosen-
sory systems are reduced or inaccurate by standing on a
Figure 10.17 Petroleum-covered glasses used to ob- foam surface and wearing petroleum-covered glasses.
scure but not completely remove visual cues for postural
control.
Treating at the Functional Task

Level
Perceived Limits of Stabilit}'
Developing adaptive capacities in the
Rehabilitation strategies involving use patient is also a critical part of retraining pos-
of postural sway biofeedback have also been tural control. The ability to perform postural
used with patients who incorrecdy perceive tasks in a natural environment requires that
that they have reduced stabOity limits. the patient modify strategies to changing task
Patients are asked to sway using larger and and environmental demands. The goal of re-
larger areas, in an effort to change percep- training at the functional level focuses on hav-
tions that they cannot move the body safely ing patients practice successfully the perfor-
in space. mance of a wide collection of functional tasks
In addition, patients may be asked to vi- in a variety of contexts.
sualize a space around them with boundaries We began our discussion of task-ori-
in which they can move safely when seated or ented retraining in the previous section fo-
standing. Patients are then asked to practice cusing on retraining strategies for postural
moving their bodies within and to those control during three tasks, self-initiated sway,
boundaries. Boundaries may be gradually ex- in response to perturbation, and anticipatory
panded with increasing sensory and motor ca- to potentially destablizing movements such as
pacities of the patient (8-10, 44). reaching, lifting, or stepping. This concept is
now broadened to include ha\ing the patient tify the problems related to function, strate-
practice a wide variet\' of functional tasks with gies, and contributing impairments, and es-
varying stabilin,' and orientation demands. tablish the goals and plan of care.
3. The plan of care for retraining posture control
This could include (a) maintaining balance
in the patient uith a neurological deficit will
with a reduced base of support, that is, with
vary widely, depending on the constellation
feet together, in tandem, or on one foot, ( b)
of underlying impairments and the degree to
maintaining balance while changing the ori-
which the patient has developed compensa-
entation of the head and trunk, for example, tory strategies that are successful in achieving
looking over one's shoulder, or leaning over, postural demands in functional tasks.
(c) maintaining balance while performing a 4. The goals of a task-oriented approach to re-
variet)-of upper extremity' tasks, such as reach- training postural control include (a) resolve or
ing, lifting, pushing, and holding objects with prevent impairments, (fa) develop effective
one or both hands. task-specific strategies, (c) retrain functional
As we mentioned Chapter 6, all tasks

in tasks, and icf) adapt task-specific strategies so
that functional tasks can be performed in
demand postural control; however, the sta-
changing environmental contexts.
bility" and orientation requirements will var\'
5. The goals of treatments aimed at the impair-
with the task and the emironment. By un-
ment level are to correct those impairments
derstanding the postural requirements inher-
that can be changed and prevent the devel-
ent in various tasks and en\ironments, the cli-
opment of secondary impairments.
nician can de\elop a hierarchy of tasks to 6. The goal of retraining at the strategy level in-
retrain postural control, beginning with tasks volves helping patients recover or develop
that ha\e relati\ely few stabiht\' demands, and sensory and motor strategies that are effective
mo\ing to those that place hea\y demands on in meeting the postural demands of functional
the postural control system. For example, tasks. This requires that the clinician under-
postural demands in\"olved in maintaining an stand the inherent requirements of the task be-
ing performed so that patients can be guided
upright seated posture while in a semisup-
in developing effective strategies for meeting
ported seated position are relatively few. In
task demands.
contrast, sitting on a mo\ing tiJt board while
7. The goal of retraining at the functional level
holding a cup of water has fairly rigorous sta-
focuses on having patients practice success-
biLitN" requirements, reflecting the changing fully the performance of a wide collection of
and unpredictable nature of the task. This task functional tasks in a variety of contexts. Since
requires constant adaptation of the postural the ability to perform postural tasks in a
system. Therefore, supported sitting would be naturalenvironment requires the ability to
a good task to begin with when working with modify strategies to changing task and envi-
a patient who has se\ere postural dyscontrol. ronmental demands, developing adaptive ca-
As the patient improves, more pacities in the patient is a critical part of re-
difficult and
training at the task level.
demandine: tasks can be introduced.
8. The development of clinical methods based
on a systems theory of motor control is just
beginning. As systems-based research pro-
Summary vides us with an increased understanding of
normal and abnormal postural control, new
1 A task-oriented approach to assessing postural methods for assessing and treating postural
control uses a variety of tests, measurements disorders will emerge.
and observations to (a) document functional
abilities related to posture and balance con-
trol, (fa) assess underlying sensor^' and motor
References
strategies, and (c) determine the underlying
sensory, motor, and cognitive systems con- 1. Mathias S, Nayak U, Issacs B. Balance in el-
tributing to postural control. derly patients: the "Get-up and Go" test.
2. Following completion of the assessment, the .\rch Phys Med RehabiJ 1986;67:387-389.
clinician must interpret the assessment, iden- 2. Podsiadlo D, Richardson S. The timed "Up
& Go": a test of basic functional mobilit)' for 18. DeFabio R, Badke MB. Relationship of sen-
elderly persons.
frail Am Geriatr Soc 1991;
J sory organization to balance tiinction in pa-
39:142-148. tients with hemiplegia. Phvs Ther 1990;
3. Mahoney RI, Barthel DW. Functional eval- 70:542-560.
uation: the Barthel Index. Md Med J 1965; 19. Cohen H, Blatchly C, Gombash L. A study
14:61-65. of the clinical test of and
sensor)- interaction
4. Duncan PW, Weiner DK, Chandler J, Stu- balance. Phys Ther 1993;73:346-354.
denski S. Functional reach: a new clinical 20. Horak F, JonesRycewicz C, Black FO,
measure of balance. J Gerontol 1990; Shumway-Cook A. Effects of vestibular re-
45:192-195. habilitation on dizziness and imbalance. Oto-
5. Tinetti ME. Performance oriented assess- lar>'ngol Head Neck Surg 1992;106:175-
ment of mobiliti,' problems in elderly patients, 180.
J Am GeriatSoc 1986 34;1 19-126. 21. Duncan PW. Stroke: physical therapy assess-
6. Tinetti ME, Ginter SF. Identifying mobility' ment and treatment. In: Contemporary man-
dysfiinctions in elderly patients: standard agement of motor control problems. Pro-
neuromuscular examination or direct assess- ceedings of the II Step Conference.
ment? JAMA 1988;259:1190-1193. Alexandria, VA: APTA, 1991:209-217.
7. Berg K. Measuring balance in the elderly: val- 22. Duncan P, Badke MB. Stroke rehabilitation:
idation of an instrment [Dissertation]. Mon- the recover},' of motor control. Chicago; Year
treal, Canada: McGill University', 1993. Book Medical Publishers, 1987.
8. Shumway-Cook A, McCollum G. Assessment 23. Stockmeyer S. Clinical decision making based
and treatment of balance disorders in the on homeostatic concepts. In: Wolf S, ed.
neurologic patient. In: Montgomery T, Con- Clinical decision making in physical therapy.
nolly B, eds. Motor control theory and prac- Philadelphia: FA Davis, 1985:79-90.
Chattanooga, TN: Chattanooga Corp.
tice. 24. Folstein MF, Folstein SE, McHugh PR.
1990:123-138. Mini-mental state: a practical method for
9. Shumvvay-Cook A, Horak F. Rehabilitation grading the cognitive states for the clinician.
strategies for patients with vestibular deficits. J Psychiatr Res 1975;12:188-198.
Neurology Clinics of North America 1990; 25. Pfeiffer E. Short portable mental status ques-
8:441-457. tionnaire. J Am Geriatr Soc 1975;23:433-
10. Shumway-Cook A, Horak F. Balance reha- 441.
bilitation in the neurologic patient: course 26. Kendall F, McCreary EK. Muscles: testing
syllabus. Seattle, NERA, 1992. and fiinction. Baltimore: Williams & Wilkins,
11. Woollacott M, Shumway-Cook A. Changes 1983.
in posture control across the life span — a sys- 27. Saunders D. Evaluation, treatment and pre-
tems approach. Phvs Ther 1990;70:799- vention of musculoskeletal disorders. Min-
807. neapolis: Viking Press, 1991.
12. Carr JH, Shepherd RB. Motor relearning 28. Magee DJ. Orthopedic physical assessment.
programme for stroke. Rock\'ille, MD: Aspen Philadelphia: WB Saunders, 1987.
Publications, 1983. 29. Kessler RM, Hertling D. Management of
13. Bobath B. Adult hemiplegia: evaluation and common musculoskeletal disorders. Philadel-
treatment. London: Wm Heinemann Medi- phia: Harper & Row, 1983.
cal Books, 1978. 30. Kaltenborn F. Mobilization of the extremity
14. Shumwav-Cook A, Horak F. Assessing the joints. Oslo: Olaf Norlis Bokhandel Univer-
influence of sensory interaction on balance. sitetsgaten, 1980.
Phys Ther 1986;66:1548-1550. 31. Leahy P. Motor control assessment. In:
15. Horak F. Clinical measurement of postural Montgomery P, Connolly BH, eds. Motor
control in adults. Phys Ther 1987; 67:1881- control and physical therapy. Hixson, TX:
1885. Chattanooga Group, 1991:69-84.
16. Nashner LM. Adaptation of human move- 32. Buchner DM, DeLateur BJ. The importance
ment to altered emironments. Trends Neu- of skeletal muscle strength to physical fiinc-
rosci 1982;5:358-361. tion in older adults. Annals of Behavioral
17. Peterka RJ, Black FO. Age-related changes in Medicine 1991;13:1-12.
human posture control: sensory organization 33. Amundsen LR. Isometric muscle strength
tests. J Vest Res 1990;1:73-85. testing with fixed-load cells. In: Amundsen
LR, ed. Muscle strength testing instrumented to therapeutic exercise in treating the lower
and non-instrumented systems. New York: extremities of hemiplegic patients. PhysTher
Churchill Livingstone, 1990:89-122. 1981;61:886-893.
34. Rogers MM. Musculoskeletal considerations 48. Baker M, Regenos E, Wolf SL, Basmajian JV.
. in production and control of movement. In: Developing strategies for biofeedback: appli-
Montgomery P, Connolly BH, eds. Motor cations in neurologically handicapped pa-
control and physical therapy. Hixson, TX: tients. Phys Ther 1977;57:402-i08.
Chattanooga Group, 1991:69-82. 49. Krebb DE. Biofeedback. In O'Sullivan S,
35. Bohannon RW. Muscle strength testing with Schmitz T, eds. Physical rehabilitation: as-
hand-held dynamometers. In: Amundsen sessment and treatment. Philadelphia: FA
LR, ed. Muscle strength testing instrumented Davis, 1988:629-645.
and non-instrumented systems. New York: 50. Bishop B. Vibration stimulation. I. Neuro-
Churchill Livingstone, 1990:69-88. physiology' of motor responses evoked by vi-
36. Bohannon RW, Andrews AW. Correlation of bratory stimulation. Phys Ther 1974;
knee extensor muscle torque and spasticit\' 54:1273.
with gait speed in patients with stroke. Arch 51. Bishop B Vibratory stimulation
. II . Vibratory
PhysMed Rehabil 1990;71:330-333. stimulation as an evaluation tool Phys Ther
37. Andrews AW. Hand held dynamometry for 1975;55:29.
measuring muscle strength. J Hum Muscle 52. Hagbarth K. Excitatory and inhibitory skin
Perform 1991;1:35-50. areas for flexorand extensor motoneurons.
38. Lynch L. Manual muscle strength testing of Acta Physiol Scand 1952;94:1-14.
the distal muscles. In: Amundsen LR, ed. 53. Eldred E, Hagbarth K. Facilitation and inhi-
Muscle strength testing instrumented and bition of gamma efFerents by stimulation of
non-instrumented systems. New York: Chur- certain skin areas. J Neurophysiol 1954;
chill Livingstone, 1990:25-68. 17:59.
39. Wilk K. Dynamic muscle strength testing. In: 54. Bobath K, Bobath B. The neurodevelop-
Amundsen LR, Muscle strength testing
ed. mental treatment. In: Man-
Scrutton D, ed.
instrumented and non-instrumented systems. agement of the motor disorders of cerebral
New York: Churchill Livingstone, 1990: palsy.Clinics in Developmental Medicine.
123-150. No 90. London: Heinemann Medical,
40. Bohannon RW, Smith MB. Interrater reli- 1984:6-18.
ability of a modified Ashworth scale of muscle 55. Merzenich MM, Kaas JH, Wall JT, Sur M,
spasticity. Phys Ther 1987;67:206-207. Nelson RJ, Felleman DJ. Progression of
41. Schmitz TJ. Coordination assessment. In: change following median nerve section in the
O'Sullivan S, Schmitz T, eds. Physical reha- of the hand in areas 3b
cortical representation
bilitation: assessment and treatment. Phila- and 1 owl and squirrel monkeys.
in adult
delphia: FA Davis, 1988:121-133. Neuroscience 1983;10:639-665.
42. Dejong RN. The neurologic examination. 56. Dejersey MC. Report on a sensory pro-
New York: Harper & Row, 1970. gramme for patients with sensory deficits.
43. Kottke FJ. Knisen's handbook of physical Aust J Phyiodier 1979;25:165-170.
medicine and rehabihtation. Philadelphia: 57. Dannenbaum RM, Dyke RW. Sensory loss in
WB Saunders, 1982. the hand after sensory stroke: therapeutic ra-
44. Shumway-Cook A, Horak FB. Vestibular re- tionale.Arch Phys Med Rehabil 1988;
habilitation: an exercise approach to manag- 69:833-839.
ing symptoms of vestibular dysfunction. Sem- 58. Carey L, Matyas T, Oke L. Sensory loss in
inars in Hearing 1989;10:196-205. stroke patients: Effective training of tactile
45. Herdman S. Vestibular rehabilitation. Phila- and proprioceptive discrimination. Arch Phys
delphia: FA Davis, 1994. Med Rehabil 1993;74:602-611.
46. Voss D, lonta M, Myers B. Proprioceptive 59. Lewis C. Phillippi L. Postural changes with
neuromuscular facilitation: patterns and tech- age and soft tissue treatment. Physical Ther-
niques. 3rd ed. Philadelphia: Harper & Row; apy Forum 1993;10:4-6.
1985. 60. Herman R. Augmented sensory feedback in
47. Binder S, Moll CB, Wolf SL. Evaluation of control of limb movement. In: Fields WS, ed.
electromyographic biofeedback as an adjunct Neural organization and its relevance to pros-
Chapter Ten Assessment AND TREATMENT OF PATIENTS WITH POSTURAL DISORDERS 235
thetics. New York: Intercontinental Medical 66. Shumway-Cook A. Unpublished observa-

Book Corp, 1973. tion.
61 Shumway-Cook A, Anson D, Haller. Postural 67. Shumway-Cook A. Retraining stability and

sway biofeedback, its effect on reestablishing mobility: translating research into clinical
stance stability in hemiplegic patients. Ai'ch practice. Presentation given at the Annual
Phys Med Rehabil 1988; 69:395-341. Meeting of the American Physical Therapy
62. Baker MP, Hudson JE, Wolf SL, A "feed- Association, Cincinnati, OH, 1993.
back" cane to improve the hemiplegic pa- 68. Horak FB, Diener HC, Nashner LM. Influ-
tient's gait. Phys Ther 1979;59:170-171. ence of central set on human postural re-
63. DeBacher G. Feedback goniometer for responses. J Neurophysiol 1989;62:841-853.
habilitation. In: Basmajian JV, ed. Biofeed- 69. Morgan MH. Ataxia and weights. Physio-
back: principles and practices for clinicians. therapy 1975;61:332-334.
Baltimore: Williams & Wilkins, 1983:359- 70. Lucy SD, Hayes KC. Postural sway profiles:
367. normal subjects and subjects with cerebellar
64. Cozean CD, Pease SW, Hubbell SL. Biofeed- ataxia. Physiotherapy Canada 1985;37:140-
back and fiinctional electric stimulation in 148.

stroke rehabilitation. Arch Phys Med Rehabil 71. Semont A, Vitte E, Freyss G. Falls in the el-
1988;69:401^05. derly: a therapeutic approach by optokinetic

65. Milezarek JJ, Kirby LM, Harrison ER, reflex stimulation. In: Vellas B, Toupet M,
MacLeod DA. Standard and four-footed Rubenstein L, Albarede JL, Christen Y, eds.
canes: their effect on the standing balance of Falls, balance and gait disorders in the elderly.
patients with hemiparesis. Arch Phys Med Paris: Elsevier, 1992:153-159.
RehabU 1993;74:281-284.
Section III
MOBILITY FUNCTIONS
Chapter 11
Control of Normal Mobility

Introduction Proactive Strategies
EssentialRequirements for Successful Non-neural Contributions to Locomotion
Locomotion Initiating Gait and Changing Speeds
Description of the Human Gait Cycle Stair Walking
Phases of the Step Cycle Ascent
Temporal Distance Factors Descent
Kinematic Description of Gait Adapting Stair Patterns to Changes in
Muscle Activation Patterns Sensory Cues
Joint Kinetics Mobility Other Than Gait
Stance Phase Transfers
Swing Phase Sit-to-Stand
Control Mechanisms for Gait Supine-to-Stand
Pattern Generators for Gait Rising-from-Bed
Descending Influences Rolling
Sensory Feedback and Adaptation of Gait Summary
Reactive Strategies for Modifyin gGait
Somatosensory Systems
Vision
Vestiijular System
Introduction consider transitions in mobility, including the

initiation of gait and transfers.
A key feature of our independence as Gait is an extraordinarily complex be-
human beings is the ability to stand up from havior. It involves the entire body and there-
abed or chair, to walk or run, and to navigate fore requires the coordination of many mus-
through ofi:en quite complex environments. cles and joints. In addition, navigating
During rehabilitation a primary goal of treat- through complex and often cluttered envi-
ment is to help patients regain as much in- ronments requires the use of multiple sensory
dependent mobility' as possible. Ofiien, re- inputs to assist in the control and adaptation
gaining mobilit}' is the primary goal of a of gait. Because of these complexities, under-
patient. This is reflected in the constantly standing both the control of normal gait and
asked question, "Will I walk again?" the mobility problems of patients with neu-
In this chapter we discuss many aspects rological impairments can seem like an over-
of mobility, including and stair
gait, transfers, whelming task.
walking, examining the contributions of the To simplify the process of understand-
individual, task, and environment to each of ing the control of gait, we describe a frame-
these tasks. We begin with a discussion of work examining gait which we have found
for
locomotion, defining the requirements for useful. The framework is built around under-
successfijl locomotion and discussing the con- standing the essential requirements of loco-
tributions of the different neural and muscu- motion and how these requirements are trans-
loskeletal systems to locomotor control. In lated into goals accomplished during the
addition, we discuss mechanisms essential for different phases of gait. Keeping in mind both
the adaptation of gait to a wide variety of task the essential requirements of gait, and the
and environmental conditions. Finally, we conditions that must be met during stance
239
240 Section III MOBILITY FUNCTIONS
and swing phases of gait to accomplish these mal subjects to meet the task requirements of
requirements, are important when examining locomotion have been well defined. Kine-
both normal and abnormal gait. matic studies describing body motions sug-
gest a similarity in movement strategies across
Essential Requirements subjects. This is consistent with intuitive ob-
FOR Successful servations that we all walk somewhat similarly.
In contrast, studies that have described the
Locomotion muscles and forces associated with gait, sug-
There major requirements for
are three gest that there is a tremendous variabiUty in
successful locomotion: {a) a basic locomotor the way these gait movements are achieved.
pattern that can move the body in the desired Thus, there appears to be a wide range of
direction, referred to as the progression muscle activation patterns used by normal
requirement; (b) the ability to maintain sta- subjects to accomplish the task requirements
bility, including the support of the body of gait.
against gravity, referred to as the stability
requirement; and (
c) the ability to adapt gait Description of the
to meet the goals of the individual and the
demands of the environment, referred to as
Human Gait Cycle
the adaptation requirement (1). These es- Let's think about the human body and
sential characteristics have been called task in- the control of gait for a moment. We have
variants, since they are minimal requirements discussed the essential requirements for nor-
for locomotion to occur (2). mal gait, that is, progression, stability, and
Human gait can be subdivided into a adaptability. The normal human perception-
stance (or support) and swing phase. Certain action system has developed elegant control
goals need to be met during each of these strategies for solving these task requirements.
phases of gait in order to achieve the three Although other gait patterns are possi-
locomotion (pro-
task invariants of successflil ble (that is, we can skip, hop, or gallop), hu-
gression, stability, and adaptability). During mans normally use symmetric alternating
a
the support phase of gait, we need to generate gait pattern, probably because it provides the
both horizontal forces against the support greatest dynamic stabflity for bipedal gait with
surface, to move the body in the desired di- minimal control demands (3). Thus, normal
rection (progression), and vertical forces, to locomotion is a bipedal gait in which the
support the body mass against gravity (stabil- limbs move in a symmetrical alternating rela-
ity). In addition, strategies used to accomplish tionship, which can be described by a phase
progression and stability must be flexible to lag of .5 (4).
accommodate changes in speed, direction, or A phase lag of .5 means that one limb
alterations in the support surface (adapta- initiates its step cycle as the opposite limb
tion). reaches the midpoint of own
cycle, as you
its
The goals to be achieved during the see in Figure 11.1. Thus, one complete cy-
if
swing phase of gait include advancement of cle is defined as the time between ipsilateral
the swing leg (progression), and repositioning foot strike (right heel contact to right heel
the limb in preparation for weight acceptance contact (Fig. 11.1), then the contralateral
(stability). Both the progression and stability limb begins its cycle midway through the ip-
goals require sufficient foot clearance so the silateral stride cycle.
toe does not drag on the supporting surface Traditionally, all descriptions of gait,
during swing. In addition, strategies used whether kinematic, EMG, or kinetic are de-
during the swing phase of gait must be suffi- scribed with reference to different aspects of
ciently flexible to allow the swing foot to the gait cycle. Thus, an understanding of the
avoid any obstacles in its path (adaptation). various phases of gait is necessary for under-
The movement strategies used by nor- standing descriptions of normal locomotion.
Chapter Eleven Control of Normal Mobility 241
Right heel contact Left toe-off Left heel contact Right toe-off Right heel contact Left toe-off
% 5 10
Double
support
<— Right single support >- Double
suppoil
-< — Left single support >h Double
support
.< Right stance phase
^
1
Figure 11.1. Temporal and distance dimensions of the gait cycle. (Adapted from Inman VT, Ralston H, ToddF. Human
walking, Baltimore: Williams & Wilkins, 1981.)
Phases of the Step Cycle contact with the ground. Single-support

phase is the period when only one foot is in
As we mentdoned earlier, the single limb contact with the ground, and in walking, this
cycle consists of two main phases: stance, consists of the time when the opposite limb is
which starts when the foot strikes the ground, in swing phase (5, 6).
and swing, which begins when the foot leaves The stance phase is often fiirther divided
the ground (Fig. 11.1). At freely chosen walk- into five subphases: [a) initial contact, {b) the
ing speeds, adults typically spend approxi- loading response (together taking up about
mately 60% of the cycle duration in stance, 10% of the step cycle, during double-support
and 40% in swing. As you see in Figure 11.1, phase), {c) mid-stance, and (rf) terminal
approximately the first and the last 10% of the stance (about 40% of stance phase, which is in
stance phase are spent in double support, that single support), and (f) pre-swing (the last
is, the period of time when both feet are in 10% of stance, in double support). The swing
phase is often divided into three subphases: than stride length. This is because you won't
initial swing, mid-swing, and terminal swing be able to note any asymmetn,' in step length
(all of which are in single support phase and ifyou evaluate only stride length.
in total make up 40% of the step cycle) (7). How fast do people normally walk?
Typically, researchers and clinicians use Normal young adults tend to walk about 1 .46
three techniques to describe different aspects m/sec or 3.26 miles per hour, and have a
of gait. Kinematic analysis allows an analysis mean cadence (step rate) of 1.9 steps/second
of joint motion; electromyography provides (112.5 steps/min) and a mean step length of
an understanding of muscle activation pat- 76.3 cm (30.05 inches) (8).
terns; and kinetic analysis describes the forces
involved in gait. For a re\dew of the technol-
og)' used to analyze gait from these various
perspectives, refer to the learning boxes on
pages 124 and 125 of Chapter 6. How do we control the speed of
walking? Let's try an experiment. Get
Temporal Distance Factors up and start walking slowly. Note your
cadence (count the number of steps/10 sec) and
estimate your step length. Now, walk as fast as you
Gait is often described with respect to
can. What happens to your step length and num-
temporal distance parameters such as velocity,
ber of steps/1 sec? As you probably found, walk-
step length, step frequency (called cadence),
ing velocity is a function of step length and step
and stride length (Fig. 11.1). Velocit\' of gait
frequency or cadence.
is defined as the average horizontal speed of
the body measured over one or more strides.
In the research literature, it is usually reported When people increase walking speed,
in the metric system (for example, cm/sec) they topically lengthen their step and increase
(7). In contrast, in the clinic, gait is usually their pace. Thus, there is a linear relationship
described in nonmetric terms (feet), and in between step length and step frequency over
either distance or time parameters. For ex- a wide range of walking speeds (9, 10). How-
ample, one might report that the patient is ever, once an upper limit to step length is
able to walk 50 feet, or the patient is able to reached, continued elevation in speed comes
walk continuously for 5 minutes. Because of from step rate.
this difference in convention between the Although normal adults have a wide
clinic and the lab, we offer information in range of walking speeds, self-selected speeds
both metric and nonmetric terms. tend to center around a small range of step
Cadence is the number of steps per unit rates, with averages of about 110 steps/min
of time, usually reported as steps per minute. for men and about 115 steps/min for women
Step length is the distance from one foot (11, 12). Preferred step rates appear to be re-
strike to the foot strike of the other foot. For lated to minimizing energ\' requirements ( 13,
example, the right step length is the distance 14). In fact, it has been found that in loco-
from the left heel to the right heel when both motion we exploit the pendular properties of
feet are in contact with the ground. Stride the leg and elastic properties of the muscles.
length is the distance covered from, for ex- Thus, in swing phase there is little energy ex-
ample, one heel-strike to the following heel- penditure. A person's
comfhtable or preferred
strike, by the same foot. Thus, right stride walking speed is at his/her point of minimal
length is defined by the distance between one energ)' expenditure. At slower or higher
right heel-strike and the next right heel-strike speeds, pendular models of gait break down,
(7). and much more energy expenditure is re-
Normal and abnormal gait are often de- quired (15).

scribed with reference to these variables. As we increase walking speed, the pro-
When performing clinical assessment, there is portion of time spent in swing and stance
an advantage to measuring step length, rather changes, with stance phase becoming pro-
gressively shorter in relation to swing (16, these large motions of the hip, and you would
17). stance/swing proportions
Finally, the see large vertical displacements of the COM.
shift from the 60/40 distribution of walking This has been called a compass jjait^ and is seen
to the 40/60 distribution as running veloci- in people who walk with
knee (22).
a stiff
ties are reached. Double support time also dis- The addition oi pelvic rotation about the
appears during running. vertical axis to motion at the hip changes the
As walking speed slows stance time in- gait pattern. Stride length increases, and
creases, while swing times remain relatively the amplitude of the sinusoidal oscillations of
constant. The double support phase of stance the COM decreases. As a result, the path
increases most. For example, double support of the COM becomes smoother and the tran-
takes up 25% of the cycle time, with step du- sition from step to step a littie less abrupt.
rations of about 1.1 sec, and 50% of the cycle With the addition oi pelvic r?7f (rotation
time when cycle duration increases to about of the pelvis about an anterior-posterior axis),
2.5 sec (16). In addition, variability increases the path of the COM
flattens even ftirther.
at lower speeds, probably due to decreased Pelvic tilt occurs during swing, when the
postural stability during the single support pe- swing hip lowers in preparation for toe-off"
riod,which also lengthens with slower speeds. (22).

Within an individual, joint angle pat- In normal gait, there is a lateral shift in
terns and EMG patterns of lower extremity the pelvis that occurs as stance is alternately
muscles are quite stable across a range of changed from one limb to another. The width
speeds, but the amplitude of muscle responses of the step contributes to the magnitude of
increases with faster speeds (12, 18, 19). In the lateral shift of the COM.
contrast, joint torque patterns appear more The addition oi knee flexion significandy
variable, though they also show gain increases improves the coordinated efficiency of gait.
as walking velocity increases. During the swing phase of gait, knee flexion
shortens the vertical length of the swing limb
Kinematic Description of Gait and allows the foot to clear the ground. Knee
flexion during stance ftirther flattens the ver-
Another way of describing normal vs. tical movements of the COM.
abnormal gait is through the kinematics of the Ankle motion also makes an important
gait cycle, that is, the movement of the joints contribution to smooth gait (Fig. 11.2). In
and segments of the body through space. Fig- particular, plantar flexion of the stance ankle
ure 1 1 .2 shows the normal movements of the allows a smooth transition from step to step
pelvis, hip, knee, and ankle in the sagittal, and contributes to the of the
initial velocit)'
frontal, and transverse planes (7). swing limb (22).

The elegant coordination of motion at Motion at the three major articulations
all the joints ensures the first requirement of within the foot is also important in the control
gait: the smooth forward progression of the of progression and stability during gait. The
center of body mass. While motion at each subtalar joint, the junction between the talus
individual joint is quite large, the coordinated and calcaneus, allows the foot to tilt medially
action of motion across all the joints results in (inversion) and laterally (eversion). Eversion
the smooth forward progression of the body, of the foot begins as part of the loading re-
with only minimal vertical displacement of the sponse, immediately after heel-strike, and
center of mass (COM) (10, 20, 21). reaches its peak by early mid-stance. Follow-
Next we consider how motion at each ing this, the motion slowly reverses, reaching
of the joints contributes to minimizing verti- the peak of inversion at the onset of pre-
cal motions of the COM. If we look at sagittal swing. During swing, the foot drifts back to
plane hip motion during gait, we see a large and then into inversion just before
neutral
amount of flexion and extension (Fig. 11.2). Subtalar motion is an essential
heel-strike.
If gait were accomplished solely through component of shock absorption during limb
these hip movements, the COM would follow loading. In addition, rigidity in this area con-
Sagittal Transverse
PLANTAR-DORSIFLEXION FOOT ROTATION
30
CO Dors
3 -10 ^^^ ~^ v\ rTVt^^

Pint
-30
KNEE FLEXION-EXTENSION
Frontal
HIPAB-ADDUCTION HIP FLEXION-EXTENSION HIP ROTATION
PELVIC OBLIQUITY PELVIC ROTATION

15
30 30
fS
Up
Ant Int
a: ^ 20 .
CD
LU
D
Down
-5
10
Post
.m Sl^ — -10
Ext
""
-30
25 50 75 25 '
50
'
% Gait Cycle
75 ' m 25*
%
so' 75 100
% Gait Cycle Gait Cycle
Figure 11.2. Normal movements of the pelvis, hip, knee, and ankle in sagittal, frontal,and transverse planes. (Adapted
from DeLuca PA, Perry |P, Ounpuu S. The fundamentals of normal walking and pathological gait. AACP & DM Inst.
Course #2. 1992.
tributes to foot stability, as weight is trans- which, when coordinated into a whole, pro-
ferred to the forefoot in terminal stance (22). vide for a smooth forward progression of the
The mid-tarsal joint is the junction of COM, with only minimal vertical displace-
the hind and forefoot. During loading, the ment. This control strateg}' reduces the en-
arch flattens quickly, allowing forefoot con- ergy cost of walking (20, 23).
tact, and thus contributes to shock absorp-
tion. Finally, motion at the metatarsophalan- Muscle Activation Patterns
geal joints allows the foot to roll over the
metatarsal heads rather than the tips of the Next, we examine the muscle responses
toes during terminal stance (22). during locomotion in terms of their fi.mction
Thus, you can see that the step cycle is at each point in the step cycle (7, 24). Despite
made up of a complex series of joint rotations the variabilit)' between subjects and condi-
LEFT FOOT-FLOOR
CONTACT k^ 1_
Vastus
lateralis
Rectus <
femorls La - i^^>
--H*^ ^#tt^ •mh ^h
Figure 11.3. Electromyographic patterns associated with the adult step cycle. (A Adapted from Murray MP, Mollinger
LA, Gardner CM, Sepic SB. Kinematic and EMC patterns during slow, free, and fast walking. Orthop Res 1 984;2:272-
J
280. 6 Adapted from Lovejoy Co. Evolution of human walking. Scientific American 1988; 5.121.)
tions in the electromyographic (EMG) pat- against the force of gravity' (stability), and (b)
terns that underlie a typical step cycle, subsequent force generation, to propel the
certain basic characteristics have been identi- body forward into the next step (progres-
fied. sion).
In general, muscles in the stance limb To accomplish the first goal, that is,
act to support the body (stability) and propel force absorption for stability, knee flexion oc-
it forward (progression). Muscle activity in curs at the initiation of stance, and there is a
the swing limb is largely confined to the be- distribution of the foot-strike impact from
ginning and end of the swing phase, since the heel contact to the foot-flat stance. At the ini-
leg swings much like a jointed pendulum un- tiation of stance, knee extensors
activity in the
der the influence of gravity (21). Typical (quadriceps) controls the small knee flexion
EMG patterns during the different phases of wave that is used to absorb the impact of foot-
the step cycle are shown in Figure 11.3. strike. Activity in the ankle dorsiflexors (an-
Remember, there are two goals to be ac- terior tibialis) decelerates the foot upon
complished during the stance phase: (a) se- touchdown, opposing and slowing the plantar
curing the stance limb against the impact flexion that results from heel-strike. Both
force of foot-strike and supporting the body muscle groups initially act to oppose the di-
rection of motion. In addition, stability' dur- Foot clearance is accomplished through

ing the stance phase involves activating exten- flexion at the hip, knee, and ankle, which re-
sor muscles at the hip, knee, and ankle, which sults in an overall shortening of the swing
keeps the body from collapsing into gravity. limb compared to the stance limb. Again, flex-
Activation of the hip extensor muscles con- ion of the hip is accomplished through acti-
trols forward motion of the head, arm, and vation of the quadriceps muscle. Flexion at
trunk segments as well. By mid-stance, the the knee is accomplished passively, since rapid
quadriceps is predominantly inactive, as are acceleration of the thigh will also produce
the pre-tibial muscles. flexion at the knee. Activation of the pre-tibial
The second goal in the stance phase is muscles produces ankle dorsiflexion late in
generating a propulsive force to keep the body swing to ensure toe clearance and in prepa-
in motion. The most common strategy used ration for the next foot- fall.
to generate propulsi\'e forces for progression
involves the concentric contraction of the Joint Kinetics
plantarflexors (gastrocnemius and soleus) at
the end of stance phase of gait. The ability of Thus far, we have examined the kine-
the body to move freely over the foot, in con- matics or movements of the body during the
junction with the concentric contraction of step cycle, and looked at the patterns of mus-
the gastrocnemius, means the COM of the cle activity in each of the phases of gait. What
body will be anterior to the supporting foot are the typical forces that these movements
by the end of stance, creating a forw'ard fall and muscle responses create during locomo-
critical to progression. The hip and knee ex- tion.' The dominant forces at a joint don't
tensors (hamstrings and quadriceps, respec- necessarily mirror the movements of the joint,
tively) may exhibit a burst of activit)' late in as you will see in the discussion that follows.
stance as a contribution to propulsion. This Determination of the forces generated

activity, however, typically is less important during the step cycle is considered a kinetic
than the observed during the force ab-
activity' analysis. The kinetic or force parameters as-
sorption phase. sociated with the normal gait pattern are less
The primary goal to be accomplished in stereotyped than the kinematic or movement
the swing phase of gait is to reposition the parameters. The active and passive muscle
limb for continued forward progression. This forces (called joint moments) that generate
requires both accelerating the limb forward locomotion are themselves quite variable.
and making sure the toe clears the ground.

Forward acceleration of the thigh in the STANCE PHASE
early swing phase is associated with a concen-
tric contraction of the quadriceps. (Fig. Remember, the goals during stance
II. 3B, part 1). By mid-swing, however, the phase include stabilizing the limb for weight
quadriceps is virtually inactive as the leg acceptance and generating propulsive forces
swings through, much like a pendulum driven for continued motion. During the stance
by an impulse force at the beginning of swing sum of
phase of the step cycle, the algebraic
phase. However, the ihopsoas contracts to aid the joint moments at the hip, knee, and ankle,
in this forward motion, as shown in Fig. called the support moment (26), is an exten-
1 1.35, parts 2 and 3. The hamstrings become sor torque (Fig. 11.4). This net extensor
active at the end of swing to slow the forward torque keeps the limb from collapsing while
rotation of the thigh, in preparation for foot- bearing weight, allowing stabilization of the
strike. (Fig. 1 1.45, part 4). Knee extension at body and thus accomplishing the stabiUt)' re-
the end of swing in preparation for loading quirements of locomotion.

the Hmb for stance phase occurs, not as the However, researchers have shown that
result of muscle activity, but as the result of people use a wide variety of force-generating
passive nonmuscular forces (25). strategies to accomplish this net extensor
Chapter Eleven CONTROL OF Normal Mobility 247
Joint angles WM22 (n = 9) and still maintain the net extensor support
moment (25-27).
Why is it important to have this flexibil-
ity in the individual contributions of joint tor-

ques to the net extensor moment? Appar-
endy, this flexibility in how torques are
generated is important to controlling balance
during gait.
David Winter, a well-known Canadian

biomechanist, and his colleagues have re-
searched gait extensively and suggest that bal-
ance during unperturbed gait is very different
from the task of balance during stance (29).
In walking, the center of gravity does not stay
within the support base of the feet and thus
the body is in a continuous state of imbalance.
WM22 Averaged The only way to prevent falling is to place the
joint moments - normal walk (n = 9)
swinging foot ahead of and lateral to the cen-
ter of gravity as it moves forward.
In addition, the mass of the head, arms,
and trunk, called the HAT segment, must be
regulated with respect to the hips, since the
HAT segment represents a large inertial load
to keep upright. Winter and colleagues pro-
pose that the dynamic balance of the HAT is
the responsibility of the hip muscles, with al-
most no involvement of the ankle muscles.

They suggest that this is because the hip has
a much smaller inerdal load to control, that
of the HAT segment,as compared to the an-
kles,which would have to control the entire

body. Thus, they propose that balance during
ongoing gait is diflferent from stance balance
control, which reUes primarily on ankle mus-
cles (29).
% of Stride They note that the hip muscles are also

involved in a separate task, that of contribut-
Figure 11.4. loint torque patterns in the hip, knee, and
ing to the extensor support moment necessary
ankle, and the net support moment associated with the
adult step cycle. (Adapted from Winter DA. Kinematic during stance, and view the muscles control-
and kinetic patterns of human gait: variability and com- ling the HAT
segment and those controlling
pensating effects. Human Movement Science 1984; the extensor support moment as two separate
3:51-76). synergies. We mentioned earlier that the net
extensor moment of the ankle, knee, and hip
joints during stance was always the same, but
torque. For example, one strategy for achiev- the individual moments were highly variable
ing a net extensor moment involves combin- fi-om stride to stride and individual to individ-
ing a dominant hip extensor moment, to ual. One reason for this variability' is to allow
counter a knee flexor moment. Alternatively, the balance control system to continuously al-
aknee and ankle extensor torque can be com- ter the anterior/posterior motor patterns on
bined to counterbalance a hip flexor torque a step-to-step basis. However, the hip balance
adjustments must be compensated for bv ap- point just after knee flexion when the ankle
propriate knee torques in order to preser\ e begins to plantarflex. The ankle joint torque
the net extensor moment essential for stance is the largest of all the torques of the lowxr
(28,29). limb and is the main contributor to the ac-
celeration of the limb into swing phase.
SWING PHASE So, in many of the pre\ious examples,
we see that the joint torque is opposite to that
The major goal during swing is to re- of the limb movement itself In other words,
position the limb, making sure the toe clears the joint torque shows us that the combined
the ground. Researchers have found that the forces may be acting to brake the movement
joint moment patterns during the swing phase or control foot faU, rather than simply accel-
are less variable than during stance phase, in- erate the limb.
dicating that adults use fairly similar force-
generating patterns to accomplish this task. Control Mechanisms
This by the large standard
is illustrated deN-i-
FOR Gait
ationsaround the mean joint torques during
stance (0 to60% of stride) as compared to the How is locomotor coordination
small standard deviations in swing (60 to achieved? What are the control mechanisms
100% of stride), shown in Figure 11.4. that ensure that the task requirements are met
For example, at normal walking speeds, for successful locomotion.' Much of the re-
early in swing, there is a flexor moment at the search examining the neural and non-neural
hip that contributes to flexion of the thigh. control mechanisms essential for locomotion
Early hip flexion is assisted by gravits', reduc- has been done with animals. It is through this
ing the need for a large flexor hip joint mo- research on locomotion in animals that sci-
ment. entists have learned about pattern formation
Once swing phase has been initiated, it in locomotion, the integration of postural
is often sustained by momentum. Then, as control to the locomotor pattern, the contri-
swing phase ends, an extensor joint torque bution of peripheral and central mechanisms
may be required to slow the thigh rotation to adaptation and modulation of gait, and the
and prepare for heelstrike (30). Thus, even role of the various senses in controlling lo-
though the thigh is still flexing, there is an comotion.
extensor torque on the thigh at this point. The following section reviews some of
What controls knee motions during the research on locomotor control in animals,
swing.' Interestingly, during swing, joint relating it to experiments examining the neu-
torque at the knee is basically used to con- ral control of locomotion in humans.
strain knee motion rather than generate mo-
tion. In early swing, an extensor torque slows Pattern Generators for Gait
knee joint flexion and contributes to reversal
of the knee joint from flexion to extension. Research in the last 25 years has greatiy
Later in swing, a flexor knee joint torque increased our understanding of the ner\'ous
slows knee extension to prepare for foot system control of the basic rh\thmic mo\'e-
placement (19, 26, 30, 31). ments underlying locomotion. Results of
At the end of swing phase and during these studies have indicated that central pat-
the initial part of stance phase, a small dorsi- tern generators within the spinal cord pla\' an
flexing ankle torque occurs at the ankle, which important role in the production of these
helps control plantarflexion at heelstrike. So movements (32, 33). A rich histor\- of re-
even though the ankle motion is one of plan- search has enhanced our understanding of the
tarflexion, the ankle joint force is a dorsiflex- neural basis of locomotion.
ion torque. In the late ISOOs, Sherrington and Mott
Moxing through the stance phase, ankle (34, 35) performed some of the first experi-
plantarflexion torque increases to a maximum ments to determine the neural control of lo-
comotion. They severed the spinal cord of an- contact, but is part of a central program. In
imals to eliminate the influence of higher addition, the spinal cat is capable of fully re-
brain centers and found that the hindlimbs cruiting motor units within the spinal cord
continued to exhibit alternating movements. when increasing gait from a walk to a gallop
In a second set of experiments, in mon- (41).
keys, they cut the sensor\' nerve roots on one Can a spinalized cat adapt the step cycle
sideof the spinal cord, eliminating senson,' in- to clear obstacles? Yes. If a glass rod touches
puts to stepping on one side of the body. the top of the cat's paw during swing phase,
They found that the monkeys didn't use the it activates a flexion response in the stimulated
deafferented limbs during v\alking. This led to leg, with simultaneous extension of the con-
the conclusion that locomotion required sen- tralateral leg. This lifts the swing leg up and
son,' input. A model of locomotor control was over the obstacle and gives postural support
created,which attributed the control of lo- in the opposite leg. Interestingly, the exact
comotion to a set of reflex chains, with the same stimulation of the dorsal surface of the
output from one phase of the step cycle acting paw during stance causes increased extension,
as a sensor}' stimulus to reflexly activate the probably to get the paw quickly out of the way
next phase. of the obstacle. Thus, the identical stimulus
Graham Brown performed an experi- to the skin activates fiinctionally separate sets
ment only a few years later (36) showing the of muscles during phases of the step
diflferent
opposite result. He found that by making bi- cycle, to appropriately compensate for differ-
lateral dorsal (sensor\') root lesions in spinal ent obstacles perturbing the movement of the
ized animals, he could see rhythmic walking paw (40). Although the spinal pattern gen-
movements. Why did the rsvo labs get difter- erators are able to produce stereot^'ped loco-
ent results.' It appears that it is because Sher- motor patterns and perform certain adaptive
rington cut only sensor)' roots on one side of functions, descending pathways from higher
the spinal cord, not both. centers and sensory feedback from the periph-
In more recent experiments, Taub and ery allow the rich variation in locomotor pat-
Berman (37) found that animals did not use terns and adaptability to task and environ-
a limb when the dorsal roots were cut on one mental conditions.
side of the body, but would begin to use the
limb again when dorsal roots on the remain- Descending Influences
ing side were sectioned. Why? Since the ani-
mal has appropriate input coming in from one Descending influences from higher
limb, and no sensation from the other, the brain centers are also important in the control
animal prefers not to use it. Interestingly, re- of locomotor activity. Much research has fo-
searchers have found that they can make ani- cused on identifying the roles of higher cen-
mals use a single deafterented limb by re- ters in controlling locomotion, through tran-
These results are the
straining the intact limb. secting the brain of animals along the neuraxis
rationale behind a therapy approach called the and observing the subsequent locomotor be-
forced-use paradigm. In this approach, havior (1). The three preparations that are
hemiplegic patients are forced to use their most often studied are the spinal, the decer-
hemiplegic arm, since the intact side is re- ebrate, and the decorticate preparations (Fig.
strained (38, 39). 11.5).
Recent studies have confirmed the re- In the spinal preparation (which can
sults of Graham Brown. These studies have be made at a level to allow the observation of
found that muscle activity' in spinalized cats is only the hind limbs or of all 4 limbs as part of
similar to that seen in normal cats walking on the preparation), one needs an external stim-
a treadmill (40), with the extensor muscles of ulus to produce locomotor behavior. This can
the knee and ankle acti\'ated prior to paw con- be either electrical or pharmacological.
tact in stance phase. This demonstrates that The decerebrate preparation leaves the
extension is not simply a reflex in response to spinal cord, brainstem, and cerebellum intact.
Brainstem ^^^^' S^^g''^
Cerebellum Cortex
Spinal
cord
Spinal Decerebrate Decorticate Intact

preparation preparation preparation system
Near normal inter/intra Improved coordination Dynamic stability Adaptable locomotor

limb rhythmic of activation patterns control system to
activation patterns Initiate reasonably meet goals of
Weight support normal goal-directed the animal in any
Functionally modulate behavior in neonatally environment
reflex action Active propulsion decorticate animal
Execute other
rhythmic movements
concurrently
Figure 11.5. The different gait capabilities of animal preparations with lesions at various points along the neuraxis.
(Adapted from Patia AE. Understanding the control of human locomotion: a prologue. In: Patia AE, ed. Adaptability of
human gait. Amsterdam: North-Holland, 1991:7.)
An area in the brainstem called the mesence- Experiments suggest that two tracts are in-
phalic locomotor repjion appears to be impor- volved in this modulation. First, the dorsal
tant in the descending control of locomotion. spinocerebellar tract is hypothesized to send
Decerebrate cats will not normally walk on a information from muscle afferents to the cer-
treadmill, but will begin to walk normally ebellum, and is phasically active during loco-
when tonic electrical stimulation is applied to motion. Second, the ventral spinocerebellar
the mesencephalic locomotor region (42). tract is hypothesized to receive information
Weight support and active propulsion are lo- from spinal neurons concerning the central
comotor characteristics seen in this prepara- pattern generator output, and to send this in-
tion. formation also to the cerebellum (44, 45).
When spinal pattern generating circuits It is also possible that the cerebellum has
are stimulated by tonic activation, they pro- an additional role in the modulation of the
duce, at best, a bad caricature of walking due step cycle. It has been hypothesized that the
to the lack of important modulating influ- cerebellum may also modulate activity, not to
ences from the brainstem and cerebellum. correct error but to alter stepping patterns
This is because normally, within each step cy- (46). For example, as an animal crosses un-
cle, the cerebellum sends modulating signals even terrain, the legs must be lifted higher or
to the brainstem that are relayed to the spinal lower depending on visual cues about the ob-
cord via the vestibulospinal, rubrospinal, and stacles encountered.The muscle response
reticulospinal pathways, which on
act direcdy patterns may be modulated through the fol-
motor neurons, to fine-tune the movements lowing steps. First, the locomotor rhythm is
according to the needs of the task (43). conveyed to the cerebellum. The cerebellum
The cerebellum also may have a ver\' im- extrapolates fonvard in time to specifi,' when
portant role in modulation of the step cvcle. the next flexion (or extension) is to occur.
Chapter Eleven Control of NORMAL MoBiLi-n' 251
The cerebellum would then facilitate de- uted to our understanding of the somatosen-
scending commands that originate from visual sor\' contributions to gait.
inputs to alter the flexion (or extension) phase
at precisely the correct time (46). Somatosensory' Systems
The decorticate preparation also leaves
the basal ganglia intact, with only the cerebral Researchers have shown that animals
cortex removed. In this preparation, an exter- that have been both spinalized and deaffer-
nal stimulus is not required to produce loco- ented can continuously generate rhythmic al-
motor behavior, and the behavior is reason- ternating contractions in muscles of all the
ably normal goal directed behavior. However, joints of the leg, with a pattern similar to that
the cortex is important in skills such as walk- Does this
seen in the normal step cycle (43).
ing over uneven terrain. mean no role
that sensor\' information plays
in the control of locomotion.' No. Though
Sensory Feedback and Adaptation these experiments have shown that animals
can still walk in the absence of sensor)' feed-
of Gait
back from the limbs, the mo\'ements show
from those in the
characteristic differences
One of the requirements of normal lo-
normal animal. These differences help us un-
comotion is the abilit\' to adapt gait to a wide
derstand the role that sensor}' input plays in
ranging set of environments. Senson,' infor-
the control of locomotion (33).
mation from all the senses is critical to our
First, sensory information from the
abilit\' to modit\' how we walk. In animals,
limbs contributes to appropriate stepping fre-
when all sensor\' information is taken away,
quency. For example, the duration of the step
stepping patterns tend to be very slow anti ste-
cycle is significantiy longer in deafferented
reot^'ped. The animal can neither maintain
cats than in a chronic spinal cat without de-
balance nor modify its stepping patterns to
afferentation (33).
make gait truly functional. Gait ataxia is a
Second, joint receptors appear to play a
common consequence among patients with
critical role in normal locomotion, with the
sensor\' loss, particularly loss of proprioceptive
position of the ipsilateral hip joint contribut-
information from the lower extremities (47).
ing to the onset of swing phase (33, 48).
There are two ways that equilibrium is
Third, cutaneous information from the
controlled during locomotion reactively — paw of the chronic spinal cat has a powerfiil
and proactixely. One uses the reacti\'e mode,
influence on the spinal pattern generator in
when, for example, there is an unexpected dis-
helping the animal navigate over obstacles, as
turbance, such as a slip or a trip. One uses the
mentioned earlier (40).
proactive mode to anticipate potential disrup-
Fourth, the Golgi tendon organ (GTO)
tions to gait and modify the way to sense and
afterents (the lb afferents) from the leg exten-
move in order to minimize the disruption.
sor muscles also can strongly influence the
Like postural control, the somatosenson,', vi-
timing of the locomotor rhythm, by inhibit-
sual, and x'estibular systems all play a role in
ing flexor burst activity' and promoting exten-
reactive and proactive control of locomotion.
sor activity. A decline in tlieir activity' at the
The next section describes how sensor^' infor-
end of the stance phase may be involved in
mation is used to modify ongoing gait.
regulating the stance to swing transition.
Note that this activity' of the GTOs is exacdy
REACTIVE STRATEGIES FOR the opposite of their activiU' when they are
MODIFYING GAIT activated passively, when the animal is at rest.
At rest, the GTOs inhibit their own muscle,
All three sensory systems, somatosen- and excite the antagonist muscles, while dur-
sory', visual, and vestibular systems, contribute ing locomotion they excite their own muscle
to reactive or feedback control of gait. Re- and inhibit antagonists (49).
search on animals and humans has contrib- Human research, similar to animal re-
search, has shown that reflexes are highly across a platform that could be perturbed at
modulated in locomotion during each phase different points in the step qxle. Results
of die step cycle, in order to adapt them func- showed that automatic postural responses
tionally to the requirements of each phase were incorporated appropriately into the dif-
(50). Stretch reflexes in the ankle extensor ferent step cycle phases (51). For example,
muscles are small in the early part of the stance postural muscle responses were activated at
phase of locomotion, since this is the time that 100 msec latencies in gastrocnemius when
the body is rotating over the foot and stretch- this muscle was stretched faster than normal
ing the ankle extensors. A large reflex at this in response to backward surface displacements
phase of the step cycle would slow or e\en pitching the body fon\'ard. This helped slow
reverse forward momentum (50). the body's rate of forward progression to re-
On the other hand, the stretch reflex is align the center of mass with the backward
large when the center of mass is in front of displaced support foot. Similarly, responses
the foot during the last part of stance phase, occurred intibialis anterior when this muscle
since this is the time when the reflex can help was shortened more slowly than normal, due
in propelling the body forward (50). This to forward surface displacements that displace
phase -appropriate modulation of the stretch the body backwards. This helped increase the
reflex is well suited to the requirements of the rate of forward progress to realign the body
task of locomotion as compared to stance. with the forward displaced foot (51).
Stretch reflex gains are fiarther reduced in run-
ning, probably because a high gain reflex re- \TSION
sponse would destabilize the gait in running.
Stretch reflex gain changes alter quickly Work with humans suggests that there
(\vithin 150 msec) as a person moves from are a variet\' of ways in which \ision modulates
stance to walking to running (50). locomotion in a feedback manner. First, visual
As was shown in research on cats, cu- flow cues help us determine our speed of lo-
taneous reflexes actually showed a complete comotion (52). Studies have shown that if
reversal from excitation to inhibition during one doubles the rate of optic flow past persons
the different phases of the step cycle. For ex- as they walk, 100% will experience that their
ample, in the first part of swing phase, when stride length has increased. In addition, about
the TA is active, the foot is in the air and little half of the subjects will perceive that the force
cutaneous input would be expected, unless e.xerted during each step is less than normal.
the foot strikes an object. If this happens, a However, other subjects will perceive that
rapid flexion would be needed to lift the foot they have nearly doubled their stepping fre-
o\er the object to prevent tripping. This is quency (53).
when the reflex is excitatoiy to the TA. How- Visual flow cues also influence the align-
ever, in the second TA burst, the foot is about ment of the body with reference to gravity
to contact the ground, which is a time when and the environment during walking (54).
a lot of cutaneous input would occur. Limb For example, when researchers tilted the
flexion wouldn't be appropriate at this time, room surrounding a treadmill on which a per-
since the limb is needed to support the body. son was running, it caused the person to in-
In addition, at this time, the reflex shows incline the trunk in the direction of the tilted
hibition of the TA (50). room to compensate for the visual illusion of
These studies ha\e shown that spinal re- body tilt in the opposite direction (54).
flexes can be appropriately integrated into dif-
ferent phases of the step c\cle to remain fianc- \'ESTIBUL.\R SYSTEM
tionally adaptive. The same outcome occurs
in the integration of compensator}' automatic An important part of controlling loco-
postural adjustments into the step cycle. Stud- motion is stabilizing the head, since it con-
ies were performed in which subjects walked tains t\\o of the most important sensors for
Chapter Eleven Control of Normal Mobility' 253
controlling motion: the vestibular and visual an icy surface, or shifting the propulsive
systems (55). The otolith organs, the saccule power fi-om ankle to hip and knee muscles
and the utricle, detect the angle of the head when climbing stairs (58).
with respect to gravit)', and the visual system Most avoidance strategies can be suc-
also provides us with the so-called visual ver- out within a step cycle. An
cessfially carried
tical. exception occurs when changing directions,

Adults appear to stabilize the heaci, and and this requires planning one step cycle in
thus gaze, by covar)'ing both pitch (forward) advance. It has been suggested that there are
rotation and vertical displacement of the head various rules associated with changing the
to give stabilit\' to the head in the sagittal placement of the foot. For example, when
plane (56, 57). The head is stabilized with a possible, step length is increased, rather than
precision (within a few degrees) that is com- shortened, and the foot is placed inside rather
patible with the efficiency of the vestibulo-oc- than outside of an obstacle, as long as the foot
ular reflex, an important mechanism for sta- doesn't need to cross the midline of the body
bilizing gaze during head movement. (58). Adapting strategies for foot placement
It has been hypothesized that during does not involve simply changing the ampli-
complex movements, like walking, postural tude of the normal locomotor patterns, but is
control is not organized from the support sur- complex and task specific.
face upward, in what is called a bottom-up
mode, but is organized in relation to the con- Non-Neural Contributions to
trol of gaze, in top-down
what is called a Locomotion
mode (55). Thus, in this mode, head move-
ments are independent from the movements So far, we have looked at neural contri-
of the trunk. It has been shown that the pro- butions to the control of locomotion, but
cess for stabilizing the head is disrupted in pa- there are also important musculoskeletal and
tients with bilateral labyrinthine lesions (55). environmental contributions. Biomechanical
analyses of locomotion in the cat have deter-
PROACTIVE STRATEGIES mined the contributions of both muscular
Proactive strategies for adapting gait fo- and nonmuscular forces to the generation of
cus on the use of sensory inputs to modif)' gait gait dynamics (59-63). This involves a t)'pe of
patterns. Proactive strategies are used to mod- kinetic analysis called inverse dynamics. To
ify and adapt gait in two different ways. First, understand more about inverse dynamics, re-
vision used proactively to identify potential

is fer to the technolog\' box on the next page.
obstacles in the environment and to navigate As we have talked about in earlier chap-
around them. Second, prediction is used to ters, nonmuscular forces, such as gravity, play
estimate the potential destabilizing effects of a role in the construction of all movement.
simultaneously performing tasks like carr)dng Wlien an inverse dynamics analysis of limb dy-
an object while walking, and anticipator)' namics is used, it is possible to determine the
modifications to the step cycle are made ac- relative importance of the muscular and non-
cordingly (58). muscular contributions. For example, during

Proactive visual control of locomotion locomotion, each segment of the cat hindlimb
has been classified into both avoidance and is subjected to a complex set of muscular and
accommodation strategies (58). Avoidance nonmuscular forces. Changes in speed lead to

strategies include («) changing the placement changes in the interactive patterns among the
of the foot, {b) increasing ground clearance to torque components (59, 63). Very often in
avoid an obstacle, (c) changing the direction cat locomotion, there are high passive exten-
of gait, when it is perceivetl that objects can't sor torques at a joint, which must be coun-
be cleared, and {d) stopping. Accommodation teracted by active flexor torques generated by
strategies involve longer term modifications, the muscles, when the animal is moving at one
such as reducing step length when walking on speed, or in one part of the step cycle. When
TECHNOLOGY BOX 1
Kinetic Analysis-Inverse Dynamics
INVERSE DYNAMICS is a process that allows researchers to calculate the joint moments of
force (torque) responsible for movement — in this case, locomotion. Researchers begin by de-
veloping a reliable model of the body using anthropometric measures such as segment masses,
center of mass, joint centers, and moments of inertia. Because these variables are difficult to
measure directly, they are usually obtained from statistical tables based on the person's height,
weight, and sex (28).
Using extremely accurate kinematic information on the limb trajectory during the step
cycle, in combination with a reliable model, researchers can calculate the torque acting on each
segment of the body. They can then partition the net torque into components due to gravity, the
mechanical interaction among segments (motion-dependent torques), and a generalized muscle
torque. This type of analysis allows researchers to assess the roles of muscular and nonmuscular
forces in the generation of the movement (27).
the speed is increased, or the animal moves to

a different part of the cycle, the passive tor- ACTIVE LEARNING MODULE
ques that must be counteracted completely
Get up and stand next to a wall,
change. How does the dialogue between the
with your shoulder touching the wall.
passive properties of the system and the neural
First try to start walking with the foot
pattern generating circuits occur.> This is still
that is next to the wall. No problem? What muscles
unclear, although the discharge from soma- did you notice contracting and relaxing? Which
tosensory receptors plays a role (61-63). way did you notice your body move in the process
What is revealed in the dynamic analysis of of preparing to take a step? Now, try to start walk-
limb movements is the intricacies of the ining, beginning with the foot that is away from the
teraction among active and passive forces. wall. What happened? Did you notice that you
The results from these studies suggest had more problems, because you couldn't easily
your weight (64)?
shift
that in normal locomotion there is a contin-
uous interaction between the central pattern
generators and descending signals. Higher
centers contribute to locomotion through Research studies confirm what you no
feedforward modulation of patterns in re- doubt noticed fi'om your own experiment: the
sponse to the goals of the individual and to initiation of gait from quiet stance begins with
environmental demands. As noted briefly the relaxation of specific postural muscles, the
above, sensory inputs are also cridcal for feed- gastrocnemius and the soleus (65, 66). In
back and feedforward modulation of loco- fact, the initiation of gait has the appearance
motor activit}' in order to adapt it to changing of a simple forward fall and regaining of one's
environmental conditions. balance by taking a step. This reduction in the
activation of the gastrocnemius and soleus is
Initiating Gait and followed by activation of the tibialis anterior,
Changing Speeds which assists dorsiflexion and moves the

COM forward in preparation for toe-off. But,
How do we initiate walking.' Before we as you noticed, and as recent research on gait
describe the initiation of gait, let's do an ex- confirms, the initiation of gait is more than a
periment. simple fall.
Chapter Eleven Control of Nojimal Mobility 255
Center of pressure What neural patterns are correlated with

these shifts in center of pressure? As the center
of pressure moves posteriorly and toward the
swing limb, both limbs are stabilized against
backward sway by acdvation of anterior leg
and thigh muscles, the tibialis anterior (TA)
and the quadriceps. Subsequent activation of
the TA then causes dorsiflexion in the stance
ankle, pulling the lower leg over the foot, as

the body moves forward in preparation for
toe-off. Anterior thigh muscles are activated
to keep the knee from flexing so that the leg
rotates forward as a unit. Activation of hip ab-
ductors counters lateral tilt of the pelvis to-
ward the swing limb side as this limb is un-
loaded. Also, activation of the peroneals
stabilizes the stance ankle. After toe-off, the
gastrocnemius and hamstrings muscles in the
Figure 1 1 .6. Trajectory of the center of pressure during
tfie initiation of gait from a balanced, symmetric stance. stance leg are used for propelling the body
movement, the center of pressure is located mid-
Prior to forward (66, 67). How long after initiation
way between the feet. (Adapted from Mann RA, Hagy )L, does it take to reach a steady velocity in
White V, Liddell D. The initiation of gait. J Bone Joint
gait? Steady state is reached within one (68)
Surg 1979;61-A:232-239.)
to three steps (67, 69) depending on the
magnitude of the velocity one is trying to
In tracing the center of pressure during achieve.
the initiation of gait in normal adults, the fol-
lowing sequence of events is evident. Prior to

Stair- Walking
movement onset, the center of pressure is po- Understanding the sensorv' and motor
sitioned just posterior to the ankle and mid- requirements associated with stair-walking is
way between both feet (Fig. 1 1.6). As the per- critical to retraining this skill. Stairs represent
son begins to move, the center of pressure a significant hazard even among the nondis-
first moves posteriorly and laterally toward the abled population. Stair- walking accounts for
nvinjj limb and then shifts toward the stance the largest percentage of falls occurring in
limb and forward (67). public places, with four out of five falls oc-
Movement of the center of pressure to- curring during stair descent (70).
ward the stance limb occurs simultaneously Stair-walking is similar to level-walking
with hip and knee flexion and ankle dorsiflex- in that it involves stereotypical reciprocal al-
ion as the swing limb prepares for toe-off. ternating movements of the lower limbs (71 ).
Then the center of pressure moves quickly to- Like locomotion, successfijl negotiation of
ward the stance limb. Toe-off of the swing stairs has three requirements: the generation
limb occurs with the center of pressure shift- of primarily concentric forces to propel the
ing from lateral to forward movement over body up stairs, or eccentric forces to control
the stance foot. Why do we first shift the cen- the body's descent down stairs (progression),
ter of pressure toward the swing limb when while controlling the center of mass within a
we initiate gait.' It has been hypothesized constantly changing base of support (stabil-
that this is a strategy for setting the center of and the capacity to adapt strategies used
it}');
mass motion, which would allow the

in for progression and stability' to accommodate
momentum of the center of mass to help changes in stair environment, such as height,
create the loss of balance leading to the first width, and the presence or absence of railings
step (67). (adaptation) (72).
256 Section III MOBILITY' FUNCTIONS
Sensor)' information is important for dorsiflexing the foot, and activation of the
controlling the body's position in space (sta- hamstrings, which flex the knee. The rectus
bility'), and to identiR- critical aspects of the femoris contracts eccentrically to reverse this
stair environment so that appropriate mo\e- motion by mid-swing. The swing leg is
ment programmed (adapta-
strategies can be brought up and forward through activation of
tion). Researchers have shown that normal the hip flexors of the swing leg, and motion
subjects change movement strategies used for of the contralateral stance leg. Final foot
negotiating stairs when sensor\' cues about placement is controlled by the hip extensors
stair characteristics are altered (70, 71). and ankle foot dorsiflexors (72).
Similar to gait, stair climbing has been
di\ided into two phases, a stance phase lasting
Descent
approximately 64% of the fijll c\'cle, and a
swing phase lasting 36% of the cycle. In ad-
dition, each phase of stair-walking has been Walking upstairs is accomplished
further subdivided to reflect the objectives through concentric contractions of the rectus
that need to be achieved during each phase. femoris, vastus lateralis, soleus, and medial
gastrocnemius. In contrast, walking down
Ascent stairs is achieved through eccentric contrac-
tions of these same muscles, which work to

During ascent, the stance phase is sub- control the body with respect to the force of
divided into weight acceptance, pull-up, and gravirs'. The stance phase of stair descent is
forward continuance, while swing is di\'ided subdi\ided into weight acceptance, forward
into foot clearance and foot placement stages. continuance, and controUed lowering, while
During stance, weight acceptance is ini- swing has two phases: leg pull-through and
tiated with the middle to front portion of the preparation for foot placement (71, 72).
foot. Pull-up occurs because of extensor ac- Weight acceptance phase is character-
tivity' at the knee and ankle, primarily concen- ized bv absorption of energ\' at the ankle and
tric contractions of the vastus lateralis and so- knee through the eccentric contraction of the
leus muscles. Stair ascent differs from level triceps surae, rectus femoris, and vastus late-
walking in two ways: (a) forces needed to ac- ralis. Energ)' absorption during this phase is
complish ascent are two times greater than critical, since forces as much as two times
those needed to control level gait, and ( b) the body weight ha\'e been recorded when the
knee extensors generate most of the energ)' to swing limb first contacts the stair. Activation
move the body forward during stair ascent of gastrocnemius prior to stair contact is re-
(72). Finally, during the forward continuance sponsible for cushioning the landing (71 ).
phase of stance, the ankle generates forward The forward continuance phase reflects
and lift forces; however, ankle force is not the the forward motion of the body, and precedes
main source of power behind forward pro- the controUed lowering phase of stance. Low-
gression in stair- walking. ering of the body is controlled primarily by
In controlling balance during stair as- the eccentric contraction of the quadriceps
cent, the greatest instabilit}' comes with muscles, and to a lesser degree, tiie eccentric
contralateral toe-oft', when the ipsilateral leg contraction of the soleus muscle.
takes the total body weight, and the hip, knee, During swing, the leg is pulled through,
and ankle joints are flexed (72). due of the hip flexor muscles.
to activation
The objectives of the swing phase of Howe\'er, by mid-swing, flexion of the hip
stair climbing are similar to level gait, and in- and knee is reversed, and all three joints ex-
clude foot clearance and placing the foot ap- tend in preparation for foot placement. Con-
propriately so weight can be accepted for the tactis made with the lateral border of the foot,
next stance phase. Foot clearance is achieved and is associated with tibialis anterior and gas-
through activation of the tibialis anterior. trocnemius activity' prior to foot contact.
Chapter Eleven Control of Normal Mobilit\' 257
Adapting Stair-Walking Patterns the recovery of these diverse mobility skills.
This requires an understanding of: («) the es-

to Changes in Sensory Cues
sential characteristics of the task, (b) the sen-
sory motor strategies that normal individuals
Researchers have shown that neurolog-
typically use to accomplish the task, and (c)
ically intact people adapt the movement strat-
the adaptations required for changing envi-
egies they use for going up and down stairs in
ronmental characteristics.
response to changes in sensory information
about the task. Thus, when normal subjects
All mobility tasks share in common
three essential task requirements: motion in a
wear large collars obstructing their view of the
desired direction (progression), postural con-
stairs, anticipator}' activation of the gastroc-
trol (stability), and the ability to adapt to
nemius prior to foot contact is reduced. This
changing task and environmental conditions
anticipator)' activity' is fiirther reduced when
the subject
(adaptation). The following sections briefly
is blindfolded (71). In this study,
subjects still managed a soft landing by chang-
review some of the research on these other
aspects of mobility ftmction. As you will see,
ing the control strategv' used to descend stairs.
compared to the tremendous number of stud-
Subjects moved slower, protracting swing
ies on normal gait, there have been relatively
time, and using the stance limb to control the
few studies examining these other aspects of
landing.
mobility fimction.
Foot clearance and placement are criti-
cal aspects of movement strategies used to
safely descend stairs. Good visual information
Transfers
about stair height is critical. When normal
Transfers represent an important aspect
subjects wear blurred vision lenses and are un-
able to clearly define the edge of the step, they
of mobility function. One cannot walkif one
cannot get out of a chair from a bed.

or rise
slow down and modify mox'cment strategies
so that foot clearance is increased and the foot
Inability to safely and independendy change
positions represents a great hindrance to the
is placed fiirther back on the step to ensure a
recovery of normal mobility.
larger margin of safety (70). Thus, informa-
Several researchers have studied transfer
tion from the visual system about the step
height appears to be necessary for optimal
skills from a biomechanical perspective. As a
programming of movement result, we know quite a bit about typical

strategies used to
negotiate stairs.
movement strategies used by neurologically
intact adults when performing diese tasks.
Howexer, use of a biomechanical approach
Mobility Other Than has provided us with little information about
Gait the perceptual strategies associated with these
various tasks. In addition, because most often
Although mobility is often thought of research subjects are constrained to carr)' out
solely in relationship to gait or locomotion, the task in a unified way, we have litde insight
there are many other aspects of mobility' that into ways in which sensor)' and movement
are essential to independence in daily life ac- strategies are modified in response to chang-
The ability to change positions,
tivities. ing task and environmental demands.
whether moving from sit to stand, rolling, ris-
ing from a bed, or moving from one chair to Sit-to-Stand
another, fundamental part of mobility'.
is a
These various types of mobilit)' activities are Sit-to-stand (STS) behaviors emerge
often grouped together and referred to as from an interaction among characteristics of
transfer tasks. the task, the individual, and constraints im-
Retraining motor function in the pa- posed by the environment. While the bio-
tient with a neurological impairment includes mechanics of STS have been described, there
are many important questions that have not ferent phases, either two, three or four, de-
yet been studied by motor control research- pending on the researcher. Each phase has its
ers. For example, how do the movements in- own unique movement and stability require-
volved in STS var)' as a flinction of the speed ments. A four-phase model of STS task is
of the task, the characteristics of the support, shown in Figure 11.7 (73, 74). This figure
including height of the chair, the compliance alsoshows the kinematic and EMG data for a
of the seat, or the presence or absence of hand normal subject completing this task.
rests? In addition, do the requirements of the The first phase, called the weight shift:,
task var)' depending on the nature of the task or flexion momentum stage, begins with the
immediately following? That is, do we stand generation of forward momentum of the up-
up differently if we are intending to walk in- per body through flexion of the trunk. The
stead of stand still? What perceptual infor- body is quite stable during this phase since the
mation is essential to establishing efficient center of mass (COM), though moving for-
movement strategies when performing STS? ward, is still within the base of support of the
The essential characteristics of the STS chair seat and the feet. Muscle activity in-
task include: (a) generating sufficient joint cludes activation of the erector spinae, which
torque needed to rise (progression), (2) en- contract eccentrically to control forward mo-
suring stability' by mo\'ing the center of mass tion of the trunk (73, 74).
from one base of support (the chair) to a base Phase 2 begins as the buttocks leave the
of support defined solely by the feet (stabil- seat,and involves the transfer of momentum
ity'), and (c) the abilit\' to modif\' movement from the upper body to the total bociy, allow-
strategies used to achieve these goals depending lift of the body (74). Phase 2 involves
ing on the environmental constraints, such as both horizontal and vertical motion of the
chair height, the presence of arm rests, and body, and is considered a critical transition
the softness of the chair (adaptation). phase. Stabilit)' requirements are precise since
The STS task has been divided into dif- it is during this phase that the COM of the
Figure 11.7. Diagram of the four phases of

the sit-to-stand movement, showing the kine-
matic and EMG patterns associated with each
phase. (Adapted from Millington PJ, Mykle-
bust BM, Shambes CM. Biomechanical anal-
ysis of the sit-to-stand motion in elderly per-
sons. Arch Phys Med Rehabil 1992;73:609-

617.)
20 40 60 100
Percent of Motion
body moves from within the base of support tory of the COM, and (c) concentric contrac-
of the chair to that of the feet. The body is tion of hip and knee muscles to generate ver-
inherendy unstable during this phase because tical propulsive forces that lift the body (74).
the COJVI is located far from the center of Accomplishing STS using a momen-
force. Because the body has developed mo- tum-transfer strategy requires a trade-oft" be-
mentum prior to lift-off, vertical rise of the tween stability and force requirements. The
body can be achieved with litde lower extrem- generation and transfer of momentum be-
muscle force (74). Muscle activity in this
it)' tween the upper body and total body reduces
phase is characterized by coactivation of hip the requirement for lower extremity force be-
and knee extensors, as you see in Figure 11.7. cause the body is already in motion as it be-
Phase 3 of the STS task is referred to as gins to lift. On the other hand, the body is in
the lift or extension phase, and is character- a precarious state of balance during the tran-
ized by extension at the hips and knees. The sition stage when momentum is transferred.
goal in this phase is primarily to move the An alternative strategy that ensures
body vertically; stability requirements are less greater stability but requires greater amount
than in phase 2 since the COM is well within of force to achieve lift-oft" includes flexing the
the base of support of the feet (74). trunk sufhcientiy to bring the COM well
The final phase of STS is the stabiliza- within the base of support of the feet ^r?or to
tion phase, and is that period following com- lift-off However, the body has zero momen-
plete extension, when task-dependent motion tum at lift-oft'. This strategy has been referred
is complete and body stability in the vertical to as a zero-momentum strategy, and requires
position is achieved. the generation of larger lower extremity
STS requires the generation of propul- forces in order to lift the body to vertical (74).
sive impulse forces in both the horizontal and Another common strategy used by
vertical directions. However, the horizontal many older adults and people with neurolog-
propulsive force responsible for moving the ical impairments involves the use of armrests
COM anterior over the base of support of the to assist in STS. Use of the arms assists in both
foot must change into a braking impulse to the stability and force generation require-
bring the body to a stop. Braking the hori- ments of the STS task.
zontal impulse begins even before lift-oft Understanding the dift'erent strategies
from the Thus, there appears to be a pre-
seat. that can be used to accomplish STS, including
programmed relationship between the gen- the trade-oft's between force and stability, will
eration and braking offerees for the STS task. help the therapist when retraining STS in the
Without this coordination between propul- patient with a neurological deficit. For ex-
sive and braking forces, the person could eas- ample, the zero-momentum strategy may be
ily fall forward upon achieving the vertical po- more appropriate to use with a patient with
sition. cerebellar pathology who has no difficult)'
Horizontal displacement of the COM with force generation, but who has a major
appears to be constant despite changes in the problem with controlling stability'. On the
speed of STS (73). Controlling the horizontal other hand, the patient with hemiparesis, who
trajectory' of the COM
is probably the invar- is ver)' weak, may need to rely more on a mo-
iant feature controlled in STS to ensure that mentum strategy to achieve the vertical posi-
stability is maintained during vertical rise of tion. The frail elderly person who is both weak
the body. and unstable may need to rely on armrests to
This strategy could be referred to as a accomplish STS.
momentum-transfer strategy, and its use re-
quires (a) adequate strength and coordina- Supine-to-Stand
tion to generate upper body movement prior
to lift-off, (b) the ability to eccentrically con- The ability' to assume a standing posi-
tract trunk and hip muscles, in order to apply tion from supine is an important milestone in
braking forces to slow the horizontal trajec- mobility skills. This skill is taught to a wide
range of patients with neurological impair- stand, the body is dixided into three compo-
ments, from young children with develop- nents, upper extremities, lower extremities,
mental disabilities first learning to stand and and axial, which includes trunk and head.
walk, to frail older people prone to fall. The Movement strategies are then described in re-
movement strategies used by normal individ- lationship to the various combinations of
uals moving from supine -to -stand have been movement patterns within each of these seg-
studied by a number of researchers. An im- ments. The research on young adults suggests
portant theoretical question addressed by that the most common pattern used involves
these researchers relates to whether rising to symmetrical movement
patterns of the trunk
stand from supine follows a developmental and extremities, and the use of a symmetrical
progression, and whether by the age of 4 or squat to achieve the vertical position (Fig.
5 years the mature, or adult-Uke, form 11.8^). However, only one-fourth of the
emerges and remains throughout life (75). subjects studied used this strategy.
Researchers have studied supine-to- The second most common movement
stand movement 4
strategies in children, ages pattern involved asymmetric squat on arising
to 7 years, and young adults, ages 20 to 35 ( 1 1.8iJ), while the third most common strat-
years (76). These researchers found that while egy involved asymmetric use of the upper ex-
there was a slight tendency towards age-spe- tremities, a partial rotation of the trunk, and
cific strategies for moving supine-to-stance, assumption of stance using a half-kneel posi-
there was also great variability among subjects tion (11. 8C).
of the same age. Their findings do not appear Additional studies have characterized
to support the traditional assumption of a sin- movement patterns used to rise from supine
gle mature supine-to-stance pattern, which in middle-aged adults, ages 30 to 39 years,
emerges after the age of 5 years. and found some differences in movement
The three most common movement strategies compared to younger adults (77).
strategies for moving from supine-to-stand In addition, this study looked at the effect of
are shown in Figure 11.8. When analyzing physical activity levels on strategies used to
strategies used for moving from supine-to- stand up. Results from the study found that
Figure 11.8. Three most common movement strategies identified among young adults for moving from supine to
VanSant AF. Rising from a supine position
stand. (Adafjted from to erect stance: description of adult movement and a
developmental hypothesis. Phys Ther 1988;68:185-192.)
Chapter Eleven Control of Normal Mobility' 261
strategies used to stand up are influenced by

life-style factors, including level of physical ac-
tivity.
Many factors probably contribute to de-

termining the type of movement strategy used
to move from supine-to-stance. Traditionally,
nervous system maturation, specifically the
maturation of the neck-on-body righting re-
actions and body-on-body righting reactions,

were considered the most significant factors
affecting the emergence of a developmentally
mature supine-to-stance strategy. Hovi'ever, a
switch from an asymmetric rotation to sym-
metric sit-up strategy may be constrained by
the ability to generate sufficient abdominal
and hip flexor strength.
Developmental changes in moving from
supine-to-stance are considered fijrther in the
chapter on age-related aspects of mobility.
RISING-FROM-BED
Clinicians are often called upon to help

patients relearn the task of getting out of bed.
In therapeutic texts on retraining motor con-
trol in the patient with neurological impair-
ments, therapists are instructed to teach pa-
tients to move from supine to side-lying, then
to push up to and from
a sitting position
there, to stand up. These instructions are
based on the assumption that this pattern rep-
resents that t)'pically used to rise from a bed
(78, 79).
To test these assumptions, researchers
examined movement patterns used by young
adults to rise from a bed (80, 81 ). These stud-
ies report that movement patterns used by
nondisabled people to rise from a bed are ex-

tremely variable. Eighty-nine patterns were
found among 60 subjects! In fact, no subject
used the same strategy consistendy in 10 trials
of getting out of bed.
Figure 1 1 .9 shows one of the most com-
mon strategies used by young adults to rise
from a bed. Essendal components of the strat-
egy include pushing with the arms (or grasp-
Figure 1 1 .9. Most common movement strategy used by
ing the side of the bed and then pushing with
young adults for getting out of bed. (Adapted from Ford-
the arms), flexing the head and trunk, pushing
Smith CD, VanSant AF. Age differences in movement
into a partial sit position, and rolling up into patterns used to rise from a bed in subjects in the third
stance. Another common strategy found was through fifth decades of age. Phys Ther 1992;73:305.)
Arm Pittem • Lift and reach above shoulder level
Haad-Think Pattern • Shoulder girdle leads
Lag Pattern • Unilateral lift
Figure 11.10. Most common movement strategy used by young adults when rolling from supine to prone. (Adapted
from RIchter RR, VanSant AF, Newton RA. Description of adult rolling movements and hypothesis of developmental
sequences. Phys Ther 1 989; 69:63-71 .)
a push-off pattern with the arms, rolling to the between movement and stability require-
side and coming to a symmetrical sitting po- ments in the different strategies. For example,
sition prior to standing up. in the roll-off strateg\', is motion achieved
While the authors of this study have not with greater efficiency at the expense of sta-
specifically stated the essential features of this bilit},'? Alternatively, the come-to-sit pattern
task, its similarity to the STS task suggests may require more force to keep the body in
they share the same invariant characteristics. motion, but stability may be inherendy
These include (a) the need to generate mo- greater.
mentum to move the body to vertical, (b) sta- This research demonstrates the tremen-
bility requirements for controlling the center dous variability of movement strategies used
of mass as it changes from within the support by neurologically intact subjects when getting
base defined by the horizontal body to that out of bed. It suggests the importance of
defined by the buttocks and feet, and finally helping patients with neurological impair-
to a base of support defined solely bv the feet; ments to learn a variety of approaches to get-
and (c) the ability to adapt how one moves to ting out of bed.
the characteristics of the environment.
In trying to better understand why peo- Rolling
ple move as they do, and in preparation for
understanding why patients move as they do, Rolling is an important part of bed mo-
it might be helpful to reexamine descriptions bilit)' skills and an essential part of many other
of movement strategies used to rise from a taskssuch as rising from bed (82 ). Movement
bed in light of these essential tasks character- strategies used by nonimpaired adults to roll
istics. In doing so, it might be possible to de- from supine to prone are ver)' variable. Figure
termine common features across dixerse strat- 11.10 shows one of the most common move-
egies that are successfiil in accomplishing ment patterns used by adults to roll from su-
invariant requirements of the task. It would pine to prone (82). Essential features of this
also be possible to examine some trade-offs strategy' include a lift-and-reach arms pattern,
Chapter Eleven CONTROL OF NORMAL MoBILin* 263
with the shoulder girdle initiating motion of used during the swing phase of gait
strategies
the head and trunk, and a unilateral lift of the must be sufficiently flexible in order to allow
the swing foot to avoid any obstacles In Its
leg.
path (adaptation).
A common assumption in the therapeu-
5. Gait Is often described with respect to tem-
tic literature is that rotation between the
poral distance parameters such as velocity,
shoulders and pelvis is an invariant character-
step length, step frequency (called cadence),
istic in rolling patterns used by normal aciults
and stride length. In addition, gait Is de-
(79); however, in this study on rolling, many scribed with reference to changes In joint an-
of the adults tested did not show this pattern. gles (kinematics), muscle activation patterns
Similar to the findings from studies on rising (EMG), and the forces used to control gait
from a bed, the great variability used by nor- (kinetics).
mal subjects to move from supine to prone 6. Many neural and non-neural elements work
suggests that therapists may use greater free- together In the control of gait. Though spinal
dom in retraining movement strategies used pattern generators are able to produce ste-
reotyped locomotor patterns and perform

by patients with neurological impairments.
certain adaptive functions, descending path-
Clearly, there is no ONE correct way to ac-
ways from higher centers and sensory feed-
complish this movement.
back from the periphery allow the rich vari-
ation In locomotor patterns and adaptability
to task and environmental conditions.
Summary 7. One normal locomo-
of the requirements of
tion Is theabllity toadapt gaittoa wide rang-
1 There are three major requirements for suc- ing set of environments, and this Involves us-
cessful locomotion: (a) progression, defined ing sensory Information from all the senses
as the ability to generate a basic locomotor both reactively and proactlvely.
pattern that can move the body in the desired 8. An Important part of controlling locomotion
direction, (b) stability, defined as the ability Is stabilizing the head, since It contains two
to support and control the body against grav- of the most important sensors for controlling
ity, and (c) adaptability, defined as the ability motion: the vestibular and visual systems. In
meet the individual's goals
to adapt gait to neurologlcally Intact adults, the head Is sta-
and the demands of the environment. bilized with great precision, allowing gaze
2. Normal locomotion is a bipedal gait in to be stabilized through the vestlbulo-ocular
which the limbs move in a symmetrical al- reflex.
ternating relationship. Gait is divided into a 9. Stair-walking Is similar to level walking In

stance and swing phase, each of which has that It involves stereotypical reciprocal alter-
its own intrinsic requirements. nating movements of the lower limbs and
3. During the support phase of gait, horizontal has three requirements: the generation of pri-
forces are generated against the support sur- marily concentric forces to propel the body
face to move the body in the desired direc- up stairs, or eccentric forces to control the
tion (progression), while vertical forces sup- body's descent down stairs (progression),
port the body mass against gravity (stability). while controlling the center of mass within
In addition, strategies used to accomplish a constantly changing base of support (sta-
both progression and stability must be flexi- and the capacity to adapt strategies
bility);
ble In order to accommodate changes in used for progression and stability to accom-
speed, direction, or alterations in the support modate changes In stair environment, such
surface (adaptation). as height, width, and the presence or ab-
4. The goals to be achieved during the swing sence of railings (adaptation).
phase of gait Include advancement of the 10. Although mobility Is often thought of In re-
swing leg (progression), and repositioning lationship to gait, many other aspects of mo-
the limb In preparation for weight accep- bility are essential to Independence. These
tance (stability). Both the progression and Include the ability to move from sit to stand,
stability goals require sufficient foot clear- rolling, rising from a bed, or moving from
ance, so the toe does not drag on the sup- one chair to another. These skills are referred
porting surface during swing. In addition. to as transfer tasks.
11. Transfer tasks are similar to locomotion in 10. InmanVT, Ralston H, Todd F. Human walk-
that they share common task requirements: ing. Baltimore: Williams & Wilkins. 1981.
motion in a desired direction (progression), 1 1 Finley FR, Cody KA. Locomotive character-
postural control (stability), and the ability to istics of urban pedestrians. Arch Phys Med
adapt to changing task and environmental Rehabil 1970;51:423-i26.
conditions (adaptation). Researchers have 12. Murray MP, Mollinger LA, Gardner GM, Se-
found great variability in the types of move- pic SB. Kinematic and EMG patterns during
ment strategies used by neurological ly intact slow, free, and fast walking. J Orthop Res
young adults when performing transfer tasks. 1984;2:272-280.
12. Understanding the stability and strength re- 13. Ralston HJ. Energetics of human walking. In:
quirements for different types of strategies Herman RM, Grillner S, Stein PSG, Stuart
used to accomplish transfer tasks has impor- DG, eds. Neural control of locomotion. New
tant implications for retraining these skills in York: Plenum, 1976:77-98.
neurologically impaired patients with differ- 14. Zarrugh MY, Todd FN, Ralston HJ. Opti-
ent types of motor constraints. mization of energy expenditure during level
walking. Eur J Appl Physiol 1974;33:293-
306.
15. Mochon S, McMahon TA. Ballistic walking.
References JBiomech 1980;13:49-57.

16. Herman R, Wirta R, Bampton S, Finley FR,
1. Patla AE. Understanding the control of hu- Human solutions for locomotion I. Single
man locomotion: a prologue. In: Patla AE, limb analysis. In: Herman R, Grillner S, Stein
ed. Adaptability of human gait. Amsterdam: P, Stuart D, of locomo-
eds. Neural control
North-Holland, 1991:3-17. tion. New York:
Plenum, 1976:13^9.
2. Das P, McCollujn G. Invariant structure in 17. Murray MP. Gait as a total pattern of move-
locomotion. Neuroscience I988;25:1023- ment. Am J Phys Med 1967;46:290-333.
1034. 18. Murray MP, Kory RC, Clarkson BH, Sepic
3. Raibert M. Symmetry in running. Science SB. Comparison of free and fast speed walk-
1986;231:1292-1294. ing patterns of normal men. Am J Phys Med
4. Grillner S. Control of locomotion in bipeds, 1966;45:8-24.
tetrapods, and fish. In: Brooks VB, ed. Hand- 19. Winter DA. Biomechanical motor patterns in
book of physiology — the nervous system. II. normal walking. Journal of Motor Behavior
Motor control. Baltimore: Williams & Wil- 1983;15:302-330.
kins, 1981:1179-1236. 20. Saunders JBdeCM, Inman VT, Eberhart
5. Murray MP, Mollinger LA, Gardner GM, Se- HD. The major determinants in normal and
pic SB. Kinematic and EMG patterns during pathological gait. J Bone Joint Surg 1953;35-
slow, free, and fast walking. J Orthop Res A:543-558,
1984;2:272-280. 21. McMahon TA. Muscles, reflexes and loco-
6. Rosenrot P, Wall JC, Charteris J. The rela- motion. Princeton, NJ: Princeton University
tionship between velocity, stride time, sup- Press. 1984.
port time and swing time during normal 22. Perry J. Gait analysis: normal and pathologi-
walking. Journal of Human Movement Stud- cal fianction. Thorofare, NJ: Slack, 1992.
ies 1980;6:323-335. 23. Eberhart HD. Physical principles of loco-
7. DeLuca PA, Perry JP, Ounpuu S. The fun- motion. In: Herman RM, Grillner S, Stein
damentals of normal walking and pathologi- PSG, Stuart DG, eds. Neural control of lo-
cal gait. AACP & DM Course #2. 1992.
Inst. comotion. New York: Plenum, 1976:1-11.
8. Craik R. Changes in locomotion in the aging 24. Basmajian JV, Deluca CJ. Muscles alive: their
adult. In:WooUacott M, Shumway-Cook A, fijnctions revealed by electromyography. 5th
eds. Development of posture and gait. ed. Baltimore:Williams & Wilkins, 1985.
Charleston, SC: Univ. of South Carolina 25. Winter DA. Kinematic and kinetic patterns of
Press, 1989:176-201. human gait: variability and compensating ef-
9. Grieve DW. Gait patterns and the speed of fects. Human Movement Science 1984;3:51-
walking. Biomedical engineering 1968; 76,

3:119-122. 26. Winter DA. Overall principle of lower limb
Chapter Eleven Control of Normal Mobilht 265
support during stance phase of gait. J Bio- dependent reflex reversal during walking in
mech 1980;13:923-927. chronic spinal cats. Brain Res 1977;85:12I-
27. Winter DA, Patla AE, Frank, JS, Walt SE. 139.
Biomechanical walking pattern changes in the 41. Smith JL, Smith LA, Dahms KL. Motor ca-
fit and healthy elderly. Phys Ther 1990; pacities of the chronic spinal cat: recruitment
70:340-347. of slow and fast extensors of the ankle. Neu-
28. Winter DA. Biomechanics and motor control roscience Abstracts 1979;5:387.
of human movement. New York: John Wiley 42. Shik ML, Severin FV, Orlovsky GN. Control
& Sons, 1990:80-84. of walking and running by means of electrical
29. Winter DA, iVIcFadyen BJ, Dickey JP. Adapt- stimulationof the mid-brain. Biophysics
ability of the CNS in human walking. In: 1966;11:756-765.
Patla AE, ed. Adaptabilit)' of human gait. 43. Grillner S, Zangger P. On the central gener-
Amsterdam: Elsevier, 1991:127-144. ation of locomotion in the low spinal cat. Exp
30. WooUacott MH, Jensen J. Stance and loco- Brain Res 1979;34:241-261.
motion. In: S Keele, H Heuer, eds. Hand- 44. Arshavsky Yu I, Berkinbht MB, Fukson OI,
book of motor skills. In press. Gelfand IM, Orlovsky GN. Recordings of
31 Cavanagh PR, Gregor RJ. Knee joint torques neurones of the dorsal spinocerebellar tract
during the swing phase of normal treadmill during evoked locomotion Brain Res 1972;
walking. J Biomech 1975;8:337-344. 43:272-275.
32. Grillner S. Lxjcomotion in the spinal cat. In: 45. Arshavsky Yu I, Berkinblit MB, Gelfand IM,
Stein RB, Pearson KG, Smith RS, Redford Orlovsky GN, Fukson OI. Activit)' of the
JB, eds. Control of posture and locomotion. neurones of the ventral spinocerebellar tract
New York: Plenum, 1973:515-535. during locomotion. Biophysics 1972;

33. Smith JL. Programming of stereotyped limb 17:926-935.
movements by spinal generators. In: Stel- 46. Keele S, Ivry R. Does the cerebellum provide
mach GE, Requin J, eds. Tutorials in motor a common computation for diverse tasks? A
behavior. Amsterdam: North-Holland. timing hypothesis. In: Diamond A, ed. De-
1980:95-115. velopmental and neural bases of higher cog-
34. Sherrington CS. Decerebrate rigidity, and re- nitive flinction. NY: New York Academy of
flex coordination of movements. J Physiol Sciences. In press.
(London) 1898;22:319-332. 47. Sudarsky L, Ronthal M. Gait disorders in the
35. Mott FW, Sherrington CS. Experiments elderly: assessing the risk for falls. In: Vellas
upon the influence of sensory nerves upon B, Toupet M, Rubenstein L, Albarede JL,
movement and nutrition of the limbs. Prelim- Christen Y, eds. Falls, balance and gait dis-
inary communication. Proc R Soc Lond orders in the elderly. Amsterdam: Elsevier
(Biol) 1895;57:481-488. 1992:117-127.
36. Brown TG. The intrinsic factors in the act of 48. Grillner S, Rossignol S. On the initiation of
progression in the mammal. Proc RSoc Lond the swing phase of locomotion in chronic spi-
(Biol) 1911;84:308-319. nal cats. Brain Res 1978;146:269-277.
37. Taub E, Berman AJ. Movement and learning 49. Pearson KG, Ramirez JM, Jiang W. Entrain-
in the absence of sensory feedback. In: Freed- ment of the locomotor by group lb aflferents
man SJ, ed. The neurophysiology of spatially from ankle extensor muscles in spinal cats.
oriented behavior. Homewood, NJ: Dorsey Exp Brain Res 1992;90:557-566.
Press, 1968:173-192. 50. Stein RB. Reflex modulation during loco-
38. Wolf S, LeCraw DE, Barton LA, Jann BB. motion: ftinctional significance. In: Patla A,
Forced use of hemiplegic upper extremities ed. Adaptability of human gait. Amsterdam:
to reverse the effect of learned nonuse among North Holland, 1991:21-36.
chronic stroke and head-injured patients. Exp 51. Nashner LM. Balance adjustment of humans
Neurol 1989;104:125-132. perturbed while walking. J Neurophysiol
39. Taub E, Miller NE, Novack TA, et al. Tech- 1980;44:650-664.
nique to improve chronic motor deficit after 52. Lackner JR, DiZio P. Visual stimulation af-
stroke.Arch Phys Med Rehabil 1993; fects the perception of voluntary leg move-
74:347-354. ments during walking. Perception 1988;
40. Forssberg H, Grillner S, Rossignol S. Phase 17:71-80.
53. Lackner JR, DiZio P. Senson' motor calibra- comotion. New York: Plenum, 1973:363-
tion processes constraining die perception of 388.
force and motion during locomotion. In: 67. Mann R.\, Hag>- JL, White V, Liddell D. The
Woollacott MH, Horak FB, eds. Posture and initiation of gait. J Bone Joint Surg 1979;61-
gait: control mechanisms. Eugene, OR; A:232-239.'"
Univ. of Oregon Books, 1992:92-96. 68. Breniere Y, Do MC. When and how does
54. Lee DN, Young DS. Gearing action to the steady state gait movement induced from up-
emironment. Exp Brain Res Series 15. Berlin: right posture begin.' J Biomech 1986;
Springer- Vedag, 1986:217-230. 19:1035-1040.
55. Berthoz A, Pozzo T. Head and body coor- 69. Cook T, Cozzens B. Human solutions for lo-
dination during locomotion and complex comotion: III. The initiation of gait. In; Her-
movements. In: Swinnen SP, Heuer H, Mas- man RM, Grillner S, Stein PSG, Stuart DG,
sion J, Casaer P, eds. Intedimb coordination: eds. Neural control of locomotion. New
neural, dynamical and cognitive constraints. York: Plenum, 1976:65-76.
San Diego: Academic Press, 1994:147-165. 70. Simoneau GG, Cavanagh PR, Ulbrecht JS,
56. Pozzo T, Berthoz A, Lefort K. Head stabili- Leibowitz HW, Tyrrell R^. The influence of
zation during various locomotor tasks in hu- visual factors on fall-related kinematic vari-
mans. 1. Normal subjects. Exp Brain Res ables during stair descent by older women. J
1990;82:97-106. Gerontol 1991;46:188-195.

57. Pozzo T, Levik Y', Berthoz A. Head stabili- 71. Craik RL, Cozzens BA, Freedman W. The
zation in the frontal plane during complex role of sensor\- conflict on stair descent per-
equilibrium tasks in humans. In: Woollacott formance in humans. Exp Brain Res 1982;
M, Horak F, eds. Posture and gait: control 45:399^09.
mechanisms. Eugene, OR; Univ. of Oregon 72 McFadyen B J, Wmter DA. i\n integrated bio-
Books, 1992:97-100. mechanical analysis of normal stair ascent and
58. Pada AE. The neural control of locomotion. descent. Biomech 1988;21:733-744.
J
In: Spivack BS, ed. Mobility* and gait. NY: 73. Millington PJ, Myklebust BM, Shambes GM.
Marcel Dekker, in press. Biomechanical analysis of the sit-to-stand
59. Hoy MG, Zernicke RF. Modulation of limb motion in elderly persons. Arch Phys Med
dynamics in the swing phase of locomotion. Rehabil 1992;73:609-617.
J Biomech 1985;18:49-60. 74. Schenkman MA, Berger RA, Riley PO, Mann
60. Hoy MG, Zernicke RF. The role of interseg- RW, Hodge WA. Whole-bodv movements
mental dynamics during rapid limb oscilla- during rising to standing from sitting. Phys
tions. I Biomech 1986;19:867-877. Ther 1990;10:638-651.
61. Hoy MG, Zernicke RF, Smith JL. Contrast- 75 VanSant AF. Rising from a supine position to
ing roles of inertia! and muscle moments at erect stance: description of adult movement
knee and ankle during paw -shake response. J and a developmental h\pothesis. Ph\'s Ther
Neurophysiol 1985;54:1282-1294. 1988;68:185-192.
62. Smith JL, Zernicke RF. Predictions for neural 76. VanSant AF. Age differences in movement
control based on limb dynamics. Trends patterns used by children to rise from a supine
Neurosci 1987;10:123-128. positions to erect stance. Phys Ther 1988;
63. Wisleder D, Zernicke RF, Smith JL. Speed- 68:1130-1138.
related changes in hindUmb intersegmental 77. Green LN, Williams K. Differences in devel-
dynamics during the swing phase of cat lo- opmental movement patterns used by active
comotion. Exp Brain Res. In press. vs sedentar\' middle-aged adults coming from
64. Larsson LE. Neural control of gait in man. a supine position to erect stance. Phys Ther
In: Eccles J, Dimitrije\ic MR, eds. Recent 1992; 72:560-568.
achievements in restorative neurolog\'. Basel: 78. Carr JH, Shepherd RB. Motor relearning
Karger, 1985:185-198. programme for stroke. Rockxille: Aspen,
65. Carlsoo, A. The initiation of walking. Acta 1983.
Anat 1966; 65:1-9. 79. Bobath B. Adult hemiplegia: evaluation and
66. Herman R, Cook T, Cozzens B, Freedman treatment. London: Heinemann, 1978.
VV. In; Stein RB, Pearson KG, Smith RS, 80. Sarnacki SJ. Rising from supine on a bed: a
Redford JB, eds. Control of posture and lo- description of adult movement and hypoth-
esis of developmental sequences. Richmond, 82, Richter RR, VanSant AF, Newton RA. De-
VA: Virginia Commonwealth University, movements and hy-
scription of adult rolling
1985. Master's thesis. pothesis of developmental sequences. Phys
McCoy AO, VanSant AF. Movement pat- Ther 1989:69:63-71.
terns of adolescents rising fi-om a bed. Phys
Ther 1993;73:182-193.
Chapter 12
A Life Span Perspective of

Mobility
Introduction Muscle Activation Patterns

Development of Locomotion Kinetic Analysis
Prenatal Development Changes in Adaptive Control
Early Stepping Behavior Proactive Adaptation
Maturation of Independent Locomotion Gait Changes in Fallers vs. Non-fallers
Run, Skip, Hop, and Gallop Role of Pathologv in Gait Changes in the
The Development of Adaptation Elderly
Reactive Strategies Cognitive Factors
Proactive Strategies Sensory Impairments
Head Stabilization During Gait Muscle Weakness
Development of Other Mobilit\' Skills Stair-Walking
Development of Rolling Age-related Changes in Other Mobility
Development of Prone Progression Skills
Development of Supine to Stand Sit-to-Stand
Locomotion in the Older Adult Rising from a Bed
Dysmobility: Aging or Pathology? Comparing Gait Characteristics of Infants and
Temporal-Distance Factors Elderly: Testing the Regression Hypothesis
Kinematic Analysis Summary
..srm
Introduction rclatcd chanjfes, which affect onl\' a few. This

chapter discusses mobility skills from a life
It is wonderflil to see children develop span perspecd\'e. We first review the devel-
their first mobilits' skills as they begin to crawl, opment of mobility skills in neurologically in-
creep, walk, and run — finally, navigating ex- tact children and summarize research from
pertly through complex environments. How different theoretical perspecdves that explore
do these skills develop? When do they first be- the factors contributing to the emergence of
gin to emerge? What key features of normal this complex abilit\'. In the latter half of the
locomotor development should we incorpo- chapter, we discuss how mobility skills change
rate into our assessment tools so that we can in the older adult.
better understand the delayed or disordered
development of the child with central ner\'ous Development of
system patholog)'? Locomotion
Falls, and the injuries that often accom-
pany them, are a serious problem in the older Independent locomotion may at first
adult. Many of these falls occur during walk- seem to be a fairly simple and automatic skill,
ing. Problems with balance and gait are con- but it is really a very intricate motor task. A
sidered major contributors to falls in the older child learning to walk needs to activate a com-
adult. Nevertheless, not all older adults have plex pattern of muscle contractions in many
difficulties with mobility' skills. Much like the body segments to produce a coordinated
study of balance control, it is important to dis- stepping movement, resulting in progression.
tinguish between aê-related chanjjes in mo- The child must be strong enough to support
bilit)' affecting all older adults and patholo/jy- body weight, and stable enough to compen-
269
270 Section III MOBILI'n' FUNCTIONS
sate for shifts in balance while walking, to ac- forms of locomotion can be elicited in 3 -day-
complish the goal of stability. Finally, the old kittens by placing them on a treadmill (4).
child must de\'elop the abilit\' to adapt gait to Howe\'er, gait in kittens is uncoordinated,
changing environmental circumstances, al- due to poor postural abilities.
lowing na\igation around and over obstacles, These results suggest that a primars'
and across uneven surfaces 1 ). In the follow-
( constraint on emerging locomotor behavior is
ing section, we summarize research evidence the immaturit}' of the postural system and
suggesting that in the development of loco- thus the inabilit}' to achieve upright stability'.
motion, these three requirements emerge se- In addition, these findings remind us to be
quentially, during the first years of life. How- careful about assuming that because a behav-
does complex behavior develop? What are
this ior is not e\ident, there is no neural circuitry'
the origins of this behavior during prenatal for it.
development?
Early Stepping Behavior
Prenatal Development
Because locomotor patterns ha\e been
Researchers have actually traced the or- developing for some months prenatallv, it is
igins of locomotor rhythms back to embr\- not surprising to find that stepping behavior
onic movements that begin to occur in the can be elicited in newborns under the right
first stages of development. Ultrasound tech- conditions (2, 7, 8). For example, when new-
niques have been used to document the born infants are held under the arms in an
movements of human infants prenatallv (2). upright position, tilted slightly forward, with
This research has shown that all movements the soles of the feet touching a surface, they
except those obsened in the earliest stages of oftren perform coordinated movements that
embr\'onic de\elopment (7 to 8 weeks) are look much like erect locomotion. Surpris-
also seen in neonates and young infants. ingly, stepping becomes progressively more
arm movements de-
Isolated leg and difficult to elicit during the first month of life,
velop in the embr\'o bv 9 weeks of age, while tending to disappear in most infants by about
alternating leg movements, similar to walking 2 months of age, and reappearing again with
movements seen after birth, develop in the in- the onset of self-generated locomotion, many
fant bv about 16 weeks of embr\-onic aa;e (2, months later.
3). This pattern of appearance and disap-

Animal research has also explored the pearance of newborn stepping was found in a
prenatal development of locomotor circuitry-. study that examined 156 children longitudi-
Detectable limb movements appear to emerge nally (7). It was found that 94 infants stepped
in a cephalocaudal sequence, with movements at I month, 18 stepped at 3 months, while
in the forelimbs preceding those in the hind- only x\\o stepped at 4 and 5 months. Then,
limbs (4). Intralimb coordination develops at 10 months, after a 4- to 8 -month period of
prior to interlimb coordination, with the first no stepping, all 156 infants stepped with sup-
detectable movements occurring at proximal port and 18 stepped without support. Thus,
joints, and moving distally with de\elopment. the stepping pattern appeared to be tempo-
Finally, interlimb coordination develops, first rarily lost in 98 to 99% of the infants.
with alternating patterns, then with synchro- WTiat causes these changes? Different
nous patterns (5). theoretical approaches explain changes in in-
Many newborn animals, such as the rat, fant beha\ior in \er\' different ways. From a
do not normally show coordinated locomotor reflex hierarchy perspective, newborn step-
movements until about I week aft:er birth (6 ). ping is thought to from a stepping re-
result
If, however, rats are placed in water at birth, flex. Its disappearance assumed to be mainly
is
they swim, demonstrating the maturit}' of the result of inhibition by maturing higher
their locomotor system. In addition, adult neural centers. Figure 12.1 iUustrates seven
Chapter Twelve A LIFE Span Perspective of MoBiLixi' 271
L-J
Figure 12.1. The seven phases of erect locomotion. (Adapted from McCraw MB. The neuromuscular maturation of
the human infant.NY: Hafner Press, 1945.)
phases in the development of infant locomo- hands gradually move from a high guard po-
tion, beginning with the observation of this sition (phase 5) down to the side (phase 6),
reflex (phase 1) and its disappearance (phase and the trunk and head become more erect
2), continuing with its reappearance (phase 3) (phase 7) (9).
and the emergence of assisted locomotion In contrast to a reflex hierarchical
(phase 4), and concluding with three phases model, researchers using a systems approach
of erect independent walking in which the have examined the emergence of stepping in
relationship to the contributions of multiple the effects of gravity, stepping increases in fre-
neural and non-neural systems. In particular, quency (10). This suggests that their weight
these studies have explored the conditions is a factor that affects the step cycle.
leading to the emergence of newborn step- Further support for the weight hypoth-
ping, and the changes that cause its disap- esis related to the disappearance of newborn
pearance. stepping comes from research examining
Esther Thelan, a psychologist, and her newborn kicking patterns. Supine kicking has
colleagues have appUed a dynamical systems the same spatial and temporal patterning as
approach to the study of locomotor devel- newborn stepping. For example, the swing
opment (8). This approach views locomotion phase of locomotion is similar to the flexion
as an emergent property of many interacting and extension phases of the kick, while the
complex processes, including sensor)', motor, stance phase is between
similar to the pause
perceptual, integrative, respiratory, cardiac, kicks. As stepping speeds up, the stance phase
and anatomical systems. According to a dy- is reduced, and as kicking speeds up, the pause
namical systems approach, moving and devel- phase is reduced (8).

oping systems have certain self-organizing This suggests that the same pattern gen-
properties, that is, they can spontaneously erator may be responsible for both supine
form patterns that arise simply from the in- kicking and newborn stepping. Yet, supine
teraction of the different parts of the system. kicking continues during the period when
A dynamical systems model stresses that newborn stepping disappears. One explana-
actions always occur within specific contexts. tion for the persistence of supine kicking is
As code will produce

a result, a given neural that it doesn't require the same strength as
ver>' different behavioral outcomes, depend- stepping, since the infants aren't working
ing on die contributions of the other elements against gravity (8, 10).
of the system, as in the position of the child Hans Forssberg, a Swedish physiologist
with relation to gravity. Thus, dynamical sys- and pecUatrician, examined the nervous sys-
tems researchers suggest that the specific leg tem contribution to the emergence of loco-
trajectory seen in newborn stepping is not motion in more detail. He postulated that hu-
coded precisely anywhere in the nervous sys- man locomotion is characterized by the
tem. Instead, the pattern emerges through interaction of many systems with certain hi-
the contributions of many elements. These in- erarchical components (7).
clude the neural substrate, anatomical link- His research suggests that an innate pat-
ages, body composition, activation or arousal tern generator creates the basic rh\TJim of the
level, and the gravitational conditions in step cycle, which can be seen in newborn step-
which the infant is kicking (8). ping. In the first year, the gradual develop-
From a dynamical systems perspective, ment of descending systems from higher neu-
the disappearance of the neonatal stepping ral centers gives the child the ability to control
pattern at about 2 months of age results irom this locomotor activity. Adaptive systems for
changes in a number of components of the equifibrium control, organized at a higher
system that reduce the likelihood of seeing level than those controlling the pattern gen-
this behavior (8). For example, body build erator, cievelop over a longer period (7).
changes greatly in the first 18 months of life. According to this research, the emer-
Infants add a lot of body fat in the first 2 gence of walking with support is not the result
months of fife and then slim down toward the of critical changes in the stepping pattern per
end of the first year. It has been suggested that se, but appears to be due to maturation of the
the stepping pattern goes away at 2 months equifibrium system. In addition, the gradual
because infants have insufficient strength to emergence of mature gait over the next year
fift the heavier leg during the step cycle (8). is hypothesized to result from a new higher
When 4-week-old infants are sub- level control system influencing the original
merged up to their trunk in water, thus mak- lower level network and modifying it (7).
ing them more buoyant and counteracting Forssberg's research, using EMG and
Chapter Twelve A Life Span Perspective of Mobility 273
motion analysis, has examined how the loco- and antagonist muscles at each joint. As with
motor pattern changes over the first 2 years the movement patterns, the EMG patterns
of development (7). Studies using motion alsobegan to look more mature during the
analysis techniques have shown a gradual of the second year, with asynchro-
latter part
transformation of the locomotor movement nous patterns emerging at the different joints
from a synchronous pattern of joint move- (7).
ments in newborn stepping to a more adult- Neonatal locomotion may be similar to
like dissociated pattern of joint motion by the that of quadrupeds who walk on the their
end of the first year of development. The toes, like cats, dogs, or horses (7). For ex-
transformation to adult-like gait patterns hap- ample, newborns show high knee/hip flexion
pens during the latter part of the second year. and do not have heel-strike. Since extensor
At this point, heel-strike begins to occur in muscle activity occurs prior to foot touch-
fi-ont of the body. Figure 12.2 shows the ki- down, it appears to be driven by an innate
nematics of neonatal vs. adult stepping move- locomotor pattern generator, as has been
ments. found in quadrupeds, rather than being re-
The EMG analysis supported the find- flexly activated by the foot in contact with the
ings of the motion analysis. For example, in ground. It has also been suggested that the
the neonate, the motor pattern was charac- neural network for stepping must be orga-
terized by a high degree of synchronized ac- nized at or below the brainstem level since
tivity. In other words, the extensor muscles of anencephalic infants (infants born without a
different joints were active simultaneously, cerebral cortex) can perform a similar pattern
and there was much coactivation of agonist of infant stepping (II).
Interestingly, some researchers believe
that the abnormal gait patterns found in many
patients with neurological impairments are
actually immature locomotor patterns. Thus,
children with cerebral palsy, mentally retarded
children, and habitual toe-walkers may persist
in using an immature locomotor pattern,
while adults with acquired neurological dis-
ease may revert to immature locomotion be-
cause of the loss of higher center modulation
over the locomotor pattern generator (7).
So what are the elements that contrib-
ute to the emergence of locomotion in the
infant? Remember that in development, some
elements of the nervous and musculoskeletal
system may be functionally ready before oth-
ers, but the system must wait for the matu-
ration of the slowest component before the
target behavior can appear. A small increase
or change in the development of the slowest
component can act as the control parameter,
becoming the impetus that drives the system
to a new behavioral form.
Adult plantigrade walk The research we just discussed shows
that many of the components that contribute
Figure 12.2. Kinematic differences in neonatal vs. adult
to independent locomotion are functional be-
gait.(Adapted from Forssburg H. Ontogeny of human
fore the child takes any independent steps.
locomotor control: 1 Infant stepping, supported loco-
.
motion and transition to independent locomotion. Exp Function of the locomotor pattern generator
Brain Res 1985;67:481 .) is present in a limited capacity at birth, and is
improved during die second half of the first Maturation of Independent

year, as the tight intralimb synergies become Locomotion
dissociated and capable of more complex
modulation and control. As we noted in the Bril and Breniere, two French research-
chapter on the development of postural con- ers, studied the emergence of locomotion and
trol, infants are able to use optic flow infor- h\,rpothesized that learning to walk is a two-
mation modulate head movements,
at birth to stage process (13). In the initial phase, infants
and at least by 5 to 6 months of age for mod- learn to control balance, while in the second
ulation of stance. Motivation to navigate to- phase, the locomotor pattern is progressively
ward a distant object is clearly present by the refined.
onset of creeping and crawling, and voluntary They studied children longitudinally
control over the hmbs is certainly present by during the first 4 years of life to see how gait
this time for many behaviors (8). patterns change as independent locomotion
So what is the constraint that keeps lo- develops. Significant changes in emerging gait
comotion fi-om emerging before 9 to 12 patterns are summarized in Figure 12.3 and
months of age? Most researchers believe that include a decrease in the double-support
it is primarily due to limitations in balance phase of gait (Fig. 12.3A), an increase in step
control, and possibly also limitations in length, and a decrease in step width (Fig.
strength (7, 8, 12). 12.3B). The greatest changes occurred in the
For example, when an infant is creeping, first4 months of independent walking (Fig.
one foot can be picked up at a time, so there 12.3^) (13).
is always a tripod stance available and, tiius, In addition, examination of the vertical
balance is much less demanding. Normal in- acceleration of the center of gravity suggested
fants who about to take their first steps
are that when infants are first learning to step,
have developed motor coordination within they fall into each successive step. By 5 to 6
the locomotor pattern generator, they have months this trend has begun to lessen, and
fiinctional visual, vestibular, and somatosen- continues to improve between the ages of 10
sory systems and the motivation to move for- to 40 months as the infant learns to integrate
ward. Infants may also have sufficient muscle balance into the step cycle (Fig. 12. 3C) (13).
strength, at least to balance, if not for use in Since the changes in step width, step
propelling the body fonvard. But they won't length, and double-support phase appear to
be able to use these processes in effective lo- relate to the master)' of balance control, their
comotion until the postural control system findings support the idea that it is during
can effectively control the shift of weight from the first phase of walking that a child learns
leg to leg, thus avoiding a fall. When these to integrate posture into locomotor move-
processes hit a particular threshold for effec- ments.
tive function, then the dynamic behavior of Studies of changes in EMG character-
independent locomotion can emerge. isticsand kinematics from the onset of walk-
When looking at the three requirements ing through the mastery of mature forms of
for successful locomotion: a rhythmic step- gait have been performed by other laborato-
ping pattern (progression), the control of bal- ries as well (14, 15).
ance (stability'), and the ability to modify gait In the first days of independent walking,
(adaptation), clearly, a rhythmic stepping pat- stepping patterns are immature. Push-off mo-
tern develops first. It is present in limited form tion in the stance phase is absent, the step
at birth, and is refined during the first year of width is ver>' wide, and the arms are held high.
life. Stance stability develops second, toward The infant appears to generate force to propel
the end of the first year and the beginning of the body fonvard by leaning forward at the
the second year of life. As we discuss in the trunk (14). The swing phase is short because
next section, it appears that adaptability is re- the infant is unable to balance on one leg.
fined in the first years after the onset of in- By 10 to 15 days of independent walk-
dependent walking. ing, the infant begins to reduce cocontrac-
Chapter Twelve A Life Span Perspective of MoBiLm- 275
Figure 12.3. Graphs of changes in different

walking parameters during the first 4 years of
walking. A, The relative duration of the double-
support phase. B, Changes in relative step length
and width. C, Changes in vertical acceleration
of the center of gravity. (Adapted from Bril B,
Breniere Y. Posture and independent locomo-
tion in childhood: learning towalk or learning
dynamic postural control? In: Savelsbergh C)P,
ed. The development of coordination in infancy.
Amsterdam: North-Holland, 1993, 337-358.)
20 30
months of I.W.
30 40 50
months of I.W.
• Xp/riOO Xp/riOO
30 40
months of I.W.
tion, and at 50 to 85 days after walking onset, and

limbs, a flexed knee during stance phase,
the muscle patterns begin to show a reciprocal an increased hip flexion, pelvic and hip tilt,
relationship. Interestingly, if infants are sup- abduction during swing phase. There is also
ported during walking, the reciprocal rela- ankle plantarflexion at foot-strike and de-
tionship between muscles emerges, but, with creased ankle flexion during swing, giving a
the additional requirement of stabilizing the relative foot-drop (15).
body while walking independendy, the coac- By 2 years of age, the pelvic tilt and ab-
tivation returns (14). duction and external rotation of the hip are
Other common gait characteristics in diminished. At foot-strike, a knee-flexion
the first year of walking include: a high step wa\'e appears, and reciprocal swing in the up-
frequency, absence of the reciprocal swinging per limb is present in about 75% of the chil-
movements between the upper and lower dren. The relative foot-drop disappears as the
ankle dorsiflexes during swing. By the end of Table 1 2.1 . Developmental Sequence for
Walking^
age 2, the infant begins to show a push-off in
stance (15). I. Walking:
During the years from 1 until 7, the A. Initial stage
1 Difficult)' maintaining upright posture
muscle amplitudes and durations gradually re-
2. Unpredictable loss of balance
duce to^\'ard adult levels. By the age of 7,
3. Rigid, halting leg action
most muscle and movement patterns during 4. Short steps
walking look ver\- similar to that of the adult 5. Flat-footed contact
(15). 6. Toes turn outward

Five important characteristics are used 7. Wide base of support
8. Flexed knee at contact followed by quick leg
to determine mature gait, including (a) du-
extension
ration of single limb stance, (
b) walking ve- B. Elementary' stage
locit)', (c) cadence, (d) step length, and (e) the 1 Gradual smoothing out of pattern
ratio of pehic span to step width (15). 2. Step length increased
Duration of single-Umb stance increases 3. Heel-toe contact

4. Arms down to sides with limited swing
steadilyfrom 32% in 1 -year-olds to 38% in 7-
5. Base of support within the lateral dimensions
year-olds (39% is a t\pical adult value ). Walk- of trunk
ing velocit^' and cadence decrease steadily, 6. Out-toeing reduced or eliminated
while step length increases. Step length is 7. Increased pelvic tilt
short in the newly walking child due to lack 8. Apparent vertical lift
C. Mature stage
of stabilin,' of the supporting limb, and
1 arm swing
Reflexive
lengthens with increasing balance abilities. Fi-
2. Narrow base of support
of pelvic span, which is defined
nally, the ratio 3. Relaxed, elongated gait
as body width at the level of the pehis, to step 4. Minimal vertical lift
width increases until age 2 1/2, after which it 5. Definite heel-toe contact
II. Common Problems
stabilizes. By 3 years of age, the gait pattern
A. Inhibited or exaggerated arm swing
is essentially mature, though small improve- B. Arms crossing midline of body
ments continue through age 7 (15). C. Improper foot placement
Table 12.1 summarizes some of the D. Exaggerated forward trunk lean
characteristic changes in the step-c\'cle from E. Arms flopping at sides or held out for balance
F. Twisting of trunk
the initiation of independent walking through
C. Poor rhythmical action
the development of mature patterns at about
H. Landing flat-footed
the age of 3 (16). These changes can be seen I. Flipping foot or lower leg in or out
more graphically in Figure 12.4.
"From Callahue DL. Understanding motor development: infants,
children, adolescents. Indianapolis: Benchmark Press, 1989:236.
Run, Skip, Hop, and Gallop
Running is often described as an exag- If central pattern generator's (CPGs)
gerated form of walking because it differs control walking, are there separate CPGs for
from the walk as the result of a brief flight hopping, galloping, and skipping? Probably
phase in each step. The flight phase that dis- not. Then why do they emerge in a fixed order
tinguishes a run is seen at about the second of appearance? It is possible to explain their
year of age. Until this time, the infant's run is emergence from the dynamical systems per-
more like a fast walk with one foot always in specti\'e (17).
contact with the ground (16). By 4 years of Remember
that walking and running
age, most children can hop (33%) and gallop are of interlimb coordination in
patterns
(43%). The development of the gallop pre- which the limbs are 50% out of phase with one
cedes the hop slightly. In one study, by 6.5 another. This is the easiest stepping pattern to
years, the children were skillful at hopping produce, and thus appears earliest. Running
and galloping. However, only 14% of 4-year- appears later than walking, probably due to its
olds could skip (step-hop) (17). increased strength and balance requirements
Figure 12.4. Body motions associated with developing gait. A, Initial forms of gait. B, Elementary' forms of gait. C,
Mature forms of gait. (Adapted from Callahue DL. Understanding motor development: infants, children, adolescents.
Indianapolis: Benchmark Press, 1989:237.)
compared with walking. Galloping requires from 1 to 2.5 years but not in older children.
that the child produce an asymmetrical gait Similar to stance perturbations, automatic
with unusual timing and a differentiation in postural responses to gait perturbations be-
force production in each limb, and it may pro- come faster with age, with mature responses
duce additional balance requirements. Hop- occurring by about 4 years. Coactivation of
ping emerges next, possibly because it re- antagonist muscles also is reduced with age.
quires the ability to balance the body's weight These changes are shown in Figure 12.5.
on one limb and it requires additional force Changes in the characteristics of compensa-
to lift: the body off the ground after landing. tory postural activity are associated with in-
Skipping (a step-hop) emerges last, possibly creased stability during gait, and increased
because one locomotor coordination pattern ability' to compensate for perturbations to gait
is imbedded into another pattern, and thus it (18). This study suggests that children as
requires additional coordination abilities young as a year old who are capable of inde-
(17). pendent locomotion can integrate compen-

sators' postural activity' into slow walking
Development of Adaptation when gait is disturbed, though their responses
are immature.
How do children learn to adapt their
walking patterns so they can navigate over and PROACTIVE STRATEGIES
around obstacles? As we mentioned in the
previous chapter, both reactive and proacti\'e Proactive strategies for adapting gait use
strategies are used to modify gait to changes sensor}' information to
modify gait patterns in
in the environment. There has been ver}' little advance of obstacles to gait. When do chil-
research examining the development of ad- dren begin integrating these strategies into
aptation in normal children. As a result, we the step cycle.' It has been suggested that chil-
know litde about how children learn to com- dren first learning to walk master feedback
pensate for disturbances to their gait, nor how control of balance, and then move on to ac-
they develop proactive strategies to modify quire feedforward control (19). The results of
gait in advance of obstacles. experiments by Bril and Breniere (13) support
this idea, since children seem to spend the first
REACTIVE STRATEGIES 4 to 5 months of walking learning to integrate
balance into the step cycle. However, there is
Reacdve strategies for adapting gait re- litde research on the development of proac-
late to the integration of compensatory pos- tive strategies to help clinicians understand
tural responses into the gait cycle. Researchers the emergence of this important aspect of mo-
have looked at compensatory postural muscle bility'.
responses to perturbations during locomo-

tion,and compared them to those during per- Head Stabilization During Gait
turbed quiet stance (18).
In response to fast velocity stance per- An important part of controlling loco-
turbations, children respond with both an au- motion is learning to stabilize the head.
tomatic postural response and a monosynap- Adults stabilize the head with great precision,
tic reflex response. As children mature, the allowing a steady gaze. Thus, control of the
stretch reflex response gets smaller in ampli- head, arm, and trunk (HAT) segments is a
tude, while the postural response gets critical part of controlling mobility'. How do
faster. In very young children, there is con- children control the trunk, arms, and head
siderable coactivation of antagonist muscles during locomotion ensuring stabilization of
(18). the head and gaze.'
Perturbations during gait produce a Assaiante and Amblard (20) performed
monosynaptic reflex response in children experiments in children from early walkers
Figure 1 2.5. Examples of the gastrocnemius EMG

responses of individual children of 1, 2.5, and 4
years of age and of an adult, when their balance is
perturbed during walking on a treadmill, by briefly

increasing treadmill speed. The left vertical line is
the onset of the treadmill acceleration, and the

dotted line to its right is the onset of the EMG re-
sponse. Note that there is a large monosynaptic re-
flex in the youngest children, before the automatic

postural response. This disappears by 4 years of
age. (Adapted from Berger W, Quintern j, Dietz V.
Stance and gait perturbations in children: devel-
opmental aspects of compensatory mechanisms.
Electroencephal Clin Neurophysiol 1985;61:385-
388.)
through children 10 years of age, to explore moves freely, minimizing movements away
changes in control of these body segments. from vertical.
Based on their findings, balance and loco- This study explored locomotor strate-
motion can be organized according to one of gies through kinematic analysis of walking in
two stable reference frames, either the support infants and children up to 8 years of age. The
surface on which the subject stands and authors found that from the acquisition of
moves, or the gravitational reference of ver- stance until about 6 years of age, children or-
tical. ganize locomotion in a bottom-up organiza-
They noted that when using the support tion, using the support surface as a reference,
surface as reference, the subject organized and controlling head movements in an en bloc
balance responses from the feet upward to- mode, which serves to reduce the degrees of
ward the head, using mainly proprioceptive freedom to be controlled. During this time
and cutaneous cues. In contrast, when the period, the children gradually learn to stabi-
subject stabilized the head using vestibular in- lize the hip, then the shoulders, and finally the
formation, balance was organized from the head. At about 7 years of age, with mastery of
head down toward the feet. These researchers control of the head, there is a transition, and
explored the changing use of these two strat- the head control changed to an articulated
is
egies in balance control during locomotor de- mode, and top-down organization of balance
velopment in children (20). during locomotion becomes dominant. The
They also noted that the head can be authors hypothesized that at 7 to 8 years of
stabilized on the trunk in one of two modes, age, information specifS'ing head position in
in an en bloc mode, where it moves with the relation to gravity becomes more available to
trunk, or in an articulated mode, where it the equilibrium control centers and thus al-
lows the child to use an articulated mode of emergence of rolling, prone progression, and
head control. They suggest that there may be the assumption of the vertical position fi-om
a transient dominance of vestibular processing supine. As we mentioned in the chapter on
in locomotor balance at this age (20). development of postural control, much of the
information we have on the emergence of
Development of Other Mobility motor behavior in children is largely the result
Skills of efforts in the 1920s and 1930s by two de-
velopmental researchers, Arnold Gesell and
The development of postural control Myrtie McGraw, who observed and recorded
underlying emergence of sitting and
the the stages of development in normal children
standing is covered in detail in Chapter 7. The (24).
first part of this chapter describes the emer-
gence of independent locomotion. We now DEVELOPMENT OF ROLLING

turn briefly to a review of some of the infor-
mation on the emergence of other mobility Rolling is an important part of mobility

behaviors during development, including skills because rotation or partial rotation is a
rolling, prone progression, and movement part of movement patterns used to achieve su-
from lying in a supine position to stance. pine-to-sit or supine-to-stand behavior. Ba-
There are two approaches to describing bies from the side-lying position to
first roll
motor development in infants and children. the supine position at 1 to 2 months of age
One approach relies on normative studies that and from supine to side-lying at 4 to 5
describe the age at which various motor be- months. Infants roll from prone to supine at
haviors emerge. Normative studies have given 4 months of age, and then from supine to
rise to norm-referenced scales that compare
prone at 6 to 8 months. Infants change their
an infant's motor behavior with the perfor- rolling pattern as they mature, from a log-roll-
mance of a group of infants of the same age.
ing pattern, where the entire body rolls as a
Normative studies can provide clinicians with unit, to a segmental pattern. By 9 months of
rough guidelines about the relative ages as- age, most infants use a segmental rotation of
sociated with specific motor milestones.
the body on the pelvis (24, 25).
However, they have universally reported that
there is incredible variability in the time at
which normal children achieve motor mile- DEVELOPMENT OF A PRONE
stones (21). PROGRESSION
Another approach to describing motor
development is with reference to the stages According to McGraw, the prone pro-
associated with the emergence of a single be- gression includes nine phases that take the in-
havior, such as rolling, or coming to stand. fant from the prone position to creeping and
Stages within the emergence of a skiU are of- crawling, and span the months from birth to
ten used by clinicians as the basis for a treat- 10 to 13 months (24). Figure 12.6 illustrates
ment progression, with the assumption that a the nine phases reported by McGraw and the
mature and stable adult-like pattern is the last relative time in which the behavior was seen.
stage in the progression. However, recent re- Graphed is the age at which the behavior was
search has raised doubts about the concept seen, and the percent of children in which the
that there is a consistent stable sequential pat- behavior was observed. The first phase is char-
tern during the emergence of a particular mo- acterized by lower extremity flexion and ex-
tor behavior (22, 23). tension in a primarily flexed posture. In phase
Given these cautions about timing, vari- 2, spinal extension begins, as does the devel-
ability, and the sequential nature of the emer- opment of head control. In the third phase,
gence of motor skills, we review some of the spinal extension continues cephalocaudally,
studies diat have examined die stages in the reaching the thoracic area. The arms can ex-
Chapter Twelve A Life Span Perspective of Mobility' 281
I *"t.î
***T..tt.
6.5 10
Age in months
-•••».
. .•b«. Phase 4
„ Phase 5
**, Phase 6
»• • •• •
• • • • •
3 6.5 10 13
Age in months
- Phase 7
.,•.-".>•-/•.,
1 .».
- Phase 8
- Phase 9
1
-..- 1* 1
6.5 10 13
Age in months
Figure 12.6. The nine phases of prone progression as reported by McCraw. Graphed which the behavior
is the age at
was seen and the percent of children in which the behavior was observed. See text for details of each stage. (Adapted
from McCraw MG. The neuromuscular maturation of the human infant. New York: Hafner Press, 1945.)
tend and suppoiT the chest oft' the surface. emphasis on the neural antecedents of matur-
Propulsion movements begin in the arms and ing motor behavior. Her emphasis was on de-
legs during phases 4 and 5. In phase 6, the scribing stages of motor development that
creeping position is assumed. Phase 7, is char- could be related to the structural growth and
acterized by fairly disorganized attempts at maturation of the central nervous system.
progression; however, by phases 8 and 9, or- Current research has shown that many factors
ganized propulsion in the creeping position contribute to the emergence of motor skill
has emerged (24). during development, including but not lim-

Keep in mind that McGraw placed great ited to maturation of the CNS (1,8).
DEVELOPMENT OF SUPINE-TO-
STAND
lust as the pattern used to roll changes
as infants age, the movement pattern used to
achieve stance from a supine position also un-
dergoes change with development. The pat-
tern initially seen in infants moving from su-
pine to stand includes rolling to prone, then

moving into an all-fours pattern, and using a
pull-to-stand method to achieve the erect po-
sition Widi development, the child
(24).
learns to move from the all-fours position to
a plantigrade position and from there to erect
stance. By the age of 2 to 3 years, the supine-
to-prone portion is modified to a partial roll
and sit-up pattern, and by ages 4 to 5, a sym-

metrical sit-up pattern emerges (Fig. 12.7).
This is considered a mature or adult-like
movement pattern used for this task. But as
you remember from the chapter on normal
mobility' skills, researchers have found tre-
mendous variability' in how adults move from
supine-to-stand (26). Just as was true for
adults, most likely, strength in the abdominals
and hip flexors plays a major roll in the t^'pe
of pattern used by infants when moving from
supine-to-stance (26).
Locomotion in the Older

Adult
Falls and the injuries that often accom-
pany them are a serious problem in the older
adult. In fact, falls are the seventh leading

cause of death in people over 75 years of age
(27). Fort\'-eight percent of adults over 75
years who have had an injurious fall acquire a
fear of falling, and 26% of these people begin
avoiding situations that require refined bal-
ance skills, thus leading to further decline in
walking and balance skills. Figure 12.7. Common pattern used to move from su-
pine to stand in children ages 4 to 5. (Adapted from

Many of the falls experienced by the el-
VanSant AF. Age differences in movement patterns used
derly occur during walking. It is thus impor-
by children to rise from a supine position to erect stance.
tant to understand the changes in the systems PhysTher 1988;68:1130-1138.)
contributing to normal gait in the elderly to
fully understand the cause of increased falls in
this population. As we stated in the first sec-

tion of this chapter, many researchers now be-
lieve that balance conti'ol is a primary contrib-
Chapter Twelve A Life Span Perspective of MoBiLirY 283
utor to stable walking. In addition, decreased unselected group of subjects from 60 to 99

balance control is a major factor affecting loss years, walking velocities were much slower
of independent mobility in many elderly. The than those for young adults, and also slower
following sections describe locomotor than other published studies on older adults
changes commonly seen in the older adult (29). It is quite possible that the subjects in
and the systems contributing to these the study were less fit, and many complained
changes. of symptoms likely to impair gait. In contrast,
84 older adults and
a study that screened 1 1
Dysmobility: Aging or Pathology? chose 32 who had no pathology, found no

changes in gait parameters tested (30).
Again, age-related changes in locomo- Thus, more recent research has begun
tion may be due to primar}' or secondary ag- to indicate that many gait disorders con-
ing phenomena. Primary' factors affecting ag- sidered to be age-related, such as gait
ing include things which result
like genetics, apraxia, hypokinetic-hypertonic syndrome,
in an inevitable decline of neuronal fianction and marche a petit pas, are really manifesta-
within a particular system. Secondary' factors tions of pathology rather than manifestations
are experiential, and include nutrition, exer- of a generalized aging process. However, as
cise, and acquired pathologies, among others. we note in the following sections, there ap-
The extent to which gait disorders in the pear to be characteristic changes in gait that
elderly are due to primary or secondary' factors occur in many, even healthy, older adults.
is a ver)' important point to consider as we
begin to look at the literature on changes in Temporal -Distance Factors
gait characteristics in the older adult.
The older clinical literature referred to Studies examining changes in walking
many different walking patterns as age-related patterns with age have used a number of dif-
gait disorders (28). These diverse gait disor- ferent experimental approaches. In one ap-
ders included: gait apraxia (slow, halting, proach, which we might call a naturalistic ap-
short-stepped, shuffling, or sliding gait), proach, adults were observed walking
hypokinetic-hypertonic syndrome (slow, de- spontaneously in a natural setting. This para-
liberate gait, but without the shuffling or slid- digm was used to try to minimize the con-
ing components described above), and straints on walking style that are often nec-
marche a petit pas (small, quick shuffling essary when quantifying gait parameters in a
steps, followed by a slow cautious, unsteady laboratory' setting.
gait), vestibular dvsflinction gait (difficulties In these studies, researchers obser\'ed
in turning) and proprioceptive dysfunction people of different ages walking along the
gait (cautious,with a tendency to watch the streets of New York Cit)' Amsterdam
(31) or
feet and make missteps) (28). (32). In the first study, of 752 pedestrians in
As was true in the postural control lit- New York City, as age increased from 20 to
erature, care must be taken when reviewing 70 was a decrease in walking ve-
years, there
studies discussing age-related changes in gait. locit)', and step rate (no statistical
step length,
When interpreting the results of a study, one analysis was reported). In the second study on
should examine carefully the population stud- 533 pedestrians in Amsterdam, similar results
ied, and ask questions such as: What type of were found. Gender differences were also
criteriawere used in selecting older subjects? found; both younger and older women
Did researchers exclude anyone with pathol- walked with slower velocity, shorter step
ogy under the assumption that pathology' is length, and higher cadence than men (32).
not a part of primar\' aging? Results will vary While there are advantages in allowing
tremendously depending on the composition subjects to walk in a natural environment, the
of older adults under study. disadvantages include being unable to control
For example, one study noted that in an for such variables as different walking goals.
such as taking a stroll \'s. hurr^'ing to work, longer, with a commensurate shortening of
and of the subjects (28).
relative health time in swing phase (34).
Laboratory studies have also repeatedly Finally hip, knee, and ankle flexion were
demonstrated that walking speed decreases less than in young adults, and the whole
with age. One of the earlier studies outlines shoulder rotation pattern was shifted to a
three stages of age-related changes in walking more extended position, with less elbow ro-
(33). Stage 1 changes were found in adults tation as well. Figure 12.8 is from their study,
between 60 and 72 years of age, and included showing the differences in the limb positions
decreases in walking speed, shorter step of a younger vs. an older man at heel-strike
move-
length, lower cadence, and less vertical (34).'
ment of the center of gravity. Subjects be- Interestingly, the researchers concluded
tween 72 and 86 years old showed stage 2 gait that the men studied did not have a patho-
changes, including the disappearance of nor- logical gait pattern. Instead, they said, walk-
mal arm-leg synergies, along with an overpro- ing was guarded, possibly with the aim to in-
duction of unnecessary movements. In stage crease stability. Gait patterns were similar to
3, in subjects ages 86 to 104 years, there was those used by someone walking on a slippery
a disintegration of the gait pattern, arrhyth- surface or someone walking in darkness.
mia and an absence of
in the stepping rate, Doesn't this sound like a postural problem?
arm swing movement (33). It was later From reading this description, one might hy-
pointed out (34) that these changes are not pothesize that gait changes in the elderly per-
typical of changes seen in healthy older adults, son relate more to the loss of balance con-
and the study probably included adults with trol than to changes in the step cycle itself
symptoms of Parkinson's disease and other (34).
motor patholog)'. In a second study (35), age-related
changes in gait patterns were investigated in
Kinematic Analysis
Later studies of age-related changes in

gait focused on a kinematic analysis of step-
ping patterns in older adults (34). In one
study, subjects were healthy men, with nor-
mal strength and range of motion, ranging in
age from 20 to 87 years of age. Those over
65 years old were given a neurological exam
to exclude the possibility of neurological def-
icits contributing to the observed changes.
Participants were photographed in the labo-
rator\' using interrupted-light photography at
20 Hz while walking at their preferred and
fast speeds.
Men over 67 years of age showed sig-
nificantly (p < .01) slower walking speeds
(118 to 123 cm/sec) than the young adults
(150 cm/sec). Stride length was also signifi-
cantly shorter, especially during fast walking.
Vertical movement of the head during the gait
cycle was smaller, while lateral movement was
Figure 1 Example of the walking pattern of a young
2.8.
larger. Stride width tended to be wider for
man. (Adapted from Murray MP,
adult vs. a healthy older
men over 74. Toeing out was also greater for Kory RC, Clarkson BH. Walking patterns in healthy old
men over 80. Beyond 65, stance phase was men. J Gerontol 1969;24:169-178.)
Chapter Twelve A LIFE Span Perspective of MoBiLi-n- 285
women, and similar changes were noted, in- among leg muscles than young adults during
cluding reduced walking speeds and shorter walking. But how do these changes in muscle
steps. These changes occurred in the 60- to acdvation patterns change the tiynamics of
70-year-old age group. gait.>
How do these slower walking speeds af- Using the method of inverse dynamics,
fect function in daily life.' Many of the previ- moments of force, as well as the mechanical
ous studies report that older adults are unable power generated and absorbed at each joint,
to walk faster than 1.4 m/min. This is the can be calculated. This process allows the
minimal speed recommendecH by the Swedish amount of power generated by muscles to be
authorit^' to safely pass an intersection. Thus, estimated. Remember from the previous
many of would not
the older adults studied chapter on locomodon that an increase in
be considered functional walkers on cit\' muscle energy is needed to initiate swing,
streets with lots of traffic. while a decrease in energy is needed to pre-
pare for heel-strike.
Muscle Activation Patterns Using inverse dynamics techniques.
Winter and colleagues compared the gait pat-
The previous studies show clear changes terns of 15 healthy older adults (age range: 62
in certain kinematic characterisdcs of the gait to 78 years) to 12 young adults (age range:
cycle in the average older adult. How do these 21 to 28 years) (38). They found that older
changes relate to changes in muscle response adultshad significantly shorter stride length
patterns? In a study comparing patterns of and longer double-support time than young
muscle activit}' in younger (ages 19 to 38 adults. In addition, in elderly subjects, plantar
years) and older (ages 64 to 86 years) women, flexors generated significandy less power at
average EMG activit}' levels in gastrocnemius, push-otf, while the quadriceps muscle ab-
tibialis anterior, biceps femoris, rectus fe- sorbed significandy less energy during late
moris, and peroneus longus were higher in stance and early swing.
the older age group than in the younger These researchers concluded that the re-
group (36). duction of plantar flexor power during push-

In addidon, there were changes in the off could explain the shorter step length,
acdvity of individual muscles at specific points flat-footed heel-strike, and increased double-
For example, at heel-strike,
in the step cycle. support stance duration. Two alternative ex-
peroneus longus and gastrocnemius were planations were proposed for a weaker push-
moderately to highly active in the older oflf in the older adult. One explanation
women, but showed litde or no activit)' in the suggested a reduction in muscle strength in
younger group. The authors suggested that the ankle plantar flexors in the older adults
this increased acdvit)' resulted from an effort could be responsible for the weaker push-off".
to improve stability during the stance phase An alternative explanation argued that re-
of gait (36). For example, increased coacti- duced push-off" could be an adaptive change
vation of agonist and antagonist muscles at a used to ensure a safer gait, since high push-
joint may be used to improve balance control, oft" power acts upward and forward and is thus
by increasing joint sdffness. This is a strateg)' destabilizing (38).
often seen in subjects who are unskilled in a In this study, an index of dynamic bal-
task, or who are performing in a situation that ance was computed to determine the ability
requires increased control (37). to coordinate the anterior/posterior balance
of the HAT segment while simultaneously
Kinetic Analysis maintaining an appropriate extensor moment
in the ankle, knee, and hip during stance
We just noted several studies indicating phase. It was found that the older adults
that older adults show higher levels of muscle showed a reduced ability to covary move-
responses and different activadon sequences ments at the hip and knee. This means that
Table 12.2. Summary Gait Changes in the postural control in response to perturbed gait
Older Adult in the ekierly.
Temporal/distance factors
Decreased velocity PROACTIVE ADAPTATION
Decreased step length
Decreased step rate Proactive adaptation depends in large
Decreased stride length
part on the abilit\' to use visual information to
Increased stride width
alter gait patterns in anticipation of upcoming
Increased stance phase
Increased time in double support obstacles (39). One group of researchers
Decreased swing phase asked whether a possible cause of poor loco-
Kinematic changes motor abilities in older adults might be a re-
Decreased vertical movement of the center of gravity
duced abilit)' to sample the visual environ-
Decreased arm swing
Decreased hip, knee, ankle flexion
ment during walking (40). They wanted to
Flatter foot on heel-strike know whether visual sampling of the environ-
Decreased ability to covary hip/knee movements ment changed with age.
Decreased dynamic stability during stance In their experiment, subjects wore
Muscle activation patterns
opaque liquid cr\'stal eyeglasses, and pressed
Increased coactivation (increased stiffness)
a sv\itch to make them transparent whenever
Kinetic changes
Decreased power generation at push-off they wanted to sample the environment. Sub-
Decreased power absorption at heel-strike jects walked across a floor that was either un-
marked, or that had footprints marked at reg-
ular intervals, on which the subjects were
supposed to walk. When subjects were con-
strained to land on the footprints, the young
older adults had trouble controlling the HAT subjects sampled frequendy, though for
segment while simultaneously maintaining an shorter intervals than older subjects, who
extensor moment in the lower stance limb. In tended to sample less often, but for longer
evaluating the older group individually, it was time periods. Thus, older adults seem to mon-
noted that two-thirds were within the normal itor the terrain much more than the young
young adult range, while one -third had ver^' adults (39,40).
low covariances of moments at the hip and What is the minimum dme required to
knee. It was concluded that some older adults implement an avoidance strategy in the
may have had problems with dynamic balance younger vs. older adult.' In a second study,
during locomotion, indicative of balance im- healthy young and older adults were asked to
pairments not detected in their medical his- walk along a walkway, and when cued by a
tor\' or simple clinical tests (38). light at specific points along the walkway, to
Numerous research studies have de- either lengthen or shorten their stride to
scribed changes in gait patterns found among match the position of the light (40).
many older adults. These changes are sum- Compared with young adults, older
marized in Table 12.2. adults had more difficult)' in modulating their
step length when the cue was given only one
Changes in Adaptive Control step duration ahead. Young adults succeeded
80% of the time, while older adults succeeded
Many falls by older adults occur while 60% of the time when lengthening the step
walking and may be due to slipping and trip- and only 38% of the time when shortening the
ping. Several research groups have examined step. Both groups were equally successfiil
proacti\e adaptive strategies during gait in the when two step durations in
the cue was given
elderly. However, there are virtually no stud- advance (40).
ies examined compensatory
to date that have The authors suggest that older adults
Chapter Twelve A LIFE SPAN PERsrECTivE OF Mobility 287
have more difficulty in shortening a step be- when they walked at a fast speed of 6 km/hr
cause of balance constraints. Shortening the when compared with subjects without a his-
step requires regulating the forward pitch of toty offalls (43). It was also noted that older
the HAT segment, which if not controlled, had balance problems unrelated to gait
fallers
could result in a fall. Remember in the review (42) because they were unable to stand as
of Winter's study presented earlier, older long as non-fallers with feet in tandem posi-
adults had more trouble than young adults tion with eyes open. Of course, it is likely that
controlling dynamic balance during gait. older adults with a histoty of falls have an un-
These results suggest that the older ciiagnosed pathological condition. Therefore,
adult may need to begin making mociifica- it is important to carefiiUy examine these sub-
tions to gait patterns in the step prior to a step jects, to determine underlying pathology that
requiring obstacle avoidance. This may be one may contribute to gait disturbances, when
cause of increased visual monitoring. performing studies on older adults who fall.
What strategies do older adults use to

avoid obstacles during walking.' To answer Role of Pathology in Gait Changes
this question, researchers analyzed the gait of in the Elderly
24 young and 24 older (mean age 71 years)
healthy adults while diey stepped over obsta- What is the role of secondaty aging fac-
cles of vatying heights (41). Obstacles were tors, particularly the role of pathology, in gait
made the height of a 1" or 2" door threshold abnormalities observed in older adults? In
or a 6" curb, and performance was compared many studies examining apparendy healthy
to a 0-mm condition (tape marked on the older adults, participants are considered pa-
walkway). No age-related changes in foot thology-free if they don't have a known neu-
clearance over the obstacles were found, but rological, cardiovascular, or musculoskeletal
older adults used a significandy more conser- disorder. Yet, when this population is exam-
vative strategy when crossing obstacles. Older ined carefiilly, many show subde pathologies.
adults used a somewhat slower approach For example, a study on idiopathic gait dis-
speed, a significandy slower crossing speed, orders among older adults found that, on
and a shorter step length. Also, four of the 24 closer medical evaluation, this t}pe of gait pat-
older adults inadvertendy stepped on an ob- tern could actually be attributed to a number
stacle, while none of the young adults did of specific disease processes (44). This sug-
(41). gests that in many instances pathological con-
ditions may be an underlying contributing
Gait Changes in Fallers vs. factor in gait pattern changes seen in older
Nonfallers adults. Pathology within a number of systems

can potentially affect locomotor skills in the
How do the walking characteristics of older adult.
older adults who fall compare to those with
no histoty of falls? While the previous studies COGNITTVE FACTORS
have shown that the gait characteristics of
healthy older adults show few differences Studies have shown that after repeated
when compared to younger adults, older in- falls, older persons develop a fear of falling,
dividuals with a histoty of falls show signifi- and this fear may contribute to changes in gait
cant differences in walking patterns (42). characteristics as well. For example, it has
Older female subjects with poor balance been shown that preferred walking pace, anx-
performance have increased step-width dur- iety level, and depression are good predictors
ing gait. Other studies reported that step- of the extent of fear of falUng in community-
width measured at the heel was significandy dwelling older adults (45). Older adults who
larger in older persons with a history of falls avoid activities because of a fear of falling tend
to walk with a slower pace, and have higher reduction in sensor\' fiinction is part of normal
levels of anxiet)' and depression compared to aging, it will be important to determine ways
adults with litde fear of falling. This has led to optimize environmental factors and use
several investigators to propose that slowed training to improve stability during walking
gait velocit)' among older adults reflects a in older adults.
conscious strategy used to ensure safe gait,
rather than the consequence of specific con- MUSCLE WEAKNESS

straints on walking speed (28, 34, 38).
In other studies examining balance con- Decreased muscle strength has been in-
trol in older adults with a fear of falling, re- dicated as a contributor to locomotor changes
searchers were not sure whether these adults in the older adult. In the section on kinetics
had real problems with balance control, or of the gait cycle, we noted that Winter and
whether the fear of falling itself was affecting colleagues (38) reported a significant decrease
stabilit}' in an artifactual way (46). Thus, it is in push-oft' power during gait in healthy older
possible that cognitive factors, such as fear of adults, which was possibly related to de-
falling, may contribute to changes in gait pat- creased muscle strength.
terns in older adults. Other researchers have studied the
strength of the ankle muscles of 1 1 1 healthy
SENSORY IMPAIRMENTS adults between the ages of 20 and 100 years
(49). They found that maximum voluntary
As noted in the chapter on changes in muscle strength of the ankle muscles began to
balance control in the older adult, pathologies drop The older sub-
in adults in their sixties.
within visual, proprioceptive, and vestibular jects also showed smaller muscle cross-sec-
systems are common among many older tional areas and lengthened twitch contrac-
adults, reducing the availability of information and half-relaxation times. During
tion Irom these senses for posture and gait. maximum voluntary effort, motor nerve stim-
Research comparing the perception of ulation caused no increase in torque in the
vertical and horizontal between six older fall- majority of the older adults. This suggests that
ers and six control subjects (ages 67 to 76 healthy older adiflts are still able to use de-
years) found that the visual perception of ver- scending motor pathways in an optimal man-
tical and horizontal showed no differences be- ner for muscle contraction (49).
tween the fallers and the controls (47). The Studies have also measured the strength
research showed, however, that half of the of upper and lower extremities using a simple
fallers showed problems with recognition of dynamometer (a modified sphygmomanom-
postural tilt when standing on a tilting plat- eter) (50). Results showed that after the age
form.The older fallers also showed a tendency of 75, age is the most significant factor pre-
to leanmore heavily on a supporting frame dicting a drop in muscle strength (other fac-
when standing on one leg when compared to tors included were height, weight, and sum of
control subjects. skinfolds). The
strength recorded for elbow
These experiments imply that older flexion, grip,knee e.xtension, and dorsiflexion
adults who fall may depend on visual cues to was the best indicator of overall limb strength.
identify postural variations; this suggests that Reductions in strength of knee extensor and
they may have proprioceptive dysfunction. flexor muscles for both concentric and eccen-
Thus, normal visual cues may be critical for tric contractions have also been reported in a
these older adults, as part of altered perceptual study comparing healthy older women (66 to
strategies to escape additional falls. However, 89 younger women (20 to 29
years old) with
it has also been reported (48) that threshold years old) (51). There were fewer age-related
levels for detection of optical flow associated differences for eccentric contractions than
with normal sway rise in the older adult. If concentric contractions.
Chapter Twelve A LIFE Span Perspective of MoBiLi-n- 289
Do these reductions in muscle strength and the subject wore a headband with a light-
relate to meaningful changes in hinction? Yes. scattering plastic shield), or (c) stairs were
It has been shown that tallers (mean age 82 painted black with a white stripe at the edge
years) with no clear patholog^' showed signif- of each tread. The stairs were surrounded by
icantly reduced abilit)' of the ankJe and knee a striped corridor.
muscles to generate peak torque and power The results of high-speed film analysis
when compared to a group of age-matched showed significandy slower cadence, larger
nonfallers (52). These results suggest that foot clearance, and more posterior foot place-
muscle weakness (primarily in the ankle mus- ment while subjects walked under the blurred
cles), is a significant contributing factor to condition as compared to the other two stair
balance dysfunction in older adults. High-in- color conditions. The authors further ob-
tensit)' resistance training has been shown to served that foot clearance was larger than that
increase knee extensor muscle strength, mus- obtained during previous pilot work from
cle size, and to enhance ftinctional mobilit\' in their laboratory' on young adults. They con-
fi-ail older adults in their nineties (53). Mean cluded that older subjects walked with larger
tandem gait speed was increased in this group foot clearance during stair descent compared
by 48% after an 8 -week training program. In to young and that gait patterns during
adults
addition, two of the frail older subjects no stair descent were affected by visual condi-
longer used canes as an aid in walk at the end tions.
of the training period.
In summary, age-related reduction in
Age-Related Changes in Other
muscle strength has been found in selected
upper and lower extremini' muscles. Concen-
Mobility Skills
tric contraction is more affected in older fe-
male subjects than eccentric contraction for SIT-TO-STAND
knee muscles. Strength training can improve
functional mobility in older frail adults. How- Research indicates that 8% of commu-
ever, since decreased mobilit\' and increased nity' dwelling older adults over 65 years of age
likelihood for falls is the result of many fac- show some problems in rising from a chair or
tors, not just weakness, strength training bed. As a result, several studies have examined
alone may not be sufficient to improve bal- the sit-to-stand (STS) task in older adults (55,
ance and mobility fiinction in many older 56).
adults with impaired balance. One study compared movement strate-
gies, forces used, and the time taken to rise
Stair-Walking fi-om sitting among young adults, older adults

able to rise without armrests (old able), and
Research has documented that walking older adults unable to from a chair (old
rise
on stairs is associated with the highest pro- unable). Average rise times from a chair were
portion of falls in public places, and that most similar in the young and old able groups 1 .56 (
of these tails occur as subjects walk down the \s 1.83 sec ) , but significantly longer in the old
stairs. To understand the physical require- unable group (3.16 sec). In addition, the
ments of stair-walking in older adults, char- hand forces used by the old able group were
acteristics of stair descent were studied in a significantly less than those used by the old
group of 36 healthy women between the ages unable group.
of 55 and 70 (55). Participants were asked to The old able were mainly different from
walk down a set of stairs under conditions of the young in the amount of time they spent
poor or distorted visual inputs. For example, phase of rising from the chair,
in the initial
(a) stairs were painted black, (b) vision of die which included the time fi-om start to lift-off
stair was blurred (stairs were painted black from the seat. They flexed their legs and
290 Section III MOBILIT\' FUNCTIONS
trunksmore during trials in which they used RISING FROM A BED

no hands to help themselves rise.
While all the elderly subjects reported Are there age-related diflferences in
no significant musculoskeletal or neurological movement patterns used in rising from a bed.'
impairment, a significantly larger proportion To answer this question, adults ranging in age
of the old unable group had a histor\' of ver- from 30 to 59 years of age were videotaped
tebral fractures, decreased vision, dizziness, while rising fi'om a bed (57). As had been re-
poor balance, and falls. Ever)' old unable sub- ported for young adults, there was consider-
had muscle weakness in the lower ex-
ject also able variabilit}' in patterns for rising from a
tremit\', decreased proprioception in the bed among the older group, aged 50 to 59.
hands and feet, and spinaland lower extremity' As was mentioned in our previous chapter, the
deformities such as kyphosis and osteoarthritis most common patterns of bed rising in the
(55,56). 30- to 39-year-old group involved a grasp and
Sequence
Figure 12.9. Frequent pattern of rising from a bed in the 30- to 39-year-olds vs. tlie 50- to 59-year-olds. (Adapted
from Ford-Smith CD, VanSant AF. Age differences in movement patterns used to rise from a bed in subjects in the third
through fifth decades of age. Phys Ther 1 993;73:305.)
Chapter Twelve A LIFE Span Perspective of Mobility 291
push pattern with the upper extremities, a ing joint stiffness, which helps in balance con-
roll-oft or come-to-sit pattern, and a synchro- trol (37).
nous lifting of the lower limbs oft' the bed with Clearly, there are many similarities in
one limb extending to the floor in front of the the gait characteristics of the young child and
other. The older group, consisting of 50- to the older adult. These similarities appear to
59-year-olds, tended to use a more synchro- relate to difficulties with balance control com-
nous lifting pattern, with both legs moved to mon to both groups. Thus, it is not necessar-
the floor simultaneously (Fig. 12.9). No stud- ily true that similarities between the very old
ies to date have been published on patterns and very young are due to a reappearance of
used by die elderly when rising from the bed. primitive reflexes. In this case, the reason is a
Since many elderly people report falls at night functional one: the two groups, for often ver\'
associated with getting out of bed, the need different reasons, have difficulties with the
for such a study is essential. balance system, but use similar strategies to
compensate for those difficulties.
Comparing Gait
Characteristics of
Infants and Elderly: Summary
Testing the Regression 1. There are three requirements for successful
locomotion: (a) the ability to generate a
Hypothesis rhythmic stepping pattern to move the body
forward (progression), (b) the control of bal-
It has been suggested that changes in
ance and (c) the ability to adapt
(stability),
the gait pattern among the elderly are related
changing task and environmental re-
gait to
to the reemergence of immature walking pat-
quirements (adaptation). In the development
terns seen in young infants. Thus, it is hy- of locomotion, these three factors emerge se-
pothesized that, as aging occurs, there is a re- quentially, with the stepping pattern appear-
gression to immature reflex patterns that ing first, equilibrium control next, followed
characterized movement in young infants. by adaptive capabilities.
This regression is thought to result from loss 2. The emergence of independent gait is char-
of higher center control over the primitive re- acterized by the development of many inter-
flexes that reemerge in the very old (54). acting systems with certain hierarchical
What components. An innate pattern generator

are the similarides and difterences be-
creates the basic rhythm of the step cycle,
tween the gait characterisdcs of the very
which can be seen in newborn stepping. In
young and the ver\' old.'
the first year, the gradual development of de-
Both groups show a shorter duration of scending systems from higher neural centers
single-limb stance and a greater relative du- gives the child increasing control over this
ration of double support. This has been in- locomotor behavior. The control of equilib-
terpreted in both groups as an indication of rium, organized at a higher level than that of
decreased balance abilities (13, 15, 30, 34). the pattern generator, develops over a longer
The gait of young walkers has also been period, as do adaptive systems essential to
described as having a wide base of support the integration of reactive and proactive
along with toeing-out, a characteristic ob- strategies into gait.
ser\'ed in the elderly as well (13, 34). It has

3. The development of locomotion behavior
begins prenatally and continues until the
been suggested in both groups that an in-
emergence of mature gait at about 7 years of
creased base of support is used to ensure bet-
age. Stepping behavior is present at birth and
ter balance control.
can be elicited in most infants if they are sup-
both young children (7) and
Finally,
ported and inclined slightly forward. This
older adults (36)show coactivation of agonist early behavior resembles quadrupedal step-
and antagonist muscles during gait. This ping, with flexion of the hip and knee, syn-
again has been described as a way of increas- chronous joint motion, and considerable
292 Section III MOBILI'n' FUNCTIONS
coactivation of agonist and antagonist mus- during the first year. Monographs of the So-
cles. ciet}' for Research in Child Development. Se-
4. In many infants, early stepping disappears at rial 223, vol 56, I99I.
about 2 months of age, possibly due to bio- 2. Prechtl HFR. Continuity- and change in early
mechanical changes in the infant's system, neural development. In; Prechd HFR, ed.
such as an increase in relative body weight. Continuit\' of neural fiinctions from prenatal
Early stepping gradually transforms into a to postnatal life. Clinics in Developmental
more mature pattern over the first 2 years of Medicine. No. 94. Oxford: Blackwell Scien-
life. tific Publications, 1984:1-15.
5. There seems to be agreement among re- 3. De Vries JIP, Visser GHA, Prechd HFR. The
searchers that the ability to integrate postural emergence of fetal beha\ior. I. Qualitative as-
control into the locomotor pattern is the most pects. Early Human Dev 1982;7:301-322.
important rate-limiting factor on the emer- 4. Bradley NS, Smith JL. Neuromuscular pat-
gence of independent walking. terns of stereotypic hindlimb behaviors in the
6. The most significant modifications to the gait first two postnatal months. I. Stepping in
pattern occur during the first 4 to 5 months normal kittens. Dev Brain Res 1988;38:37-
of independent walking. Most of these 52.
changes reflect the child's growing ability to 5. Stehouwer DJ, Farel PB. Development of
integrate balance control with locomotion in hindlimb locomotor behavior in the frog.
these first months. Dev Psychobiol 1984;17:217-232.
7. Studies characterizing gait patterns in older 6. Bradley NS, Bekoff A. Development of lo-
adults have consistently shown that healthy comotion: animal models. In: Woollacott
older adults have reduced walking speed, MH, Shumway-Cook A, eds. Development
shorter stride length, and shorter step length of posture and gait across the lifespan. Co-
than young adults. lumbia, SC: Univ. of South Carolina Press,
8. Proactive locomotor abilities also change 1989:48-73.
with age, with older adults taking more time 7. Forssberg H. Ontogeny of human locomotor
to monitor the visual environment, more control: 1. Infant stepping, supported loco-
time to an upcoming step to avoid an
alter motion and transition to independent loco-
obstacle, and using strategies such as slow- motion. Exp Brain Res 1985;67:480-193.
ing of approach and cross-over time when 8. Thelen E, Ulrich B, Jensen J. The develop-
stepping over obstacles. mental origins of locomotion. In: Woollacott
9. Changes in the characteristics of gait patterns MH, Shumway-Cook A, eds. Development
in older adults are influenced by balance of posture and gait across the lifespan. Co-
ability, leg muscle and changes in
strength, lumbia, SC; Univ. of South Carolina Press,
the availability of sensory information. Cog- 1989:25^7.
nitive factors such as fear of falling and in- 9. McGraw MB. The neuromuscular matura-
attention may also be important contribution of the human infant. NV': Columbia Uni-
tors. 1945.
versity- Press,
10. When evaluating gait patterns of older peo- 10. Thelen E, Fisher DM, Ridley- Johnson R. The
ple, consideration must be given to the un- relationship between physical growth and a
derlying mechanisms contributing to these newborn reflex. Infant Behavior and Devel-
changes. In this way, one can differentiate opment 1984;7:479-93.
between contributions related to pathology 1 1 Peiper A. Cerebral fiinctions in infancy and
vs.aging per se. Only after the systems con- childhood. New York: Consultants Bureau,
tributing towalking pattern dysfunction are 1961.
identified can a clinician design effective 12. WooUacon MH, Shumway-Cook A, WilUams
and appropriate interventions to improve H. The development of posture and balance
gait and thus help older adults achieve a safe control in children. In; Woollacott MH,
and independent life-style. Shumway-Cook A, Development of pos-
eds.
ture and gait across the lifespan. Columbia,
References SC: Univ. of Soutii Carolina Press, 1989:77-

96.
1 . Thelen E, Ulrich BD. Hidden skills: a dy- 13. Bril B, Breniere Y. Posture and independent
namic systems analysis of treadmill stepping locomotion in childhood: learning to walk or
learning dynamic postural control? In: Sav- SW. Neural and vestibular aging associated
elsbergh GJP, ed. The development of coor- with falls. In: Birren JE, Schaie KW, eds.
dination in infancy. Amsterdam: North-Hol- Handbook of the psychology of aging, NY:
land, 1993, 337-358. Van Nostrand & Reinholdt I985;378-399.
14. Okamoto T, Kumamoto M. Electromyo- 28. Craik R. Changes in locomotion in the aging
graphic study of the learning process of walk- adult. In: Woollacott MH, Shumway-Cook
ing in infants. Electromyography 1972; A, eds. Development of posture and gait
12:149-158. across the lifespan. Columbia, SC: Univ. of
15. Sutherland Olshen R, Cooper L, Woo
DH, South Carolina Press, 1989:176-201.
S. The development of mature gait. J Bone 29. Imms FJ, Edholm OG. Studies of gait and
Joint Surg 1980;62-A:336-353. mobility in the elderly. Age Ageing 1981;
16. Gallahue DL. Understanding motor devel- 10:147-156.
opment: infants, children, adolescents. Indi- 30. Gabell A, Nayak USL. The effect of age on
anapolis:Benchmark Press, 1989. variability in gait. J Gerontol I984;39:662-
17. Clark JE, Whitall J. Changing patterns of lo- 666.
comotion: from walking to skipping. In: 31. Drillis R. The influence of aging on the ki-
Woollacott MH, Shumway-Cook A, eds. De- nematics of gait. The geriatric amputee, pub-
velopment of posture and gait across the life- lication 919. National Academy of Science,
span. Columbia, SC: Univ. of South Carolina National Research Council, 1961.
Press, 1989:128-151. 32. Molen HH. Problems on the evaluation of
18. Berger W, Quintern J, Dietz V. Stance and gait. [Dissertation]. Amsterdam: Free Uni-
gait perturbations in children: developmental versity, The Institute of Biomechanics and
aspects of compensatory mechanisms. Elec- Experimental Rehabilitation, 1973.
troencephalography and clinical Neurophys- 33. Spielberg PI. Walking patterns of old people:
iology 1985;61:385-395. cyclographic analysis. In: Bernstein NA, ed.
19. Hass G, Diener HC. Development of stance Investigations on the biodynamics of walk-
control in children. In: Amblard B, Berthoz ing, running, and jumping. Moscow: Cen-
A, Clarac F, eds. Development, adaptation tral Scientific Institute of Physical Culture,
and modulation of posture and gait. Amster- 1940.
dam: Elsevier, 1988:49-58. 34. Murray MP, Kory RC, Clarkson BH. Walk-
20. Assaiante C, Amblard B. An ontogenetic ing patterns in healthy old men. J Gerontol
model for the sensorimotor organization of 1969;24:169-178.
balance control in humans. In press. 35. Murray MP, Kory RC, Sepic SB. Walking
2 1 Palisano RJ. Neuromotor and developmental patterns of normal women. Arch Phys Med
assessment. In: Wilhelm IJ, ed. Physical ther- Rehabil 1970;51:637-650.
apy assessment in early infancy. New York: 36. Finley FR, Cody KA, Finizie RV. Locomo-
Churchill Livingstone, 1993:173-224. tion patterns in elderly women. Arch Phys
22. Fishkind M, Haley SM. Independent sitting Med 1969;50:140-146.
development and die emergence of associ- 37. Woollacott M. Gait and postural control in
ated motor components. Phys Ther 1986; the aging adult. In: Bles W, Brandt T, eds.
66:1509-1514. Disorders of posture and gait. Amsterdam:
23. Horowitz L, Sharby N. Development of Elsevier, 1986:325-336.
prone extension postures in healthy infants. 38. Winter DA, Pada AE, Frank JS, Walt SE. Bio-
Phys Ther 1988;68:32-39. mechanical walking pattern changes in the fit
24. McGraw MG. The neuromuscular matura- and healdiy elderiy. Phys Ther 1990;70:340-
tion of the human infant. New York: Hafner 347.
Press, 1945. 39. Patla AE. Age-related changes in visually
25. Touwen B. Neurological development in in- guided locomotion over different terrains:
fancy. Clinics in Developmental Medicine, major issues. In: Stelmach G, Homberg V,
No 58, Philadelphia: JB Lippincott, 1976. eds. Sensorimotor impairment in the elderly.
26. VanSant AF. Age differences in movement Dordrecht: Kluwer, 1993:231-252.
patterns used by children to rise from a supine 40. Pada AE, Prentice SD, Martin C, Rietdyk S.
position to erect stance. Phys Ther 1988; The bases of selection of alternate foot place-
68:1130-1138. ment during locomotion in humans. In: Pos-
27. Ochs AL, Newberry J, Lenhardt ML, Harkins nare and gait: control mechanisms. WooUa-
cott MH, Horak F, eds. Eugene, OR: Univ. 50. Rice CL, Cunningham DA, Paterson DH,
of Oregon Books, 1992: 226-229. Rechnitzer PA. Strength in an elderly popu-
41. Chen H, Ashton-Miller JA, Alexander NB, lation. Arch Phys Med Rehabil 1989;70:391-
Schultz AB. Stepping over obstacles: gait pat- 397.
terns of healthy young and old adults. J Ger- 5 1 Vandervoort AA, Kramer JF, Wharram ER.
ontol1991;46:M196-M203. Eccentric knee strength of elderly females. J
42. Heitmann DK, Gossman MR_, Shaddeau SA, Gerontol 1990;45:B125-B128.

Jackson JR. Balance performance and step 52. Whipple RH, Wolfson LI, Amerman PM.
widtli in non-institutionalized elderly female The relationship of knee and ankle weakness
fallersand nonfallers. Phys Ther 1989; to falls in nursing home residents: an isoki-
69:923-931. netic study. J Am Geriatr Soc 1987;35:13-
43. Gehlsen GM, Whaley MH. Falls in the el- 20.
derly: Part I, gait. Arch Phys Med Rehabil 53. Fiatarone MA, Marks EC, Ryan ND, Mere-
1990:71:735-738. dith CN, Lipsitz LA, Evans WJ. High-inten-
44. Sudarsk)' L, Ronthal M. Gait disorders sity' strength training in nonagenarians.
among of 50
elderly patients: a survey study JAMA 1990;263:3029-3034.
patients. Arch Neurol 1983;40:740-743. 54. Shaltenbrand G. The development of human
45. Tinetti ME, Richman D, Powell L. Falls ef- motility and motor disturbances. Arch Neu-
ficacy as a measure of fear of falling. J Ger- rol Psychiatr 1928;20:720.
ontol 1990;45:P239-P243. 55. Simoneau GG, Cavanagh PR, Ulbrecht JS,
46. Maki BE, Holliday PJ, Topper AK. Fear of Leibowitz HW^ Tyrrell RA. The influence of
falling and postural performance in the el- visual factors on fall-related kinematic vari-
J Gerontol 1991;46:M123-M131.
derly. ables during stair descent by older women. J
47. Brownlee MG, Banks MA, Crosbie WJ, Gerontol 1991;46:M188-M195.

Meldrum F, Nimmo MA. Consideration of 55. Alexander NB, Schultz AB, Warwick DN.
spatial orientation mechanisms as related to Rising from a chair: effect of age and fiinc-
elderly fallers. Gerontolog>' 1989;35:323- tional ability on performance biomechanics. J
331. Gerontol 1991;46:M91-M98.
48. Warren WH, Blacb\ell AW, Morris MW. Age 56. Millington PJ, Myklebust BM, Shambes GM.
differences in perceiving the direction of self- Biomechanical analysis of the sit-to-stand
motion lirom optical flow. J Gerontol 1989; motion in elderly persons. Arch Phys Med
44:P147-P153. Rehabil 1992;73:609-617.
49. Vander\'oort AA, McComas AJ. Contractile 57. Ford-Smith CD, VanSant AF. Age differ-
changes in opposing muscles of the human ences in movement patterns used to rise from
ankle joint with aging. J Appl Physiol 1986; a bed in subjects in the third through fifth
61:361-367. decades of age. Phys Ther 1993;73:300-309.

Chapter 13
Abnormal Mobility
Introduction Terminal Stance
Abnormal Gait Problem 1 Lack of Hip
.
Musculoskeletal Limitations Hyperextension

Neuromuscular Impairments Problem 2. Inadequate Toe-Off
Weakness Problem 3. Pelvic Retraction
Muscle Tone and Changes in Stiffness Swing
Control Problems Swing
Initial
Sensory Impairments Problem 1. Inadequate Hip Flexion
Somatosensory Deficits Mid-Swing
Visual Deficits Problem 1. Inadequate Knee Flexion
Vestibular Deficits Problem 2. Excessive Adduction
Misrepresentation of Stability Limits Terminal Swing
Adaptation Problems Problem 1 Inadequate Knee
.
Pain Extension
Effects of Impairments on Phases of Gait Stair-Walking
Stance Pathology-Based Descriptions of Abnormal
Foot Contact Loading Gait J
Problem 1. Impaired Heel-Strike Stroke

Problem 2. Coronal Plane Deviations Parkinson's Disease
Mid-Stance Cerebral Palsy
Problem 1 Excessive Knee Extension
. Disorders of Mobility Other Than Gait :
Problem 2. Persisting Knee Flexion Sit-to-Stand (STS)

Problem 3. Excessive Hip Flexion Bed Mobility Skills -
Problem 4. Backvi/ard Lean of Trunk Summary ,^
Problem 5. Lateral Lean of Trunk

Drop in Pelvis
;
Problem 6.
Problem 7. Scissors Gait
.
Introduction lems from a diagnostic perspective, presenting

the kinds of mobility problems often found in
This chapter describes impaired mobil- patients with stroke, Parkinson's disease, and
ity function in the patient with neurological cerebral palsy.
impairments, including abnormalities of gait, Understanding the effects of sensory,
stair-walking, and Using the frame-
transfers. motor, and cognitive impairments on mobil-
work established in our previous chapter on ity fijnction, as well as the types of patients
abnormal postural control, we examine prob- likely to have these problems, will assist the
lems that affect mobility' fimction in the pa- clinician in assessing and planning treatments
tient with an upper motor neuron lesion that are effective in retraining mobility.
(UMN). We examine the various t}'pes of
first
impairments resulting from a lesion, UMN Abnormal Gait

then look at how impairments affect move-
ment strategies within the stance and swing While abnormal gait is a common char-
phases of gait. We also consider how impair- acteristicof many neurological pathologies,
ments affect the abilit}' to adapt strategies to the constellation of underlying problems that
changes in goals and/or environmental de- produce disordered gait varies from patient to
mands. Finally, we list various mobility' prob- patient, even within the same general area of
295
pathology. A patient's problems with gait will creased joint mobility during stance restricts
depend on both the type of impairment and forward motion of the body over the sup-
the extent to which the patient is able to com- porting foot, thus affecting progression. In
pensate for that impairment. Understanding swing, decreased joint mobihty reduces foot
the contribution of these two elements to gait clearance, affecting progression, and appropri-
dysfiinction in the patient with neurological ate foot placement for weight acceptance, af-
dysfimction can be ver)' difficult. As a result, fecting stability. Limited range of motion also
technology such as EMG, kinematic, and/or limits a patient's ability to modify movement
kinetic analysis is often necessary to distin- strategies, thus affecting adaptation. For ex-
guish impairment from compensation. Tech- ample, a patient with limited ankle and knee
nology is thus used extensively in studies ex- flexion will be unable to increase limb flexion
amining gait dysfunction in the patient with during the swing phase of gait to step over an
neurological impairments. obstacle.
There are many people whose research As we mentioned in the chapter on ab-
has contributed to our understanding of normal postural control, musculoskeletal lim-
pathological gait. Among them is Dr. Jacque- itations found in the patient with neurological
lin Perr^', a noted orthopedic surgeon, who dysfunction most often develop secondary to
began her work in gait analysis as a practicing a UMN lesion. Musculoskeletal impairments
physical therapist. Dr. Perry and her col- that particularly affect gait include ankle ex-
leagues at Rancho Los Amigos Gait labora- tensor contractures, knee and hip flexor con-
tory at UCLA have studied gait in many types tractures, and reduced pelvic and spinal mo-
of patient populations. She has published ex- bility.
tensively, including a vet}' comprehensive and

detailed book on normal and pathological gait Neuromuscular Impairments
function (1). Much of the knowledge found
in this section on abnormal gait is based on Many categories neuromuscular
of
research by Dr. Perry and her colleagues (2, problems with neu-
affect gait in the patient
3). rological impairments. These range from

During the recovery of mobility ftinc- force control problems, including both weak-
tion following a neurological lesion, a primary ness and abnormalities of muscle tone, to tim-
focus of therapy will be on helping the patient ing problems.
to regain progression, stability, and adaptive
functions underlying the control of gait. A key WEAKNESS
to developing effective mobility function is
understanding the musculoskeletal and neural Upper motor neuron lesions affect both
constraints or impairments that affect the pa- the non-neural and neural components of
tient's ability to walk and perform other mo- force production. Neural lesions can produce
bility skills. In the following sections we dis- a primary neuromuscular impairment affect-
cuss the constraints on motor control ing the number, type, and discharge fre-
resulting from dysfunction in the different quency of motor neurons recruited during a
systems contributing to mobility function. voluntary contraction, as well as during gait
(1,4). In addition, secondary changes in the
Musculoskeletal Limitations muscle fibers themselves affect the patient's
ability to generate tension. Muscles act in gait
Both soft tissue contractures and bony both concentrically to generate motion, and
constrictions limit joint range of motion eccentrically to control motion. Thus, weak-
(ROM). This movement and po-
constrains ness can result in both the inability to gener-
tentially increases the workload on the mus- ate forces to move the body, as well as unre-
cles, thus affecting a patient's ability to meet strained motions such as foot-slap following
the requirements of gait. In general, de- heel-strike. Foot-slap results from loss of ec-
Chapter Thirteen Abnormal Mobili-H' 297
centric control by the tibialis anterior and iin- creased stiffness of a joint, can affect the free-
controUed plantarflexion (1 ). dom of body segments to mo\'e rapidly with
How much does weakness affect the regard to one another. This limits the transfer
ability to walk independendy? This depencHs of momentum during gait, affecting the pro-
on what muscles are weak, and the capacity' of gression requirements of locomotion (5).
other muscles to substitute for weak muscles Spasticity, or increased stiffness, also
in achieving the requirements of gait. For ex- manifests as excessive activation of muscles in
ample, trunk strength is needed to keep the response to stretch. Stretch-dependent gait
HAT segment upright. However, no signifi- abnormalities are primarily apparent in those
cant trunk deviations occur in gait unless phases of gait in which the spastic muscle is
weakness in the trunk muscles is significant, being lengthened rapidly. For example, rapid
that is, less than a grade 3 on a manual muscle knee flexion, just following weight acceptance
test (1). at the beginning of stance, can result in ex-
Nevertheless, weak hip extensors will cessive activationof a spastic quadriceps, due
have a tremendous impact on the patient's to the rapid lengthening of the muscle.
ability to walk when that patient also has a hip
flexion contracture, requiring a forward lean CONTROL PROBLEMS
posture of the trunk. Strong hip and trunk
extensors will be needed to keep the patient's We define control problems as muscle
trunk ft'om fiirther flexing under the effects of activation problems that manifest as task-spe-
gravity (1). cific locomotor difficulties, such as the inabil-
The inability to recruit motor units, ity' to recruit a muscle during gait, even
whether due to weakness or control prob- though voluntary force generation may be in-
lems, affects the patient's ability' to meet all tact. It is important to note that there is often
three requirements of locomotion, that is, no clear distinction between problems such as
progression, stabilit)', and adaptation. weakness, tone, and control, since they are in-
terrelated in the patient with a neurological

MUSCLE TONE AND CHANGES IN disability'.
STIFFNESS The inabilit)' to recruit a muscle during

an automatic task such posture or gait was
as
As we mentioned in the chapter on ab- discussed in the chapter on abnormal posture
normal postural control, abnormalities of control. Because a patient has some ability to
muscle tone are characteristic of most UMN generate force voluntarily during a manual
lesions. The t)'pe and severity of muscle tone muscle test does not mean that the muscle will
problems vary, depending on the location and perform normally during movement strate-
extent of the neural lesion. Spasticity', which gies needed for gait. Therefore, strength is
is a velocity-dependent increase in the stretch not always a good predictor of performance

is the most common manifestation of
reflex, in locomotion.
abnormal muscle tone seen in the patient with Control problems can also be seen as an
a UMN lesion. Spasticity affects gait in a num- inappropriate activation of a muscle during
ber of ways. gaitwhich is not related to stretch of the mus-
The generation of momentum and the cleand therefore not defined as spasticity. An
transferof momentum to adjacent body seg- example is overactivity' of the hamstrings in
ments is an important aspect of progression, phases of gait when the muscle is not being
and results in an energy-efficient gait. Trans- lengthened (6).
fer of momentum requires that one joint be Coactivation of agonist and antagonist
able tomove fi-eely relative to another joint, muscles around a joint can increase stiffness
changing directions rapidly (5). Therefore, and decrease motion, thereby affecting pro-
changes in the mechanical properties of the gression during gait (1, 6, 7). Some research-
musculoskeletal system, in particular in- ers believe that coactivation of muscles during
gait does not always represent an impairment, cles in the environment. Visually impaired and
but may be a compensatory strategy to dis- blind patients tend to walk more slowly. In
ordered control (7, 8). addition, they appear able to utilize auditory
Control problems in patients with neu- cues to assist in locating obstacles in space
rological lesions, particularly in the cerebel- (U).
lum, can produce problems related to scaling
the amplitude of muscle activit)' during gait, VESTIBULAR DEFICITS
producing what is called an ataxic gait pattern.
Scaling problems were discussed in more de- Patients with vestibular deficits may
tail in the chapter on abnormal postural con- walk more slowly than normal subjects. Other
trol. changes include a prolonged double-support
phase, and a 6.5% longer cycle time than nor-
Sensory Disorders mal subjects (12). Interestingly, when vestib-
ular patients were asked to walk at a normal
Sensation is a critical determinant for velocity, using a metronome to establish the
maintaining gait environments
in natural
pace, their double-support phase duration be-
where we are required to constandy modify came more normal. It is not clear why vestib-
how we move in response to changes in our ular patients seem to prefer a slower gait and
surroundings. Sensory inputs play several im-
whether practicing at faster speeds would im-
portant roles in the control of locomotion.
prove the kinematics of their gait cycle (12).
They are critical to signaling terminal stance
It has been reported that these patients
and serve as a trigger for the inidation of may also show impairments in head stabili-
swing (9). In addition, sensory inputs are nec- when walking
zation during gait, especially in
essary in adapting locomotor patterns to
the dark (13-15). Surprisingly, gaze is equally
changes in environmental demands. This in- stable for vestibular deficit patients and nor-
cludes signaling unanticipated disruptions to
mal subjects during sitting and standing.
gait, as well as the ability to predict and antic-
However, when walking, the ability to stabi-
ipate upcoming obstacles.
lize gaze is impaired and thus patients have
complaints of impaired vision and oscillopsia.
SOMATOSENSORY DEFICITS In addition, eye movements compensate for
Abnormal somatosensory inputs result

head movements more effectively during ac-
tive head rotations than during similar move-
in gait ataxia (10). Gait problems in patients
with sensory ataxia can be due to interruption ments made while walking. It has been sug-
of either peripheral or central proprioceptive gested that this may be due to the predictable
pathways. When this occurs, the patient is

nature of active voluntary head movements vs.
the passive head movements made during lo-
usually no longer aware of the position of the
legs in space, or even of the position of the comotion (14).
body itself With mild sensory dysfiinction, When normal subjects walk or run in the
walking may not appear dark, the amplitude and velocity of head ro-
to be obviously ab-
tation are decreased compared to head move-
normal, if the patient can use vision. How-
ever, ataxia is worse when visual cues are re-
ments during normal walking. However,
duced or tiiese parameters increase for subjects with bi-
inappropriate. Staggering and
unsteadiness increase, and some patients lose
lateral vestibular deficits when they walk in the
the ability to walk (10). dark (15).
VISUAL DEFICITS MISREPRESENTATION OF STABILITY

LIMITS
Loss of vision affects primarily stability
aspects of gait, reducing the patient's ability Many patients with neurological dys-
to modify gait patterns in response to obsta- function have problems with impaired body
Chapter Thirteen Abnormal Mobility 299
image. This can result in a number of gait de- cussion highlights common problems found
\iations, including ipsilateral trunk lean to- in the patient with neurological impairments,
wards the stance leg, resulting in loss of sta- and is not intended as an exhaustive review of
bilit}'. Impaired body image can also result in all gait abnormalities. The reader is referred
inappropriate foot placement, and difficult}' in to the extensive review by Perr\' (
I ) for further
controlling the center of body mass relative to information.
the changing base of support of die feet (1). This method of analysis is important
preparation for our next chapter, which dis-
ADAPTATION PROBLEMS cusses assessment and treatment of mobility
problems. It is the format used for obserx'a-
few studies have focused on the
Vêry tional gait analysis, the primar)' clinical tool
adaptability' of the gait cycle in patients with used to evaluate gait abnormalities. In addi-
neurological impairments. Adaptability in- tion, treatment strategies useful in retraining
cludes the abillDi' to avoid obstacles and to gait are presented within a problem-based
var)' the step cycle in unusual terrains to pre- framework.
vent accidents. Clinicians working with pa-
tients with neurological impairments are well Stance
aware that such problems exist, but there is FOOT CONTACT/LOADING
virtually no research examining exactly how The position of the foot when it meets
adaptation becomes impaired.
the floor at the beginning of the stance phase
of gait has a great impact on both stabilit)' and
Pain progression. Normally, contact is made with
the heel of the foot. This is followed by an
Pain can also cause the patient to alter
eccentric activation of the tibialis anterior
movement patterns used for gait. For exam-
(TA), allowing a controlled plantarflexion of
ple, rapid motion of the knee of a patient with
the ankle, and a smooth transference of
intrinsic joint patholog\' can result in pain.
weight to the entire foot. Heel-strike and the
This will limit knee flexion and affect gait in
subsequent smooth transference of weight to
much the same way as when the patient has
the whole foot are essential to redirecting mo-
weakened quadriceps muscles (1 ).
mentum important to forward progression. In
In addition, pain at the hip is a primar\'
addition, a proper heel-strike foot position se-
cause of persistent hip flexion and inadequate
cures a stable base of support during weight
hip extension during gait. Intra-articular pres-
acceptance, and therefore is important to sta-
sure is least at 30° to 40° flexion. As a result,
bility'.
this is the position most often assumed by pa-
tients who are experiencing pain (
I ). Problem I . Impaired Heel-Strike
Abnormalities at foot-strike and loading

Effects of Impairments are shown in Figure 13.1, and can include:
ON Phases of Gait low heel contact, foot-flat contact, forefoot
contact, contact made with the medial (or lat-
we reviewed
In the previous section, the eral) border of the foot, and foot-slap during
range of impairments commonly found in pa- loading. A wide range of problems occurring
tients with neurological dysfiinction that can at the ankle, knee, and hip joints can reduce
potentially constrain gait. We now look more a patient's abilit}' to use a heel-strike strategy
specifically at how impairments affect pro- during the initiation of stance.
gression, and adaptation require-
stability,
ments within each of the phases of gait. Using Causes
a problem-based format, we first describe the Ankle joint plantarflexion contrac-
problem, then discuss the range of impair- tures. During the stance phase of gait, ankle
ments that can produce the problem. The dis- plantarflexor contractions can impair a pa-
Figure 13.1. Abnormal foot position at heel-strike. Abnormalities at foot-strike include A, low-heel contact, B, forefoot
contact, and C, flat-foot contact. (Adapted from Perry J. Gait analysis: normal and pathological function. Thorofare,
NJ: Slack, Inc., 1992:315,324.)
dent's ability to move the limb over the foot. foot-strike, or alternatively, the heel may
The effect of a plantarflexion contracture on strike, but the foot drops quickly (foot-slap)
foot-strike will vary, depending on its severity. due to inadequate eccentric contraction of the
Fifteen degrees of plantarflexion contracture anterior tibialis. The presence of a rapid foot-
is common in adults with acquired disability, drop following heel-strike suggests that the
and usually results in a low heel contact and underlying impairment is an inactive TA
early flat-foot during loading. A 30° contrac- rather than a spastic or contracted G or soleus
ture is not uncommon in children with spastic (1,3).
cerebral palsy, and produces a forefoot con- Inadequate knee extension durin£[ ter-
tact at foot-strike (1). minal swing. Inadequate knee extension,
Gastrocnemius/soleus spasticity. Se- particularly at the end of the swing phase of
vere spasticity in the gastrocnemius (G) and gait, will keep the knee from fiilly extending,
soleus muscles results in the continued acti- thus affecting placement of the foot in prep-
vation of these muscles during most of gait. aration for stance. Inadequate knee extension
Clinically, it can be difficult to distinguish be- can be due to knee flexion contractures, or
tween a spastic G and a contracted G 1 Of- ( ). fi"om overactivity' of the hamstrings (1, 3).
ten, excessive G activity part of an extensor
is
synergy gait pattern. The strategy for initial Problem 2. Coronal Plane Deviations
contact depends on the knee position. Low-
heel or flat-foot contact is made with an ex- Coronal plane deviations at foot-strike
tended knee, while a flexed knee is associated include excessive inversion (varus foot) or ex-
with a forefoot contact. cessive eversion (valgus foot).
If there is no motion available at the an-
kle, forefoot support continues. Alternatively, Causes
if motion at the ankle is possible, the heel
drops, but the tibia is driven backwards, re- Excessive varus: spastic invertors.
sulting in hyperextension or back-kneeing ( 1 Excessive inversion, referred to clinically as a
3). varus foot, is common in patients with ankle-
Inactivity of tibialis anterior. Inabil- joint spasticity. Varus is seen clinically as the
ity to activate the TA results in a flat-foot at elevation of the first metatarsal head from the
Chapter Thirteen ABNORMAL Mobility 301
floor with the subsequent foot contact made

on the lateral border of the foot only. Exces-
si\e varus foot position tends to be seen pri-
marily in spastic patients.

Because of the many muscles that cross
the ankle joint, there are potentially many dif-
ferent causes of excessive varus during gait.

Inappropriate action of the soleus muscle dur-
ing terminal swing is one of the most com-
mon causes, and results in excessive varus at
heel-strike. Excessive inversion in swing is
often associated with activation of the anterior

tibialis muscle (part of a total flexion synergy)
and inactivity of the toe extensors (1, 3).
Causes
Va{gtis: inactive invertors. Valgus is
seen clinically as excessive eversion, with most Figure 1 3.2. Knee hyperextension is a common gait de-
of the support carried on the medial portions viation in the patient with neurological dysfunction, and
is seen during the stance phase of gait. Knee hyperexten-
of the foot, in particular, the first metatarsal
sion can be the direct result of pathology, such as plan-
head. The most common cause of valgus is
tarflexor spasticity, or alternatively used as a compen-
v\'eakness or inaction by the ankle in\'ertors, satory strategy to control the knee in the presence of
for example, a weak or inactive soleus. Thus, impaired force control in the quadriceps muscle.
a flaccid paralysis tends to lead to a valgus foot (Adapted from Perry J. Gait analysis: normal and patho-
logical function. Thorofare, NJ: Slack, Inc., 1992:324.)
posture (1, 3).
MID-STANCE
has suflRcient mobility to move posteriorly
In mid-stance, the foot is stationary', and past neutral. This is callecH recurvatum. Knee
tibial advance is controlled by the graded ec- hyperextension can occur quickly or slowly,
centric contraction of the soleus muscle. A and usually begins in mid- or terminal stance
smooth progression over the supporting foot and continues into pre-swing. Excessive knee
requires a minimum of 5° of ankle dorsiflex- extension means the tibia cannot advance
ion. As the stance phase continues, the center o\'er the stationan' foot in the stance phase
of pressure mo\'es forward to the metatarsal (1).
heads, and the heel rises. Finally, the center
of mass falls beyond the base of support of the Causes
stance leg, resulting in an acceleration of the
body through free-fall ( 1 ). Plantarflexor contractures. Plantar-
flexor contractures limit tibial advancement
Problem 1. Excessive Knee Extension over the stationary foot during stance. If the
contracture is elastic, that is, able to lengthen
Excessive knee extension can be mani- in response to body weight, the only result
fested as either extensor thrust (primitive ex- may be an inappropriate foot position at foot
tensor synergy or spasticit>' pattern) or hyper- contact, since body weight will lengthen the
extension of the knee. Extensor thrust is plantarflexors, allowing the tibia to advance
defined as a rapid extension of the knee, but (!)•
not into a h^'perextension range, and it usually Plantarflexor spasticity. Spasticity' can
occurs during loading. Hyperextension, prevent knee flexion during loading, and pro-
shown in Figure 13.2, occurs when the knee duce knee hyperextension during the stance
phase of gait. Major compensations for loss of Problem 2. Persisting Knee Flexion
progression due to excessive plantar flexion
Causes
include hyperextension of the knee, and/or
fonvard trunk lean. Patients also compensate
Inadequate activation of plantarflex-
ors. Inadequate gastrocnemius/soleus acti-
by shortening the step length of the other
vation causes excessive ankle dorsiflexion and
limb. Which compensator^' strategy' is used
can increase knee flexion from 15 to 30°. Per-
will depend on a number of factors. Knee mo-
bilitTi' is critical to the hyperextension strategi,-.
sisting knee flexion increases the demands on
In contrast, good hip and trunk extensors are

the Q muscle, which eccentrically contracts to
necessary' for the tnmk flexion strategy (1,3).
stabilize the knee. In addition, weak or inac-
In stance, quad- tive plantarflexors cannot restrain forward

Quadriceps spasticity.
riceps spasticity' will have its greatest effect motion of the tibia through an eccentric con-
traction.
during loading. Remember that during
weight acceptance, there is a brief flexion of Inadequate activation of the soleus also
the knee that assists in absorbing the shock of

results in lossof heel-rise at terminal stance,
loading. Quadriceps spasticit)' results in an ex-
and a loss of terminal stance knee extension.
cessive response toknee flexion and subse- As a result, knee flexion persists.
quent lengthening of the quadriceps, trigger- Inadequate activation of the plantarflex-
ing a stretch reflex response that can limit ors can be caused by: weakness, surgical over-
flexion and result in premature extension of lengthening of the acliilles tendon, or the use
the knee (1, 3).

of a fixed ankle-foot orthosis (1, 3).
Compensation for weak quadriceps. A Knee flexioti contracture. Knee flex-
weak quadriceps (grades 3-1- to 4) will lead to ion contracture can also result in a persisting
difficult)' controfling knee flexion during knee flexion posture during the stance phase
loading. A very weak quadriceps (Q) (grades of gait.
to 3 ) will lead to trouble stabilizing the knee Hamstrings overactivity. Hamstrings
during mid-stance. The primary' compensa- hyperactivity' can manifest as either premature
tion for this is hyperextension of the knee dur- or prolonged activation of the hamstrings ( 1 ).
ing mid-stance, since the forward movement It was originally thought that hamstrings
of the body weight wiU serve as the knee ex- overactivity' was the result of spasticity, that is,
velocit)'- dependent hyperactivity of the
tensor force. When hyperextension is contin-
ued into pre-swing, it prevents the knee from stretch reflex. But researchers have subse-
freely moving during the swing phase. This quentiy found that performing a dorsal rhi-
can slow progression and result in toe-drag zotomy, which involves selectively cutting the
sensor\' nerve roots, does not decrease ham-
(1,3).
There are several disadvantages to the strings hyperactivity in children with cerebral
use of a knee hyperextension strateg)' as a palsy. This suggests that the basis for ham-
compensation for weak Q. First, it limits knee strings hyperactivity is abnormal coordina-
flexion during loading and thus increases the tion, not a simple hyperactive stretch reflex
impact of body weight on the structures of the (1).
stance limb. In addition, it traumatizes the in- Hamstrings activity is ofi:en used to sub-
ternal structure of the knee and can damage stitute for a weak gluteus maximus and ad-
these structures in the long term. The advan- ductor magnus. This helps to stabilize the
tage of the knee hyperextension strategy is
trunk but leads to a mild loss of knee exten-
that it allows a more stable posture, and there- sion in stance.
fore may be a reasonable and appropriate Problem 3. Excessive Hip Flexion

strategy for patients with a \'er)' weak Q ( 1 ).
Pain. Knee flexion during the stance Causes

phase of gait may be avoided in patients who Hip flexor spasticity. Spasticit)' of the
have joint pathology to minimize compres- hip flexor muscles will result in excessive hip
sion of painfiil joints. flexion and a forward trunk posture during
the stance phase of gait, shown in Figure the hip and knees is called a crouch gait and
13.3. is often seen in spastic cerebral palsy as a com-
Hip flexor contracture. Hip flexion pensator}' gait pattern for inadequate hip ex-
contractures result in inadequate hip exten- tension (1, 3).
sion, which can affect both stabilit)' and pro- However, this compensatory strateg)'
gression. During mid-stance, if the hip can't has its own limitations since it increases the
extend to neutral, the trunk will flex forward, demands on the quadriceps muscle to control
bringing the center of mass anterior to the hip the knee. Increased knee flexion also requires
joint. Gravit)' \\\\\ pull the tmnk forward into either excessive ankle dorsiflexion or heel-rise
more flexion, and this places an additional de- onto the forefoot during stance, and thus
mand on the hip extensors to prevent collapse constrains progression (1, 3).
of the forward trunk, and loss of stabilit}' ( 1 ). Compensation for weak quadri-
Lumbar lordosis is a compensator^' pos- ceps. Forward trunk lean brings the body
ture used to reduce the workload of the hip vector anterior to the knee, and is an effective
extensors. Thus, a hip flexion contracture of compensatory strateg)' for stabilizing the knee
15° can be compensated for with increased in response to weak quadriceps. However, it
lordosis, unless there is associated loss of spi- threatens stabilit)'.
nal flexibihty. A 40° hip flexion contracture in Hip extensor weakness. Hip extensor
children can be compensated for with in- weakness can also produce a forward trunk
creased lumbar lordosis due to the flexibility lean that threatens stability.
of the growing spine ( 1 ). Pain. Pathologies of the hip joint
An alternative \\'ay to compensate for a such as osteoarthritis, producing pain, lead to
hip flexion contracture is to flex the knees. a forward-flexed posture that minimizes intra-
This allows the pelvis to be normally aligned articular pressure (16).
despite the hip flexion contracture. Flexion of
Problem 4. Backward Lean of Trunk
Causes
Compensation for weak hip exten-
sors. Backward lean in stance compensates
for hip extensor weakness by bringing the
center of mass behind the hips; it is used for
stabilit)'. However, TA activit>' is needed to

prevent falls in the backward direction 1 ). (
Problem 5. Lateral Lean of Trunk
Lateral lean refers to the lateral lean of

the trunk toward the stance leg.
Causes
Causes of lateral lean of the trunk in-
clude weak hip abductors and adductor mus-

cle contractures.
Figure 13.3. Hip flexion and a forward lean trunk pos-

Problem 6. Drop in Pelvis
ture can result from primary impairments such as hip

flexor spasticity or contracture, or from hip extensor Causes
weakness. Alternatively, this posture can be an effective
strategy to compensate for weak quadriceps, since it
Contralateral hip abductor weak-
brings the line of gravity anterior to the knee joint, thus ness. Hip abductor weakness results in a
stabilizing the knee. drop of the pehds on the side contralateral to
the weakness. This is shown in Figure 13.4. displacement of the thigh past vertical, which
This can also be seen clinically as excessive ad- will appear as medial displacement of the en-
duction of the leg contralateral to the side tire limb.
with the weakness. Medial displacement of Abductor weakness. Weak abductor
the contralateral leg reduces the base of sup- muscles (gluteus medius) result in a contra-
port, impacting stability'. lateral pelvic drop and medial displacement of

Contralateral hip adductor spasti- the entire limb.
city. Hip adductor spasticity' or contracture
can also produce a contralateral drop in the TERMINAL STANCE
pelvis during stance, as the femur is drawn in
medially. Problem 1 . Lack of Hip Hyperextension
Problem 7. Scissors Gait Causes

Scissors gait is characterized by excessive
Hip flexion contractures. Lack of hip
adduction. During the swing phase of gait, as
extension has a great effect on terminal stance,
the hip flexes, excessive adduction produces a
since it is during this phase that the hip is nor-
severe medial displacement of the entire limb.
mally hyperextended. Lack of hip extension
This results in a reduced base of support af-
produces an anterior pelvic tilt and an inability
fecting stability. In severe cases, the adducted
to move the thigh posterior to the hip. This
swing leg catches on the stance limb and im-
results in a shortened step length and reduces
pedes progression (1, 3).
forward progression of the body.
Causes
Problem 2. Inadequate Toe-Off
Adductor Adductor spas-
spasticity.
ticity produces adduction on the ipsUateral

Causes
side. Excessive adduction can result in medial
Spastic £astrocnemius/soleus. Inade-
quate toe-off is often the result of an extended
knee position into terminal stance. Inability to
adequately flex the knee makes toe-off more
difficult,and requires the hip and knee flexors

to work harder to lift the limb and clear the
foot during swing. Toe-drag is the conse-
quence of inadequate knee flexion at pre- and
initial swing (1).
Problem 3. Pelvic Retraction
Pelvic retraction is defined as excessive
backward rotation of the pelvis (I). Pelvic re-

traction during terminal stance is shown in
Figure 13.5.
Causes
Weak plantarflexors. Dynamic back-

ward rotation occurs in terminal stance and is
Figure 13.4. Drop in pelvic position on the contralat-
eral side in response to abductor weakness. (Adapted
usually associated with persistent heel contact
from Perry). Gait analysis: normal and pathological func- due to calf muscle weakness. To maintain a
tion. Thorofare, N); Slack, Inc., 1992:270.) reasonable gait velocity, the pelvis is rotated
INITIAL SWING
Problem 1. Inadequate Hip Flexion
Causes
Hip flexor weakness or inability to ac-
tivate muscles. Normal gait requires only a
grade 2 + , (poor plus) muscle strength in the
hip flexors (1). Hip flexor weakness, produc-
ing inadequate hip flexion, primarily affects the
swing phase of gait. Knee flexion is lost in
swing when there is inadequate hip flexion;
thus, the patient is unable to develop sufficient
momentum at the hip to indirectly flex the
knee. As a result, toe clearance is reduced or
lost. A shortened step is also associated with
inadequate hip flexion. A shortened step can
affect the position of the foot at heel-strike.
When the hip can't be flexed at the initiation
Figure 13.5. Pelvic rotation in terminal stance usually
is
of swing, limb advancement and thus pro-
associated with persistent heel contact due to calf muscle
gression are hampered. At the same time,
weakness. To maintain a reasonable gait velocity, the
pelvis is rotated backward to lengthen the limb and to
placement of the foot in preparation for weight
avoid a shortened step. (Adapted from Perry J. Gait anal- acceptance is affected, challenging stabilit)'.
ysis; normal and pathological function. Thorofare, NJ: There are several compensatory strate-
Slack, Inc., 1992:271.)
gies patients use to achieve foot clearance dur-
ing swing, despite inadequate hip flexion, and
these are shown in Figure 13.6. The first uses
a posterior of the pelvis and activation of
tilt
backward to lengthen the limb and avoid a

the abdominal muscles to advance the swing
shortened step. limb (Fig. 13.6^). The second uses circum-

duction, defined as hip-hike, forward rotation
Swing of the pelvis, and abduction of the hip, to ad-

vance the limb (Fig. 13.62?). The other strat-
The goals to be achieved in swing phase egies used to advance the limb despite hip
include advancement of the swing leg for pro- flexor weakness include contralateral vaulting
gression,and repositioning the limb in prep- (Fig. I3.6C), involving coming up onto the
aration for weight acceptance for stabilit)'. forefoot of the stance limb, or leaning the
Both goals require adequate foot clearance trunk laterally toward the opposite limb (1,3)
during swing, which requires that the swing (Fig. 13.6D)'.
leg be shorter than the stance leg. This is nor- Decreased proprioception. Delayed in-
mally accomplished through the action of the itiadon of swing phase can also result from
hip flexors, which generate enough velocit)' at decreased proprioceptive cues signaling hy-
initiation of swing to flex the knee, allowing perextension in the hip and the termination
the limb to shorten enough so the foot clears of stance.
the support surface. An
important strategy
used to accomplish foot placement involves MID-SWING
the transfer of momentum from the forward Problem I . Inadequate Knee Flexion
moving thigh segment to the shank segment.
This allows the knee to extend in preparation Causes
for the next foot placement with relatively lit-
tle muscle activity' (1, 3).

Knee extension contractures. Con-
tractures of the knee extensors result in an in-
Figure 13.6. Compensatory strategies used to advance the swing leg despite inadequate hip flexion include A, acti-
vation of the abdominal muscles in conjunction with a posterior tip of the pelvis, B, circumduction, C, contralateral
vaulting, or D, leaning the trunk laterally toward the opposite limb. (Adapted from Perry J. Gait analysis: normal and
pathological function. Thorofare, NJ: Slack, Inc., 1992:268.)
ability to freely flex the knee in response to

the momentum generated, limiting foot
clearance and producing toe-drag. This prob-
lem is illustrated in Figure 13.7.
Plantarflexor contractures. Plantar-
flexor contractures also affect foot clearance
during swing by preventing sufficient ankle
flexion to allow toe clearance.
Plantarflexor spasticity. Spastic plan-
tarflexors, like contractures, affect forward
foot clearance during swing. Compensator^'
strategies include a shortened stride length
and reduced gait velocit^•.
Quadriceps spasticity. During swing,
knee flexion can be inhibited by hyperactivity
of the stretched quadriceps.
Problem 2. Excessive Adduction
Causes Figure 13.7. Inadequate foot clearance and toe-drag
Compensatory to hip flexor inactiv- are common gait deviations seen in the patient with neu-
rological dysfunction and can result from many types of
ity. Overuse of the adductors to assist with
impairments. (Adapted from Perry ). Gait analysis: nor-
hip flexion can result in excessive adduction mal and pathological function. Thorofare, N): Slack, Inc.,
of the limb during swing. 1992:315.)
TERMINAL SWING
Problem 1. Inadequate Knee Extension
Causes
Knee flexion contractures. Knee flex-
ion contractures prevent the knee from frilly
extending at the end of swing. This afreets the

patient's ability' to place the fr)ot appropriately
fr)r weight transfer, reducing stability and in-
creasing the need for muscular action to con-

trol the knee (I).
Hamstrings overactivity. Inappro-
priate activation of the hamstrings muscles
during the swing phase of gait
is a major cause
of inadequate knee extension during terminal

swing. While this overactivity occurs during B
lengthening of the hamstrings, researchers
Figure 1 3.8. Pathological gait patterns resulting from A,
have determined that it is not always due to
a total extensor synergy pattern used in stance, and B, a
hyperactivity of the stretch reflex (1).
flexor synergy pattern used in swing. (Adapted from Cage
Flexor synergy. Persistent flexion of JR. Gait analysis in cerebral palsy. New York: MacKeith
the knee throughout the swing cycle is ofren Press, 1991:134.)
associated with the use of a flexor synergy, or
total flexor pattern at all three joints. Use of
the flexor synergy results in an inability to ex- achieved largely through eccentric contrac-
tend the knee while flexing the hip during ter- tions of the hip, knee, and ankle extensors,
minal swing. This is shown in Figure 13. 8B. which control the body position in response
Figure I3.8v4 illustrates an extensor synergy. to the accelerating force of gravity. Energy ab-
This concludes our review of some typ- sorption and a controlled landing are ensured
ical problems afr'ecting gait in the neurologi- through anticipatory activation of the gas-
cally impaired patient, and their potential trocnemius prior to foot contact with the step
causes. Later in this chapter we summarize (17).
gait problems from a diagnostic perspective, This means that in the patient with a
reviewing studies that describe gait patterns neurological deficit, decreased concentric
in stroke, Parkinson's disease, and cerebral control will primarily afreet stair ascent, while
palsy. decreased eccentric control will primarily af-
fect stair descent.

Stair- Walking Patients with an UMN lesion tend to
stair-walk slowly, require the use of rails for
Like level-walking, stair-walking in- support and progression, and in severe cases
volves reciprocal movements of the legs of dyscontrol, are unable to use a reciprocal
through alternating stance and swing phases. pattern for stair-walking. Instead, they bring
Climbing up stairs requires the generation of both feet to the same step prior to progressing
concentric forces at the and ankle
knee to the next step.
(mosdy the knee) for forward and vertical Impaired visual sensation affects antici-
progression. Stability demands are greatest patory aspects of this task. For example, gas-
during the single limb stance phase, when the trocnemius activity, which precedes foot
swing leg is advancing to the next step (17). contact, is less when visual cues are reduced
In contrast to stair ascent, descent is (18).
progression due to inadequate flexion at the

Pathology-Based
hip, knee, and ankle; and (b) inappropriate
Descriptions of Abnormal foot placement, due to incomplete knee ex-
Gait tension and ankle dorsiflexion at the end of
swing (3).
Stroke Researchers have studied muscle pat-
terns in stroke patients to determine the un-
An of temporal distance factors
analysis derlying cause of abnormal gait patterns. De-
shows that patients who have had a stroke spite the similarity' in how stroke patients
walk 50% more slowly than healthy adults, av- walk, the muscle activation patterns underly-
eraging 37 m/min, compared with 82 m/ ing this characteristic gait var\' considerably
min However, the speed of
for healthy adults. (6). Researchers have classified muscle acti-
walking depends greatly on the degree of re- vation problems into the following categories:
cover)'. Researchers have shown a relationship (rt) a mass synerg)' pattern, characterized by
between walking speed and Brunnstrom's flill limb flexion during swing and extension
stages of recovery following stroke (19). Pa- during stance, (b) premature or prolonged ac-
tients in stage 6, defined by Brunnstrom as the tivity' that is nonstretch related, (c) stretch-de-
ability to perform isolated joint movements pendent overactivity' (spasticity'), (d) cocon-
fi-eely in a well-coordinated manner, walk traction, and (e) increased musculoskeletal
considerably faster (.65 m/sec) than do pa- restraint, that is, impaired motion in the ab-
tients in stage 3, defined as the ability to vol- sence of increased EMG activity (3, 6, 20).
untarily initiate movements only within a fiiU- A study examining energy expenditure
limb synerg}' and with marked spasticity (.16 associated with hemiplegic gait suggests that
m/sec). Patients who were at stage 1 or 2 despite its abnormal appearance, the oxygen
were unable to walk (19). cost is quite low. The inefficiency associated
Double-support time in stroke patients with abnormal gait patterns appears to be off-
is increased, with a decrease in stance time by set by slow gait speed. This suggests that am-
the involved leg, and a shortened step by the bulation is not physiologically stressful for the
noninvolved leg. This results in a significant t)'pical stroke patient unless there are cardio-
step asymmetry'. vascular problems as well (2). But this study
Many researchers have described gait examined energy costs in stroke patients
patterns typically seen in hemiparetic patients. based on time walked. When distance walked
Several problems are quite common among is considered, energy expenditure associated
stroke patients; however, EMG studies sug- with hemiparetic gait is twice as much as that
gest that the underlying cause may var)' from of normal gait, because it takes stroke pa-
patient to patient (3). tients, who walk at half the velocity' of normal
Characteristics of the stance phase of adults, twice as long to cover that same dis-
hemiplegic gait include (a) equinovarus foot tance (3).
position, leading to a forefoot or flat-foot
strike during loading; (^) knee hyperextension Parkinson's Disease
in mid-stance with a forward lean of the trunk;
(c) inability to place the hemiparetic leg in a Patients with Parkinson's disease (PD)
trailing positionduring terminal stance, al- characteristicallv show a stooped posture and
lowing the body weight to advance over the shuffling gait pattern. They take smaller steps,
forefoot with subsequent heel-rise. Often, the and usually there is decreased amplitude and
pelvis is retracted on the stance leg and drops speed of movement throughout the body.
on the swing side due to abductor weakness Studies examining temporal distance factors
(3). report that patients with PD walk slower,
The most common problems in the (33.6 m/min compared with 81.6 m/min in
swing phase include (a) toe-drag, impeding age-matched normals). Step length is asym-
Chapter Thirteen Abnormal Mobiliit 309
metric and decreased, while cycle duration is sonism. It is h^'pothesized instead that disor-
increased; thus, despite their rapid steps, dered locomotion results from impairments in
walking speed is slower than in normals (21- postural control and/or the ability to control
24). locomotion. These impairments are consid-
As was true for stroke patients, the de- ered the result of problems in the generation
gree of gait impairment is related to the se- and control of force (29).
verityof the disease progression. Researchers EMG studies suggest that gait patterns
have found a significant relationship between in PD are associated with three types of mus-
walking velocity in patients with Parkinson's cle activation patterns: {a) continuous EMG
disease and the stages of disability as described activity instead of cyclical activity, (b) reduced
by the Schwab Classification of Progression amplitude of muscle activation, and (c) ab-
(25) or the Hoehn and Yahr Classification normal coactivation of muscles.
(26,27). PD patients often have difficulty in ini-
In addition, joint angular displacement tiating gait. Researchers point out that poor
patterns show a smaller range through out the gait initiation is often the result of inadequate
body (21, 28). Specific alterations in the weight shift laterally. However, some PD pa-
stance phase of gait include: (a) lack of heel- tients are unable to initiate stepping, despite
strike; instead, patients make contact with the the ability to shift weight (29).
foot flat or with the forefoot; (
b) incomplete
knee extension during mid-stance; (c) inabil-
Cerebral Palsy
ity to extend the knee and plantarflex the an-
kle in terminal stance, resulting in decreased
forward thrust of the body; (d) forward trunk Cerebral palsy (CP) is a developmental
lean; (e) diminished trunk motion; and (J) re- disorder characterized by a wide range of dis-
duced or absent arm swing. abilities, all having in common central ner-
Decreased motion of the joints is appar- vous system pathology occurring pre- or per-
ent in swing phase as well. While dorsiflexion inatally. CP is classified into several categories
may be exaggerated during the swing phase, of motor abnormalities, including spastic,
decreased hip and knee flexion lead to dimin- athetoid, and ataxic, based on the location of
ished toe clearance. In addition, reduced the lesion and type of motor abnormality
speed and amplitude of motion of the swing (30). Abnormalities can affect one side of the
leg also affect forward thrust of the body (21, body (for example, hemiplegia or hemi-
28). ataxia), all four extremities equally (as in spas-
Gait in PD patients is characterized by tic quadriplegia), or primarily the legs (for ex-
an inability to control momentum. If a patient ample, spastic diplegia). The types of sensor\',
is unable to generate sufficient momentum, motor, and cognitive impairments found in
forward progression is arrested. This is often the patient with cerebral palsy vary widely in
referred to as a frozen gait pattern. In con- each category. In addition, there is much vari-
trast, unrestrained momentum leads to un- ation among individuals within categories.
controlled progression, called a propulsive Much of the work examining patholog-
gait pattern. Propulsive gait disorders may be ical gait in CP has been with children who
due to an exaggerated forward incfination of have spastic hemiplegia or diplegia (7). Tem-
the body, resulting in an anterior displace- poral distance analysis of CP gait indicates
ment of the center of mass (COM) beyond that walking speed is slower, averaging 40 m/
the supporting foot. In some instances, how- min.
ever, propulsive gait is seen in patients who Two of
gait patterns are characteristic
have normal vertical posture but seem unable spastic CP: crouch gait and a genu recur-
a
to oppose forward momentum (21 ). vatum gait pattern ( 1, 31 ). A crouch gait pat-
Many researchers consider that the step- tern is often associated with a bilateral motor
ping mechanism itself is unimpaired in parkin- control impairment, such as spastic diplegia.
310 Section III MOBILI-R' FUNCTIONS
Crouch gait results from excessive hip and Thus, researchers are now considering the
knee flexion, excessive ankle plantarflexion, possibility' that cocontraction represents a
and anterior pelvic tilt during stance and compensatory strategy' aimed at stiffening a
swing phases of gait. Foot-strike is abnormal, joint to compensate for postural instability' or
with an equino\'arus foot posture and most paresis (7, 32, 33).
often forefoot contact. This foot position is The particular gait profile seen in an in-
continued through the stance phase of gait. dividual will reflect a combination of the fac-
Excessive plantarflexion and knee and hip tors just listed. Thus, each individual with CP
flexion are seen during foading and continued will present a slightiy different gait pattern.
through the stance phase of gait. Excessive Heart rate and oxTgen rates are higher
flexion persists into terminal stance, and the in CP children than for age-matched normals.
pre-swing phase is minimal or absent due to Researchers believe that this is because the
an inabilit)- to extend the hip and knee. flexed posture, which of the crouch
is t}'pical
The swing phase of gait also shows ex- gait pattern, requires additional muscle activ-
cessive ankle, knee, and hip flexion. Often, ity' for stability'. Interestingly, the physiologi-
foot-floor clearance is greater than normal, cal costs of walking decrease in normal chil-
due to excessive flexion of the swing limb (7, dren as they get older. In contrast, the
31). physiological costs of walking increase as chil-
A genu recur\'atum gait pattern presents dren with CP get older. Why does this hap-
the opposite clinical picture to the crouch gait pen? Increased physiological costs of walking
pattern. This gait pattern is characterized by are not due to an increase in motor abnor-
knee hyperextension during stance and exces- malities Ln CP, since it is a nonprogressive dis-
sive ankle plantarflexion. Hip flexion and for- ease. Instead, researchers believe that oxygen
ward lean of the trunk may occur as the pa- rates associated with walking increase as CP
tient leans forward to balance over a children get older because changes in body
plantarflexed foot. Loading is onto the fore- morphologN', including increased body
foot due to inadequate knee extension and ex- weight and size, interact with impaired motor
cessive plantarflexion during swing. During control. This results in an increase in the phys-
swing, toe-drag constrains progression, re- iological cost of gait in older children. As a
quiring contralateral trunk lean to free the result, the older CP child may walk less and
foot and advance the thigh. The genu recur- increasingly rely on a wheelchair (2).
vatum gait pattern is more common in uni-
lateral motor impairments such as in spastic Disorders of Mobility
hemiplegia (7, 31). Other Than Gait
EMG studies suggest that abnormal gait
in spastic CP may be classified into the follow- During the performance of transfer ac-
ing problems: (a) defective recruitment of tivities such as sit-to-stand (STS), rolling, and
motor units, referred to as a paresis or weak- rising from a bed, healthy voung adults tend
ness pattern; ( b) abnormal velocit}'-dependent to use momentum to move the body
recruitment during muscle stretch, the so- smoothly and efficiendy from one position to
called spasticit)' pattern; (c) nonselective another. A momentum strateg\' requires the
activation of antagonist muscles with a loss of generation of concentric forces to propel the
a normal reciprocal inhibitor\- pattern, called body and eccentric forces to control motion,
the cocontraction pattern, and (d) problems thus ensuring stabilit}'. Momentum generated
associated with musculoskeletal restraint due through movement of the trunk is transferred
to changes in mechanical properties of mus- to the legs, and the body moves smoothly,
cles, the non-neural problem pattern (7). without stopping, to the new position. The
Interestingly, in children with spastic abiUt}' to transfer momentum from trunk to
hemiplegic cerebral palsy, a cocontraction lower extremities, which is a characteristic
pattern of muscle activity' was found in both common to most transfer tasks, requires un-
the hemiplegic leg and the noninvolved leg. restrained motion of the trunk (34, 35).
An alternative strategy that can be used acteristic patterns used by stroke patients to
when performing transfer tasks is a force-con- achieve these skills (36-39).
trol strategy'. This strategy' is characterized by
frequent stops. Forces are generated in one Sit-to-Stand
body segment move the body to an interim
to
position of stability. Then force is generated Many patients with UMN lesions tend
in an adjacent body segment to fiarther propel to use a force-control strategy to accomplish
the body to the new position. For example, sit-to-stand (STS) position. Often, this is be-
when using a force-control strategy to move cause of a combination of impairments affect-
from STS, the trunk moves forward, bringing ing both stability and progression aspects of
the COM over the feet. Then forces are gen- themovement. In addition, this is the strategy
erated to lift the body to the vertical position. most commonly taught by clinicians when re-
The force-control strategy ensures stability, training transfer tasks.
but requires greater forces for progression. In When using a force-control strategy to
some cases, the arms are used to generate move from STS, the trunk moves forward,
force, assisting with progression and stability' bringing the COM over the feet; then, forces
(34,35). are generated to lift the body to the vertical
Limitations in the ability to activate position. This strategy emphasizes die control
muscles concentrically generally affect the of stability, but cannot make use of momen-
progression requirement of mobility skills, tum because of the breaks in the movement.
that is, the ability to move the body. Inability Thus, it is a less efficient approach (34, 35).
to activate muscles eccentrically impairs Impaired force control affects the STS
the ability to control motion, affecting sta- task in two ways. The inability to activate
bility. trunk, hip, and knee muscles concentrically
There are many reasons why the neu- limits the generation of propulsive forces to
rologically impaired patient tends to use a move and lift the body. Loss of eccentric con-
force-control strategy during transfers. Pos- trol limits the patient's ability to control hor-
tural control problems limiting stability, izontal motions of the center of mass and thus
cardiovascular problems such as orthostatic impairs stability (34, 35).
hypotension, and dizziness complaints, all Decreased spinal mobility and dimin-
may require a patient to move slowly and ished motion in the hips, knees, and ankles
make interim stops during the task. For ex- will restrict a patient's ability to move freely.
ample, when arising from a bed, a patient with This affects momentum and force-control
orthostatic hypotension would need to sit for strategies, but primarily the momentum con-
a moment on the side of the bed before stand- trol strategy. This is because the ability to
ing up, or risk a sudden drop in blood pres- transfer momentum from one body segment
sure and loss of balance. The overreliance on to another requires freedom of motion in the
a force-control strategy and upper extremity joints (34, 35).
control during transfer tasks, however, can Decreased postural control impairs the
limit these patients' adaptability in response movements of the
ability to effectively control
to changing environmental conditions. For COM, and represents a major constraint on
example, they may find it difficult to stand up the STS task. One of the most frequendy seen
independently from a chair without arms (34, problems in patients with impaired stability is
35). falling in the backward direction when trying
There have been many studies examin- to stand up. This results when the patient pre-
ing pathological gait in neurologically dys- maturely generates propulsive forces to lift the
few studies
fiinctional patients. In contrast, body, before the COM is adequately posi-
have systematically explored problems con- tioned within the base of support of the feet.
straining other mobiht)' skills in these pa- Sensory impairments affect a patient's
tients. Much of the information available ability to determine the position of the body
comes from anecdotal descriptions of char- in space, particularly placement of the COM
312 Section III MOBILm' FUNCTIONS
with respect to the supporting surface. Force-

control problems limit the patient's abiiit\' to Summary
control horizontal mo\ements of the COM
from over the buttocks to the new base of 1. while abnormal gait is a common character-
support, the feet. istic of many neurological pathologies, the
Perceptual impairments including im- constellation of underlying problems that pro-
paired body image, inappropriate internal duce disordered gait will vary from patient to
representarions of stabilin,' limits, and abnor- patient depending on (a) primary impairments
such as Inadequate activation of a muscle,
mal motion perception (dizziness) also affect (fa)
secondary impairments, such as contractures,

a patient's ability' to safely accomplish STS.
and (c) compensatory strategies developed to
Finally, cognitive impairments can also
meet the requirements of mobility in the face
significandy limit a patient's abilit}' to perform
of persisting impairments.
a task safely. For example, a patient may tr\'
2. Musculoskeletal impairments constrain move-
to sit down before he or she is appropriately ment and increase the workload on the mus-
positioned with respect to a chair. meet the
cles, affecting a patient's ability to
requirements of gait. Decreased joint mobility
Bed MobUity Skills during stance restricts forward motion of the

body over the supporting foot, affecting pro-
Bed mobilit}' skills include changing po- gression. In swing, decreased joint mobility
sition while in bed (rolling supine to side-ly- reduces foot clearance, affecting progression,
and appropriate foot placement for weight ac-
ing or prone), and getting out of bed, either
ceptance, affecting stability.
to a chair or standing up. Researchers have
3. Neuromuscular impairments affecting gait in-
found that normal young adults use a varieri'
clude weakness, abnormalities of muscle
of momentum-related strategies when per-
tone, and task-specific control problems.
forming bed mobilitv' skills. There is incredi- Task-specific control problems consist of (a)
ble variet)' in how people move; in fact, none the inability to recruit a muscle during an au-
of the young adults tested used exactiy the tomatic task such as posture or gait; (fa) inap-
same strategy' twice! In contrast, force-control propriate activation of a muscle during gait,
mo\ement strategies are frequentiy used by which is not related to stretch of the muscle;
neurologically impaired patients, and are (c) coactivation of agonist and antagonist
characterized by frequent starts and stops. As muscles around a which increases stiff-
joint,
ness and decreases motion; (di problems re-

mentioned previously, there are many reasons
lated to scaling the amplitude of muscle activ-
why a force-control strateg\' may be more ap-
ity during gait.
propriate in the neurologically impaired pa-
4. Sensory disorders can lead to problems in the
tient than momentum strategy' (36).
a
following areas of locomotor control: (a) sig-
The most common approach to rolling naling terminal stance and thus triggering the
shown normal young adults imohes
b\- initiation of swing, (fa) signaling unanticipated
reaching and lifting with the upper extremity', disruptions to gait, and (c) detecting upcom-
flexing the head and upper trunk, and lifting ing obstacles important for modifying gait to
the leg to roll onto the side, then over to changes In task and environmental condi-
prone. Most healthy young adults did not tions.
show rotation between the shoulders and pel- 5. Impairments can manifest as problems affect-
vis, assumed by many clinicians to be an ining the patient's abilities to meet the progres-
variant feature of rolling (36).
sion, stability, and adaptation goals inherent
in both the stance and the swing phase of gait.
Because bed mobilit}- skills are primarily
A careful analysis of movement patterns can
initiated by movement of the head, upper
lead the clinician to generate multiple hy-
trunk, and shoulders, impairments that afreet
potheses about the potential underlying
these structures (such as weakness and or causes of gait problems.
range of motion limitations) will limit perfor- 6. During the performance of transfer activities
mance of these skills. such as sit-to-stand (STS), rolling, and rising
Chapter Thirteen ABNORMAL Mobility 313
from a bed, healthy young adults tend to use of posture and gait. Amsterdam: Elsevier
a momentum strategy, which requires the gen- 1986, 177-183.
eration of concentric and eccentric contrac- 1 1 Ashmead DH, HQl EW, Talor CR. Obstacle
tions to control motion, and ensures stability. perception by congenitally blind children.
In contrast, a force-control strategy, charac- Perception and Psychophysiology 1989;
terized by frequent starts and stops, is fre- 46:425-433.
quently used by neurologically dysfunctional 12. Kirkpatrick R, Tucker C, Ramirez J, et al.
patients. This is related to impairments affect- Center of gravity' control in normal and ves-
ing both stability and progression aspects of WooUacott M, Horak F,
tibulopathic gait. In:
the movement. This is also the strategy most eds. Posture andcontrol mechanisms
gait:
commonly taught by clinicians when retrain- Eugene, OR: Universit)' of Oregon Books,
ing transfer tasks. 1992:260-263.
13. Takahashi M, Hoshikawa H, Tjujita N, Aki-
yama I. Effect of labyrinthine dysfunction
References upon head oscillation and gaze during step-
ping and running. Acta-Otolaryngol (Stockh)
1. Perry J. Gait analysis: normal and pathologi-
1988;106:348-353.
Thorofare, NJ: Slack Inc., 1992.
cal fiinction.
14. Grossman GE, Leigh RJ. Instability of gaze
2. Waters RL. Energ)' expenditure. In: Perry J.
during locomotion in patients with deficient
Gait analysis: normal and pathological func-
vestibular fiinction. Ann Neurol 1990;
tion. Thorofare, NJ: Slack Inc., 1992.
27:528-532.
3. Montgomery' J. Assessment and treatment of
locomotor deficits in stroke. In: Duncan PW,
15. Pozzo T, Berthoz A, Lefort L, Vitte E. Head
stabilization during various locomotor tasks
Badke MB, eds. Stroke rehabilitation: the re-
in humans. II. Patients with bilateral periph-
cover}' of motor control, Chicago: Year Book
eral vestibular deficits. Exp Brain Res 1991;
Medical Publishers, 1987:223-259.
85:208-217.
4. Duncan P, Badke MB. Stroke rehabilitadon:
the recovery of motor control. Chicago: Year
16. Eyring EJ, Murray W. The effect of joint po-
Book Medical Publishers, 1987. sition on the pressure of intra-articular effu-
5. Oatis CA, Perspectives on the evaluation and
sion. J Bone Joint Surg 1965;47A:313-
322.
treatment of gait disorders. In: Montgomery
PC, Connolly, BH, eds. Motor control and 1 7. McFadyen BJ, Winter DA. An integrated bio-
framework and mechanical analysis of normal stair ascent and

physical therapy: theoretical
practical applications. Hi.\son, TN: Chatta- descent. J Biomechanics 1988;21:733-744.
nooga, 1990:141-155. 18. Simoneau GG, Cavanagh PR, Ulbrecht JS,
6. Knutsson E, Richards C. Different t\pes of Leibowitz HW, Tyrrell RA. The influence of
visual factors on fall-related kinematic vari-
disturbed motor control in gait of hemipa-
retic patients. Brain 1989;102:405-i30. ables during stair descent by older women, J
7. Crenna P, Inverno M, Frigo C, Palmieri R, Gerontol 1991;46:M188-M195.

Fedrizzi E. Pathophysiological profile of gait 19. BrandstaterM,deBruinH, GowlandC,etal.:
in children with cerebral palsy. In: Forssberg Hemiplegic gait: analysis of temporal vari-
H, Hirshfeld H, eds. Movement disorders in ables. Arch Phys Med Rehabil 1983;64:583-
children. Basel: Karger, 1991:186-198. 587.

8. Shumway-Cook A. Equilibrium deficits in 20. Berger W Altenmueller E, Dietz V. Normal
children. WooUacott MH, Shumway-
In: and impaired development of children's gait.
Cook A. Development of posture and gait Human Neurobiology 1984;3:163-170.
across the lifespan. Charleston, SC: Univer- 21 Knuttson E. An analysis of Parkinsonian gait.
of South Carolina Press, 1989:229-252.

sit)' Brain1972;95:475^86.
9. Smith JL. Programming of stereotyped limb 22. Murray MP, Sepic SB, Gardner GM, et al.
movements by spinal generators. In: Stel- Walking patterns of men with parkinsonism.
mach GE, Requin J, eds. Tutorials in motor Am J Phys Med 1978;57:278-294.
behavior. Amsterdam: North -Holland. 1980: 23. Blin O, Ferrandez AM, Serratrice G. Quan-
95-115. titati\'e analysis of gait in Parkinson patients:
10. Katoka S, CroU GA, Bles W. Somatosensory' increased variability of stride length. J Neurol
ataxia. In: Bles W, Brandt T, eds. Disorders Sci 1990;98:91-97.
24. Martin JP, Hurwitz LJ. Locomotion and the Electroencephalogr Clin Neurophysiol 1982;
basal ganglia. Brain I962;26I-289. 53:538-548.
25. Schwab RS. Progression and prognosis in 33. Leonard CT, Hirshfeld H, Forssberg H. The
Parkinson's disease. J Nerv Ment Dis I960; development of independent walking in chil-
130:556-572. dren with cerebral palsy. Dev Med Child
26. Hoehn MM, Yahr MD. Parkinsonism: onset, Neurol 1991;33:567-577.
progression and mortality. Neurology 1967; 34. Schenkman MA, Berger RA, Riley PO, Mann
17:427^35. RW, Hodge WA. Whole-body movements
27. O'Sullivan S. Parkinson's disease: physical re- during rising to standing from sitting. Phys
habilitation. In: O'Sullivan S, Schmitz T. Ther 1990;10:638-651.
Physical rehabilitation: assessment and treat- 35. Carr J, Shepard R. Motor relearning pro-
ment. 2nd ed. Philadelphia: FA Davis, gramme for stroke. Rockville, MD: Aspen
1988:481-493. Systems, 1987.
28. Stern GM Franklyn SE, Imms FJ, Prestidge 36. Davies P. Steps to follow. London: Heinne-
SP. Quantitative assessments ot gait and mo- man, 1985.
bility' in Parkinson's disease. ] Neural Transm 37. Charness A. Stroke/head injury: a guide to
Park DisDement Sect 1983;19:201-214. frmctional outcomes in physical therapy man-
29. Rogers M. Motor control problems in Par- agement. Rockville, MD: Aspen Systems,
kinson's disease. In: Contemporary manage- 1986.
ment of motor control problems: proceed 38. Bobath B. Adult hemiplegia. London: Hein-
ings of the II Step Conference. Alexandria neman, 1978.
VA: APTA, 1991:195-208. 35. Millington PJ, Myklebust BM, Shambes GM.
30. Bobath B. The very early treatment of cere- Biomechanical analysis of the sit-to-stand
bral palsy. Dev Med Child Neurol 1967; motion in elderly persons. Arch Phys Med
9:373-390. Rehabil 1992;73:609-617.
31. Gage JR. Gait analysis in cerebral palsy. New 36. Richter RR, VanSant AF, Newton RA. De-
York: Mac Keidi Press, 1991. scription of adult rolling movements and hy-
32. Berger W, Quintern J, Dietz V. Pathophysi- pothesis of developmental sequences. Phys
olog)' of gait in children with cerebral palsy. Ther 1989;69:63-71.
Chapter 14

THE Patient with Mobility
Disorders
Introduction Treatment of Gait

Assessment Treatment at the Impairment Level
Assessing at the Functional Level Preambulation Skill Training
Three-Minute Walk Test Limitations on Impairment Level
Self-Paced Velocity Training
Energy Efficiency Treatment at the Strategy Level
1^ Measuring Stance Phase
^
H^^^
Stability
Quantifying Temporal/Distance
Factors
Impaired Heel-strike
Plantarflexion Contractures
^^^^âies for Assessing Mobility Plantarflexor Spasticity
BBjPBbuke Mobility Skills Profile Inactivity of the TA
Functional Independence Measure Coronal Plane Foot Problems
Dynamic Gait Index Mid-Stance Problems
Limitations of Performance-Based Knee Hyperextension
Measures Excessive Knee Flexion
Assessing at the Strategy Level Forw/ard Flexed Trunk
Observational Gait Analysis Contralateral Pelvic Drop
Stance Phase of Gait Terminal Stance
Swing Phase Swing
Forms for Observational Gait Analysis Pre-swing
Rancho Los Amigos Gait Analysis Mid-swing
Form Terminal Swing
Mobility Assessment Form Assistive Devices
Gait Assessment Rating Scale (GARS) Treatment at the Functional Task Level
Limitations to Observational Gait Task-Oriented Gait Training
Analysis Stair-Walking
Assessing at the Impairment Level Retraining Other Mobility Skills
Transition to Treatment Setting Goals Sit-to-Stand (STS)
^
B<
Long-Term Goals
Short-Term Goals
Retraining a Force-Control Strategy
Retraining a Momentum Strategy ^
^ Defining Goals Based on Task Bed Mobility Skills ^
Requirements of Gait Rolling »
Progression Rising from a Bed j
Stability Summary
Adaptation
Introduction clinical decision-making process, hypothesis-

oriented clinical practice, a model of disable-
This chapter presents a task-oriented ap- ment, and a theory of motor control.
proach to assessing and treating the patient In a task-oriented approach, assessment
with mobility' dysfunction. In previous chap- targets three levels of behavior: (a) the func-
ters we defined a task-oriented approach with tional task level, (b) the sensor)' and motor
reference to four important concepts: the strategies used to perform functional tasks.
315
316 Section III iVIOBILI'R' FUNCTIONS
and (c) the underh'ing senson', motor, and By what criteria should clinicians determine
cognitive impairments that constrain motor differences in ambulaton- status.' WTiat are the
control. The goals of a task-oriented approach minimum distance and velocity- requirements
to retraining include (a) resolve or prevent for independent communit\' ambulation? As
underlying impairments, ( b) develop eftecti\'e noted in the chapter on normal mobilit}- func-
task-specific strategies, (c) retrain functional tion, several researchers have examined walk-
tasks, and (
d) adapt task-specific strategies so ing characteristics in neurologicalh- intact
that txinctional tasks can be performed in These studies have found that normal
adults.
changing emironmental contexts. men benveen the ages of 20 to 60 years walk
An important part of the clinical inter- about 82 m/min, or 270 ft/min, which is
vention process is the abilit\- to generate mul- 3.06 miles per hour (mph). Normal women
tiple h\potheses about the potential causes of the same age walk slighd\- slower, between 74
dysfiinction in the patient, and to systemati- and 78 m/min, or 244 to 257 fi;/min, which
cally test those h\potheses in order to refine is mph (1-3).
2.76 to 2.91
one's understanding of the problems contrib- Researchers suggest that in order to be
uting to loss of fiinction. a communit}- ambulator, patients need to be
We now apply this approach to assess- able to walk at greater than 33% of a normal
ment and treatment of mobilin,' problems in adult's velocity-, or
about 1.0 mph. In addi-
patients with upper motor neuron lesions. We tion, theyneed to be able to walk a minimum
begin with a review of some of the tests and of 300 m, or about 1000 fi:. This suggests diat
measurements that can be used to document the abilit}- to cover 1000 fi: in approximately
functional abilities related to mobilit\-. We 11 '72 minutes is a minimum requirement for
then look at the process of obsenational gait communit}' ambulation (2, 3).
analysis, an approach to assessing gait strate- One study examined the minimum re-
gies. (Assessment of senson-, motor, and cog- quirements for a range of instrumental activ-
nitive systems is described in the chapter on
ities of daily li\ing (L^DL) in the Los Angeles
assessment and treatment of the patient with area to determinewhether the criteria used by
postural disorders.) The last half of the chap- clinicians to
judge independence in the com-
ter addresses issues related to retraining mo- munit}- were consistent with actual distances
bilit\- skills in the patient with neurological and velocities needed to function indepen-
impairments. dendy (4). The results suggest that the fol-
lowing minimum standards may be required
Assessment to be considered an independent communin-
ambulator:
Assessing at the Functional Level
1. The abilit\' to walk 300 meters, or 1000
Performance -based measures quantif,' feet;
the patient's functional walking abilities, but
2. The abilitA- to achie\-e 80 m/min veloc-
do not address the qualit\' of movement pat-
ity- for approximately 13 to 27 m in or-
terns,nor the underhing senson-, motor, and
der to cross a street safel\- in the normal
cognitive determinants of function. Perfor-
time allotted by stoplights;
mance-based functional assessment can be ex-
3. The abilit>- to negotiate 7- to 8 -inch
pressed witii reference to die le\-el of assis-
curbs independentiy (with assistive de-
tance a patient requires when performing
vices as needed);
mobilit>' skills, or in relationship to the
tem- 4. The abilit>- to turn the head while walk-
poral and distance characteristics of a
patient's ing, without losing balance.
gait.
Often, a goal of mobiIit>' retraining is to

In addition, the study found that clini-
help the patient become a functional
com- cians generally underestimated the distance
munit\- ambulator. But what does this mean.- and speed needed to fijnction independentiy
i
Chapter Fourteen Assessment and Treatment of the Patient with Mobility Disorders 317
within a residential or commercial community cause the 12-minute test is time-consuming
(4). This may be because tests of normal ac- and fatiguing to the patient, several research-
tivities of daily living (ADL) skills, for exam- ers examined the validity of a shortened ver-
ple, the Functional Independence Measure, sion of the test, determining the performance
often define complete independence in loco- of patients at 2 and 6 minutes (8). These re-
motor skills as being able to walk 150 ft safely searchers concluded that, while the 12-min-
(5). However, this standard may underesti- ute test has excellent test-retest reliability, the
mate the requirements for being truly inde- 2-minute test is equally reliable, though
pendent within the community. slightly less sensitive in discriminating a pa-
In carrying out ADL and lADL tasks, tient's level of exercise tolerance. They con-
the average person walks approximately 300 cluded that the use of 12 minutes when
m (or about 1000 ft) per day, with an average assessing exercise tolerance is not critical
walking speed of 80 m/min, or approximately (8).

3.0 mph (4). Thus, the process of carrying out In our own clinic, this research has been
ADL requires about 4 minutes of walking per applied to the development of a 3-minute
day for the average person (6). walk test, assessing walking tolerance in the
Walking must not carry too high a price patient with neurological dysfianction. We use
with respect to cardiovascular output and en- a premeasured established path in corridors
ergy expenditure. The patient who is able to that are not subject to hea\T amounts of traf-
walk 1000 ft but is fatigued to the point of fic. Standardized instructions are given to pa-
exhaustion afterwards, cannot be considered tients to walk at a comfortable pace, using
a ftmctionai community' ambulator (3). whatever assistive device they would use when
Finally, safet)' is an attribute of fiinc- walking outside their homes, and stopping to
tional locomotion skills. A patient must be restwhenever they need to. Patients are al-
able to meet the stabilit)' requirements of lo- lowed to rest as needed, though the clock is
comotion to be fianctionally independent in not stopped during the rest period. Thus, an
gait. This includes (a) the integration of re- increased number of rest periods is reflected
active balance strategies into gait to recover in a shorter distance traveled.
stability in response to unexpected perturba- Variables collected during the 3-minute

tions; and (b) the abilit)' to use anticipatory walk test include: distance covered, calculated
strategies to avoid or accommodate upcom- as self-paced velocit)', number of rests re-
ing obstacles. Thus, an essential part of fijnc- quired, number of deviations from a 15-inch
tional locomotion is the ability to adapt gait path, and heart rate before and after the walk.
to both unexpected as well as anticipated dis-
ruptions. Self- Paced Velocity
The following series of tests are exam-
ples of ways in which fiinctional mobility' skills Self-selected velocit}' represents a cu-
can be documented using parameters such as mulative quality score of a patient's ability and
distance, speed, or velocity, cardiovascular ef- confidence in walking (9). Converting the pa-
ficiency, stability, and adaptabilit}'. tient's self-selected gait velocit)' to a percent-
age of normal (based on a normal score of 82

THREE-MINUTE WALK TEST m/min) can be an effective way to commu-
nicate locomotor abilities to patients, their
In our clinic, patients who have some families, and insurers (10). Patients who walk
locomotor ability are asked to walk at a self- at less than 30% of normal do not usually be-
selected speed for a period of 3 minutes. The come community ambulators because it takes
3-minute walk test is a variation of the 12- too long to cover the required distances in-
minute walking test, a tool designed to ex- volved in LADL (10).

amine exercise tolerance in patients with As part of a study examining the effects
chronic respiratory disease (7). However, be- of exercise on balance mobilit\' and the like-
318 Section III i\IOBILIT\' FUNCTIONS
lihood for falls in older adults with a histon, rologically intact indi\-idual are determined by
of recurrent falls, scores on the 3-minute walk using published tables and charts (3).
test were compared bet\\een healthy older
adults ages 65 to 90 with no neurological im- Measuring Stability-

pairments and older adults the same age with
a histon,- of imbalance and falls. Results from Stabiht\- during nonperturbed gait is
this study found that neurologically intact documented by reporting the number of de-
older adults were able to walk 727±148 ft in viations from a 15-inch path and/or changes
3 minutes v\ith no loss of balance, compared in velocity- the patient demonstrates during
with 323±166 ft in the group of fallers. In the 3-minute walk test, and any physical as-
addition, the older adults with a histon.- of im- sistance required to prevent a fall. Figure 14.1
balance lost balance an average of four times gives an example of the scoring for a patient
during the 3-minute test (11). who has completed a 3-minute walk test. The
resultsof his gait assessment suggest that the
Energ\- Efficiency- patient is not a ftinctional communir\' ambu-
lator at this point.
Energy- cost measurements can quantif\'
the physiological costs of walking (6). Heart QUANTIFnNG TEMPORAL/
rate is a standard indicator for relative exercise DISTANCE FACTORS
intensity-and work rate. Heart rate can be
monitored through manual palpation or com- A number of authors have advocated the
mercially available heart rate monitors. En- inclusion of other temporal/distance factors
erg)- costs associated with walking in the neu- such as cadence, step, and stride length, in-
Dislance = 80°
Velocity= 26.67min
Loss of Balance = 8
Figure 14.1. Scoring distance, velocity and stability in a 3-minute walk test. Shown is a patient's path along a pre-
measured course, and the number of times he deviates from a 1 5-inch path.
Chapter Fourteen Assessment and Treatment of the Patient with MoBiLiTi' Disorders 319
eluding right/left asymmetry in step and formation to determine fiinctional mobility'

stride length in the assessment of gait (12, capacit)'.
13). These factors are usually documented

during ambulation over a short distance, e.g., Scales for Assessing Mobility
20 to 30 ft. Characteristics of constant veloc-
it)' gait are determined, and thus the first and
Walking in a natural environment is
characterized by starts and stops, changes in

last 5 ft are not used in the calculations. Pa-
tients are usually given one practice trial, fol-

direction and speed, stepping over and
lowed by two data collection trials, separated around obstacles, and the integration of mul-
by a rest period. tiple tasks such as talking, turning to look at
A number of methods for quantifying something, or carr»'ing objects during gait.
temporal/distance factors in the clinic have Thus, assessment of functional gait must in-
been suggested, including clude not only the evaluation of unimpeded
a footprint analysis
using either inked feet and white butcher pa- gait, defined as a closed-skill task (18), but
per (12) or floor grids (Fig. 14.2) (13). Many and adapt gait to
also the abilit\' to modif)'
clinicians do not have the time to subse-

both expected and unexpected disturbances
to locomotion. Several tests are available to
quendy analyze the results of these tests, and
thus unanalyzed rolls of inked butcher block examine the patient's ability' to adapt gait to
paper are found in closets and storage areas

changing locomotor demands.
ever\'\vhere (14).
Other researchers have suggested that
DUKE MOBILITY SKILLS PROFILE
velocity'alone can be used as a single measure
The Duke Mobilit>- Skills Profile (19)
of fiinctional gait, since it is simple and quick,
quantifies performance on multiple mobility'
and appears to be a composite measure of the
tasks, including unimpeded gait, transfers,
other temporal/distance variables (9, 15-18).
and stairs. Skills are scored 0, 1, or 2 based on
However, as we discussed earlier, functional
the criteria shown in Table 14.1. This test is
measures limited to velocity' alone do not ap-
a highly reliable and valid predictor of fall
pear to proNide the clinician with sufficient in-
probability' in older adults.
FUNCTIONAL INDEPENDENCE
MEASURE L f^f^
The Functional Independence Measure
1, li\\% (FIM), is an ADL test that includes measures
of mobility function, including locomotion

i
and transfers (5). Patient performance is
^ \~
?k=^ ffl
graded based on the level of assistance re-
quired to perform mobilit\' tasks using a 7-
point scale. Scores range irom 1 for total
assistance, to 7, defined as complete indepen-
dence. Performance is defined relative to dis-
tance, effort, and assistance required. For ex-
ample, complete independence is defined as
the abilitii' to walk a minimum of 150 ft with-
out assistive devices. The patient needing to-
Figure 1 4.2. An example of a floor grid that can be used
tal assistance is defined as one who performs
to visually guide patients towards better foot placement
during gait. (Adapted from Jims C. Foot placement pat-
with less than 25% effort (meaning that ap-
tern, an aid in gait training. Suggestions from the field. proximately 25% of the work related to this
PhysTher 1977;57:286.) activin,' is being performed bv the patient, and
320 Section III MOBILITi' FLTNCTIONS
Table 14.1. Duke Mobility Skills Profile^
1 . Can this person walk?

Yes —go to question three yes=1
No —go to question two no=0
2. Can this person sit upright without human assistance for 60 seconds? yes=1
no=0
3. Does gait meet criteria? yes = 1
(no device, symmetric, step length twice foot length) no=0

4. Can this person descend stairs step-over-step without holding railing? yes=1
no=0
Total—
Soc. Sec.»:
dav :
Mobility Skills Protocol

(modified 10/26/89)
Equipment - Straight-back hard-seated chair, tape measure, 12-inch ruler, shoebox, stopwatch, pencil, white tape
1 Sitting balance
Will you sit forward in the chair, arms folded across chest, for 1 minute? (patient sits in
standardized chair (straight back/kitchen-type chair), without leaning back for 1 minute.
2 = can sit upright, unsupported for 60 sec.
1 = can sit upright, independently with support for 60 sec. (holding onto arm of chair or leaning
against back of chair)
= cannot sit upright independently for 60 sec.
2. Sitting reach
Will you reach forward and get this ruler out of my hand? (45° plane-forward — put ruler 1 2 inches
beyond dominant hand reach).
2 = reaches forward and successfully grasps item
1 = cannot grasp or requires arm support
= does not attempt reach
C = contraindicated
R = refused
3. Transfer
Will you show me how to get from your chair to the bed (or to another chair)?
2 = performs independently (without help from a person), appears steady and safe
1 = performs independently (without help from a person), but appears unsteady
= cannot do or requires help from a person to complete the task
C = contraindicated
R = refused
4. Rising from a chair
a. Will you get up from the chair without using your arms to push up?
(patient seated in hardback/kitchen chair, arms folded across chest or out in front)
1= done on 1st try

= not done on st 1 try
C = contraindicated
R = refused
b. Will you get up from the chair using your arms to push up? (subject can put hands on arms of
chair or on chair seat for assistance; subject can hold onto assistive device if desired as he stands
up)
***** NOTE (score = - 1 if scored 1 on 4a)
1 = can do independently
= can't do independently
C = contraindicated
R = refused
Table 14.1 — continued

5. Standing balance
a. Will you stand the way you usually stand for 1 minute?
2 = steady, without holding onto walking aid or other object for support for 60 sec.
1 = steady, but uses walking aid or other object for support for 60 sec.
= cannot stand upright for 60 sec
C = contraindicated
R = refused
6. Picliing up object off the floor
(drop pencil 1 ft in front of subject (out of base of support), Will you pick this pencil up from the
floor?
2 = performs independently
(without help from object/person)
1 = performs with some help (holds on to a table, chair, assistive device, person, etc.) or is
unsteady (staggers or sways, has to catch self to to keep from falling, etc.)
= unable to pick up object and return to standing

C = contraindicated
R = refused
7. Walking
(measure 10-ft pathway)
Will you walk in your usual way (with or without assistive device) from here to here (indicate
distance to patient — allow 3'-5' for warm-up).
2 = meets all standards for gait characteristics
1 = fails any standard or uses assistive device
= unstable, can't do (requires intervention to keep from falling or staggers, trips)
C = contraindicated
R = refused
standards /or ga/f (tested only in subject's preferred manner!
1. symmetrical step length
2. walks along straight path
3. distance between stance toe and heel of swing foot at least 1 ft length
8. Turning
Will you walk along the path, then turn and come back?
2 = no more than three continuous steps, no assistive device
1 = fails criteria for a score of 2 but completes task without intervention
= unable to turn, requires intervention to prevent falling
C = contraindicated
R = refused
9. Abrupt stop
Will you walk as fast as you can and stop when I say stop? (walks with subject and announce
"stop" after 6-8 steps)
2 = stops within one step without stumbling or grabbing
1 = cannot stop within one step or stumbles, uses assistive device
= requires intervention to avoid fall
C = contraindicated
R = refused
10. Obstacle
(place shoebox in walking path)
Will you walk at your normal pace and step over the shoebox that is in the way?
2 = steps over without interrupting stride
1 = catches foot, interrupts stride, uses assistive device
= cannot step over box
C = contraindicated
R = refused
Table 14.1. continued
1 1 Standing Reacli
(45° plane — forward — put ruler 12 inches beyond dominant hand reach)
Will you reach forward and get this ruler from me?
2 = reaches forward and successfully grasps ruler without stepping or holding on
1 = reaches forward but cannot grasp ruler without stepping or holding on to device
= does not attempt to shift weight
C = contraindicated
R = refused
12. Stairs (must have at least 2 steps)
Try to go up and down these stairs without holding on to the railing.
Ascending
2 = steps over step, does not hold on to railing or device
1 = one step at a time, or must hold on to railing or device
= unsteady, can't do
C = contraindicated
R = refused
Descending
2 = steps over step, does not hold on to railing or device
1 = one step at a time, or must hold on to railing or device
= unsteady, can't do
C = contraindicated
R = refused
1 3. Preferred assistive device:
wheelchair (= 1)
walker (= 2)
quad cane (= 3)
straight cane (= 4)
other (= 5)
none (= 0)
Reprinted with permission: Duncan P. Duke Mobility Skills Profile. Center for Human Aging, Duke University.
75% is being performed by the therapist), or agnosis such as stroice or Parkinson's disease,
requires assistance of tvvo people, or does not received a mean score of 11 ±4 (21 ).
walk a minimum of 50 ft (5).

LIMITATIONS OF PERFORMANCE-
DYNAMIC GAIT INDEX BASED TEMPORAL/DISTANCE GAIT
MEASURES
The Dynamic Gait Index (20) evaluates
and documents modify
a patient's abilit)' to All performance-based measures, wheth-
gait in response tochanging task demands. er of mobility', balance, or general motor con-
This index was developed as part of a profile trol, are indicators of end-product only, and
of tests and measurements effective in pre- do not provide information about the way
dieting likelihood for falls in older adults (21). performance is achieved. Thus, these mea-
The test isshown in Table 14.2. Preliminary sures do not provide insight into underlying
research has shown that the test has good in- impairments that require treatment. How-
ter-raterand test-retcst reliability and is a valid ever, performance-based measures are good
predictor of falls among the elderly. For ex- indicators of overall fiinction and, therefore,
ample, a population of 1 5 healthy older adults important indices of change,
with no neurological impairments or histon'
of imbalance received a mean score of 2 1 ±3 Assessing at the Strategy Level
on the Dynamic Gait Index. In contrast, an
equal number of older adults with a histon,' of Assessment of gait also includes a sys-
falls and imbalance, but no neurological di- tematic description of the strategies used by
Chapter Fourteen ASSESSMENT AND Treatment of the Patient with Mobility Disorders 323
Table 14.2. Dynamic Gait Index
1 . Gait level surface

InstructionsiWalk at your normal speed from here to the next mark (20')
Grading: Mark the lowest category which applies.

(3) Normal: Walks 20', no assistive devices, good speed, no evidence for imbalance, normal gait pattern.
(2) Mild impairment: Walks 20', uses assistive devices, slower speed, mild gait deviations.
(1) Moderate impairment: Walks 20', slow speed, abnormal gait pattern, evidence for imbalance.
(0) Severe impairment: Cannot walk 20' without assistance, severe gait deviations, or imbalance.
2. Change in gait speed
Instructions: Begin walking at your normal pace (for 5'), when I tell you "go," walk as fast as you can (for 5'). When
I tell you "slow," walk as slowly as you can (for 5').
Grading:
Mark the lowest category that applies.
(3) Normal: Able to smoothly change walking speed without loss of balance or gait deviation. Shows a significant
difference in walking speeds between normal, fast, and slow speeds.
(2) Mild impairment: Is able to change speed but demonstrates mild gait deviations, or no gait deviations but
unable to achieve a significant change in velocity, or uses an assistive device.
(1) Moderate impairment: Makes only minor adjustments to walking speed, or accomplishes a change in speed
with significant gait deviations, or changes speed but loses significant gait deviations, or changes speed but
loses balance but is able to recover and continue walking.
(0) Severe impairment: Cannot change speeds, or loses balance and has to reach for wall or be caught.
3. Gait with horizontal head turns
Instructions: Begin walking at your normal pace. When I tell you to "look right," keep walking straight, but turn
your head to the Keep looking to the

right. right until I tell you, "look left," then keep walking straight and turn your
head to the left. Keep your head to the left until I tell you, "look straight," then keep walking straight, but return
your head to the center.
Grading: Mark the lowest category which applies.
(3) Normal: Performs head turns smoothly with no change in gait
(2) Mild impairment: Performs head turns smoothly with slight change in gait velocity, i.e., minor disruption to
smooth gait path or uses walking aid.
(1) Moderate impairment: Performs head turns with moderate change in gait velocity, slows down, staggers but
recovers, can continue to walk.
(0) Severe impairment: Performs task with severe disruption of gait, i.e., staggers outside 1
5" path, loses balance,
stops, reaches for wall.

4. Gait with vertical head turns
Instructions: Begin walking at your normal pace. When tell you to "look up," keep walking straight, but tip your
I
head and look up. Keep looking up until tell you, "look down." Then keep walking straight and turn your head
I
down. Keep looking down until tell you, "look straight," then keep walking straight, but return your head to the
I
center.
Grading: Mark the lowest category that applies.
(3) Normal: Performs head turns with no change in gait.
(2) Mild impairment: Performs task with slight change in gait velocity i.e., minor disruption to smooth gait path or
uses walking aid.
(1 Moderate impairment: Performs task with moderate change in gait velocity, slows down, staggers but recovers,
can continue to walk.
(0) Severe impairment: Performs task with severe disruption of gait, i.e., staggers outside 1
5" path, loses balance,
stops, reaches for wall.

5. Gait and pivot turn
Instructions: Begin walking at your normal pace. When I tell you, "turn and stop," turn as quickly as you can to
face the opposite direction and stop.
(3) Normal: Pivot turns safely within 3 seconds and stops quickly with no loss of balance.
(2) Mild impairment: Pivot turns safely in > 3 seconds and stops with no loss of balance.
(1) Moderate impairment: Turns slowly, requires verbal cueing, requires several small steps to catch balance
following turn and stop.
(0) Severe impairment: Cannot turn safely, requires assistance to turn and stop.
Table 14.2. — continued

6. Step over obstacle
Instructions: Begin walking at your normal speed. When you come to the shoe box, step over it, not around it, and
keep walking.
(3) Normal: Is able to step over box without changing gait speed; no evidence for imbalance.
(2) Is able to step over box, but must slow down and adjust steps to clear box safely.
Mild impairment:
(1Moderate impairment: Is able to step over box but must stop, then step over. May require verbal cuing.
(0) Severe impairment: Cannot perform without assistance.
7. Step around obstacles
Instructions: Begin walking at your normal speed. When you come to the first cone (about 6' away), walk around
the right side of it. When you come to the second cone (6' past first cone), walk around it to the left.
(3) Normal: Is able to walk around cones safely without changing gait speed; no evidence of imbalance.
(2) Mild impairment: Is able to step around both cones, but must slow down and adjust steps to clear cones.
(1) Moderate impairment: Is able to clear cones but must significantly slow, speed to accomplish task, or requires
verbal cueing.
(0) Severe impairment: Unable to clear cones, walks into one or both cones, or requires physical assistance.
8. Steps
/ns(ruct/ons: Walk up these stairs as you would at home (i.e., using the rail if necessary. At the top, turn around and
walk down.
(3) Normal: Alternating feet, no rail.
(2) feet, must use rail.

Mild impairment: Alternating
(1 Moderate impairment: Two feet to a stair; must use rail.
(0) Severe impairment: Cannot do safely.
patients to meet the requirements inherent in Wlien performing an obser\'ational gait

locomotion. analysis using the framework suggested in this
table, the clinician obser\'es one phase of gait
OBSERVATIONAL GAIT ANALYSIS at a time, noting the action performed at each

of the joints, and describing any deviation
Obserxational gait analysis is the tool from a normal gait strateg)'. The analysis of
most frequently used in the clinic to qualita- stance uould begin by obser\'ing the position
tively evaluate gait patterns (22). Documen- of the foot at heel-strike. For example, the cli-
tation forms have been developed to assist the nician might note that the patient with neu-
clinician in observational gait analysis; these rological dysfrinction makes initial contact at
forms van' in their degree of complexity' and heel-strike with a flat foot.
detail. The possible causes hst guides the cli-
Prior to discussing some of these forms, nician forming preliminary hypotheses

in
we need to understand some key points that about the underlying causes of the at\'pical
are important to note in each of the phases of pattern. In our example, foot-flat position at
gait during an observational gait analysis. Ta- heel-strike could be the result of plantarflexor
ble 14.3 summarizes these points. Also sum- contracture, or inactiviU' of the tib-
spasticity,',
marized are deviations commonly seen in the ialis These hypotheses guide
anterior (TA).
patient with neurological impairments and a the next step of assessment, which is to test
list of the possible impairments that could re- these hypotheses and thus determine the
sult in these deviations. This list is not in- cause of the patient's gait deviation. In our
tended to be exhaustive, but rather summa- example, the patient uses a foot-flat heel-
rizes some of the major problems commonly strike, due to inactivit}' of the TA.
found in such patients. With this information, the clinician is
Table 14.3. Problem-Oriented Observational Gait Analysis
Stance Phase
Ankle
Normal: Dorsiflexed at heel-strike, neutral with respect to eversion/inversion, smooth progression into plantar flexion
until the foot is flat, controlled advancement of the tibia over the stable foot (ankle dorsiflexion)
Abnormal Possible Causes

Low heel/foot flat/ or forefoot contact — at heel-strike Plantarflexor contractures
Plantarflexor spasticity
Inactivity of TA
Poor knee extension-terminal svv'ing
Use of an extensor synergy

Foot-slap Inactivity of TA
Lateral border (varus) Inventor spasticity
Medial border (valgus) Inactivity of TA
Excessive plantarflexion As above
Excessive dorsiflexion midstance Plantarflexor weakness
No heel-rise, no toe-off terminal stance Plantarflexor spasticity
Plantarflexor contracture
Knee
Norma/; extension followed by brief flexion followed by knee extension until terminal stance when knee begins to
flex

Excessive knee extension Plantar flexor contracture
Plantar flexor spasticity
Quadriceps spasticity
Compensate weak quads
Use of an extensor synergy
Persistent knee flexion Plantarflexor inactivity

Hamstring overactivity
Trunk/Hip
Normal: initial hip flexion with smooth progression to extension by midstance, and hyperextension by terminal
stance; trunk remains vertical

Excessive hip flexion/Trunk forward lean Hip flexor contracture
Hip flexor spasticity
Hip extensor weakness
Compensate weak quads
Hip extension/Trunk backward lean Compensate weak flexors

Compensate weak hip extensors
Trunk lateral lean Weak abductors
Scissors Gait Abductor weakness —contralateral
Adductor spasticity — ipsilateral
Normal: neutral with respect to vertical displacement and anterior/posterior tip; forward rotation at heel-strike,
smooth progression into posterior rotation through stance phase

Elevated Abductor weakness —contralateral
Dropped
Retracted
Table 14.3. continued
Swing phase
Ankle
Normal: Plantarflexion at toe-off, into dorsiflexion by midswing, continued dorsiflexion into terminal swing

Plantarflexion at ankle Plantar flexor contracture
Inadequate activation of TA
Varus foot position Spastic inverters
Valgus foot position Inadequate activation of TA
Knee
Normal: Initial flexion, increased flexion by mid swing, into extension by terminal swing.

Inadequate knee flexion — Initial to mid swing Knee extensor contracture
Quadriceps spasticity
Plantar flexor contracture
Use of Extensor synergy
Inadequate knee extension —Terminal swing Knee flexor contractures

Hamstring hyperactivity
Use of flexor synergy
Hip
Normal: General motion is extension into flexion; begins with hyperextension, neutral by midswing, and flexion by
terminal swing
Inadequate hip flexion Inability to activate hip flexors
Decreased hip proprioception
Circumduction (hip hike/abduction) Compensate weak hip flexors

Contralateral vaulting Compensate weak hip flexors
Pelvis
Normal: Pelvis drops slightly during swing, remains neutral with respect to anterior/posterior tilt, rotates from
backward to forward position

Posterior tip Compensate weak hip flexors
Anterior rotation Compensate weak hip flexors
Elevation Compensate weak hip flexors
able to develop appropriate treatment strate- Stance Phase of Gait

gies to remediate the underlying impairment
and improve the gait pattern. In our example, Key elements within a normal strategy
the clinician could use the following treat- for controlling the and foot during
ankle
ments: («) a strengthening program targeting stance include {a) smooth unimpeded motion
eccentric and concentric contractions of the of the body over the foot, (b) the foot leading
TA; (b) facilitation techniques, like icing the the body at heel-strike but trailing the body
TA immediately before and during gait train- at toe-oft'. This trailing toe -oft" posture is im-
ing; (c) electrical stimulation during gait to portant for eft'ective generation of propulsive
improve activation of the TA, and (
d) use of forces.
an ankle-foot orthosis to control plantarflex- A clinician might ask several questions
ion during gait until the patient is able to ac- when examining ankle-foot control during
tivate the TA appropriately. the stance phase of gait. For example, what is
the position of the foot when it contacts the What is the position of the pelvis at heel-
floor at the beginning of the stance phase of strike? Is it tipped backward, or alternatively,
gait? During loading and progression to mid- forward with excessive lordosis? Does the pel-
stance, does the tibia advance over the stable vis move smoothly from a forwardly rotated
foot, reaching vertical by mid-stance? During position to a posterior position as stance pro-
terminal stance, does the tibia continue its gresses? Does the pelvis drop or hike exces-
forward advancement, with the ankle dorsi- sively?
flexing, even though the hip and knee are ex-

tended? Does the heel rise at the end of stance Swing Phase
phase as the body rises up onto the forefoot
in preparation for toe-off? Key elements within a normal strategy
Key elements normally observed at the for controlling the ankle and foot during
knee are (a) the position of the knee during swing include (a) plantarflexion of the foot at
weight acceptance, and (b) the position and toe-oft^, followed by dorsiflexion of the ankle,
stability of the knee during the course of the which is maintained throughout the rest of
stance phase. Normally, the knee is extended swing, and (b) foot clearance is normally
at the initiation of stance. Just after contact achieved by less than I inch (23).
with the surface, there is a brief flexion, then Is the foot traihng the stance limb at the
extension at the knee, which absorbs some of initiation of swing? Is the heel oft' the ground
the impact of weight acceptance. The knee re- during toe-oft) Is toe-off" accomplished from
mains extended, but not hyperextended, dur- a forefoot support position? Does the foot
ing stance, beginning to flex at toe-oft' (23). clear theground with no catches during the
Is the patient's knee flexed or extended swing phase? Does the foot move in front of
at heel-strike? Is there a brief period of flexion the swing leg toward the end of swing in prep-
foUowed by extension as the patient shifiis aration for heel-strike?
weight to the stance limb? Does the knee Normal control of the knee during
move into extension smoothly or abrupdy? At swing includes (a) flexion of the knee at the
mid-stance, is the knee extended to neutral, beginning of swing, continuing through mid-
or does the knee thrust back into hyperexten- swing, and (b) extension of the knee by ter-
sion? minal swing in preparation for heel-strike
Key elements of normal hip control dur- (23).
ing stance include {a) the smooth progression Is there a smooth transition from flexion
from flexion into extension by mid-stance and into extension at the knee during the swing
hyperextension by terminal stance, and (b) the phase? Is the patient unable to flex the knee
trunk remains vertical (23). during swing? Is the patient using a flexor syn-
What is the position of the hip at heel- ergy pattern to achieve flexion of the swing
strike? Is it flexed? Does the hip move limb? Is the patient able to extend the knee
smoothly into extension, or remain flexed? Is (in conjunction with hip flexion) at terminal
the hip extended to neutral by mid-stance swing?
with the trunk upright and aligned above it? Key points to observe in normal hip and
Is the patient able to hyperextend the hip, trunk control include («) the general motion
while at the same time dorsiflexing the ankle, at the hip during swing is extension to flexion,
as the stance limb moves posterior? {b) the hip begins from a hyperextended po-
Normal pelvic control includes the fol- sition at the beginning of gait, moves to neu-
lowing key elements: {a) of
a slight elevation tral by mid-swing, and into flexion by the end
the pelvis during movement to single limb of swing, and (
c) the trunk remains vertical
stance; (b) smooth rotation of the pelvis from (23).
forward to backward position during stance; How is the patient advancing the limb?
and (c) neutral position of the pelvis with re- Is foot clearance being achieved? Is the pa-
spect to anterior and posterior tilt. tient using a flexor strategy to bring the Hmb
328 Section III MOBILITi' FUNCTIONS
forward? If flexion is being used, is it widiin a weight of the body during loading may be an
total flexor synerg\' pattern? Alternatively, is effective strategy' for stability during stance,
the patient forced to hike and circumduct the but at the same time, an extensor synergy dur-
hip to advance the limb? Is vaulting of the ing swing will prexent the forward ad\'ance-
contralateral stance limb being used to effect ment of the limb, limiting progression 10). (
toe clearance? Does the hip start from an ex- The clinician may use obserxational gait
tended posidon? Is the hip flexing with analysis to document the presence of a hip
enough force to produce an associated flexion hike/circumduction movement strategy to
at the knee? Is the trunk vertical, or is it in- advance the limb during the swing phase of
clined forward, backward, or to one side? gait. But what are the potential effects of us-
Normal control of the pelvis during ing this strategy on this patient's functional
swing involves (a) a slight drop of the pelvis performance? First, use of a hip-hike, circum-
during swing, but the pelvis tends to remain duction strategy to advance the swing leg will
neutral with respect to anterior/posterior tilt. slow the patient's gait speed. This will be re-
Pern,' (23) states that if upwards or down- flected in decreased distance traveled and ve-
wards tilt of the pelvis is apparent, it is always locity, both variables measured on the 3-min-
abnormal, since it is not possible to distin- ute walk test. In addition, it will affect his or
guish this motion in normal gait; {b) at the her score on the Dynamic Gait Index, since
beginning of swing, the pelvis is rotated in the he/she may be unable to adapt this pattern
backward direction, and rotates smoothly for- sufficientiy to complete the tasks involving
ward during the swing phase (23). walking on uneven surfaces.
Does the pelvis remain relatively vertical The clinician can use information from
during swing? Is there an obvious tilting mo- both functional assessment and gait analysis
tion of the pelvis, either anteriorly or poste- to determine whether treatment is warranted.
riorly? In the beginning of swing, is the pelvis Can treatment improx'e the strategies a pa-
rotated backwards? Is there a smooth pro- tient is using to achieve the demands of lo-
gression of the pelvis to an anterior position comotion, thereby increasing functional ca-
during swing? pability?
Some examples of obser\'ational gait
FORMS FOR OBSERVATIONAL GAIT analysis forms that are helpfial in guiding a
ANALYSIS clinical assessment of gait patterns in the pa-

tient with neurological impairments follow.
There are forms available to help clini-
cians structure their approach to observ'a- Rancho Los Amigos Gait Analysis Form
tional gait analysis. Why would a clinician
need to perform both a functional mobilit\' The Gait Analysis Form, shown in Table
assessment and a gait analysis? Each provides 14.4, from the physical therapy department,
information helphil in establishing a plan of Rancho Los Amigos Hospital, Downey, Cal-
care for die patient. For example, an obser- is a comprehensive approach to move-
ifornia,
vational gait analysis can help a clinician de- ment analysis during gait (24).
termine the extent to which the current strat-
egies a patient is using meet the inherent Mobility Assessment Form

progression, stability', and adaptation de-
mands of locomotion. A more simple and less extensi\'e ap-
What are the costs and benefits of cur- proach to obsenational gait analysis is repre-
rent strategies; that is, are they eJfective at sented by the Mobility Assessment Form, part
meeting demands in one phase of gait, but of the Balance and Mobility Assessment Pro-
limit the patient's options in another phase of file proposed by Tinetti (25). This is provided
gait? For example, the use of an extensor syn- in the chapter on assessment and treatment of
ergy' pattern during stance to support the postural control.
Chapter Fourteen Assessment and Treatment of the Patient with MoBiLiri' Disorders 329
Table 14.4. Rancho Los Amigos Gait Analysis Form'"
Gait Analysis: Full Body

r.anc;ho Los amicos Medical Center
Physical therapy department
Reference Limb
LI R \
MAJOR
PROBLEMS:
Weight
Acceptance
Swing Limb
Advancement
Toes i[
Inadequate Lxrcnsim
Diagnosis
c:a 90242
'Reprinted with permission of Rancho Los Amigos Medical Center's Physical Therapy Department and Pathokinesiology Laboratory,
Downey, California.
Gait Assessment Rating Scale (GARS) Gait analysis is an integral part of almost
ever\' motor control evaluation in the patient
The Gait Assessment Rating Scale with neurological dysfijnction. Gait itself is
(GARS), developed by Wolfson and col- complex, and understanding the complica-
leagues, is shown in Table 14.5 (26). The tions of gait is even more difficult. Therefore,
scale also allows the quantification and doc- it is essential that a clinician have a systematic
umentation of four categories of gait abnor- and consistent approach to observing and an-
malities. The scale has been used to document alyzing gait. Despite its limitations, an obser-
gait problems in healthy elderly, as well as in \'ational gait analysis form provides a frame-
older adults with a histon' of falls. The test work for s\'stematically observing gait, and is
has been shown to have high inter-rater reli- therefore an essential part of the gait assess-
ability', and is a sensitive indicator of changes ment process.
in gait function among older adults.
Finally, clinicians can develop their own
Assessing at the Impairment Level
form to guide form can vary
gait analysis. This
in its complexity' and depth of analysis, as was
shown in the form in Table 14.2. Regardless Assessing the impairment level of fiinc-
of the D,pe of form used, the process of gait tion was discussed in detail in Chapter 10, and
analysis can be facilitated by \'ideotaping the
thus is not discussed fiirther in this chapter.
patient's gait. Though potentially time con-

The reader encouraged to refer back to
is
suming to perform, in the long run, Nideotap- Chapter 10 for a review of principles of as-
ing provides the clinician more time to ob-

sessment at the impairment level of frinction.
serve gait, and therefore increases the A task-oriented approach to assessing

mobility fiinction includes performance-
reliabilit}' and validity' of the obser\'ational gait
analysis. Replaying the tape, particularly with based measures of mobility', assessment of gait
strategies, and evaluation and documentation
stop-frame and slow-motion features, allows
repeated viewing of gait patterns \\ithout fa-
of underlying impairments. There are a range
tiguing a patient.
of tests and measurements from which a cli-
nician can draw, in order to assemble a com-
prehensive mobility assessment tool.
LIMITATIONS TO OBSERVATIONAL
GAIT ANALYSIS
Transition to Treatment
Studies ha\'e shown that a major limi-
tation of most obser\adonal gait analysis is Setting Goals
poor reliabilit>' even among highly trained and
experienced clinicians (22 ). In addition, a de- As is true for goal setting related to
tailed qualitative gait analysis is ven,' time-con- other physical skills, clinicians need to estab-
suming, and often unrealistic in a busy clinical lish both long- and short-term goals during
environment. Finally, strong evidence does mobility' retraining yvhich are objective, mea-
not exist to indicate that most obser\'ational surable, and meaningful.
gait analysis forms are sensitive to changes in
gait patterns in response to therapy.
An alternative to obsen'ational gait anal- LONG-TERM GOALS
ysis is the use of technological systems to
quantify' movement patterns, muscle activa- Long-term goals are ofiien stated in
tion patterns, and forces used in gait. How- terms of functional performance. They usually
ever, this technology' is beyond the reach of reflect ambulation outcomes yyith respect to
the average clinician. In addition to being ex- level of independence. An example of a long-
tremely expensive, this equipment requires term goal might be: the patient yvill be able
considerable time and technical expertise to to walk a minimum of 1000 ft independentiy
use. y\ith the use of a cane and orthosis.
Table 14.5. Components of the Gait Assessment Rating Score (GARS)^
A. General Categories
1 Variability — a measure of inconsistency and arrhythmiclty of stepping and of arm movements.
= fluid and predictably paced limb movements;
1 = occasional interruptions (changes in velocity), approximately <25% of time;
2 = unpredictability of rhythm approximately 25-75% of time;
3 = random timing of limb movements.
2. Guardedness — hesitancy, slowness, diminished propulsion and lack of commitment in stepping and arm
swing
= good forward momentum and lack of apprehension in propulsion;
1 = center of gravity of head, arms and trunk (HAT) projects only slightly in front of push-off, but still good
arm-leg coordination;
2 = HAT held over anterior aspect of foot, and some moderate loss of smooth reciprocation;
3 = HAT held over rear aspect of stance-phase foot, and great tentativity in stepping.
3. Weaving — an irregular and wavering line of progression
= straight line of progression on frontal viewing;
1 = a single deviation from straight (line of best fit) line of progression:
2 = two to three deviations from line of progression;

3 = four or more deviations from line of progression.
4. Waddling — a broad-based gait characterized by excessive truncal crossing of the midline and side-bending
= narrow base of support and body held nearly vertically over feet;
1 = slight separation of medial aspects of feet and just perceptible lateral movement of head and trunk;
2 = 3-4" separation feet and obvious bending of trunk to side so that COG of head lies well over ipsilateral
stance foot;
3 = extreme pendular deviations of head and trunk (head passes lateral to ipsilateral stance foot), and further
widening of base of support.
5. Staggering — sudden and unexpected laterally directed partial losses of balance
= no losses of balance to side;
1 = a single lurch to side;
2 = two lurches to side;
3 = three or more lurches to side.
B. Lower Extremity Categories
1 % Time in Swing a loss — in the percentage of the gait cycle constituted by the swing phase
= approximately 3:2 ratio of duration of stance to swing phase;
1 = a :1 or slightly less ratio of stance to swing;
1
2 = markedly prolonged stance phase, but with some obvious swing time remaining;
3 = barely perceptible portion of cycle spent in swing.
2. Foot Contact — the degree to which heel strikes the ground before the forefoot
= very obvious angle of impact of heel on ground;
1 = barely visible contact of heel before forefoot;
2 = entire foot lands flat on ground;
3 = anterior aspect of foot strikes ground before heel.
3. Hip ROM —the degree of loss of hip range of motion seen during a gait cycle
= obvious angulation of thigh backwards during double support (10 deg);
1 = just barely visible angulation backwards from vertical;
2 = thigh in line with vertical projection from ground;
3 = thigh angled forwards from vertical at maximum posterior excursion.
4. —
Knee Range of Motion the degree of loss of knee range of motion seen during a gait cycle
= knee moves from complete extension at heel-strike (and late-stance) to almost 90° (@ 70°) during swing
phase;
1 = slight bend in knee seen at heel-strike and late-stance and maximal flexion at midswing is closer to 45°
than 90°;
2 = knee flexion at late stance more obvious than at heel-strike, very little clearance seen for toe during
swing;
3 = toe appears to touch ground during swing, knee flexion appears constant during stance, and knee angle
during swing appears 45° or less.
Table 14.5. — continued

C. Trunk, Head, and Upper Extremity Categories
1 —
Elbow Extension a measure of the decrease of elbow range of motion
= large peak-to-peak excursion of forearm (approximately 20 deg), with distinct maximal flexion at end of
anterior trajectory;
1 = 25% decrement of extension during maximal posterior excursion of upper extremity;
2 = almost no change in elbow angle;
3 = no apparent change in elbow angle (held in flexion);
2. Shoulder Extension —a measure of the decrease of shoulder range of motion
= clearly seen movement of upper arm anterior (15 deg) and pKJSterior (20 deg) to vertical axis of trunk;
1 = shoulder flexes slightly anterior to vertical axis;

2 = shoulder comes only to vertical axis, or slightly posterior to It during flexion;
3 = shoulder stays well behind vertical axis during entire excursion.
3. Shoulder Abduction — a measure of pathological increase in shoulder range of motion laterally
= shoulders held almost parallel to trunk;
1 = shoulders held 5-10 deg to side;
2 = shoulders held 10-20 deg to side;
3 = shoudlers held greater than 20 deg to side.
4. Arm-Heelstrike Synchrony — the extent to which the contralateral movements of an arm and leg are out of
phase
= good temporal conjunction of arm and contralateral leg at apex of shoulder and hip excursions all of the
time;
1 = arm and leg slightly out of phase 25% of the time;
2 = arm and leg moderately out of phase 25-50% of time;
3 = little or no temporal coherence of arm and leg.
5. Head Held Forward —a measure of the pathological forward projection of the head relative to the trunk
= ear-lobe vertically aligned with shoulder tip;
1 = ear-lobe vertical projection falls 1

" anterior to shoulder tip;
2 = ear-lobe vertical projection falls 2" anterior to shoulder tip;
3 = ear-lobe vertical projection falls 3" or

more anterior to shoulder tip.
6. Shoulders Held Elevated —the degree
which the scapular girdle is held higher than normal
to
= tip of shoulder (acromion) markedly below level of chin (1-2");
1 = tip of shoulder slightly below level of chin;
2 = tip of shoulder at level of chin:
3 = of shoulder above level of chin.
tip
7. Upper Trunk Flexed Forward a measure — of kyphotic involvement of the trunk

= very gentle thoracic convexity, cervical spine flat, or almost flat;
1 = emerging cervical curve, more distant thoracic convexity;

2 = anterior concavity at mid-chest level apparent;
3 = anterior concavity at mid-chest level very obvious.
'From Wolfsan L, Whipple R. Amerman P, Tobin JN. Gait assessment in the elderly: ! gait abnormality rating scale and its relation to
falls. I Gerontal 1990: 45:M12-M19.
flexion by 20° and thereby improve up-

SHORT-TERM GO.\LS
right posture during stance and swing
Short-term goals for mobilit>' retraining phase of gait.
can be expressed in terms of: Accomplishing interim steps towards
long-term goals. Examples include (a)
1. Changing underhing impairments. to increase distance walked, with only
One example would be to decrease flex- stand-by assist, from 10 ft to 25 ft; (Z')
ion contractures at the hip by 20°, at the to increase speed; patient will be able to
knee by 15°, and at iJie ankle by 20°. walk 200 ft, stand-by assist only, in 45
2. ImproNing gait patterns. One example sees; (c) to become independent in the
would be to decrease forward trunk use of a quadruped cane.
Short-term goals usually lead to treat- flexor synergy pattern to accomplish the
ment aimed at resolving underlying
strategies goals of swing.
impairments, and improving the quality of 3. Work on the goal of helping the patient
gait strategies used to achieve the three task to a vertical body posture,
achieve
requirements of gait, that is, progression, sta- which allows the center of body mass to
bility, and adaptation. Long-term goals often move anterior to the stance leg. This
lead to treatment strategies related to improv- will create a fon\'ard fall position essen-
ing the overall performance of ambulation, tial for generating forward momentum.
such as increasing the distance walked or the This begins during initial stance with fa-
speed of ambulation. Often, the two are in- ciUtation of graded plantarflexor activ-
terrelated, as when the goal is to improve a ity, so that the tibia can advance
particular aspect of the locomotor pattern to smoothly over the stationary foot. In
increase the velocity of gait. addition, during single limb stance, one
In addition, goals related to retraining can encourage the patient to maintain a
mobihty fimction can be defined in relation- vertical trunk, with the hip and knee ex-
ship to the three requirements of gait (20). tended. If the trunk is leaning forward
and hips are flexed, this position will
DEFINING GOALS BASED ON THE keep the body vector within the base of
TASK REQUIREMENTS OF GAIT support of the foot. One can then prog-
ress to helping the patient develop a
Progression trailing limb position during terminal
stance, facilitating hip and knee exten-
Treatment goals related to progression
sion, knee extension, lifting the heel,
concern helping the patient develop the ca-
and rolling the body weight onto the
pacity to generate momentum to facilitate
forefoot.
forward propulsion of the body. Specific ex-
4. Help the patient develop a strategy that
amples include:
ensures adequate toe clearance during
1. Improve the range and freedom of mo- the swing phase of gait. Patients need
tion so that momentum can be trans- to learn to advance the swing leg with
ferred freely between body segments. minimal contact with the surface, since
This encompasses improving range of this decreases, or halts, forward mo-
motion, decreasing contractures, re- mentum.

ducing spasticity, which limits the ve-
locity of motion, and reducing coacti- StabiUty
vation of muscles, which increases joint
stiffness. Treatment goals related to stability re-
2. Increase the patient's speed of walking flect [a) the need for a good foot placement,
because generation of momentum re- to facilitate weightbearing during initiation of
quires, a minimum speed of movement. stance, {b) the presence of sufficient extensor
This includes increasing the speed at torque to support the body against gravity
which segments moved. For exam-
are during single limb stance, and (c) facilitation
ple, increasing the speed and amplitude of hip and trunk extensors to control the hip-
of hip flexion during the swing phase of arm-trunk (HAT) segment. Thus, working
gait will advance the thigh segment toward stability during gait includes helping
quickly. This will facilitate passive knee the patient:
flexion for toe clearance, and subse-
quent knee extension for foot place- 1 . To accomplish a heel first foot-strike, in
ment. This goal requires facilitation of the absence of coronal plane deviations.
hip flexion in conjunction with knee ex- This position will allow the body to
tension, thus decreasing reliance on a move smoothly over the foot, and to
take advantage of the full weightbearing develop the most eflfective and efficient strat-
surface of the foot, enhancing stabilitv'; egies for locomotion possible in the face of
2. To develop coordinated extension at persisting sensory, motor, and cognitive im-
the hip and knee, in order to generate pairments. Treatment is directed at (a) im-
an extensor moment to support body proving or preventing impairments, as possi-
weight during single limb stance; ble, (b) developing gait strategies that meet
3. To develop a vertical posture of the the requirements for progression and stability
trunk, with good hip and back exten- during the stance and swing phases of gait,
sion to control the HAT segment, and and (c) developing adaptive strategies appro-
adequate activation of the abductors to changing task and environmental de-
priate to
control the pelvis; and mands. In addition, treatment may be geared
4. To facilitate extensor moments at the towards preventing the development of sec-
hip and knee while maintaining the ca- ondary' impairments.
pacity to dorsiflex the ankle, thus avoid-
ing use of a total extensor synergy pat- Treating at the Impairment Level
tern during stance.
Adaptation The goal of treatment aimed at the im-

pairment level is to correct those impairments
Treatment goals related to adaptation
that can be changed, and prevent the devel-
require the patient to modify movement and
opment of secondary' impairments. Alleviat-
sensory strategies for locomotor control in re-
ing underlving impairments enables the pa-
sponse to changing task and environmental
tient to resume using pre\'iously developed
demands.
strategies for gait. When permanent impair-
Adaptation goals include helping the
ments make resumption of previously used
patient:
strategies impossible, new strategies will have
1 To develop the ability' to integrate com- to be developed.
pensatory aspects of postural control In the chapter on assessment and treat-
into the ongoing gait cycle; ment of the patient with postural disorders,
2. To develop the ability to utilize visual we discussed at length various treatment strat-
cues to identify upcoming disturbances egies aimed at resolving or preventing under-

to mobility and modify gait strategies in lying musculoskeletal and neuromuscular im-
order to minimize their effect. Specific pairments; thus, this information will not be
examples include scanning strategies to repeated in this chapter. However, we will
identify potential obstacles, and step- briefly discuss the role of training pre-ambu-
ping over an obstacle safely, without lation skills as a part of gait training.
modifying the gait path;
3. To develop the abilit\' to utilize visual
PRE-AMBULATION SKILL TRAINING
cues to identify potential obstacles and
alter walking path appropriately so as to
Several treatment approaches recom-
avoid obstacles.
mend that during gait training, the patient
With comprehensive and realistic goals should begin by practicing skills that are con-
established, based on the patient's desires and sidered precursors to ambulation. These skills
problems, the clinician can move ahead to are thought to lead to successfiil ambulation,
planning treatments designed to meet these and are thus considered pre-ambulation skills
goals. (27-30).
Many of the pre-ambulation gait train-
Treatment of Gait ing sequences are based on having the patient
A task-oriented approach to retraining repeat activities that are part of a normal de-
mobility fianction focuses on helping a patient velopmental sequence (30, 31 ). The sequence
Chapter Fourteen ASSESSMENT AND TREATMENT OF THE PATIENT WITH Mobility Disorders 335
begins by having patients practice mobility Ostensibly, as the number of gait devi-
and stability skills in prone and supine posi- ations increases, the patient's level of func-
tions. This inclucles such activities as rolling, tional performance should decrease. How-
maintaining prone on elbows or hands, supine ever, research has notdemonstrated a strong
bridging, and practicing counter-rotation relationship between number and types of im-
trunk motions, that is, movements in which pairments and functional gait performance.
the shoulders rotate in the opposite direction For example, several researchers have shown
from the hips. As motor control in supine and there is no strong relationship between tem-
prone positions is recovered, patients begin to poral distance factors such as velocity, used to
practice activities on all fours, then sitting, measure functional performance, and number
kneeling, half-kneeling, modified plantigrade of gait deviations (9, 17). This has led re-
position, and finally standing (28-30, 32, searchers to suggest that, while there are in-
33). stances when sensory and motor impairments

There are a growing number of re- predict gait performance, they are not valid
searchers and clinicians who are questioning predictors of fijnctional ambulation outcomes
the requirement that patients regain mobility in patients with concomitant cognitive, per-
skills according to a developmental sequence ceptual, and pain impairments (9, 17).
(34-38). In this book, we present a task-ori- Considerable time and effort are di-
ented approach to retraining mobility that is rected at modifying underlying impairments

not based on a developmental sequence for on the assumption that the result will be im-
approach focuses on
retraining. Instead, this proved patterns of gait, and thus increased
identifying and remediating underlying im- levels of independence in ambulation. Yet re-
pairments, retraining locomotor strategies, search has raised questions about the effect of
and integrating those strategies into the per- retraining components on flinctional perfor-
formance of functional mobility tasks. Thus, mance of gait, at least as expressed by tem-
in this approach to retraining, the patient poral and distance factors. Thus, resolution of
would not be required to practice maintaining impairments alone may not be enough to en-
all fours unless this represents a necessary part sure recovery of ambulation skills.
of their everyday activities. Perhaps when retraining mobility skills,
more time should be spent on helping pa-

LIMITATIONS ON IMPAIRMENT tients develop efficient gait strategies that
LEVEL TRAINING compensate for their impairments, rather than
on resolving the impairments themselves.
Though most clinicians develop treat- This is especially important when impair-
ment strategies to remediate underlying im- ments are permanent, making the return to
pairments, the extent to which these types of normal gait strategies unreasonable. More re-
improvements carry over to gait is still unde- search on these issues is critically needed.
termined. There is mounting evidence to sug-
gest that this type of training does not carry
over to gait. For example, researchers have
Treatment at the Strategy Level
found that, while therapeutic strategies were
effective in significantiy increasing hip flexion This section discusses therapeutic strat-
range motion and improving trunk
of egies for retraining gait patterns in the patient
strength, improvements in these areas did not with neurological dysfiinction according to a
significantiy improve gait speed (39). problem-based framework. Therapeutic strat-
Several studies have shown that postural egies are presented with reference to prob-
sway biofeedback is an effective way to train lems that occur during the stance and swing
patients to control center of massmotions in phase of gait and that limit a patient's ability
the standing position (40), but there was no to achieve the task demands. Techniques sug-
carry-over to improved locomotion (41). gested are drawn from cUnical efforts related
to the application of a systems theon' of mo- prone position to reduce plantarflexor con-
tor control and motor learning, and a wide tractures. Surgen' and/or nerve blocks may
variers' of sources including the neurofacili- be needed in response to excessively tight heel
tation approaches (28, 29, 32, 38, 42). It is cords that cannot be changed in any other
important to remember that while these tech- way (10,23).
niques are commonly used by clinicians to re- Plantarflexor spasticity. Impaired
train gait, they have not been validated heel-strikedue to overactivity' of the plantar-
through controlled research involving pa- flexors may respond to facilitation techniques
tientswith neurological impairments. to decrease abnormal muscle tone these were (
Finally, remember the goal of retraining discussed in Chapter 10). An ankle-foot or-
at the strategy' level is to assist the patient in thosis can be used to prevent plantarflexion at
developing movement strategies that are ef- the ankle. However, the loss of plantarflexion
fective inmeeting the inherent demands of results in sustained knee flexion, and thus re-
the stance and swing phase of gait and thus quires good quadriceps activation to prevent
the overall demands for progression, stability', collapse of the stance limb during weight ac-
and adaptation. Wliile much of gait retraining ceptance (10).
strives to assist patients in the recoven,' of pre- Electrical stimulation of the anterior tib-
viously used normal gait patterns, this may ialis muscle can be used to reciprocally inhibit
not be a realistic goal in the face of permanent activity of the gastrocnemius/soleus. Finally,
sensory and motor impairments. Thus, a bet- EMG biofeedback to the gastrocnemius may
ter standard for judging the efficacy of a pa- help the patient to reduce overactivir\' during
tient's movement strategies, is to ask, "Are the stance and swing phase of gait (44-50).
they effective in meeting the demands of the Inactivity of the tibialis anterior.
task?" Inability' to activate the tibialis anterior (TA)
muscle is a common cause of impaired heel-
STANCE PHASE strike in the patient with neurological impair-
ments. Strengthening exercises to increase
Impaired Heel-strike force production of the TA are important in
making sure the TA is capable of generating
The
goals of therapeutic strategies are force in response to descending commands.
(a) toimprove tlie foot position at heel-strike, Unfortunately, the capacit}' to generate force
(b) to improve weight acceptance during does not ensure that the muscle will be re-
loading, (c) to improve motion at the ankle cruited during the automatic process of gait.
and foot, and d) to facilitate smooth motion
(
Nonetheless, strengthening is necessary to
of the body over the stationary' foot. Heel- ensure that force generation capabilin' is at
strike foot contact with a smooth transition to least present.

a flat foot position facilitates fonvard progres- Biofeedback and/or electrical stimula-
sion and a stable base of support important tion of the TA in conjunction with a foot
for stabilit)'. The treatment of problems im- switch placed inside the patient's shoe has
pairing heel-strike depends on the underlying been used effectively to increase activation of
cause. the TA at heel-strike (44, 49, 50).
Plantarflexion contractures. Plantar- SensoPi' stimulation of the tibialis ante-
flexion contractures can be treated with man- rior, such as icing or tapping, during manually
ual stretch to increase range in a tight heel assisted step initiation has also been suggested
cord. In addition, fixed plaster casts can be as an approach to facilitate anterior tibialis ac-
used to place the muscle on sustained stretch, tivation just prior to heel-strike (10). Manu-
thus increasing range (43). Eventually, plaster ally assisted step initiation with heel-strike is
casts may be bivalved, padded, and used as shown in Figure 14.3.

resting splints during the night and da\time Use of an orthotic that has motion at
rest periods (10). the ankle joint, but a posterior stop is an ef-
Patients are often positioned in the fecti\e way to control foot-drop in the patient
Chapter Fourteen ASSESSMENT .\nd Treatment of the Patient with Mobility' Disorders 337
who isunable to recruit the TA. The ankle Mid-stance Problems

joint motion of the orthotic allows some dor-
siflexion, thus allowing the tibia to advance The goals of treatments to improve gait
over the supporting foot (10). patterns for mid-stance include {a) activation
Foot placement can also be assisted by of hip and knee extension, to facilitate an ex-
making a grid on the floor, which helps to tensor support moment for single limb stance,
visually guide patients in establishing a better and (
b) maintenance of a vertical posture and
toot placement pattern on the floor. This was control of the HAT segment.
shown in Figure 14.2. The grid is especially Knee hyperextension. A common mid-
helpfiil for patients who have difficult}' with stance problem involves hyperextension of the
reduced stride length and/or adduction of knee during loading at mid-stance (23).
the leg. Distance bet\veen horizontal stripes Treatment depends on whether the hyperex-
can be individualized to the patient's desired tension is the primar)' impairment or compen-
stride length (51). sator)' to another problem elsewhere.
Coronal plane foot problems. A varus If hyperextension is due to hyperactivit)'

foot position during gait can be controlled of the plantarflexors, then an AFO with plan-
with an ankle-foot orthosis (AFO) with a tarflexion stop can be used (10, 43). Alter-
slight buildup of the lateral border of the foot natively, techniques to decrease muscle tone
(10). In addition, electrical stimulation of the in the plantarflexors, such as having the pa-
toe extensors can reduce varus positioning tient practice weightbearing with the ankle
during stance. A valgus foot'position can be dorsiflexed, have been recommended ( 10, 28,
controlled \\ith an AFO with a slight buildup 38).
of the medial border (10). When hyperextension is compensator)'
to weak quadriceps, preventing knee collapse
during loading and mid-stance, strengthening
exercises, electrical stimulation, and EMG
biofeedback have been recommended to fa-
cilitate acdvation of the quadriceps. Knee hy-
perextension block orthoses have also been

proposed to prevent hyperextension of the
knee during gait.
In patients who do not have sufficient
knee control to prevent collapse during load-
ing, the knee can be externally supported to
prevent collapse, such as with a posterior plas-
ter shell applied to the leg with an ace ban-
dage during stance weightbearing activities or
knee brace, as shown in Figure 14.4. While
the benefit of bracing the knee into extension
is to facilitate weightbearing in the stance
phase of gait, the benefit is lost during the
swing phase when the padent is required to
flex the leg to advance the limb.
Excessive knee flexion. In some pa-

dents with neurological dysfunction, prob-
lems in mid-stance relate to excessive knee
flexion rather than extension. When knee flex-
ion results from weak quadriceps, funcdonal
Figure 1 4.3. Helping the patient to accomplish a heel-
electrical stimulation can be used to activate
first foot strike pattern. This position will allow the body
to move smoothly over the foot with a good weightbear- the quadriceps. Alternatively, (or in combi-
ing surface, enhancing stability and progression. nation), a positional feedback device such as
Forward-flexed trunk. A forward-

flexed trunk posture often accompanies hy-
perextension of the knee. To facilitate a ver-
tical posture and control of HAT stability

t)ver the extended manual cues are given
hips,
to the padent at the shoulders (shown in Fig.
14.5) or at the hips. Manual assistance can
progress from light manual guidance to verbal
cuing.
Contralateral pelvic drop. Inade-
quate activation of the abductors results in
a contralateral drop of the pelvis during
mid-stance. Treatment suggestions include
strengthening exercises for the abductors and
having the patients practice maintaining sym-
metrical pelvic height during single limb
stance.
Figure 1 4.4. In patients who do not have sufficient knee

control to prevent collapse during loading, the knee can
be externally supported to prevent collapse with a knee
brace.
an ELGON can be used to provide kinematic

feedback. External devices can also be used to
prevent knee flexion (10).
It has also been recommended that pa-
tients practice generating an extensor support

moment in other tasks such as moving from
sit to stand position, or alternatively, standing
against a wall and flexing, then extending the
knees and hips (27, 28). Again, it is not
known whether practicing the generadon of
an extensor support moment in a task other
than gait will carry over to improvements in Figure 14.5. Assisting a patient learning to maintain a
stability during the stance phase of gait. vertical trunk posture during gait with manual cues.
Chapter Fourteen ASSESSMENT AND TREATMENT of the Patient with Mobility Disorders 339
Terminal Stance
The goals of terminal stance activities

are to generate the propulsive forces that are
necessary to advance the body, and to provide
the momentum to advance the leg.
Inability to advance the weight to the
forefoot due to clawing of the toes can be
treated with a shoe insert that spreads and ex-
tends the toes (10).
The generation of propulsive forces dur-
ing terminal stance depends on a trailing leg
position of the stance hmb in which the center
of mass is forward of the trailing foot. This
requires hip extension in conjunction with
knee extension, and plantarflexion of the an-
kle, with only a small degree of posterior ro-
tation of the pelvis. Patients are often asked
to practice assuming and maintaining a trail-
ing leg position initially in a static stance po-

sition, as shown in Figure 14.6. This includes
practicing lifi:ing the heel and moving the
body weight anteriorly over the forefoot,
while the hip and knee are extended. Manual
cues and assistance are given by the clinician
as needed to facilitate this component of gait
(27,28,38).
SWING
Figure 14.6. Manually assisting a patient to master a
Pre -swing trailing limb posture, important in the generation of mo-
mentum for progression.
The goals of pre-swing treatment strat-
egies are (a) to improve the ability to generate
force in the ankle plantarflexors and hip flex- comitant activation of the TA to achieve foot
ors to power the swing limb and (b) to im- clearance.
prove activation of hip flexors, and associated Early in gait training, when the patient
knee and ankle flexion for foot clearance. Sug- does not have sufficient control to advance
gestions for improving the patient's abilit)' to the swing limb using hip and knee flexion, a
flex the hip during the initiation of swing in- towel can be placed under the patient's foot
clude manually supporting the swing limb at to facilitate advancement of the foot (28).
the foot while the patient generates the force This is shown in Figure 14.7. Alternatively,
necessary to pull the thigh segment forward. an ace bandage (Fig. 14.8) can be used to pre-
vent ankle plantarflexion and subsequent toe-
Mid-swing drag during the swing phase of gait (28).
The goal is to advance the swing limb, Terminal Swing

clearing the supporting surface with the foot.
This requires facihtation of the hip flexors, to The goals of terminal swing treatment
draw the thigh segment forward with suffi- are (a) to improve mobility in the knee, (b)
cient force to passively flex the knee, and con- to improve the ability to extend the knee with
A number of factors are considered when pre-

scribing an assistive device for a patient with
neurological impairments, including extent of
physical disability, cognitive impairment, and
the patient's personal motivations and desires
(30).
Assistive devices contribute to stability
by widening the base of support and provid-
ing additional support against gravity. Canes
are usually held in the hand opposite to the
invoK'ed extremity, allowing a reciprocal gait
pattern, with the opposite arm and involved
leg moving together. Base of support can be
hnther widened by choosing a cane with
more points of contact with the floor, such as
a small- or a large-based quad cane.
Walkers provide the greatest degree of

stability' during ambulation, further widening
the base of support, and improving both lat-
eral and anterior stability. A variet\' of walkers

exist, including straight walkers and rolling
walkers with two or four casters. An advan-
Figure 14.7. Placing a towel under the hemiparetic leg

to facilitate advancement of the swing limb without lat-
eral trunk lean or hip hike.
hip flexed, in preparation for foot-strike, and

(c) to improve activation of the ankle dorsi-
flexors, in preparation for heel-strike.
When inadequate knee extension is the
result of knee flexion contractures, manual
stretching, casting, and splinting can be used
to alter mechanical constraints (10). How-
ever, contractures will simply recur if the un-
derlying cause is chronic overactivity of the
hamstrings muscle during swing. Electrical
stimulation of the quadriceps has been used
to reciprocally inhibit an overactive ham-
strings muscle (10, 48).
Assistive Devices
A variety of assistive devices are available

to provide support to patients who are unable
to locomote independently. These include
standard walkers, rolling walkers, quad canes, Figure 14.8. Using an ace bandage to maintain the an-
straight canes, and various types of crutches. kle in dorsiflexion and to facilitate toe clearance.
Chapter Fourteen ASSESSMENT AND TREATMENT of THE Patient with Mobility Disorders 341
tage of rolling walkers is that they allow effective in meeting the task requirement of
patients to maintain speed, facilitating the locomotion on a level surface, training is
generation of momentum, and thereby facil- broadened to include locomotor training (a)
itating progression. However, the disadvan- at different speeds of walking, (b) on different
tage of rolling walkers is that they have a re- surfaces, for example, inclines, uneven sur-
duced stability compared with standard faces, and carpeted surfaces, and (c) in a va-
pick-up walkers. Further information describ- riety of visual conditions such as reduced
ing types of assistive devices, procedures for lighting, or in the presence of visual motion
measuring, and techniques for training gait cues in the environment. Adapting gait in an-
withassistive devices on level surface, curbs, ticipation of upcoming obstacles is also prac-
and steps may be found in detail in other ticed, including the ability to step over obsta-
sources (30). cles of various heights, such as those shown
It is important to consider the effects of in Figure 14.9, or to step around obstacles,
using various assistive devices on factors other shown in Figure 14.10.
than gait skills, such as attentional resources. Patients also practice walking under a
There appear to be anentional costs associated variety of task conditions such as walking with
with using an assistive device. Attentional abrupt stops, walking with quick change in
costs refer to the demand for attentional redirection, walking with head turns (Fig.
sources for information processing during the 14.11), and walking while carr^'ing a variety
performance of The attentional de-
a task. of objects (54, 55). In this way, patients learn
mands can var)' depending on the type of as-
sistive device used, and the patient's familiar-
ity with the device. A recent study examined
the attentional demands associated with two
types of walkers, a standard pick-up walker vs.
a rolling walker (52). This study found that,
while both a rolling and a standard walker are
attention-demanding, a rolling walker is less
attentionally demanding than a standard pick-

up walker.
Since there is some evidence that com-
peting demands for attentional resources by
postural and cognitive systems contribute to
instability in the elderly, understanding the
attentional requirements of the assistive de-
vices we give patients is an important consid-
eration during gait training (53).
Treatment at the Functional

Task Level
The goal of retraining at the functional

level focuses on having patients practice walk-
ing in a variety of contexts. The ability to walk
in a nattiral environment requires modifying
gait strategies to changing task and environ-
mental demands. Thus, developing the abilit)'
to adapt is a critical part of retraining mobilit)'
Figure 1 4.9. Stepping over obstacles of various heights,
skills. such as a shoe box, assists patients in learning to modify
As patients learn to develop strategies strategies for gait in anticipation of upcoming obstacles.
physical therapy approaches in helping stroke

patients learn to walk has found that the task-
oriented approach helped patients achieve
faster velocities of gait speed. However, the
differences between the two groups disap-
peared after several months (56-58). More
research is needed on
approach to ther-
this
apy. There are increasing numbers of available

harness and pulley devices that can be used
with and without a treadmill to facilitate early
gait training, even in patients who are unable
to support the body against gravity' (56-58).
An example of such a harness and support sys-
tem is shown in Figure 14.12.
Figure 14.10. The patient is asked to walk around

cones placed on the floor to help patients learn to modify
gait to avoid obstacles in the gait path.
to modify gait in anticipation of potentially

destabilizing threats to balance during gait.
Strategies to retrain compensatory re-
sponses to unexpected perturbations to gait

have not been idendfied. One possible ap-
proach might be to unexpectedly change the
speed of a treadmill while the patient is walk-
ing. A harness system would be necessary to
safely carr\' out perturbations to ongoing gait.
TASK-ORIENTED GAIT TRAINING
Recently, there has been a great deal of

interest in a strict task-oriented gait retraining
program. In this approach, littie time is spent
on pre-ambulation skills or on retraining gait
patterns. The approach is based on the phi-
losophy that, if you want to improve gait, you
should practice gait. Thus, patients are ex-
posed early in rehabilitation to a program of

ambulation training using reciprocal leg activ-
ities on a kinetron and treadmill walking with
support provided as needed (56, 57).
Figure 14.11. The patient practices walking while turn-
Preliminan' evidence comparing an ining her head to look from side to side in order to retrain
tensive task-oriented approach to traditional gait under altered task conditions.
Chapter Fourteen ASSESSMENT AND TREATMENT of the Patient with Mobility Disorders 343
control the involved leg (27, 28, 38). This is
shown in Figure 14.13. The clinician helps to

control the knee to prevent collapse during
the single limb stance phase, and assists with
knee and ankle flexion to ensure foot clear-
ance in the swing leg (Fig. 14.14).
During stair descent, shown in Figure
14.15, the stroke patient is taught to advance
the hemiplegic leg first. The therapist assists
as needed with foot placement and in knee
control, to prevent collapse of the leg when
the noninvolved leg is advanced during swing
(28).
Research has shown that certain stair
features are critical in establishing effective

movement strategies for stair-walking. Thus,
it is possible that accentuating stair features,
such as the edge of the step, or the height of
the step, and drawing the patient's attention
to these features, may enhance the patient's
ability to develop effective stair-walking strat-
egies.
Retraining Other
Mobility Skills
When retraining transfers and other
types of mobility skills, it is important to re-
Figure 14.12. An example ot a harness and pulley sys- member that there is no single correct strat-
tem used in conjunction with a treadmill for assisted task- egy for patients to learn. Research suggests
oriented gait training.
that healthy young people perform such tasks
as rising from a bed, standing up from the
floor, or rolling, in many different ways. Vari-
Stair- Walking ability' characterizes the movement patterns

used by neurologically intact individuals to
The patient with neurological impair- perform everyday mobility skills. In fact, often
ments with decreased concentric control will the same exact strateg}' is never repeated. In-
have primary problems climbing up stairs, stead, normal young adults seem to learn the
while the patient with difficulty controlling rules for performing a task. This means that
eccentric forces will have primary problems they learn what the essential or invariant re-
descending stairs. In addition, sensory im- quirements of a task are, and develop a variety
pairments will affect the patient's ability to of strategies to accomplish these require-
clear the step during swing, and place the foot ments. This suggests that the goal when re-
appropriately for the next step. Published training transfer skills in the patient with neu-
strategies for retraining stair-walking have fo- rological disabilities is to help the patient
cused primarily on therapeutic strategies with develop sensory and motor strategies that are
stroke patients. effective in meeting the task requirements,
During stair ascent, the patient is taught despite persisting impairments.
to advance the nonhemiplegic leg first. Man- Patients need to learn new rules for
ual assistance is given as needed to guide and moving and sensing, given their impairments.
344 Section III MOBILm' FUNCTIONS
the extent to which these principles can be

applied to the patient with neurological dys-
function.
To assist in the process of exploring
moN'ement strategies that are effective in
meeting the task demands, conditions of
training are varied. For example, during the
process of learning to move from sit-to-stand
position (STS), the patient may practice
standing up from a wheelchair, from the bed,
from a chair without arms, and a chair with
arms. In addition, patients may learn to em-
bed STS of other tasks such as
in a variety
stand-up and stop, stand-up and walk, or
stand-up and lean-over. This t\'pe of
variability encourages the patient to adapt the
strategy used to stand up in response to
changes in the demands of the task and the
environment.
Figure 14.13. Assisting stair-walking, controlling the

involved leg for single limb stance in stair ascent.
rather t±ian learning to use a "normal''' pat-

tern of movement. Research with neurologi-
cally intact subjects suggests that there is no
single template for moving that can be used

to train patients. Instead, guided by the cli-
nician, patients learn to explore the possibili-

ties for performing a task. Patients learn the
boundaries of what is possible, given the de-
mands of the task and the current constella-
tion of impairments. Again, it is important to
remember that the therapeutic strategies sug-
gested in the following sections are attempts
to apply a systems theory of motor control
and current research in motor learning to re-
training mobility skills. These suggestions
have not yet been validated through research.
In addition, motor learning principles related
to the acquisition of a learned skill are largely
Figure 14.14. Manually assisting with knee and ankle
drawn from research with neurologically in- flexion to ensure foot clearance in the swing leg during
tact subjects. Research has not \'et determined stair ascent.
Chapter Fourteen Assessment .\nd Treatment of the Patient with Mobility- Disori")ERS 345
Sit-to-Stand Position
Remember from the chapter on normal

mobility skills that there are two basic strate-
gies that can be used separately or in combi-
nation to stand up: a momentum strateg)' and
a force-control strategy. The momentum
strategy uses concentric forces to propel the
body for\\'ard and upward, while eccentric
forces control the horizontal displacement of
the center of mass, necessar\' for stability. A
momentum strategy requires both a minimal
speed of movement and that there be no
breaks in the motion. This allows the transfer
of momentum from the trtink to the whole
body.
In contrast, a force-control strateg)' is
characterized by frequent stops. Forces are

generated in the trunk to bring the center of
mass over the base of support of the feet.
Then, forces are generated to lift: the body to
the vertical position. The force-control strat-
egy ensures stabilit)' but requires greater
forces for progression. Often, the arms are
used to generate force, assisting with pro-
gression and stability.
Figure 14.15. Manually assisting control of the knee
during stair descent.
RETRAINING A FORCE-CONTROL
STRATEGY
This concept of trial and error explora-
tion in the learning of perceptual and motor As we mentioned in the chapter on ab-
strategies, which may be quite etFective in normal mobility, patients with neurological
meeting task goals, has a number of important dysfijnction are most often taught a force-
implications for clinicians. Initial performance control strategy for STS. In many cases, this
may be quite poor as patients learn to explore is an appropriate response to various t)'pes of
and to find their own solutions. Patients may impairments such as decreased postural con-
not progress as fast as if they were taught a trol limiting center of mass (COM) control,
single solution to the task being learned. orthostatic hypotension, and/or vestibular
Short- and long-term therapy goals need to problems that produce dizziness if the patient
be rewritten to reflect the importance placed assumes a vertical position too quickly. When
on multiple solutions to a task. An example of teaching a patient to use a force-control strat-
a new goal would be: the patient will dem- eg)', the patient is taught to bring the but-
onstrate the ability to adapt motor responses tocks forward towards the edge of the chair.
by performing the sit-to-stand task in three The trunk is brought forwarcH, bringing the
different ways. "nose over the toes." This is shown in Figure

With these principles in mind, we will 14.16. This brings the COM
over the base of
explore some ideas for retraining mobility support of the feet. The patient is then cued
skills other than gait including STS, rolling, to stand up. The patient can be encouraged
and rising from a bed. to bring the arms forward, with or without a
control, forward motion of the knee is not

blocked as the patient stands up.

You can try this for yourself. Ask a
partner to sit on the edge of a chair.
Place your knee so that it is touching
the front of your partner's knee, and apply pressure
so the knee cannot move forward during the mo-
tion. Now ask your partner to stand up. What hap-
pens? in most cases, the person will be unable to
stand. If your partner succeeds in getting the but-
tocks off the ground, he or she will often fall back-
wards. When you're standing, the knee must be
allowed to move forward as the patient assumes a
vertical position, bringing the center of mass over
the feet.
In patients with unilateral motor con-

trol problems such as hemiplegia, sensory
stimulation techniques have been suggested
as an approach to assisting the patient in
learning symmetrical weightbearing. Ideas in-

clude (a) pressure downward applied at the
Figure 14.16. Ttviching .1 Ion e-i untrol strategy tor ac-
knee to increase the weightbearing sensation
complishing sit-to-stand position involves asking the pa-
tient to move forward to the edge of the chair, incline
of the foot in contact with the support surface
trunk until the "nose is over the toes," then stand up. (refer back to Fig. 14.19); (b) lifting the pa-
support, to assist in bringing the trunk mass

forward; this isshown in Figure 14.17.
Patients whose weakness makes it difh-
cult to achieve a vertical position from a nor-
mal-height chair, can begin learning STS from
a raised chair, reducing the strength require-
ment for lifting the body (Fig. 14.18). As the
patient improves, seat height can be lowered.
In padents with asymmetrical force pro-
duction problems, facilitating symmetry is im-
portant when symmetrical
possible, since
weightbearing enhances both progression and
stability during the task. A symmetrical
weightbearing posture is only possible in pa-
tients who can generate sufficient force to
control the knee, and prevent collapse of the
body when the impaired limb is loaded. When
this not the case, the clinician will have to
is
manually control the knee for the patient, for

Figure 14.17. Encouraging forward inclination of the
example as illustrated in Figure 14.19. It is trunk during sit-to-stand position (STS) by bringing the
important that when manually assisting knee arms forward.
performing a transfer task, since this strategy
ismost efficient and requires the least amount

of muscular activity. Essential elements of
teaching a momentum strategy include en-
couraging the patient to move quickly, and
avoiding breaks in the motion. Figure 14.21
shows a patient using a momentum strategy.
The clinician can verbally instruct the
patient to quickly, with no stops. An
move
appropriate prompt might be, "Now we are
going to try standing up again, but this time
I want you to move quickly with no stops."
Manual cues can be used at the shoulders to

set the pace. Padents who have trouble gen-
erating force quickly with the trunk can try
swinging their arms freely while standing up,
to increase momentum generated in upper
body segments.
When teaching a momentum strategy,
clinicians should be aware of the stringent sta-
bility requirements of this strategy,and ade-

quately safeguard a patient with poor postural
control to prevent a fall. The risk for a back-
ward fall will be greatest at the beginning of
the movement if the patient tries to transfer
momentum from the trunk to the legs for a
vertical lift before the COM is sufficiendy for-
Figure 14.18. Teaching STS from a raised chair reduces
the strength requirement for Mfting the body and allows
the weak patient to accomplish the task.
tdent'sfoot and rubbing the heel on the

ground (Fig. 14.20); (c) lightly pounding the
heel on the ground; or (d) using pressure on
the dorsum of the foot during the STS move-
ment (27, 28, 32, 38).
It is often easier for a patient to learn to
sit down than to stand up because eccentric
force control is often gained prior to concen-
tric force control (38, 59). When teaching pa-
tients to sit down, the therapist asks the pa-
tient to practice flexing the knees in
preparation for sitting. This requires eccentric
contraction of the quadriceps to control pre-
mature collapse of the knee.
RETRAINING A MOMENTUM
STRATEGY
A patient should be allowed to explore Figure 14.19. Manually controlling the knee when as-
the possibilities for using momentum when sisting a patient who is moving from sit-to-stand position.
Figure 14.20. Facilitating symmetrical weightbear-

ing in a hemiplegic patient by providing pressure
downward applied at the knee
and lilling the patient's
foot and rubbing the heel on the ground.
ward over the feet. This is often characteristic Momentum is lost each time the patient stops
of STS in a hemiparetic patient (55). In con- to change position.
trast, the risk for a forward fall will be greatest In contrast, in a squat-pivot transfer, the
at the end of the movement in patients who patient positions the wheelchair at an angle to
are unable to control horizontal forces affect- the bed (or chair), moves to the edge of the
ing the COM. When this occurs, the COM chair,and in one swift motion moves the but-
continues to accelerate forward of the base of tocks from one surface to another, keeping
support of the feet after the patient reaches a the hips, knees, and ankles flexed. This re-
vertical position, resulting in a fall forward. quires good eccentric control of the quadri-
This is often characteristic of STS in cerebellar ceps and hip extensors.
patients who have difficulty scaling forces for It is often surprising to see the number
movement (55). of patients who find it easier to perform a
The same principles for retraining STS squat-pivot transfer, instead of a standing-
apply to retraining other types of transfers. pivot transfer, which is often the preferred
For example, when learning to transfer from strateg\' taught in therapy. For example, a 24-
a chair to a chair (or bed, or mat), a patient year-old traumatic brain injur)' patient with
can learn a standing pivot transfer, which is cerebellar ataxia was being taught to transfer
more consistent with a force-control strategy, so that she could go home on leave from the
or a squat-pivot transfer, which is more con- rehabilitation center. She was being taught a
sistent with a momentum strategy. In the standing-pivot transfer, and was unable to
standing-pivot transfer, the patient moves perform this independently, due to instability
from the seated position to a vertical stance on rising to a vertical position. As a result, she
position, then pivots around, and sits down. required moderate assistance to transfer
Chapter Fourteen Assessment and Treatment of the Patient with Mobility' Disorders 349
Figure 14.21. Training a momentum strategy involves asking the patient to move quicl<ly and avoid stopping during
the move from (A) sit to (C and D) stand positions.
Figure 14.22. Learning lu rise Iruni a bed using a force-control strategy breaks the movement into three stages,
beginning with rolling to side-lying.
Figure 14.23. Second stage of a force-control strategy to move from supine-to-stand position is moving into a sitting
position.
Figure 14.24. The final stage of a force-control strategy is moving from sit-to-stand position.
Chapter Fourteen ASSESSMENT AND Treatment of the Patient with Mobility Disorders 351
safely.Since she had fairly good eccentric con- on the bed. Pressure downward by the leg
trolof the knees, but poor balance, it was de- propels the body to side -lying and on to
cided to try letting her experiment with a mo- prone. Flexion of the head and trunk and
mentum-driven squat-pivot transfer. To her reaching movements of the arms can assist in
surpriseand ours, she learned to transfer from generating force to over (28, 29, 38).
roll
the bed to her wheelchair independendy in When rolling towards their involved
one therapy session. side, stroke patients with hemiparetic arms

should be shown how to protect the shoulder
Bed MobUity Skills by manually protracting and abducting the

arm prior to rolling (28).
Retraining bed mobility' skills includes
such tasks changing position while in bed
as
RISING FROM A BED
(for example, rolling from the supine position
Research examining movement patterns
to side-lying or prone), and getting out of
used by normal adults to get out of bed again
bed, either to a chair or standing up. As noted
suggests great variability' in how this is accom-
earlier, researchers have found that normal
plished. Still, momentum-based strategies are
young adults use a variety of momentum-re-
most often used. For example, the movement
lated strategies when performing bed mobilit)' begins as the person pushes the trunk into
skills. In contrast, force-control movement
flexion with the arms, or alternatively grasps
strategies are frequentiy used by patients with
the side of the bed and pulls and pushes into
neurological impairments, and are character-
flexion and immediately into a partial sitting
ized by frequent starts and stops.
position with the weight on one side of the
buttocks. Without stopping, the person con-
ROLLING tinues to roll off" the bed into a standing po-
sition. This strategy has stringent stability' re-
The most common approach to rolling quirements, but because there are no breaks
in normal young adults involves reaching and
in the movement, uses momentimi to move
lifi:ing with the upper extremity, flexing the
the body efficiently. As noted in the chapter
head and upper trunk, and lifting the leg to
on abnormal mobilit\' skills, there are many
roll onto the side, then over to prone. How-
reasons why this strategy is not appropriate for
ever, there were many variations to this pat-
a patient with neurological dysfunction.
tern. Most healthy young adults did not show
An alternative force-control strategy in-
rotation between the shoulders and pelvis, as-
volves teaching a patient to roll to side-lying
sumed by many clinicians to be an invariant push up to po-
(Fig. 14.22), then to a sitting
feature of rolling.
sition (Fig. 14.23). After the patient is stable
There are at least two ways to teach a
in a symmetrical sitting position with the feet
patient to roll over. The first relies more heav-
flat on the floor, he/she is taught to stand up
ilyon the generation of momentum to propel
(Fig. 14.24) (28,29,38).
the body from supine to prone. Motion is ini-
tiated with flexion of the head and trunk. At
the same time, the patient reaches over the Summary
body with the upper extremity. In addition,
1. The key to recovery of mobility skills follow-
the leg is lifted and rotated over the opposite
ing neurological injury is learning to meet the
leg, to assist with the generation of momen-
task requirements of progression, stability,
tum, to roll the body to side-lying and on to
and adaptability, despite persisting sensory,
prone. motor, and cognitive impairments. Research
An alternative to a momentum-based examining mobility strategies in neurologi-
strategy is a force-control (or combination) cally intact subjects suggests that there is no
strategy. In this approach, the patient is one right strategy that can, or should, be used
taught to lift one leg and place the foot flat to meet these requirements.
2. Retraining the patient with impaired mobility terns of normal women. Arch Phys Med Re-
skills begins with an assessment of functional habil 1970;51:637-650.
mobility skills, strategies used to accomplish 3. Waters RL, Lunsford BR, Pern,- J, Byrd R.
stance and swing requirements of gait, and Energ\-speed relationship of w alking: stan-
underlying sensory, motor, and cognitive im- dard tables. J Orthop Res 1988;6:215-222.
pairments. 4. Lerner-Frankiel MB, Vargas S, Bro«n MB,
3. Observational gait analysis is the most com- Krusell L, Schoneberger W. Functional com-
monly used clinical tool to aid therapists in munir\' ambulation: what are your criteria?
systematically analyzing a patient's gait strat- Clinical Management 1990;6:12-15.

egies. 5. Keith RA, Granger CV, Hamilton BB, Sher-
4. Treatment focuses on helping patients to re- win FS. The functional independence mea-
solve impairments, to develop strategies effec- sure: a new tool for rehabilitation. In: Eisent-
tive in meeting mobility requirements, and to berg MG, Grzesiak RC, eds. Advances in
learn how to adapt and modify these strategies clinical rehabilitation, vol 1. New York:
so performance can be sustained in a wide Springer Verlag, 1987:6-18.
variety' of settings. 6. Waters RL. Energy expenditure. In: Perri' J,
5. There are many approaches to retraining the ed. Gait analysis: normal and pathological
patient with impaired mobility skills, particu- fiinction. Thorofare, NJ: Slack Incorporated,
larly gait. Some approaches stress the impor- 1992:443^90.
tance of following a strict developmental se- 7. McGavin CR, Gupta SP, McHardy GJR.
quence. This involves learning mobility and Twehe minute walking test for assessing dis-
stability in developmental postures such as all abilit>' in chronic bronchitis. Br Med J 1976;
fours and upright kneel, prior to working on 1:822-823.
ambulation itself. 8. Butiand RJA, Pang J, Gross ER, Woodcock
6. At the other end of the retraining spectrum are A,\, Geddes DM. Two-, sLx-, and 12-minute
gait retraining approaches that focus solely on walking tests in respirator)' disease. Br Med J
retraining locomotion, with no emphasis on 1982;284:1607-1608.

training potentially unrelated skills. Often, 9. Brandstater M, deBruin H, Gowland C,
therapy is started with task-oriented gait train- Clark B. Hemiplegic gait: analysis of temporal
ing before the patient can sustain weight. Pa- variable. Arch Phys Med Rehabil 1983;
tients are supported by harnesses, and tread- 65:583-587.

mills are used to facilitate walking. 10. Montgomery' J. Assessment and treatment of
7. Other approaches, such as the one presented locomotor deficits in stroke. In: Duncan PW,
in this chapter, stress a balance between ther- Badke MB. Stroke rehabilitation: the recov-
apeutic strategies focused on resolving im- er\' of motor control. Chicago: Year Book
pairments, retraining movement strategies for Publications, 1987:223-259.^
and having the patient practice walking
gait, 1 1 Shumway-Cook A, Gruber W, Rigby P, Bald-
under various task and environmental condi- win P. Unpublished obser\ations.

tions. 12. Holden MK, Gill KM, Magliozzi MR, Na-
8. More research is needed to determine the than J, Piehl-BakerL. Clinical gait assessment
comparative effectiveness of these ap- and
in the neurologically impaired: reliabilit)'
proaches in retraining mobility skills in vari- meaningflilness. PhysTher 1984;64:35^0.

ous t>'pes of patients with neurological im- 13. Robinson JL, Smidt GL. Quantitative gait
pairments. evaluation in the clinic. Phys Ther 1981;
61:351-353.
14. Shumway-Cook, unpublished obsenations.
1 5 Murray- M, Kory R, Sepic S. Walking patterns
References
of normal women. Arch Phys Med Rehabil
1 Craik R. Changes in locomotion in the aging 1970;51:637-650.
adult. In: Woollacott MH, Shumway-Cook 16. Smidt GL, Mommens lsL\. System of report-
A, eds. Development of posture and gait ing and comparing influence of ambulator)'
across the lifespan. Columbia, SC: Universit)' aids on gait. Phys Ther 1971; 51:9-21.
of South Carolina Press, 1989:176-201. 17. Holden MK, Gill KM, Magliozzi MR. Gait
2. Finley PR, Cody KA, Sepic SB. Walking pat- assessment for neurologically impaired pa-
Chapter Fourteen .\ssessment .\nd TREATMENT OF THE PATIENT v\ith Mobility' Disorders 353
tients: standards for outcome assessment. 33. Brunnstrom S. Movement therapy in henii-
Phys Ther 1986;66:1530-1539. plegia: a neurophysiological approach. Hag-
18. Gentile A. The nature of skill acquisition: erstown, MD: Harper & Row, 1970.
therapeutic implications for children with 34. Forssberg H. Motor learning: A neurophysi-
movement disorders. In; Forssberg H, ological review. In: Berg K, Eriksson B, eds.
Flirschfeld H, eds. Movement disorders in Children and exercise, vol 9. Baltimore: Uni-
children. Med Sport Sci. Basel: Karger, versity Park Press, 1980:13-22.
1992:31^0. 35. Mayston M. The Bobath concept: evolution
19. Duncan P. Balance dysfunction and motor and application. In: Forssberg H, Hirschfeld
control theon,-. S\ilabus from a talk. Wash- H, eds. Moxement disorders in children. Med
ington State APTA annual conference, 1993. Sport Sci, Basel: Karger, 1992:1-6.
20. Shumwav-Cook A. Retraining stabilit)' and 36. Shumway-Cook A. Equilibrium deficits in
mobilit)'. Cincinnati. Annual conference, WooUacott M, Shumwav-Cook
children. In:
APTA, 1993. A. Development of posture and gait across
21. Shumwav-Cook A, Baldwin M, Kems K, the lifespan. Columbia, SC: Universit\' of
WooUacott M. Reducing the likelihood for Soudi Carolina Press, 1989:229-252.
falls in the elderlv, the effect of exercise. Ab- 37. WooUacott M, Shumway-Cook A, Williams
stract. Orthop Trans, J Bone Joint SurgePi', H. The development of posture and balance
in press. control. In: WooUacott M, Shumway-Cook
22. Krebs DE, Edelstein JE, Fishman S. Reliabil- A, eds. Development of posture and gait
of observational kinematic gait
it)- analysis. across the lifespan. Columbia SC: Universit\'
Phys Ther 1985;65:1027-1033. of Soutii Carolina Press, 1989:77-96.
23. Perr\' J. Gait analvsis: normal and pathologi- 38. Carr JH, Shepard RB. A motor releaming
cal fiinction. Thorofare, NJ: Slack Incorpo- programme for stroke. RockviUe, MD: Aspen
rated, 1992. Systems, 1983.
24. Rancho Los Amigos Hospital (Gait Analvsis 39. Godges JJ, MacRae PG, Engelke KA. Effects
Form) of exercise on hip range of motion, trunk
25. Tinetti ME. Performance -oriented assess- muscle performance and gait
" economy. Phys
ment of mobilin,' problems in elderlv patients. Ther 1993;73:468-477.
J Am Geriatr Soc 1986;34:119-126. 40. Shumway-Cook A, Anson D, Haller S. Pos-
26. Wolfson L, Whipple R, Amerman P, Tobin tural sway biofeedback for pretraining pos-
JN. Gait assessment in the elderly: a gait ab- tural control following hemiplegia. Arch Phvs
normality rating scale and its relation to falls. Med RehabU 1988;69:395-4ob.
J Gerontol 1990;45:MI2-MI9. 41 . Winstein C, Gardner ER, McNeal DR, Barto
27. Bobath, B. Adult hemiplegia: evaluation and PS, Nicholson DE. Standing balance train-
treatment. London: Heinemann, 1978. ing: effect on balance and locomotion in
28. Da\ies, PM. Steps to follow. Berlin: Springer- hemiparetic adults. Arch Phys Med Rehabil
Verlag, 1985. 1989;70:755-762.
29. Voss D, lonta M, Myers B. Propriocepti\e 42. Bobath K, Bobath B. The neurodevelop-
neuromuscular facilitation: patterns and tech- mental treatment. In: Scrutton D, ed. Man-
niques. 3rd ed. Philadelphia: Harper & Row, agement of the motor disorders of cerebral
1985. palsv. CUnics in Developmental Medicine,
30. Schmitz, TJ. Gait training with assistive de- no. 90. London, Heinemann Medical, 1984.
\ices. In: O'SuUivan S, Schmitz TM, eds. 43. Rosentiial RB, Deutsch SD, Miller W, Schu-
Physical rehabilitation: assessment and treat- mann, Hall JE. A fixed-ankle, below-the-knee
ment. Philadelphia: FA Davis 1988:281-306. orthosis for the management of genu recur-
31. Stockmyer S. ^\n interpretation of the ap- vatum in spastic cerebral palsv. J Bone Joint
proach of rood to the treatment of neuro- Surg 1975;57A:545-547.
muscular dysfunction. Am J Phvs Med 1967; 44. Basmajian JV. Kukulka CG, Narayan MD,
46:950-955. Takebe K. Biofeedback treatment of foot-
32. Charness, A. Stroke/head injurx': a guide to drop after stroke compared with standard re-
fimctional outcomes in physical therapy man- habilitation technique: effects on voluntary'
agement. Rock\ille, MD: Aspen Systems, control and strength. Arch Phys Med Rehabil
1986. 1975;56:231-236.
45. Binder S, Moll CB, Wolf SL. Evaluation of mands of ambulation with walking devices.
electromyographic biofeedback as an adjunct Phys Ther 1992;72:306-315.

to therapeutic exercise in treating the lower 53. Shumway-Cook A, Baldwin M, Kerns K,
extremities of hemiplegic patients. Phys Ther Woollacott M. The effects of cognitive de-
1981;6I;886-893. mands on postural control in elderly fallers
46. Baker M, Regenos E, Wolf SL, Basmajian JV. and non-fallers. Abstract. Society for Neuro-
Developing strategies for biofeedback: appli- science, 1993.
cations in neurologically handicapped pa- 54. Shumway-Cook A, Horak F. Rehabilitation
tients. Phys Ther I977;57;402^08. strategies for patients with vestibular deficits.
47. Baker MP, Hudson JE, Wolf SL. A "Feed- Neurol CUn North Amer 1990;8:44I^57.
back" cane to improve the hemiplegic 55. Shumway-Cook A, Horak F. Balance reha-
patient's gait. Phys Ther I979;59:I70- bilitation in the neurologic patient: Course
171. Syllabus. Seattie, NERA, 1992.

48. Bogataj U, Gros N, Malezic M, Kelih B, Klja- 56. Richards CL, Malouin F, Wood-Dauphinee
jic M, Acimo\'ic R. Restoration of gait during S, Williams JI, Bouchard JP, Brunei D. Task-
t\^'o to three weeks of therapy with multi- specific physical therapy for optimization of
channel electrical stimulation. Phys Ther gait reco\en' in acute stroke patients. Arch
I989;69;3I9-327. Phys Med Rehabil 1993;74:612-620.
49. Takebe D, Kukulka C, Narayan G, Milner M, 57. Malouin F, Pouin M, Prevost J, Richards C,
Basmajian JV. Peroneal nerx'e stimulator in Wood-Dauphinee S. Use of an intensive task-
rehabilitation of hemiplegic patients. Arch oriented gait training program in a series of

Phys Med Rehab 1975;56:237-239. patients with acute cerebrovascular accidents.
50. Waters R, McNeal DR. Tasto J. Peroneal Phys Ther 1992;72:781-789.
nene conduction velocit>' after chronic elec- 58. Harburn K, HiU K, Kramer J, Noh S, Van-
trical stimulation. Arch Phys Med Rehabil der\oort A, Matheson J. An overhead harness
1975;56:240-243. and trollev system for balance and ambulation
51. Jims, C. Foot placement pattern, an aid in assessment and training. Arch Phys Med Reh-
gait training. Suggestions fi-om the field. Phys abil 1993; 74:220-223.
Ther I977;57:286. 59. Duncan PW, Badke MB. Stroke rehabilita-
52. Wright DL, Kemp TL. The dual-task meth- tion: the recovers' of motor control. Chicago:
odology and assessing the attentional de- Year Book Publications, 1987.
Section IV
UPPER EXTREMITY CONTROL

Chapter 15
Upper Extremity Manipulation

Skills
Introduction Basic Characteristics of Reaching Tasks

Locating a Target — Eye-Head Coordination Fitts' Law
Reach and Grasp Complex Reaching and Bimanual Tasks

1 Motor Components Theories on the Control of Reaching
Role of Senses Distance vs. Location Programming
Visual Guidance in Reaching Distance Theories
Visually Controlled Reaches Across the Multiple Corrections Theory
Si
Midline Schmidt's Impulse Variability Model
i
Somatosensory Contributions to Hybrid Model: Optimized Initial
i
Reaching Impulse Model
'i Sensory Contribution to Anticipatory Location Programming
5
Aspects of Reaching Summary
r'
Adaptation of Reach and Grasp
Precision Grip
Postural Control
JffMI
Introduction spectnmi of treatment strategies available

when retraining either fl.mction. Clearly, re-
Mrs. Poirot has just had a stroke and is covery of upper extremity fiinction can be rel-
now in rehabilitation. She has partial paralysis evant to many aspects of retraining motor
of her right arm with some spasticity'. She is control.
having difficulty with many reaching and Knowing this, what is the best way to
manipulative activities, including dressing, determine Mrs. Poirot's main areas of upper
brushing her teeth, and feeding herself meals. extremity dysfunction, and how can we create
She has difficultii' with other tasks that rely on an optimal rehabilitation program for retrain-
arm function such as controlling her wheel- ing Mrs. Poirot's problems with upper ex-
chair, rising from a bed, or pushing herself up tremity fimction?
from sit-to-stand position. In addition, arm Before we can answer this question, we
fi.mction is affected by movements in other need to understand the basic requirements of
parts of her body. For example, when she manipulatory fimction. This will provide a
walks, her hemiplegic arm draws up into flex- framework for discussing normal control and
ion. the effect of neurological patholog)' on ma-
Upper extremity, posture, and mobilit)' nipulatory skills. In addition, it will provide
skills are interwoven. The
upper ex-lack of the structure for approaching clinical man-
tremity' fimction can affect many aspects of agement of upper extremit)' dysfunction in
posture and mobility' fiinction. Recover)' of the patient with neurological impairments.
posture and gait can be facilitated through We suggest that the following compo-
upper extremity training strategies. In the nents could be considerecH key elements of up-
same way, the recovery of upper extremitv per extremitv manipulator}' functions: («) lo-
fimction can be enhanced by retraining pos- cating a target, requiring the coordination of
ture and gait. This interaction broadens the eye-head movements, {b) reaching, involving
357
358 Section W UPPER EXTREMITY' CONTROL
transportation of the arm and hand in space, that activation of neck muscles usually occurs
(c) manipulation, including grip formation, 20 to 40 msec prior to activation of the mus-
grasp, and release, and (rf) postural control. cles controlling eye movements. Because the
As we mentioned in earlier chapters, the eyes have less inertia than that of the head,
systems theor\' of motor control predicts that the eyes move first, even though the neural
there are specific neural and non-neural sub- signal occurs first in the neck muscles.
systems that contribute to the control of these When the head is moved to look at an
components of manipulatoPi' fianction. Mus- object, mo\ed all the way to focus directiv
is it
culoskeletal components include such things on the object? Not necessarily. The amplitude
as joint range of motion, spinal flexibility', of head movements is usually only about 60
muscle properties, and biomechanical rela- to 75% of the distance to the target (2, 3).
tionships among linked body segments. Neu- However, when arm mo\'ements requiring
ral components encompass (a) motor pro- great accuracy are performed, this behavior
cesses, including the coordination of the eye, may be modified. It has been shown that peo-
head, trunk and arm movements; (b) sensor\' ple trained to thro\\' with great accuracy make
processes, including the coordination of \i- combined eye-head movements that go most
sual, vestibular, and somatosensory' systems; of the distance to the target (4).
(c) internal representations important for the Some tasks require eye movements
mapping of sensation to action; and (rf) alone, while others require a combination of
higher-level processes essential for adaptive eye-head movement, and still other tasks re-
and anticipaton' aspects of manipulator*' quire a combination of eye -head -trunk move-
fianctions. ments. Because of this variability', researchers
We begin our discussion of components have argued that eye-head coordination is not
of normal upper extremit}' function with a de- controlled by a unitar\' mechanism, but rather
scription of the manner in which the eye and emerges ft-om an interaction of several differ-
head are coupled during target location. We ent neural mechanisms. These could include
then discuss the components of reach and one neural mechanism that subser\'es the abil-
grasp, including motor aspects, the role of the ity' to locate objects in the near peripher\', re-
senson,' systems, and higher level adaptive quiring primarily eye movements, with litde
abilities. Finally, we review some of the the- head motion; a second mechanism to locate
ories of the control of reaching movements. objects in the fiarther periphen,', controlling
combined eye-head movements; and a third
Locating a Target-Eye- mechanism to locate objects in the far peri-
Head Coordination pheral, controlling the mo\'ements of eye-

head and trunk together { 1 ).
In a normal reaching movement, the Based on these findings, there are many
eyes, head, and arm mo\'e sequentially. How control mechanisms involved in a normal
are reachingmovements of the arm coordi- reach, and thus many possible contributing
nated with the movements of the eye and problems with reach-
factors to Mrs. Poirot's
head? Do we move our eyes first to a target, ing. Though the research just noted has fo-
then our head, and finally our hand? Kine- cused primarily on understanding normal
matic studies have shown that when an object mechanisms involved in eye-head coordina-
to be grasped appears in the peripheral \isual tion, we might make the following hypothesis
field, there is normally the following sequence regarding its application to retraining patients
of movements. The eye mo\'ement onset has with upper extremity' problems. For example,
the shortest latency, so it begins first, even be- when retraining Mrs. Poirot's problems re-
fore the head. The eyes reach the target first lated to tile coordination of eye-head move-
because they move ven,' quickly, so thev focus ments needed for upper extremity' fiinction,
on the target before the head even stops mov- the clinician might focus on training the dif-
ing (1). However, EiVIG studies have shown ferent control systems separately. For exam-
Chapter Fifteen Upper ExTB£Mi"n' Manipul.\tion Skills 359
pie, the clinician might begin by retraining flexion to approximately 100° to 120°, the
eye movements to targets located within the abilit)' to extend the wrist to slightly beyond
central visual field, then progress to retraining neutral, and sufficient mobility' in the hand to
eye head movements to targets located in the allow grasp and release (5).
peripheral visual field. Finally, movements in- Neuromotor aspects of reaching include
volving eye, head, and trunk motions could appropriate muscle tone, discussed earlier in
be practiced as patients learn to locate targets detail in Chapter 6, muscle strength, and co-
oriented in the far peripher)'. ordination. More specifically, this involves ap-
From a kinematic perspective, coordi- propriate activation of muscles to stabilize the

nation in reaching appears to be characterized scapula, rib cage, and humeral head during
by the sequential activation of eye, head, then upper extremity reaching movements, and ac-
hand movements. However, an EMG analysis tivation of muscles at the shoulder, elbow,
of reaching found that muscles in the eye, and wrist joint for transport of the arm.
head, and arm were activated almost synchro- In a patient with neurological deficits it
nously, not sequentially. In addition, the pat- is often not easy to determine the relative con-
tern of muscle activity' did not change with tribution of neural vs. musculoskeletal prob-
regard to target location. What can we learn lems to abnormal reaching. Motor control
from these experiments? Apparently, eye, problems that affect the inertial characteristics
head, and arm movement sequencing are of the system will give rise to coordination
strongly affected by inertia. problems, even when the patterns of activa-
tion are normal. For example, an increase in
Reach and Grasp stiffness will change the inertial characteristics
of the head, arm, and/or trunk, making the

Motor Components initiation of motion more difficult. Thus, we
see the important interaction between the
When the arm is used to point to an ob- biomechanics of movement and the neural
ject, all the segments of the arm are controlled control mechanisms.
as a unit. But when the arm is used to reach Two important requirements are nec-
for and grasp an object, the hand appears to essary for successfully grasping an object.
be controlled independently of the other arm First, the hand must be adapted to the shape,
segments, with the arm carrying out move- size, and use of the object. Second, the finger
ments related to transport, and the hand car- movements must be timed appropriately in
rying out movements related to grasping the relation to transport so that they close on the
object. Thus, reaching for an object can be object just at the appropriate moment. If they
divided into two subcomponents, the reach close too early or too late, the grasp will be
vs. component, which appear to be
the grasp inappropriate (1).
controlled by separate areas of the brain, as During reach and grasp, the shaping of
you will see in the research described below. the hand for grasping occurs during the trans-
Both reach and grasp involve a complex portation component of the reach (Fig.
interaction of musculoskeletal and neural sys- 15.1^) (6). This pre-grasp hand shaping ap-
tems. Musculoskeletal components include pears to be under visual control. There are
such things as joint range of motion, spinal two different categories of properties of ob-
flexibility, muscle properties, and biomechan- jects that affect pre-grasp hand shaping: in-
ical relationships among linked body seg- trinsic properties, such as the object's size,
ments. In particular, it has been suggested shape and texture, and extrinsic or contextual
that the following types of joint motion are properties, such as the object's orientation,
essential to the ability to move the arm nor- distance from the body, and location with re-
mally: scapular rotation, appropriate move- spect to the body ( 1 ).
ment of the humeral head, the ability to su- Remember that grip formation takes
pinate the forearm, shoulder, and elbow place during the transportation phase (Fig.
360 Section W UPPER EXTREMm' CONTROL
200 400 600

Time (msec)
1
^ ^cyl
_ /
I \
\
'J \
^ 'rod
"•oo Time (msec'
Figure 15.1. Characteristics of the transport phase of reaching. A, Changes in hand movement velocity \soM line)
and grip size {dotted line) as a function of time during a reach. (Adapted from Brooks VB. The neural basis of motor
control.NY: Oxford University Press. 1986:133.) B, Grip size differences for a 2-mm vs. a 10-cm cylinder. (From
Jeannerod M. The neural and behavioural organization of goal-directed movements, Oxford: Clarendon Press,
1990:61.)
15.1^) and is anticipator\' of the characteris- for difterent shapes of objects as well. The dis-
ticsof the object to be grasped. The size of tance between the thumb and index finger is
the maximum grip opening is proportional to usually largest during the final slow approach
the size of the object. This relationship is phase. It has been shown that adults with
shown in Figure 15. IB (1) with a subject prosthetic hands show this same relationship
reaching for a 2-mm rod vs. a 10-cm cylinder. ber\\'een grasp and transport phases. Appar-
When subjects change the grip opening, they endy, this relationship isn't due to biome-
do it almost entirely with finger movements, chanical or neural constraints, but may be the
while the thumb one place 1 ).
stays in ( When most efficient way to reach (1, 7).
reaching for an object, as the arm is trans- components, reach and grasp,
If the t\vo
ported forward, the fingers begin to stretch, are truly controlled independently through
and the grip size increases rapidly to a maxi- different motor programs, then one should be
mum, and then is reduced to match the size able to modify one component and not affect
of the object. the other. To test if this were possible, exper-
Subjects show difterential hand shaping iments were performed in which a spherical
Chapter Fifteen UPPER ExTREMiTi' M\nipul.-\tion Skills 361
object was transformed into an ellipsoid ob- lems with the grasp component of reaching,
ject, after a reach had begun. Figure 15.2yl although the transport component may be
shows the normal differences in hand shape normal (1 ).
when an ellipse vs. a sphere is the object to be Though research indicates that these
grasped, while Figures 15.25 and C show an two components of reaching may be con-
unperturbed reach for a sphere vs. a perturbed trolled independendy, they share a common
reach, where the subject reshapes the grasp to time course, with synchronous changes in the
accommodate an ellipsoid. It was noted that kinematics of the trajectories of the rsvo com-
the reach component was not affected by the ponents. For example, the time of maximal
change in shape of the target; however, the grip size (grasp component) closely corre-
grip shape began to change within 540 msec sponds to that of the beginning of arm reac-
after onset of the perturbation 1 ). ( celeration (transport component) ( 1 ).
Studies on both reaching in infants and Based on this research, we could hy-
in patients with specific neural lesions also pothesize that in the case of Mrs. Poirot, who
support the concept that reach and grasp has upper extremin,' paresis complicated by
components are controlled separately. Infants both reach and grasp will be af-
spasticity',
of 1 week may intercept mo\ing objects and fected. We

might predict that she will recover
contact them, but this is done with a hand the reach phase earlier and more completely
that is wide open without any grip formation. than the grasp phase (10). Because the two
Grip formation appears to develop at about aspects appear to be controlled separately,
10 to 22 weeks (1, 8). monkey this is
In the they could potentially be trained both sepa-
and it has
also the case, been shown that the and together. For example, Mrs. Poirot
rately
appearance of the grasp component, at 8 may beginpracticing the reach component by
months of age, is correlated with the matu- mo\ing her hand towards an object but not
ration of connections bet^veen the corticospi- actually grasping it. Because even the reach
nal tract and the motor neurons (9). Children phase is task-dependent, it is important to
with pyramidal lesions also show similar prob- practice reaching within the context of many
"^
^--i
^^»
Figure 1 5.2. hand shaping. A, Normal differences in hand shape when an ellipse vs. a sphere
Task-specific changes in
is An unperturbed reach for a sphere. C, A perturbed reach, where the subject reshapes the grasp to
grasped. B,
accommodate an ellipsoid. The reach begins to change shape at 540 msec after the perturbation onset. (From Jeannerod
M. The neural and behavioural organization of goal -directed movements, Oxford; Clarendon Press, 1990:65.)
362 Section IV UPPER EXTREMITY CONTROL
different t\pes of functional tasks, such as and brainstem. Descending pathways from
reach and point, reach in preparation for a the motor cortex and brainstem then activate
grasp, reach in preparation for a grasp and lift, spinal cord networks, and spinal motor neu-
or grasp and move. rons activate the muscles, and you reach for
She may also work on grasp and release the milk. If the milk carton is fiall, when you
of an object close to her hand, alleviating the thought it was almost empt)', spinal reflex
need for controlling reach. Finally, she may pathways will compensate for the extra weight
work on combining reach and grasp compo- that you didn't expect and activate more mo-
nents. tor neurons. Then the sensor)' consequences
of your reach will be evaluated, and the cer-
Role of the Senses ebellum will update the movement in this —
case, to accommodate a heavier milk carton.
What is the role of sensory information From this description, you can see that
in controlling reach and grasp.' You may recall sensor)' information plays many roles during
that in Chapter 3, to better understand the the control of reaching. Sensor)' information
function of the different levels of the nervous is used to correct errors during the execution
system, we took upper extremity
a specific of the movement itself, ensuring accuracy
function task and walked through the path- during the final portions of the movement. In
ways of the nervous system that contributed addition, sensor)' information is used proac-
to its planning and execution. We gave the tively in helping to make the movement plan.
example of being thirst)' and wanting to pour For a detailed review of the pathways involved
some milk from the milk carton in front of in these movements, please see Chapter 3.
you into a glass.
Sensory inputscome in from the periph-
ery to you what is happening around you,
tell
VISUAL GUIDANCE IN REACHING
where you are in space and where your joints
are relative to each other: they give you a map The primary function of visual feedback
of your body in space. Sensor)' inputs from the in reaching appears to be related to the at-
visual system go through two parallel path- tainment of final accuracy. It has been hy-
ways involved in goal-directed reaching: one pothesized that the constancy of thumb po-
related to "what" is being reached for (object sition with relation to the wrist during
recognition) and the other related to reaching may be part of a strategy of provid-
"where" the object is in extrapersonal space ing clear visual feedback information regard-
(localization). The "what" pathway goes ing the endpoint of the limb (II).
fi-om visual cortex to temporal cortex, while To determine the fiinction of visual
the "where" pathway goes from visual cortex feedback in reaching, studies have been per-
to the parietal lobe. formed to compare reaches made with and
Higher centers in the cortex take this without vision. Reaches with visual feedback
information (using, possibly, the parietal showed a longer duration than those per-
lobes and pre-motor cortex) and make a plan formed without feedback. Absence of visual
to act on this information in relation to the feedback didn't alter the grasp component of
goal: reaching for the carton of milk. You the reach (1).
make a specific movement plan: you're going Can reaching still occur in the absence
to reach over the box of corn flakes in fi-ont of visual cortex function? It is usually accepted
of you. This plan is sent to the motor cortex, that destruction of the visual cortex in hu-
and muscle groups are specified. The plan is mans produces total blindness, except for ver)'
also sent to the cerebellum and basal ganglia, poor visual perception of illumination
it to refine the movement.
and they modify changes (1). However, research on monkeys
The cerebellum sends an update of the with visual cortex lesions has shown some \'er\'
movement output plan to the motor cortex interesting results related to visual-motor
Chapter Fifteen UPPER Extremity Manipulation Skills 363
control (12). Though these monkeys appear was made, as long as vision was available. Re-
to be blind when their visual behavior is searchers noted that the monkeys' move-
tested, they can still reach for objects that ap- ments were awkward at first, with animals
pear in or move across their visual field. It has only sweeping objects along the floor. Mon-
been hypothesized that the superior coiliculus keys then developed a primitive grasp with
in the midbrain contributes to this residual four fingers together and no thumb, and fi-
reaching behavior. nally redeveloped a crude pincer grasp a few

Since the monkey studies were per- months after the lesion was made (1). Other
formed, human studies have verified these re- experiments discussed later in this chapter
sults (13, 14). In extending the monkey stud- have shown that deatferented monkeys can
ies to humans, these researchers used a new still make reasonably accurate single joint
experimental paradigm that had not been pointing movements, even when vision of the
used before in humans. Instead of asking hu- arm is occluded, when the pointing task was
mans with visual cortex lesions if they could learned before deafterentation (17). In this
see an object, they asked them to try to point case, even displacing the arm before the
to where they "guessed" the target would be. movement didn't aft'ect terminal accuracy,
It was shown that subjects did not point ran- even though the monkeys couldn't see or feel
domly; there was a significant correlation be- their arm position! Thus, it was concluded
tween pointing and target position. However, that the monkey is capable of using a central
they did show larger constant errors when motor program to perform previously learned
reaching within their blind visual field. They reaching movements and that kinesthetic
typicallyovershot targets when they were feedback isn't required for achieving reason-
within 30° of midline, and undershot them able accuracy when performing well-learned
when they were beyond 30° (1). movements.
E.xperiments performed with humans
with severe peripheral sensory neuropathy in
VISUALLY CONTROLLED REACHES
four limbs have shown similar results (18).
ACROSS THE MIDLINE all
One patient was able to perform a wide variety

of hand movements, such as tapping move-
Researchers have consistently found
ments and drawing figures in the air, even
that reaching movements across the midline
with the eyes closed. However, when he was
(toward targets in the visual hemifield of the
asked to repeat the movement many times
opposite arm) are slower and less accurate
with the eyes closed, the performance deteri-
than movements to targets on the same side
orated quickly. Thus, it appears that somato-
as the arm. Ipsilateral (uncrossed) reaches in
sensory informadon isn't required for arm
these studies were shorter in latency, made
movement initiation or execution, as long as
with higher maximum velocity and completed
the movements are simple or nonrepetitive.
more quickly, and made significandy more ac-
However, if subjects have to make complex
curately than contralateral (crossed) reaches
movements requiring coordination of many
(15).
joints, or repeat movements, without visual
feedback, they are unable to update their cen-
SOMATOSENSORY CONTRIBUTIONS tral representations of body space and show
TO REACHING considerable movement "drift" and problems
with coordination (1).
somatosensory input essential for the
Is These experiments suggest that certain
production of reaching movements.^ Consid- movements may be carried out without so-
erable research (1, 16) has shown that mon- matosensory feedback. Nevertheless, consid-
keys that were deatferented were still able to erablework has also shown the important
perform adequate reaching and grasping contributions of sensory feedback to the fine
movements as soon as 2 weeks after the lesion reguladon of movement.
Researchers originally thought that it It has been hypothesized that visual and
was mainly joint receptors that controlled po- somatosensory information is also used to up-
sition sense during reaching. However, more date proprioceptive and visual body maps that
recent research suggests that joint receptors allow the accurate programming of reaching
are mainly active at the extremes of joint mo- movements. To determine the influence of
tion, but not at mid-position. This would thus updated maps of the body workspace on the
make it impossible for these receptors to sig- accuracy of a reaching movement, experi-
nal limb position in the mid-working range of ments were performed to manipulate visual
joints (1). information regarding hand and target posi-
More recent work has begun to build tions prior to movement. It was shown that
evidence for a strong role for muscle spindles when a subject could not see the hand prior
in position sense. Experiments have been per- to movement, there were large errors in
formed in which tendons were vibrated, spe- reaching the target. It was thus concluded
cifically activating muscle spindle la afterents. that a proprioceptive map of the hand, by it-
Subjects consistently had the illusion that the self, was not adequate to appropriately code
joint was moving in the direction that it the hand position in the reaching workspace.
would have been moving if the muscle were This suggests that somatosensory inputs must
being stretched. For e.xample, if the biceps be calibrated by vision in order for the pro-
tendon was vibrated, it produced the illusion prioceptive map and the visual map to be
of elbow extension (19). matched (1). No experiments have yet been
Cutaneous afFerents are also important performed to determine how often the pro-
contributors to position sense. Mechanore- prioceptive map needs to be updated by visual
ceptors in the glabrous area of the hand are inputs to ensure accurate movements.
strongly activated by isotonic movements of
the fingers (20). Adaptation of Reach and Grasp
Interestingly, subjects who are recover-
ing from paralysis report that when the muscle Studies have also been performed I (
is still completely paralyzed, they have no feel- 21) to better understand the task-dependent
ing of heaviness in the limb. But as they begin adaptation of reach anti grasp. As you will see,
to regain movement ability, they feel as if the this research suggests that the ability' to adapt
limb is being helddown by weights. These how we reach is a critical part of upper ex-
sensations of heaviness are reduced as move- tremity' fiinction, since reaching movements
ments become easier and strength increases. vat)' according to the goals and constraints of
This could be due to an internal perception the task.
of the intensity of motor commands (1). Researchers have shown that the veloc-
it>' and movement durations of a reach
profiles
SENSORY CONTRIBUTION TO vary, depending on the goal of the task. If the
ANTICIPATORY ASPECTS OF subject was asked to grasp the object, the
REACHING movement duration of the reach was much
longer than if the subject pointed and hit the
An essential component of all reaching target. Also, when preparing to grasp an ob-
movements is proactive visual and somatosen- ject, the acceleration phase of the reaching
sory control, which is responsible for the cor- movement was much shorter than the decel-
rect initial direction of the limb toward the eration phase, but if the subject was asked to
target and the initial coordination between hit the target with the index finger, the ac-
limb segments. In addition, visual informa- celeration phase was longer than the deceler-
tion about the characteristics of the object to ation phase, with the subject hitting the target
be grasped is used proactively to preprogram at a relati\'cly high velocity (21). This is shown
the forces used in precision grip. in Figure 15.3.
Chapter Fifteen Upper ExTREMiTi' M\nipulation Skills 365
Figure 15.3. The different velocity pro-

files arm for grasping vs. pointing
of the
movements. Note that in the grasp, the ac-
celeration phase is shorter than the decel-
eration phase, while in the point, the re-
verse is true. (From Jeannerod M. The
neural and behavioural organization of

goal-directed movements, Oxford: Clar-
endon Press, 1990:19.)
40 60
NORMALIZED TIME
In addition, if the subject grasped tlie been shown that there are discrete phases to
target, then either fit it in a small box, or al- any lifting task. These phases are associated
ternati\ely, threw it, movement times and ve- with responses in sensor)' receptors of the
locity' profiles were also different. Movement hand. The first phase of a lift: starts with con-
times were shorter for grasp and throw vs. tact between the fingers and the object to be
grasp and fit. In addition, the acceleration lifted. When contact has been established, the
phase of the mo\'ement was longer for grasp second phase begins, with the grip force and
and throv\' than for grasp and fit. Clearly, the the load force ( load on the fingers starting to
)
task constraints and goals affect the reaching increase. The third phase begins \\'hen the
phase of the movement. This finding has im- load force has overcome the weight of the ob-
plications for the clinician engaged in retrain- ject and it starts to move. The fourth phase
ing the patient with problems related to reach occurs at the end of the lifting task, when
and grasp. Since movements used during there is a decrease in the grip and load force
reaching for an object van, depending on the shortly after the object makes contact with the
nature of the task, reaching movements need table. (22).
to be practiced within a variet\- of tasks. For This t)'pe of an organizational control
example, these tasks might include practicing scheme has many advantages. For example, it
reach during reach and point, reach and grasp, allows great flexibility' in lifting objects of dif-
reach, grasp, and throw, or reach, grasp and ferent weights. Thus, the duration of the
manipulate. loading phase depends on the object's weight:
heavier objects require higher load forces be-
Precision Grip fore they move. This also ensures that proper
grip forces are used during the load phase.
The t\pes of objects that are picked up Tills scheme also requires limited senson,' pro-
during a gi\-en da\' may van' from a light pen cessing, since the end of one phase senes as
to a hea\-\' slick bottie of oil. The nen'ous sys- the trigger for the next.
tem is capable of adapting precision grip so To ensure a safe grip, the grip-to-load
that it accommodates to objects of manv dif- force ratio has to be above a certain level; oth-
ferent weights and surface characteristics. The enN-ise, slipping will occur. One cannot as-
control mechanisms underlying these abilities sume that two objects of the same weight will
have been carefully investigated (22). It has require the same grip force, since one may be
366 Section IV UPPER EXTREMITi' CONTROL
much more slippen' than the other. How does Basic Characteristics of
the nervous system choose the correct param-
and load force? It
eters for grip appears to use
Reaching Tasks
both previous experience and afferent infor-
Until now, we have described the bio-
mation during the task. If there is a mismatch
mechanical and neural contributions to the
between the expected and actual properties of
different components of reach and grasp.
an object, then receptors in the finger pads are
However, another approach to studying the
activated. Pacinian corpuscles are ver\' sensi-
control of reaching has come from the field of
tive and capable of easily detecting that an ob-
psychology', where researchers have focused
ject has started to move earlier than expected.
on describing basic characteristics of reaching,
In addition, visual and other t\'pes of cuta-
and formulated theories about the neural con-
neous cues are important in determining the
trol of reaching based on these characteristics.
choice of grip parameters (22).
Fitts' Law
Postural Control Some basic characteristics of arm move-
ments that you may find intuitively ob\'ious
As was discussed in Chapter 6, postural are that whenever arm movement precision is
control, defined as the ability' to control the movement distance is increased,
increased or
body's position in space for the purpose of movement time becomes longer. In the
stabilit)' and orientation, has a strong influ-
1950s, Fitts (24) quantified these character-
ence on upper extremity' function. The abilit\'
istics, in the following experiments. He asked
to control the body's position in space is es-
subjects to move a pointer back and forth be-
sential to being able to move one part of the
tsveen an initial position and a target position
body, in this case the arms, without destabi- as quickly as possible. In the set of experi-
lizing the rest of the body. ments, he s\'Stematically varied the movement
Just as manipulator\' control is task-de- He
distance and the width of the target.
pendent, postural requirements also varA' ac-
found that he could create a simple equation
cording to the task. For example, postural re-
relatingmovement time to the distance
quirements involved in a seated reaching task moved and the target width. This equation,
will be less stringent than those in a standing which has become known as Fitts' law, is
task and thus may require only muscles in the
shown below:
trunk. In contrast, postural demands during
reaching while standing are greater, requiring MT= a + b log, 2D/W
more extensive activation of muscles in both
the legs and trunk to prevent instability'. a and b are empirically determined constants,
Postural demands can affect the speed MTis movement time, D is distance moved,
and accurac)' of an upper extremin,' move- and Wis the width of the target (25). The
ment. When postural demands are decreased term /q/f, 2 D/W his been called the index of
by providing external support, upper extrem- difficult^'. Movement time increases linearly
it)' movements are faster, since prior postural with the index of difficulty, that is, the more
stabilization is not necessar>' (23). difficult the task, the longer it takes to make
Helping a patient to regain sufficient the movement.
postural control to meet the postural require- This equation has come to be known as
ments inherent in a reaching task is essential Fitts' law because move-

its abilit}' to relate
to retraining that task. The reader is urged to ment time to movement accuracy and dis-
review Chapters 6-10 discussing postural tance applies to many different kinds of tasks,
control, its relationship to reaching, and issues including discrete aiming movements, mov-
related to retraining the patient with postural ing objects to insert them in a hole, moving a
disorders. cursor on a screen, small finger movements
Chapter Fifteen Upper Extremity Manipulation Skills 367
under a microscope, and even throwing darts. tures. For example, it was shown (28) that
Fitts' law has proven accurate in describing when subjects were asked to point at two tar-
movements macie by subjects of all ages, from gets, they moved the hands simultaneously,
infants to older adults (25, 26). even if the reaching tasks were ver\' different
What are the constraints of the individ- example, one was near and
in difficulty (for
ual and the task that lead to this particular law large, and the other one was far away and
regarding movement? It has been suggested small). Other researchers have noted this
that movement time increases with distance same tight bimanual coordination when sub-
and accuracy in part due to the constraints of jects reached forward to manipulate an object
our visual system. to translate our
It is difficult with two hands. Thus, it has been suggested
visual perception of the distance to be covered that independent body segments become
precisely into an actualmovement; thus, as functionally linked for the execution of a
the hand approaches the target, time is common task (1).
needed to further update the movement tra- How does the ner\'ous system control
jecton,' (25). complex arm movements to reach targets with
speed and elegant precision? This is a complex
Complex Reaching and problem that could be solved in different
Bimanual Tasks ways. For example, the nervous system could
plan reaching movements with respect to the
In Chapter when we discussed
1, theo- activation sequences of individual muscles;
ries of motor control, we mentioned Bern- this has been referred to as a muscle coordi-
stein's contributions to systems theor\' (27). nate strategy (29). Alternatively, reaching
Remember that he proposed that a given ner- could be planned in relationship to joint angle
vous system program will produce different coordinates, that is, planning the movements
outcomes in different situations because the of shoulder, elbow and wrist joints to arrive
response of the body will depend on the initial at the target. This would mean that the ner-
position of the limbs and on outside forces vous system was planning the movement
such as gravit)' and inertia. When body seg- around a set of intrinsic coordinates of the
ments act together, the nervous system must body, expressed in terms of the joint angles.
also take into account the forces they generate Finally, the nervous system could plan arm
with respect to each other. Bernstein hypoth- movements in terms of the final endpoint co-
esized that the nervous system possessed a ordinates, using extrinsic coordinates in space
central representation of the movement that (29).
was form of a "motor image," repre-
in the Levels of planning could also be consid-
senting the form of the movement to be ered in terms of a hierarchy, with, for exam-
achieved, not the impulses needed to achieve ple, both kinematic and kinetic levels of plan-
it. He believed that proprioception was im- ning. Kinematic levels of planning would be
portant to the final achievement of the move- organized around geometry, such as joint an-
ment, not in a reflex-triggering sense, but as gle variables and endpoint variables. Kinetic
it contributed to the central representation of levels of planning would be organized around
the movement (I). He also suggested that forces, such as the forces of muscle activation
one way of controlling the high number of and joint torques (29).
degrees of freedom involved in any complex On the one hand, it seems intuitively
movement was to organize the actions in obvious that we would need to use some vari-
terms of synergies, or groups of muscles or ation on endpoint coordinates planning in or-

joints that were constrained to act as a unit der to do something like picking up a glass of
(27). water. If we plan a movement using intrinsic
In fact, many researchers have now coordinates alone, without regard to the ac-
shown that hand movements are organized tual position of the object in space, the ac-
synergically or through coordinative struc- curacy of the movement with respect to the
end position needed is likely to be decreased. guided modern physiologists in their experi-
But when the nen'ous system plans according ments exploring the control of reaching
to endpoint coordinates, it needs to make a movements.
complex mathematical transformation called
If the spatial shape of a trajectory is invariant
an inverse kinematics transformation, which
of the muscle scheme or the joint
irrespecti\e
would transform endpoint coordinates into
scheme, then the motor plan must be closely
joint angle coordinates. Then it has to create
related to the topologi,' of the trajecton,-and
this trajector>' by producing the appropriate
considerabh' removed from joints and muscles.
muscle activation patterns (29).
It has also been proposed that move- Thus, experimenters have begun to look
ments are planned in terms of joint angle co- for invariant characteristics in different vari-
ordinates, which has the advantage of not re- ables related to the reach. If invariances are
quiring an inverse kinematics transformation. found across different conditions, this could
This would mean that the organization of be considered e\idence that the ner\'ous sys-
mo\'ement by the nervous system would be tem uses this variable to plan movements.
simplified. However, the nervous system It shown (30) that the path of
has been
would still have to do an imerse d\'namics the wrist in an arm mo\'ement is unaffected
transformation that would transform joint an- by movement speed, or load (weights held in
gle coordinates into muscle torques and mus- the hand). In addition, the velocity' profiles of
cle activation patterns required to make the a movement are also unaffected by mo\ement
movement (29). speed or load. These findings support the
If trajectories were planned in terms of concept that the ner\ous system uses kine-
muscle activation patterns, it would ha\'e the matic variables for planning.
advantage of simplif\'ing the in\'erse kinemat- Remember that there are two t\'pes of
icsand inverse dynamics problems, but we kinematic variables that could be used for
have also mentioned that muscle activation movement planning: joint angle coordinates
patterns are only indirectly related to final and endpoint coordinates. If the nervous sys-
joint positions. Thus, programming move- tem controls movements in joint angle coor-
ments in this manner could cause large inac- dinates, the hand should move in a curved
curacies (29). line, because the movements will be about the
How does one go about answering the axis of a joint, as you see in Figure 15.4^.
question of how the nervous system plans Howe\'er, if it plans mo\'ements with respect
movements.' Hollerbach, in an excellent re- to extrapersonal space or endpoint coordi-
of the research on arm movement plan-
vie\\' nates, the hand would be expected to move
ning (29), mentions that Bernstein actually in a straight line (Figure 15.45) (I, 26, 29).
made the foUowing statement (27), which has To answer this question, researchers
Hand space Joint space Hand space Joint space
Figure 15.4. Different variables that can be used for planning arm movements. A, If movements are controlled in
joint coordinates,hand trajectories are curved. B, If movements are controlled in endpoint coordinates, joint space is
cur\'ed (acomplex elbow and shoulder movement is required). (Adapted from Hollerbach JM. Planning of arm move-
ments. In: Osherson DN, Kosslyn SM, Hollerbach JM, eds. Visual cognition and action: an invitation to cognitive
science, vol 2. Cambridge, MA: MIT Press, 1990:187.)
Chapter Fifteen Upper ExxREMi-n- Manipulation Skills 369
(31) asked subjects to point to targets in two- nal location is the parameter being pro-
dimensional space (on a surface) and recorded grammed.
tlieir handtrajectories. They found that sub-
jects tended to move the hand in straight Distance vs. Location
lines, with their joints going through complex Programming
angular changes. Even when they were asked
to draw cur\'ed lines, the subjects tended to What do we mean by programming dis-
draw a series of straight line subunits. These tance vs. location? According to the distance
support the concept that the CNS pro-

results programming theory, when making an arm
grams movements according to endpoint co- movement toward a target, people visually
ordinates. perceive the distance to be covered. Then,
Other researchers (29) have explored they activate a particular set of agonist muscles
arm mo\'cment control hirther and have to propel the arm the proper distance to the
shown that the nervous system can direcdy target. At a particular point, they turn off the
control the joints and still produce straight agonist muscles and activate antagonist mus-
line movements. This is done by varying the cles at the joint in order to provide a braking
onset dmes for the joint movements, with all force to stop the movement (33).
joints stopping at the same time. This method According to the location programming
of control gives movements with almost theor\', the ner\ous system programs the rel-
straight line paths. This suggests that straight ative balance of tensions (or stiflihess) of two
line trajectories can occur even when the CNS opposing (agonist and antagonist) muscle
is using joint angle coordinates to program sets. According to this theor\', even,' location
movements. Thus, it is not clear whether the in space corresponds to a family of stiffness
CNS programs movements exclusively by one relations berv\'een opposing muscles, as we ex-
method or the other. plain later in the chapter. Let's first look at
Russian researchers (32) have shown distance programming theories.
that the elbow and wrist joints are controlled

as a synergic unit. When subjects were asked
DISTANCE THEORIES
to move the elbow and wrist joint congru- Multiple Corrections Theory
endy both together), the subjects
(flexing
could perform this task with ease, with joint It has been shown repeatedly that ac-
motions starting and stopping as a unit. When curacy of arm mo\'ements decreases when vi-
asked to move the joints incongruendy (flex- sion is For example, when subjects
absent.
ing one and extending the other), they per- were asked to make arm movements of diflfer-
formed the task with considerable difficult!,', ent durations to a target, movements of 190
moving the joints much less smoothly. This is msec or less were unaffected by loss of vision,
additional evidence for joint-based planning while movements of 260 msec or more were
(26). affected by loss of visual feedback (34). Thus,
it appears that mo\'ement trajectories are cor-
Theories on the Control rected based on visual feedback, and that it
OF Reaching takes about 200 to 250 msec for vision to be

able to update a movement trajector\'. Con-
A number of theories on the control of sidering that some movement time must oc-
reaching have been formulated to explain cur before the limb is close enough to the tar-
some of the reaching characteristics just de- get to use visual feedback, one realizes that
scribed. The following section explores these the visual processing time is slighdy shorter.
theories. The first group of theories tends to It has been shov^'n (35) that subjects need to
assume that the nen-ous system is program- see their hand for at least 135 msec during a
ming distance in making movements, while movement to use vision to improve move-
the second group of theories suggests that fi- ment accuracv.
In the 1960s, researchers (36, 37) pro- when he asked subjects to make a fast but ac-
posed that aiming movements consisted of a curate movement, the large forces required
series of submovements, each responding to caused increased force variability. This in-
and reducing visual error. Thus, an initial creased variability resulted in a decreased
movement, before any visual correction takes movement accuracy (38). These movement
place, covers most of the distance to a target characteristics were described in the following
and is independent of final precision. This equation:
model predicts a constant b for Fitts' law,
which is almost identical to the one that Fitts W, = a+b D/MT
and Peterson calculated originally (25).
There are, however, some problems where We is variation in movement endpoint
with this model. Typically, aiming move- expressed in standard deviation units, D is dis-
ments to a target have only one correction, if tance moved, and MT\s movement time. This
any, and when corrections are made, they do equation is similar to Fitts' law. It indicates
not have constant durations or proportions of that simply taking into account that faster
the distance to the target (26). movement requires more force can explain
How
might this theory be used to ex- Fitts' law,without having to factor in a need
plain problems related to inaccurate reaching for visual feedback for movement accuracy
movements commonly found in patients with (25).
neurological deficits? The multiple correc- This theory alone cannot be used to ex-
tions theory stresses the importance of visual plain aiming movements, since as we have
when making corrections during a
feedback seen earlier, many movements, particularly
movement to increase accuracy. Thus, inac- those lasting longer than 250 msec, do use
curate movements could be the result of loss visual feedback for accuracy.
of \dsual feedback. When retraining a patient Nonetheless, this theory does have rel-
using a multiple corrections theory, the cli- evance for the clinician involved in retraining
nician could have the patient practice slow upper extremity control. It suggests the im-
movements, requiring a high degree of accu- portance of practicing fast movements of
racy, drawing the patient's attention to visual varying amplitudes during therapy sessions.
cues relating hand movement to target loca- In this way, the patient learns to program
tion. forces appropriately for quick and accurate
movements.
Schmidt's Impulse Variability Model
Hybrid Model: Optimized Initial Impulse
Another way of explaining the charac- Model
teristics of arm movement seen in Fitts' equa-
tion is to h\'pothesize that the initial phase of The previous two models deal with two
the movement, involving the generation of a extremes of movement control (a) the use of
force impulse is more important than later visual feedback to improve accuracy during
phases of the movement dealing with ongoing ongoing portions of slower movements, and
control. This would be particularly true in (
b) very fast movements that cannot easily use
cases where the movement is too fast to utilize visual feedback, and thus are controlled only
visual feedback to aid in accuracy. through the amplitude of the initial impulse.
Schmidt performed research in which In an attempt to create a model to explain the
subjects were asked to make movements
fast entire range of possible aiming movements,
over a fixed distance. These movements re- more recent studies (39) have described a hy-
quired large amounts of force, since high-ve- brid model that combines elements of both of
locit}' movements require large forces to gen- these models. This hybrid model is referred to
erate the movement. He showed that the size as the optimized initial impulse model.
of the subject's error increased in proportion Researchers involved in studying this
to the magnitude of the force used. Thus, model hypothesized that a subject makes a
Chapter Fifteen Upper Extremity Manipulation Skills 371
first movement toward a target, which, if suc-

cessfiil, is the sole movement. However, if it
is inaccurate, for example, it undershoots or

overshoots the target, another movement will
be required involving visual feedback during

ongoing movement control. Clearly, the sub-
ject needs to find a balance between moving
quickly, which requires a large initial force,
and moving slowly enough to allow correc-

tions to the ongoing movement, thereby en-
suring accuracy.
It was found that an equation taking
these issues into account was similar to Fitts'
law:
Figure 15.5. The cafe-door model. Simplified expla-
T= a+ b(n(D/W)'^" nation of the mass-spring model of motor control. A,
When a cafe door is at rest, it resembles a joint at mid-
where Tis movement time, D is distance, and point, with both muscles at midlength. B, When one
spring of a cafe door is shortened and the other is length-
Wîs width of the target, and « is the number
ened, the door is open, analogous to one muscle con-
of submovements used to reach the tai-get and the other relaxing
tracting to allow the joint to flex.
(26).
Since functional activities require a va-
riety of movements, both fast and slow, with
the springs of the cafe door. We can change
the position of the joint simply by chang-
varying degrees of accuracy, it is important to
ing the relative stiffness of the two muscles,
retrain a patient's ability to perform a contin-
through higher or lower relative activation
uum of movements that vary in both speed
levels. Though this may sound like an unusual
and accuracy.
way for the nervous system to program reach-
LOCATION PROGRAMMING ing movements, experiments have shown that
this occurs in many circumstances.
As we mentioned two
earlier, there are For example, experiments performed on
ways that the nervous system could program monkeys (17) suggest that many movements
arm movements, through distance program- may be controlled through location rather
ming, or through programming the endpoint than distance programming. In these experi-
location of the movement (25, 40). The ex- ments, the monkeys were trained to make el-
ample of a cafe-door swinging on springs has bow movements to different targets whenever
sometimes been used to explain the location lights above those targets were turned on
programming model (33). Figure 15.5v4 (Fig. 15.6D). The monkeys wore a large col-
shows the door in a closed position. The lar that blocked sight of the arm, eliminating
movement of the cafe door is described as oc- visual feedback. In addition, in certain exper-
curring when there is a reduction in length of iments, the dorsal roots of the spinal cord
one spring and the lengthening of the other were severed, eliminating kinesthetic feed-
spring. When the door is released, the imbal- back from the arm. The accuracy of the mon-
ance between the springs causes the door to keys' arm movements was measured with and
return to its closed position, where the springs without visual and kinesthetic feedback. Re-
are at their resting length. If you want to keep searchers found that the monkeys' reaching
the door open, you can simply change one was normal, despite a loss of visual and kin-
spring for another of a different stiflFness, and esthetic feedback (Fig. 15.6^) (17).
then it will have a new resting position (Fig. They then gave a perturbation to the
15.5B). deaff'erented monkeys' arm, moving it fi-om
It has been suggested that the agonist/ its original position, just after the target light
antagonist muscle pairs at the joints are like was turned on, but before the monkey began
Figure 15.6. Experimental set-up to test the

mass-spring model of control. The deaffer-
ented monkey is pointing to a target, but is
unable to see its hand. The hand can be

moved to a new position by a torque motor
after the target is illuminated, but before the
hand starts to move. As you see from the
movement traces, the monkey was able to
successfully point to the target, even when the
hand was perturbed. (From Brooks VB. The
neural basis of motor control. NY: Oxford
University Press. 1986;138.)
to move. Remember,monkeys couldn't

the where the limb was perturbed, because, just
feel when it was per-
or see their arm position like the cafe door, the limb would swing to its
turbed. Nevertheless, the monkeys reached new spring-setting. Thus, the monkey didn't
for the target with reasonable accuracy (Fig. have to know its starting point in order to go
15.65 and C). monkeys were using dis-
If the to the correct endpoint.
tance programming for reaching, this would It is interesting that in these experi-
have been impossible, because they would ments (17), the monkeys were not able to
have applied a fixed force pulse in the elbow continue to make accurate movements when
muscles to move their arm to the new posi- the shoulder position was changed. It appears
tion. Since the arm had already been per- that without visual or somatosensory feed-
turbed, they should have ended up in the back from the arm, they could not update a
wrong place! central reference concerning shoulder posi-
The only way these results can be ex- tion changes. These changes then threw off
plained through the use of endpoint loca-
is the elbow location programming.
tion programming. In this case, what the ner- More recent work (41) with humans
vous system would program is the stiffness (or produced similar results. In this study, sub-
background activity level) settings on the ag- jects were blindfolded, and their fingers were
onist and antagonist muscles of the arm. For anesthetized by using a pressure cuft'. Before
example, if the arm was originally in a flexed testing began, they were trained to move their
position, they would have high background fingers to a specific position in space. They
activity levels in the elbow flexors and low lev- were then given brief finger perturbations
els in the extensors. To move the arm pre- during the course of their finger movement.
cisely to the new location, they would simply With complete loss of finger sensation, there
change the background activity (stiffness) lev- was ven,' little difference in terminal error be-
els so that the spring constant of the elbow tween the perturbed and unperturbed move-
flexors was at a specific lower level, and that ments.
of the extensors was at a predetermined These results suggest that the nervous
higher level (Fig. 15.6£). Once this new system is able to encode the location of body
spring setting was made, it wouldn't matter segments in space, in relation to a base body
Chapter Fifteen UprER Extremit\' Manipul.'vtion Skjlls 373
position asvandng activation levels of agonist tant part of retraining accurate upper extrem-
and antagonist muscles. What does this mean? it}' movements.
It has been suggested {S3) that this could ex-
plain why we can perform a skill (such as

reaching for a cup or throwing a ball) hun- Summary
dreds of times without repeating exactiy the
same movement. According to classic pro- 1 From a kinematic perspective, coordination in
gramming theory, one would have to make a reaching is characterized by the sequential
new program for each movement variation, activation of eye, head, then hand move-
ments. However, muscle responses in these
but according to the mass-spring model, all
segments tend to be activated synchronously,
one would have to do is program the appro-
not sequentially. Thus, inertial characteristics
priate muscle activity ratios, and the limb
play an important part in the final movement
would move appropriately to its final position.
characteristics.
Do these results suggest that distance 2. Reach and grasp represent two distinct com-
programming iswrong? No. Most likely, both ponents that appear to be controlled by dif-
strategies are used for arm movements, de- ferent neuralmechanisms. Thus, patients with
pending on the task and the context. For ex- motor control problems can have difficulties
ample, it has been shown that when humans in one or both aspects. This has implications
make rapid elbow flexion movements (42), for retraining.
they show a triphasic burst of contraction: first 3. Certain aspects of the grasp component, such
the biceps is activated, followed by the triceps as force of the grasp, are based on the person's
perception of the characteristics of the object
(braking the movement), and then the biceps
to be grasped, and thus are programmed in
again. This same pattern was found in patients
advance.
with loss of kinesthetic sensation. However,
4. Visual and somatosensory information are
when subjects uere asked to move more also used reactively for error correction during
slowly and smoothly, they showed continuous reaching and grasping.
biceps acti\'it\' and no triceps activit}'. This has 5. Fitts' law expresses the relationship between
led some researchers (25) to argue that the movement time, distance, and accuracy, stat-
subjects are using mass-spring or location ing that when the demands for accuracy in-
programming for slow movements, and a crease, movement time will also increase.
combination of distance programming and 6. There are two theories regarding the neural
location programming for faster movements. control of reaching: distance programming vs.
location theories.
There are also limitations to the mass-
7. According to the distance programming the-
spring model. The model only holds with sin-
ory, when people make an arm movement to-
gle-joint, one-plane movements. Most move-
ward a target, they visually perceive the dis-
ments involve many joints, are carried out in tance to be covered, and then they activate a
three-dimensional space, and have to take particular set of agonist muscles to propel the
into account gravity (1). arm the proper distance to the target. At a par-
In summar\', research studies appear to ticular point, they turn off the agonist muscles
indicate that single-joint movements that are and activate antagonist muscles at the joint to
shorter than .25 seconds are too short to take provide a braking force to stop the movement.
advantage of visual feedback, while those 8. According to the location programming the-
longer than about .25 seconds involve visual ory, the nervous system programs the relative
balance of tensions (or stiffness) of two op-
feedback in the homing-in phase. Slower
posing (agonist and antagonist) muscle sets.
movements may involve location program-
According to this theory, every location in
ming, while faster movements may involve a
space corresponds to a family of stiffness re-
combination of distance and location pro- lations between opposing muscles.
gramming. 9. It is probably the case that both strategies are
This model would suggest that the caused for arm movements, depending on the
modulate stiffness levels between the
pacity' to task and the context.
agonist and antagonist muscles is an impor-
References 16. Taub E, Berman AJ. Movement and learning

in the absence of sensor\' feedback. In: Freed-
1. Jeannerod M. The neuraJ and beha\ioral or- man SJ, ed. The neurophysiology of spatially
ganization of goal-directed movements. Clar- oriented behavior. Homewood, NJ: Dorsey
endon Press; Oxford, 1990. Press, 1968:173-192.
2. Grest\' MA. Coordination of head and eye 17. Polit A, Bizzi E. Characteristics of motor pro-
movements to fixate continuous and inter-
grams underhing arm movements in mon-
mittent targets. Vision research 1974; keys. J Neurophysiol 1979;42:183-194.
14:395-403. 18. Rodiwell JC, Traub MM, Day BL, Obeso JA,
3. Biguer B, Prablanc C, Jeannerod M. The Thomas PK, Marsden CD. Manual motor
contribution of coordinated eye and head performance in a deafFerented man. Brain
movements in hand pointing accuracy. Exp 1982;105:515-542.
Brain Res 1984;55:462-169. 19. Goodwin GM, McCloskey DI, Matthews
4. Roll IP, Bard C, Paillard, J. Head orienting PBC. The contribution of muscle afferents to
contributes to directional accuracy' of aiming kinaesthesia shown by vibration induced il-
at distant targets. Hum Mov Sci 1986;5:359- lusions of movement and by the effects of par-
371. alysing joint afferents. Brain 1972;95:705-
5. Charness AL. Management of the upper ex- 748.
tremity in the patient with hemiplegia. 20. HulligerM, Nordh E, Thelin AE, VaUbo AB.
Course syllabus, Annual Meeting, Washing- The responses of afferent fibers from the gla-
ton Physical Therapy Association, 1994. brous skin of the hand during voluntary fin-
6. Jeannerod M. The timing of natural prehen- ger movements in man. Physiol 1979;
J
sion movements. J Motor Behav 1984; 291:233-249.
16:235-254. 21 Marteniuk RG, Mackenzie CL, Jeannerod M,
7. Eraser C, Wing A. A case study of reaching Athenes S, Dugas C. Constraints on human
bv a user of a manually-operated artificial
arm movements trajectories. Can Psychol
J
hand. Prosdiet Orthot Int 1981;5:151-156. 1987;41:365-368.
8. Bruner JS, Koslowsld B. Visually pre-adapted 22. Johansson RS, Edin BB. Neural control of
constituents of manipulator!' action. Percep- manipulation and grasp. In: Forssberg H,
tion 1972;1:3-14. Hirschfeld H, eds. Movement disorders in
9. Kuypers HGJM. Corticospinal connections: children. Basel: Karger,1992:107-112.
postnatal development in rhesus monkey. Sci-
23. Cordo P, Nashner LM. Properties of postural
ence 1962;138:678-680. adjustments associated with rapid arm move-
10. De Souza LH, Hewer RL, Miller S. Assess- ments. J Neurophysiologv' 1982;47:287-
ment of reco\er\' of arm control in hemiplegic 302.
stroke patients. 1. Arm fiinction tests. Int Re- 24. PM. The information capacitv- of the hu-
Fitts
hab Med 1980;2:3-9. man motor system in controlling the ampli-
1 1 Wing ^\M, Frazer C. The contribution of the tude of movement J Exp Psychol 1954;
thumb to reaching movements. Q J Exp Psy- 47:381-391.
chol 1983;35A:297-309. 25. Keele SW. Behavioral analysis of movement.
12. Humphrey NK, Weiskrantz L. V'ision in
In: Brooks \T5, ed. Handbook of physiologv':
monkeys after removal of the striate cortex. section I: The nervous system, vol 2. Motor
Nature 1969;215:595-597. control, part 2. Baltimore: Williams & Wil-
13. Weiskrantz L, Warrington ER, Sanders MD, kins, 1981:1391-1414.
Marshall J. Visual capacity- in the hemianopic 26. Rosenbaum DA. Human motor control. San
field following a restricted occipital ablation. Diego: Academic Press. 1991.
Brain 1974;97:709-728. 27. Bernstein N. The coordination and regula-
14. Perenin MT, Jeannerod M. Residual \ision in
tion of movements. Oxford: Pergamon Press,
cortically blind hemifields. Neuropsychologia 1967.
1975;13:1-7. 28. Kelso JAS, Southard DL, Goodman D. On
15. Fisk JD, Goodale MA. The organization of the coordination of two-handed movements.
eye and limb movements during unrestricted Exp Psychol [Hum Percept] 1979;5:229-
J
reaching to targets in contralateral and ipsi-
238.
lateral visual space. Exp Brain Res 1985; 29. HoUerbach JM. Planning of arm movements.
60:159-178. In: Osherson DN, Kosslyn SM, Hollerbach
Chapter Fifteen Upper ExTRF.Miri' Manipulation Skills 375
JM, eds. Visual cognition and action: an in- 37. Keele SW. Movement control in skilled mo-
vitation to cognitive science, vol 2. Cam- tor performance. Psychol Bull 1968;70:387-
bridge, Mass: IVIIT Press. 1990:183-211. 403.
30. Atkeson CG, Hollerbach JM. Kinematic fea- 38. Schmidt RA, Zelaznik HN, Hawkins B,
tures of unrestrained vertical arm movements. Frank JS, Quinn JT, Jr. Motor output vari-
J Neuroscience 1985;5:2318-2330. ability: a theory for the accuracy of rapid mo-
31. Morasso P. Spatial control of arm move- tor acts. Psychol Rev I979;86:4I5^52.
ments. Exp Brain Res 1981;42:223-227. 39. Meyer DE, Abrams RA, Kornblum S, Wright
32. Kots YM, Syrovegin AV. Fi.xed set of variants CE, Smith JEK. Optimality in human motor
of interactions of the muscles to two joints in performance: ideal control of rapid aimed
the execution of simple voluntary move- movements. Psychol Rev 1988;95:340-370.
ments. Biophysics 1966;11:1212-1219. 40. Feldman AG. Change in the length of the
33. Keele SW. Motor control. In: Kaufman L, muscle as a consequence of a shift in equilib-
Thomas J, Boff K, eds. Handbook of percep- rium in the muscle-load system. Biofizika
tion and performance. New York: John Wiley 1974;19:534-538.
&Sons, 1986:30.1-30.60. 41. Kelso JAS, Holt KJ. Exploring a vibratory sys-
34. Keele SW, Posner MI. Processing visual feed- tems analysis of human movement produc-
back in rapicH movement. J Exp Psychol 1968; tion. J Neurophysiol 1980;43:1183-1196.
77:155-158. 42. Hallett M, Shahani BT, Young RR. EMG
35. Carlton LG. Processing visual feedback infor- analysis of stereotyped voluntary movements
mation for movement control. J Exp Psychol in man. J Neurol Neurosurg Psychiatry' 1975;
[Hum Percept] 1981;7:1019-1030. 38:1154-1162.

36. Grossman ERFW, Goodeve PJ. Feedback
control of hand-movement and Fitts' law. Q
J Exp Psychol I983;35A:25I-278.
Chapter 16
Upper Extremity Manipulation

Skills: Changes Across the Life
Span
Introduction Role of Experience in Development of Eye-

Role of Reflexes in Development of Hand Coordination
Reaching Behaviors Eye-Head-Hand Coordination Development
Reaching Behaviors: Innate or Learned? Reaction-Time Reaching Tasks
Innate Hypothesis Fitts' Law
Early Development of Eye-Hand Coordination Movement Accuracy

Visually Triggered vs. Visually Guided Kinematics of Reaching Movements
Reaching Changes in Older Adults
Emergence of Hand Orientation Changes in Reaching Movement Time with
Learning to Grasp Moving Objects Age
(Catching) Compensation and Reversibility of
Development of Pincer Grasp Decrements in Reaching Performance
Emergence of Object Exploration Summary
Introduction integration of those reflexes into voluntary

movement (1). We also discuss the relative
The development of reaching and ma- contributions of genetics vs. experience to the
nipulation complex and actually in-
skills is emergence of reaching in the neonate. We
volves the development of many behaviors, then review more recent studies that come
each of which emerges progressively over time fi"om newer theories of motor control such as
in association with maturation of different the ecological, dynamical, and systems ap-
parts of the nervous and musculoskeletal sys- proaches.
tems and with experience. For example, the
infant's ability to transport the arm towards Role of Reflexes in Development
an object precedes the ability to grasp. The of Reaching Behaviors
ability to grasp emerges at 4 to 5 months, pre-
ceding the infant's ability to explore objects, Is early reaching reflexly controlled.'
which does not emerge until about the first This is a question that has been debated in the
year of life. Thus, the development of mature developmental literature for many years. Early
reaching and manipulation occurs gradually theories of the development of reaching ar-
over the first few years of life. gued that reflexes provide the physiological
This chapter explores the research on substrate for complex voluntary movements
the development of reaching abilities in in- such as reaching (1). According to these the-
fants and children as well as the changes in ories, the transition from reflexes to voluntary
reaching abilities that occur in older adults. reaching is a continuous process, with new-
We first discuss some of the early hypotheses born reflexes gradually being incorporated
concerning the development of reaching, into a hierarchy of complex coordinated ac-
which propose that reaching either results tions (2). A review of eye-hand coordination
fi-om the inhibition of primitive reflexes or the development mentions that early develop-
377
mental theoreticians may have overlooked an- showed both and manual activity in the
visual
other possibility' regarding the development first few weeks but these move-
after birth,
of reaching: that eye-hand coordination may ments were apparentiy unrelated (4). Thus, in
emerge concurrentiy with the maturation of the 1960s, many researchers in development
reflex fiinction rather than emerging from the supported the theory that visual and hand
modification of reflex function (2). Thus, control systems are unrelated at birth.
such reflexes as the grasp reflex may develop In the 1970s, a group of scientists (5,
separately from the eye-hand coordination 6) presented interesting evidence that they
system, and may underlie different fiinctions. believed supported the opposite concept: that
there was clear coordination of eye andhand
Reaching Behaviors: Innate or in thenewborn. They reported that infants
Learned? between 7 and 14 days of age showed arm
movements that were clearly directed toward
A second question that has intrigued re- the object in the visual field. They said a sig-
which the in-
searchers concerns the extent to nificant proportion of reaches were within 5°
tegration of sensory and motor systems un- to 10° of the object and that in 30 to 40% of
derlying eye-hand coordinations is genetically the reaches, the hand closed around the ob-
predetermined and/or experientially deter- ject. They also obser\'ed that infants diflFer-
mined. entiated between graspable (small object) and
If the integration of eye-hand coordi- nongraspable (large object at large distance)
nation were completely genetically predeter- objects: they reached for the first but not the
mined, it would imply that the nervous system second.
has a ready-made map of visual space and one Many had diflftculty
researchers initiaUy
of manipulative space out in a one-to-one
laid replicating these experiments, and thus the re-
correspondence. Thus, just by seeing an ob- sults were questioned (7). However, more re-
ject, an infant would know exacdy where to cent studies indicate that an early form of eye-
reach. If it were completely experientially de- hand coordination does exist in the neonate,
termined, experience would be required to although reaching doesn't seem to be as ac-
"map" visual space onto motor space. curate or coordinated as originally indicated
The first hypothesis implies that once (8,9).
the nervous system's sensory and motor path- In 1980, Amiel-Tison and Grenier, two
ways for visually guided reaching have ma- researchers from France, wrote a surprising
tured, the infant will be able to reach accu- article on neonatal abilities (10). They
rately for an object, with little or no prior showed that when the heads of neonates were
experience. The second hypothesis predicts stabflized, giving them postural support,
that a learning period is required in develop- amazing coordination of other behaviors was
ment, during which the infant creates, seen. For example, they showed that chaotic
through trial and error, the visual map that movements of the arms became still and the
overlays the motor map for reaching. infants appeared to be able to reach forward
In the 1950s, Piaget's research on child toward objects (Fig. 16.1). Their article is one
development led him to beUeve that though example of recent research that supports the
nervous system maturation is a requirement hypothesis that infants are born with certain
for the appearance of a behavior, experience innate abilities or behaviors, which have
is responsible for its coordination with the sometimes been termed pre-reaching behav-
senses. He believed that only through repeat- iors (10).
edly and simultaneously looking at and touch- In the late 1970s and 1980s, Claes von
ing an object would the visual and manipu- Hofsten, a psychologist from Sweden, began
lative impressions be associated (3). exploring the development of eye-hand co-
Other researchers gave fiarther support ordination in the neonate (11). He placed in-
to this concept when they noted that neonates fants in an infant seat and moved an object in
Chapter Sixteen Upper Extremity Manipulation Skills: Changes Across Life Span 379
nates also have proprioceptive control of hand

movements: they reach toward their mouth
without vision, in a goal-directed way. If they
miss at first, they move to the mouth using
proprioceptive feedback (11, 12).
Thus, this research suggests that some
aspects of reaching, in particular, the ability to
locate objects in space and transport the arm,
may be present in rudimentary form at birth,
while other components, such as grasp, de-
velop later in the first year of life. These find-
ings suggest support for the hypothesis that
at least some aspects of reaching are innate.
In the next sections we follow the pro-
gression of the development of reaching and
manipulation skills through the first few years
of life, exploring the emergence of various as-
pects of reaching and manipulation behaviors.

Figure 16.1. The release of reaching movements in a We have already seen that location of an ob-
neonate by stabilizing the head. (Adapted from Amiel- ject in space is possible in the neonate, and
Tison C, Crenier A. Evaluation neurologiquedu nuveau-
that the ability to transport the arm toward
ne et du nourrisson. Paris: Masson, 1980:95.)
the object is also available at birth. However,
as you will see, the grasp component of reach-
ing does not develop until 4 to 5 months of
front of them, as you see in Figure 16.2, and age, with pincer grasp developing at 9 to 13
carefully documented the number and accu- months. Higher cognitive aspects of reaching
racy of reaches that he observed. He observed begin to emerge at about 1 year of age.
the infants' arm movements with and without Throughout development, there appears to
an object present. He showed that the num- be a repetitive shift between visually triggered
ber of extended movements performed when (or proactively guided reaching) and visually
the infants were visually fixating on the object guided (or feedback-controlled) reaching.
was twice as high as when the object wasn't
fixated. The reaching movements weren't Early Development of
very accurate. However, those that were made
while the infants fixated on the target were Eye -Hand Coordination
aimed an average of 32° laterally and 25° ver-
tically toward the target, while those that were During the first year of life, there are a
made without fixation were only within 52° number of clear transitions in the infant's eye-
laterally and 37° vertically. Though these hand coordination abilities. The first transfor-
reaching movements weren't as accurate as mation in reaching skills appears to occur at
had previously been postulated, they were about 2 months of age (11, 12). Until this
clearly aimed at the target, since they were sig- time, whenever the infant extends the arm,
nificantly more accurate than the nonvisually the hand opens in extension at the same time,
fixated movements.
These results thus so that it is difficult to grasp an object. At 2
showed a clear effect of vision on forward di- months, the extension synergy is broken up,
rected movements (11). so that the fingers flex as the arm extends: the
Von Hofsten noted that the system probability of seeing this behavior goes from
works from hand to eye as well. Several times 1 to 70% of reaches, from shortly after birth
the infant accidentally touched the object and to 2 months. At 2 months, head-arm move-
immediately turned the eyes toward it. Neo- ments become coupled very strongly as the
380 Section I\' UPPER EXTREMm' CONTROL
Figure 16.2. A, Experimental set-up used to study reaching in neonates. The infant is placed in an infant seat (50°
angle! that has head support on the back and arms freedom to move. B, Diagram of the outline
sides, but allows the
of the infant as it touched the object, taken from single frames from the two video cameras seen in A. (Adapted from
Hofsten C von. Eye-hand coordination in the newborn. Developmental Psychology 1982;1 8:452.)
infant gains control over the neck muscles thus converge at about 4 months of age, all
(11,12). of which are essential for the emergence of
Over the next 2 months, there is an in- successtlil reaching. This supports the concept
creased uncoupling of head and arm move- that the emergence of successful reaching is
ments, which allows more flexibilit}- in eye- not due to the maturation of a single system,
hand coordination. At about 4 months, but to contributions of multiple maturing sys-
infants begin to gain trunk stabilit^', so they tems (11, 12).
have a more stable base for reaching mo\e- At about 4 months, infants enter a new
ments. developmental phase, imolving integration of
A number of developmental changes the newly developed skill of eye-hand coor-
Chapter Sixteen Upper ExrR£Mm NL\NiPi.T_\"noN Skills: Ch.axges Across Life Span 381
dination. Reaches of 4-momh-olds npicalh" infent's hand comes into viev^-, he/she is able
consist and the final approach
of several srej>s, to perceive the discrepancv- between hand fxi-
toward the object is crooked and a^xicward. In sirion and target and correct the tra-
jxjsition
the next 2 months, tlie approach path jecton.-. 5^ months, vi-
This suggests that by
straightens and the number of steps in the sually guided reaching is evident in most
reach are reduced in number, with the first infants. Msually guided reaching, or the abil-
part of the reach getting longer and more itv- to make corrections to a trajectorv- based
powerful. By 6 months of age, the trajector\ on visual inibnnation, p>eaks at around 7
of most reaches appears to be adult-like dl months, and then is graduaUy replaced by a
12 I. ballistic stvie of reach, though infants can still
use visual guidance when needed. In a ballistic

Visually Triggered vs. Visually stvie of reach, corrections are made at the end
Guided Reaching of the movement instead of during the on-

going movement. Once the movement is
Remember from the last chapter that completed, the error between hand jxjsition
reaching movements in adults have rô dif- and target f>osition is used to correct the p>o-
l^rent phases: the transpon phase and the sition of the hand in space.
grasp phase. It has been hypothesized that the
beginning of the reach is \isually triggered. Emergence of Hand Orientation
That is, visual location of the target is used to
initiate the movement. Thus, the position of ^"hen do infants first begin to orient
the object is defined \isually, while the posi- their hands to the position and shape of the
tion of the arm is defined propriocepriveh". In object: To answer this question, researchers
contrast, the last part of the reach is consid- placed brighdy colored rods either horizon-
ered visually guided. In this case, the position or vertically in fi-ont of the infant and re-
tally
of the arm is defined visually with reference to corded the characteristics of their reaching
the target, allowing precise adjustments to be movements, as you see in Figure 16.3. Pre-
made to ensure the accuracv" of the reach 13 ( ). paraton- of hand orientation
adjustments
Newixjms seem able to use the visually depending on object
(vertical vs. horizontal,
triggered mode reasonably w ell, since they are orientation occurred wlien infants first began
l
able to initiate a reach aimed tow ard the tar- to grasp objects, as early as 4^^ to 5 months
get ( 8 ). How ev er, do not appear to be
they of age 14 However, the adjustments of the
1 1.
proficient in the visually guided mode, since hand to the orientation of the object became
they are still verv" inaccurate in their reaches. more precise with age. Adjustments of the
Msually guided reaching requires the abilitv hand were often done before or during the
to attend to the hand as it mov es tow ard the early pan of the reach, though they could also
object, as well as the abilitv- to attend to the be seen during the approach phase.
object. It also requires the abilitv- to anticipate To reach smoothly for an object, the in-
possible errors. fant must time the grasp appropriately with
Research indicates that the visually relationship to encountering the object. If the
guided mode of reaching emerges between hand closes too late, the object will bounce
the 4th and the 5th month of life just as trunk off the palm of the inl^t, and if the hand
control and arm coordination are also im- closes too early, the object will hit the knuck-
proving (2, 12 1. les. This t\-pe of plaiuiing requires visual con-
In order to study the development of trol, since tactile control would not allow the
visuallv- guided reaching in infants, researchers hand to close until after touching the object
have fitted infants with special glasses with (14 ). In experiments in which the kinematics
prism lenses to give an apparent lateral shift of reaching of 5-, 6-, 9-, and 13-month-olds
in the target position as the infants reached were compared to those of adults, it was
for small tov-s (2 1. By SVz months, when the shown that infant grasping was visually con-
382 Section r\' UPPER EXTREMI'n' CONTROL
with research performed on the anatomic de-

velopment of the primate motor svstem. In
primates, it has been shown that neural
^ «. « i pathways controlling mo\ements of the
arm are from those that control
different
the fine movements of the fingers and hand.
) The two systems develop at different times.

1 Arm control, which appears to be mainlv
coordinated at the brainstem level, develops
earlier than hand and finger control, which
appears to be coordinated at the cortical level
\ (16,17).
fn Researchers found that infant monkeys

show arm mo\ements to\\ard objects earlv in
development, but do not show independent
finger and hand movements until they are 3
months old 12, 18). It has also been shown
(
that at about 9 to 13 months of age, with the

de\'elopment of the pyramidal tract, infants
are able to control fractionated finger move-

ments and thus develop more diffrcult grasp-
ing skills such as the pincer grasp (12). At
about 14 to 16 months of age, the infant de-
velops the abiliri' to adapt reaching to the
weight of objects, using shape and size as in-
dicators of weight (19).
Learning to Grasp Moving Objects

Figure 1 6.3. Infant reaching for a horizontally oriented
(Catching)
bar (Adapted from Hofsten C von, Fazel-Zandy S. De-
velopment of visually guided hand orientation in reach-
Studies have also been performed to de-
ing. J Exp Child Psychol 1984;38:210.)
termine the emergence of the abilit}' of infants
to reach for and grasp a moving object; this
could be considered a rudimentan' form of
trolJed as early as 5 to 6 moniJis of age, witJi catching behavior (20). Researchers have
the hand starting to close in anticipation of shown that by the time infants could reach
reaching the object. Also, the opening of the successfiilly for nonmoving objects, they were
hand was related to the size of the object for also successfiil at reaching for moving objects.
the 9- and 13-month-olds, but not in the Infants as young as 18 weeks could catch ob-
younger group. Finally, the 13-month-olds jects moving at 30 cm/sec. Fifteen-week-olds
initiated the grasp farther away from the tar- could intercept the object, but were not yet
get than the younger groups, with timing of able to grasp it. These results suggest that in-
the grasp similar to that seen in adults. The fants are able to predict where the object will
grasp component of the reach is still not ma- be at a future point in time because they must
ture in the 13-month-old, however, since, un- start reaching early to intercept it in its path.
like adults, they do not yet correlate the onset It was noted that the infants didn't automat-
of closing of the hand with the size of the ically reach toward ever\' object that passed
object to be grasped ( 15 ). by. Rather, they seemed to be able to detect
These developmental changes related to in ad\ance whether they had a reasonable
reaching and grasping skills correlate well chance to reach it (20).
Chapter SLxteen Upper EXTREMITY MANIPULATION Skills: Changes Across Life Span 383
Development of the Pincer Grasp as long in 10-month-olds, and two times as

long up to 3 years of age, com-
in children
There are t\\o different ways that ob- pared with adults. It was common in the
jects can be grasped. They can be grasped in younger children to have several touches by
a power grip, using the palm and palmar sur- the thumb and index finger before the object
face of the fingers, with the thumb reinforcing was properly gripped. Also, any finger could
this grip, or they can be grasped in a precision be the first to contact the object (22).
grip, between the terminal pads of the finger
and the thumb. The precision grip requires Emergence of Object Exploration
that the fingers be moved independendy, and
is a prerequisite for accurate and skilled move- When do infants first begin to change
ment of objects (21, 22). their manipulative activities in relation to the
In the first months after birth, infant characteristics of the objects grasped? During
grasping movements are controlled by tactile the first year, the actions infants perform with
and proprioceptive reflexes. Thus, when an objects tend to be mouthing, waving, shak-
object contacts the palm, the fingers close. ing, or banging. Rigid objects tend to be
Also, when the arm flexes, the hand closes, as banged, while spongy objects are squeezed or
part of a flexor synerg\'. At about 4 months of rubbed (23). In studies on 6-, 9-, and 12-
age, with the onset of fiinctional reaching, the month-olds, it was noted that mouthing ac-
palmer grasp is used exclusively by the infant. tivity decreased with age and that object ro-
With subsequent development, first the tation, transferring the object bet\\'een hands
thumb and then the fingers begin to operate and looking at and fingering the object, in-
independently, and at about 9 to 10 months creased (19, 24).
of age, pincer grasp develops (22). At about 1 year of age, infants begin to
Recent experiments have followed the acquire the understanding of how to use ob-
development and refinement of precision jects, but even before this age, they can dis-
grasp in human infants and children ranging cover simple tiinctional relationships if these
in age from 8 months to 15 years. Remember require litde precision. Thus, an infant first
from our last chapter that when an adult is uses a spoon for banging or shaking before
asked to lift an object, as soon as his/her fin- using it for eating. The infant establishes the
gers touch the object, cutaneous receptors ac- between spoon and hand, spoon
relationships
tivate a centrally programmed response that and mouth, and spoon and plate as subrou-
consists of an increase in grip forces and load tines before putting them together for the act
forces, designed to lift the object without let- of eating, in which the spoon is filled at the
ting it slip through the fingers. In adults, these plate, and transported to the mouth with an
two forces are always programmed in parallel, anticipatory mouth opening (25).
to prevent slips and to avoid squeezing the At about 16 to 19 months of age, in-
object too hard (22). fants begin to understand that certain objects
This parallel programming of grip and go together culturally, such as a cup in a sau-
load forces was not found in human infints. cer. Finally, at the end of the second year, they
In fact, until 5 years of age, the children begin to perform symbolic actions like pre-
pushed the object into the table as they intending to eat or drink ( 19).
creased the grip force, showing a reversed co- After 1 year of age, infants begin to de-
ordination between the two forces. In these velop skills requiring more precision of move-
children, the grip force had to be very high ment and between ob-
closer relationships
before the load force increase occurred. In ad- jects, such one object into another.
as fitting
dition, the timing and sequencing of the dif- At 1 3 to 15 months, infants begin piling two
ferent phases of lifting were much longer in cubes on top of each other; at 18 months,
the infants. For example, the time between three cubes; at 21 months, five cubes; and at
first and second finger contact was three times 23 to 24 months, six cubes. This shows that
the infant is gradually developing coordinated

reaching and manipulation, so that objects
can be placed and released carefliUy (19, 26).
Role of Experience in the

Development of Eye-
Hand Coordination
Remember that in humans, reaching be-
havior has two aspects, a visually triggered
portion and a visually guided portion. These
two aspects of eye-limb coordination are also
found in cats. Elegant studies on the devel-
opment of these two aspects of eye-limb co-
ordination have shown that movement-pro-
duced visual feedback experience is essential
for the visually guided portion to develop
(27).
In these experiments, kittens were raised
in the dark until 4 weeks of age and then al-
lowed to move freely for 6 hours each day in
a normal environment. But during this time,
they wore lightweight opaque collars that
kept them from seeing their Umbs and torso.
This is shown in Figure 16.4^. For the rest
of the day, they remained in the dark. After
12 days of this treatment, the animals were
tested for the presence of visually triggered vs.
guided placing reactions. This was ac-
visually Figure 16.4. A, Experimental collar worn by kittens to
block their view of their paws during early development.

complished by lowering the kitten toward a
B, Pronged apparatus for testing visually guided reach-
continuous surface (requires only visually
ing. (Adapted from Hein A, Held R. Dissociation of the
triggered placing, since accuracy is not re- visual placing response into elicited and guided com-
quired) vs. a discontinuous surface, made up ponents. Science 1967;1 58:391.)
of prongs (requires visually guided placing to
hit the prong). All animals showed a visually
triggered placing reaction, in which they au- sufficient, or must it be active (28)? To answer
tomatically extended the forelimb toward a this question, they tested 10 pairs of kittens.
continuous surface. But they showed no One of each pair was able to walk freely
kitten
greater than chance hits for a placing reaction in a circularroom, pulling a gondola, and the
to a pronged surface (Fig. 16.45). However, other kitten was placed in the gondola and
after removal of the collar, the animals only was passively pulled around the room. This is
required 18 hours in a normal environment shown in Figure 16.5. Thus, both kittens had
before showing visually guided placing. It was similar visual feedback and motion cues, but
thus concluded that visually triggered paw ex- for the kittens who walked, the cues were ac-
tension develops without sight, but visually tive and for the kittens who rode, they \\ere
guided paw placing requires prolonged view- passive.
ing of the limbs (27). The kittens had experience with the ap-
The researchers then asked: what kind paratus for 3 hours a day. At the end of the
of contact with the environment is important experiment, the active animals showed normal
for visually guided behavior.' Is passive contact visually guided placing reactions and re-
Chapter Sixteen UPPER Extremity Manipulation Skills: Changes Across Life Span 385
Figure 16.5. Experimental apparatus in which one cat actively pulls the second cat, which is passively pulled in the
gondola. (Adapted from Held R, Hein A. Movement-produced stimulation in the development of visually guided
behavior. Journal of Comparative and Physiological Psychology 1963;56:873.)
sponses to a visual clifFtest, in which a normal from Marseille, France, has begun to explore
animal does not walk out over an illusory cliff, these developmental changes in children from
but the passive animals dici not. Thus, the re- 6 to 1 1 years of age. Remember from our last
searchers concluded that self-produced move- chapter that, in adults, when a target is placed
ment is necessary for the development of vi- to the side, both eye and hand reaction time
sually guided behavior. However, once again, increase, compared to when it is at midline.
after 48 hours in a normal environment, the This is of reaction times for the eye
also true
passive group of animals showed normal vi- and hand for children 6 to 1 1 years of age
sually guided paw placement (28). (29). However, in children under 8 years of
age, when the head must also turn to look at
Eye-Head-Hand Coordination the target as the child reaches, the head move-
Development ment seems hand move-
to interfere with the
ment, and slows the reaction time, compared
to movements with the head held fixed. Head
In our last chapter, we mentioned that
movements also seemed to interfere with the
the eye, head, and hand are coordinatedwhen
abifity to intercept a moving target in children
adults reach, such that the eyes move first, fol-
of this age (29).
lowed by the head, and then the arm. How
does eye, head, and hand coordination de-
Reaction -Time Reaching Tasks
velop in children? Little research has been per-
formed in this area. However, research by A great deal of research has been per-
Laurette Hay, a developmental psychologist formed on developmental changes in reac-
386 Section IV UPPER EXTREMIT\' CONTROL
tion -time (RT) tasks. In general, it has been ID = los2{2A/W)

shown that for simple RT tasks, reaction times
become faster as children mature. The great-
where A = amplitude of the movement; = W
width of the target, and ID = index of diffi-
est changes occur about 8 to 9 years of
until
cult)' (29, 31).
age, with slower changes occurring subse-
Studies testing the extent to which Fitts'
quendy, until reaction times reach adult levels
law applies to children have found that move-
at 16 to 1 7 years. However, when children are
asked to perform more complex movements

ment time decreases with age. This decrease
is in general a linear change, except for a re-
as part of the RT task, these developmental
gression, which appears to occur at about 7
changes vary according to the task. For ex-
ample, in a study in which 2- to 8-year-old
years of age (29, 33). Remember that in the
development of postural control, there is also
children were asked to make target aiming
a regression, as indicated by an increase in
movements, a decrease in RT was obsen'ed
postural response latencies, between 4 and 6
from 2 to 5 years of age, followed by a sta-
years of age (34). A study examining 5- to 9-
bilization in RT (29, 30).
Movement year-olds has shown that these developmental
time in these reaction time
tasks also changes as a fianction of age. Re-
decreases and regressions in movement time
are not related to any changes in biomechan-
member from our last chapter that movement
ical factors, such as growth of the bones of the
time depends on the accuracy and distance re-
arm (35).
quirements of a task (31). Strategies for pro-
Using Fitts' law, one can plot move-
gramming movements also vary, depending
on whether the movement requires an accu-
ment time as a fiinction of index of difficulty
for different age groups. This relationship is
rate stop or not. If an accurate stop is re-
shown in Figure 16.6. The intercept of the
quired, the individual must use a braking ac-
line with the v axis reflects the general effi-
tion controlled by antagonist muscles.
ciency of the motor system, while the slope of
Alternatively, if the movement can be stopped
the line reflects the amount of information
automatically by hitting a target, antagonist
that can be processed per second by the motor
muscle activation isn't required. Studies ana-
system (29). Almost all studies have shown
lyzing movement time in children from 6 to
that the r-intercept decreases with age, indi-
10 years of age, for either type of movement,
However, age-re-
cating increased efficiency.
have shown a reduction in movement time
lated improvements depend
in slope appear to
with increased age. As might be expected,
on the task involved, and appear to be more
movements that require an accurate stop are
evident in discrete rather than serial move-
slower at all However, the difference be-
ages.
ments (36).
tween the speed of the two types of move-
ments is about three times higher at 6 years
Movement Accuracy
of age than at 8 to 10 years of age. It has been
hypothesized that this could be due to a dif- To determine developmental changes in
ficult)' experienced by the 6-year-olds in mod- children's use of visual feedback in making
ulating the braking action of the antagonist reaching movements, studies were performed
muscle system (29, 32). in which they were asked to make movements
with or without visual feedback. Laurette Hay
Fitts' Law has shown that there are interesting changes
Remember from our last chapter that in the use of visual information by children
Fitts' law shows a specific relationship be- between 4 and 11 years of age (37). Children
tween the time to make a movement and the between 4 and 6 years of age can make move-
amplitude and accuracy of that movement. ments without visual feedback with reason-
The difficult)' of the task is related both to the able accuracy, as you see in Figure 16.7.
accuracy and the amplitude requirements, and (Note that, although 5 -year-olds may appear
is represented by the following equation: to be more accurate than adults, there are no
Chapter Sixteen UrPER Extremity Manipulation Skills: Changes Across Life Span 387
Figure 16.6. Graph showing the relationship between movement time (y axis) and the index of difficulty (ID) of a
task, for four age groups of children. (Adapted from Hay L. Developmental changes in eye-hand coordination behaviors:
preprogramming versus feedback control. In: Bard C, Fleury M, Hay L, eds. Development of eye-hand coordination
across the lifespan. Columbia, SC: University of South Carolina Press, 1 990:227.)
significant differences between these groups.) phase remains constant in duration as children
However, at age 7, tliere is an abrupt reduc- develop. However, the accuracy or closeness
tion in this ability, as seen in the increased of approach to the target at the end of this
errors made in reaching without visual feed- phase increases (29, 38). This increase in ac-
back. The accuracy then begins to increase curacy results in a reduction in the number of
again, reaching adult levels by 10 to 11 years corrections required in the homing-in phase.
of age. As we describe in the next section, this However, on discrete reaching tasks, this
reduction in accuracy is reflected in an in- developmental change starts only at the age
creased dependence on visual feedback at the of 7 to 8 years, while the opposite develop-
age of 7 years. mental trend occurs between 6 and 7 to 8
years (39). This is thus one more piece of re-
Kinematics of Reaching search to support the hypothesis that the age
Movements of 7 is a transition time in the development of
reaching (29).
In our last chapter, we described studies Other studies analyzing the kinematics
on the kinematics of reaching movements in of reaching movements without visual feed-
adults, and showed that reaches consist of an back in children ages 5 to 1 1 also support this
initial ballistic, distance-covering phase, fol- hypothesis (40). Figure 16.8 shows that 5-
lowed by a homing-in phase, which uses visual year-olds produce mainly ballistic move-
feedback. In studies of children from 5 to 9 ments, with sharp decelerations at the end of
years of age performing reciprocal tappinjj the movement {black bars), while this pattern
tasks, it was determined that the ballistic shows a sharp decrease at age 7. At this age.
Age
(yrs)
Figure 16.7. Graph showing pointing errors when visual feedback was not present for children from 4 to 1 1 years of
age, compared to adults. (Adapted from Hay L. Developmental changes in eye-hand coordination behaviors: prepro-
gramming versus feedback control. In: Bard C, Fleury M, Hay L, eds. Development of eye-hand coordination across
the lifespan. Columbia, SC: University of South Carolina Press, 1990:228.)
a ramp and step pattern increases {stippled kinematics of the hand movement show a
bars). At the same time, ballistic patterns with curved, rather than a straight line trajectory
asmooth deceleration at the end of the move- toward the object. This occurs as the hand
ment increase and continue to increase shifts fi'om an initially incorrect path, due to
through 9 years of age (striped bars). It has the shift in the visual image caused by the pris-
been hypothesized that this could be due to matic lenses, to a correct path when the hand
the increased use of proprioceptive feedback comes into view, based on visual information
control in 7-year-olds, and the progressive re- of the relative hand and target positions. The
striction of feedback control to the final hom- length of the \'isually corrected path indicates
ing-in phase in older children, possibly the re- the amount of \isual feedback used in the
sultof increased efficiency of the movement movement.
braking system (29). As evident in Figure 16.9^ and B, 5-
For a closer look at de\'elopmental year-old children corrected the movement
changes in the use of visual feedback in reach- late in its trajectory, and in fact, the majority
ing movements in children, experiments were of these children did not make a correction
performed in which children ages 5 to 11 until they reached the virtual target, indicat-
were asked to make reaches while wearing ing minimal use of visual feedback. Thus, in
prismatic lenses, which make an illusory shift this age group, visual control occurs mainly
in the image of the object (40). These exper- after, rather than during, reaching move-
iments are similar to those described earlier, ments. This is correlated with highly stereo-
examining the use of visual feedback in reach- typed movement times seen in this aê group
ing in neonates and infants. As you see in Fig- (29).
ure 16.9^, as the children make a reach, the The 7-vear-old children corrected the
Chapter Sixteen UPPER EXTREMITY' Manipulation Skills: Changes Across Life Span 389
s^lL AÛ
^\\~ v^ wvw
9 11 AGE
Figure 16.8. Percentage of time three different reaching movement patterns were seen in children from 5 to 1 1 years
of age. 1 = ballistic pattern with sharp accelerations/decelerations; 2 = ballistic patterns with smooth decelerations;
3 = step and ramp patterns. (Adapted from Hay L. Developmental changes in eye-hand coordination behaviors:
preprogramming versus feedback control. In: Bard C, Fleury M, Hay L, eds. Development of eye-hand coordination
across the lifespan. Columbia, SC: University of South Carolina Press, 1990:231.)
movements earlier than any other group, in- fast, accurate movements by 9 years of age
dicating a strong use of visual feedback. While (29).
this gives rise to an increased flexibilit)' in
reaching behavior, it is coupled v\'ith increased Changes in Older Adults
movement times, and decreased
variability' in
accuracy when visual feedback is not present. As we have noted in our previous chap-
The 9- and 11 -year-olds showed an in- ters on age-related changes in postural control
termediate level of trajectory correction, in- and mobilit}' skills, there are specific changes
dicating a shift in the use of visual control to- in these skills with age. These can be divided
ward the
final phase of the movement into (a) time-related changes, such as slowing
Thus, between 5 and 9 years of age
trajector}'. of onset latencies for postural response or de-
there appears to be a reorganization in the creased movement speed in locomotion; (b)
programming of reaching movements from coordination factors, related to changes in
mainly feed-forward or anticipator^' activation movement or muscle activation patterns; and
of reaching, to predominant feedback con- (
c) changes of feedback and feed-
in the use
trol, and finally to an integration of the feedforward control of both postural and mobility
forward and feedback control, resulting in skills.
REAL POSITION
OF THE TARGET
13
12 .
10
8 .
0)0}
6 .
;9Q
5 7 9 ,j
age i f»s,
Figure 16.9. A, Diagram of reaching movements of children who wore prismatic lenses, displacing the apparent
position of the target in the visual field. B, Corrected portion of the reaching trajectory for 5-, 7-, 9-, and 1 1 -year-olds.
(Adapted from Hay L. Spatial-temporal analysis of movements in children: motor programs versus feedback in the
development of reaching. Journal of Motor Behavior 1979;! 1:1 96, 198.)
Chapter Sixteen Upper Extremity Manipulation Skills: Changes Across Life Span 391
Unfortunately, the literature on changes creased, the subjects could follow it less easUy,
in eye-hand coordination characteristics in the until at some point it was impossible to fol-
older adult is compared to the

very limited low.
extensive research on the development of eye- However, there was a difference be-
hand coordination in children. Research on tween the older and younger subjects. As you
age-related changes in eye-hand coordination see in Figure 16.10, the older adults dropped
has focused more on changes in reaction time off in their ability to follow the movements
and movement time in reaching tasks of var- sooner than the young adults. Welford hy-
ious levels of complexity, rather than on pothesized that the limitation in the perfor-
changes in the kinetics of reaching and grasp- mance of the older adults wasn't due to prob-
ing movements. In the sections that follow, lems with the motor system because they
we review the literature in these research ar- could move faster if they were not following
the target. He hypothesized that the limita-
tion wasn't sensory because the older adults
Changes in Reaching Movement could easily see the target. Therefore, he con-
Time with Age cluded that the limitation was in central pro-
cessing abilities, that is, in the older adults'
A review of studies examining changes ability to match the target and pointer and
in the speed of reaching movements with age react quickly to changes in target direction.
has shown that discrete reaching movements This implies that the time spent in actual
show a range of 30 to 90% reduction in ve- movement itself slows little compared to the
locity with aging, depending on the ages time taken to make decisions about the next
compared and the task performed. For ex- part of the movement sequence (43).
ample, one study examining changes in the It has also beenshown that hand stead-
speed of discrete arm movements showed a iness decreases with age during reaching tasks
32% reduction between the ages of 50 and 90 (42). When older adults were asked to insert
years, while another showed a reduction in a small stylus in slots of different diameters
movement speed of 90% when comparing (V2- to Vs-inch), steadiness dropped by 77%
subjects from 20 to 69 years performing a re- from the 50s to the 90s. Steadiness deterio-
petitive tapping task (41, 42). rated faster in the nonpreferred hand than in
What are some of the age-related the preferred hand.
changes in different systems of the body that Based on the literattire, there appears to
might contribute to this slowing in reaching be little change
performance speed for
in
movements.' Different systems that could reaching movements with age, if subjects are
contribute to the slowing include (a) sensoiy asked to repeat the same simple action, like
and perceptual systems, such as the visual sys- tapping a pencil between two targets, or per-
tem's ability to detect the target, {b) central forming a simple reaction time (SRT) task
processing systems, (c) motor systems, and (41, 43). In this case, the slowing may be as
(d) arousal and motivational systems (41). little as 16%. But if the complexity of the task
Welford, an English psychologist, per- is increased, by making the target smaller, us-
formed an experiment to determine if changes ing successive targets, or using a choice reac-
in central mechanisms contribute to the slow- tion time (CRT) task, then slowing can range
ing in reaching speed in older adults. In these from 86 to 276%. Table 16.1 gives examples
experiments, subjects were asked to keep a of these differences in slowing of the perfor-
pointer (which they could move with a han- mance of reaching movements with complex-
dle) in line with a target that continuously ity of the task. The largest slowing in perfor-
moved from side to side, in an irregular si- mance was in tasks involving symbolic
nusoidal fashion, with the movement varying translations (using a code to relate a stimulus
in both speed and extent. He found that as to a response) or spatial transpositions (for ex-
the speed of the target movement was in- ample, a light cue on the left requiring a reach
392 Section I\' UPPER EXTREMm^ CONTROL
150 --
^ Subjects over 30
X Subjects under 30
• Target pointer
19
Speed in seconds per revolution of the cam
Figure 16.10. Graph showing the ability' of young vs. older adults to follow unpredictable target movements of
different speeds. (Adapted from Welford AT. Motor skills and aging. In: Mortimer )A, Pirozzolo FJ, Maletta CJ, eds.
The aging motor system. NY: Praeger, 1982:159.)
Table 1 6.1 Age-Related Slowing

. in the Performance of Reaching Movements as a Function of Task
Complexity'
Task Age-Groups Compared Percentage Increase"
Simple key press or release sound

to light or
Average of 1 1 studies listed by Welford twenties with sixties
Ten-choice (Birren, Riegel, and Morrison, 1962) 18-33 with 65-72
Straightforward relationship 27
With numerical code, mean of 5 studies 50
With verbal code, mean of two studies 45
With color code 94
With part color and part letter code 86
Ten-choice (Kay 1954, 1955) 25-34 with 65-72
a. Signal lights immediately above response keys -1 3 (no errors made)
b. Signal lights 3 ft from keys 26 (-43)
c. As b, but signal lights arranged so that leftmost responded 46 (-19)
to with rightmost key, and so forth
d. With numerical code 56(-t-138)
e. The difficulties of d and b combined 299 (+464)
"Adapted from Welford AT. Motor skills and aging. In: Mortimer JA, Pirozzolo FJ, Maletta CJ, eds. The aging motor system. NY: Praeger,
1962:163.
'Percentage changes of errors are shown in parentheses
Chapter Sixteen Upper Extremity' ^L\NIPULATION Skills: Changes Across Life Span 393
to the right). Though decrements have been pressing tasks, one that had few subunits (12,
found in performance on many RT tasks, a 12, 12, etc.) and one that was more complex
recent study has also show'n that when older (1234, 32, 1234, etc). They found that the
adults are not instructed to worn- about ac- older adults were slower than the young
curacy on such a task, they demonstrate no adults, particularlv with the second series
decrease in reaching speed (44). (46).
The primary source of the slowing in
complex RT tasks is in the first phase of per- Compensation and Reversibility of
formance, the time to obserxe the signals and Decrements in Reaching
relate them to action, rather than in the sec- Performance
ond phase, the time to execute the movement
(41, 43 WTien performing more continuous
). Although decrements in reaching per-
tasks, the second phase, that of movement ex- formance may be found in older adults in ex-
ecution, can overlap to some extent with the perimental conditions, they are often not ob-
first. For example, a person may process the sened in the workplace or in Acd\ities of
information relating to the next signal while Daily Living (41). It has been suggested that
making the first response. This t\'pe of task performance is preserx-ed because many com-
appears to be more difficult for older adults pensator)' strategies are used to improve reach
possibly because they need more time to and grasp skills. Many of these compensatory
monitor their responses, and thus have diffi- strategies used by older adults appear to be
culr\' processing other signals simultaneouslv unconscious, automatic processes (41). For
(41). example, older adults may increase the effort
For example, older adults (63 to 76 the\- put into the mo\ement. In the work-
years old) were compared to \'ounger adults place, they ma\' work more continuously with
(19 to 29 years) on a task where they moved fewer brief pauses. They may also prepare for
as quickly as possible toone of t%vo alternate movements that require speed and accuracy'
endpoints, with one farther away than the in advance, thus allowing anticipator)' pro-
other, in the same direction (45). The cesses to aid in pertbrmance. In many tasks,
younger subjects could overlap the time re- they may also make a trade-off" betxveen speed
quired to choose the endpoint with the initial and accuracy. Finallv, it has also been shown
stages of the movement itself, while the older that older adults set higher criteria for re-
subjects were less able to do this. Although sponding to RT signals in sensor\' discrimi-
there is no evidence that the time taken for nation tasks (41 ).
monitoring increases with age, older adults Can the changes in reaching skills that
seem less able to suppress monitoring (41). occur with aging be counteracted bv practice
What might be reasons for this lack of or training? Yes! Clear impro\'ement has been
suppression? It has been hypothesized that reported for eye-hand coordination skills in
suppression of monitoring occurs when the older adults with practice (42, 47). Greatest
outcome of a task is certain; thus, if there is a improvement is seen in more complex tasks.
possibilit\' of error, monitoring will be more Interestingly, older adults show more im-
probable. In addition, suppression of moni- provement with practice than do younger
toring may be possible when movement sub- adults on performance of RT tasks (48). This
units are coordinated into higher units of per- may occur because young adults are closer to
formance (41). However, to do this often their ceiling of performance when starting to
requires that the subject hold the movement learn the task. However, practice does not
subunits together in working memon' while eliminate the age differences in the perfor-
performing the task. mance of these tasks.
A study tested this ability- in older (ages Practice also improves performance in
60 to 81) vs. young (ages 17 to 28) adults. older adults related to the perceptual pro-
Subjects were asked to perform tw-o serial key cesses involved in eve-hand coordination
394 Section IV UPPER EXTREMITi^ CONTROL
tasks, such as \isual acuit\', signal detection, strations. However, in this case, the pace of
and auditor}' discrimination (42). the demonstration should be under the lear-
In addition, the effects of practice re- ner's control. Thus, using slow-morion, self-
main high, even 1 month after practice on paced \ideos in training may help (41).
eye-hand coordination tasks has ended. One Acti\e decision making is also an im-
study compared the performance of yoimg portant factor in learning at any age. In a
(ages 19 to 27) and older (ages 62 to 73) maze study with adults, it was shown that
adults on a task that invoked fine movements learning took place much faster if the correct
of the hands, signal detection, memor\- scan- pathway was marked, but the subject had to
ning, \isual discrimination, and anticipation make an active choice (50). This helped sub-
riming, called "Space Trek." Subjects were jects of all ages, but it especially helped older
given 51 1-hour practice sessions over a pe- adults (41, 50).
riod of 2 to 5 months. One month after train- It was also shown that using a mixture
ing ended, there was only a small decrease in of mental practice and physical practice when
performance levels (42). learning a pursuit rotor task was as good as
In another study, older adults (ages 57 physical practice alone for 65- to 69- and 80-
to 83) were given practice in eye -hand coor- to 100-year-olds (51).
dination skills by pla\ing \ideo games, such as Thus, learning of eye-hand coordina-
PacMan These games involved making
(49). tion tasks by older adults can be facilitated by
fast decisions about changes in the speed and using a t\'pe of discover)' learning, which in-
direction of hand movements. Over a 7-week volves demonstrations that can be self-paced,
period, scores tripled on the task. In addition, and a combination of physical
active learning,
practice on the \ideo games transferred over and mental practice (41).
to other RT tasks that required subjects to
quickly select a motor response.
Summary
These studies suggest that older adults
learn as much, if not more, with practice than 1. Infants as young as a week old show pre-
young adults, and that they retain the learned reaching behaviors, where they reach to-
skills as well as young adults. In addition, the ward objects that are in front of them. These
way subjects improved with practice was sim- reaches are not accurate, and the infants do
not grasp the object, since an extension syn-
ilar for the \'oung and older adults; however,
ergy controls the arm/hand movements.
the older adults simph- learned more slowly.
When the arm is extended, the hand is of>en.
This slower rate of learning of eye-hand co-
But the reaches are clearly aimed at the ob-
ordination skills in older subjects may be due more accu-
since they are significantly
ject,
to material taking longer to register in long-
rate than arm movements where the eyes
term memor\- (41 ). aren't fixated on the object.
What does this mean in terms of deter- 2. At about 2 months, the extension svnergy is
mining the best strategies for teaching eye- broken up, so that the fingers flex as the ami
hand coordination skills to older adults.- Since extends. At this time, head-arm movements
the time required for registering information become coupled as the infant gains control
in long-term memon' lengthens with age, over the neck muscles.
learning needs to be unhurried. Otherwise ex- 3. At about 4 months, infants begin to gain
trunk stability, along with a progressive un-
tra information to be processed during the
coupling of head-arm-hand synergies. These
time required to register information in
changes allow the emergence of functional
longer-term memor>- will simply disrupt the
reach and grasp behavior.
memor\- process (41 ).
From 4 months onward, reaching becomes
3.
In teaching eye-hand coordination more refined, with the approach path
skills, there are sometimes problems in trans- straightening and the number of segments of
lating verbal instruction into motor perfor- the reach being reduced.
mance. To avoid this, one can use demon- 6. Visually triggered reaching is dominant in
Chapter Sixteen UPPER Extremht Manipulation Skills: Changes Across Life Span 395
the newborn, changing to visually guided 5. Bower TGR, Broughton JM, Moore MK.
reaching at about 5 months of age, and re- The coordination of visual and tactual input
turning to visually triggered reaching by 1 in infants. Perception Psychophysics 1970;
year of age, though guided reaching is still 8:51-53.
available. 6. Bovver TGR, Broughton JM, Moore MK.
7. The development of hand orientation begins Demonstration of intention in the reaching
to occur at the onset of successful reaching, behavior of neonate humans. Nature. 1970;
at about 5 months of age. 228:679-681.
8. The pincer grasp develops at about 9 to 10 7. Dodwell PC, Muir D, Difranco D. Responses
months of age, along with the development of infants to visual presented objects. Science
of the pyramidal tract. 1976;194:209-211.
9. Reaction time shows a progressive reduction 8. Hofsten C von. Eye-hand coordination in the
with age, with sharper changes occurring newborn. Developmental Psychology' 1982;
until 8 to 9 years, followed by slower 18:450-461.
changes until 1 6 to 17 years. 9. Vinter A. Manual imitations and reaching be-
10. Children from 4 to 6 years make predomi- haviors: an illustration of action control in in-
nantly visually triggered (feed-forward) fancy. In: Bard C, Fleury M, Hay L, eds. De-
movements, using little visual feedback. At 7 velopment of eye-hand coordination across
to 8 years, visual feedback is dominant, lead- the lifespan. Columbia, SC: University of
ing to poor reaching in the dark, but more South Carolina Press 1990:157-187.
accurate reaching with vision present. By 9 10. Amiel-Tison C, Grenier A. Evaluation neu-
to 11 years, there is an integration between rologique du nouveau-ne et du nourrison.
feed-forward and feedback movements. Paris: Masson, 1980.
11. Older adults show a slowing in reaching 1 1 Hofsten C von. Studjang the development of
movements, with much of this due to central goal-directed behavior. In: Kalverboer AF,
processing slowing. The slowing in perfor- Hopkins B, Geuze R, eds. Motor develop-
mance on reaching movements is greater for ment in early and later childhood: longitudi-
more complex tasks. nal approaches. Cambridge: Cambridge Uni-
12. Part of the slowing may result from an in- versity Press 1993:109-124.
ability to suppress monitoring of movements, 12. Hofsten C von. Developmental changes in
due to either uncertainty concerning the ac- the organization of pre-reaching movements.
curacy of the movement, or an inability to
Developmental Psychology 1984;3;378-
integrate movement subunits into larger
388.
chunks in working memory. 13. Paillard The contribution of peripheral and
J.
13. Most age-related decrements in reaching central vision to visually guided reaching. In:
performance can be improved with training. Ingle DJ, Goodale MA, Mansfield RfW, eds.
Training effects remain high for at least a
Analysis of visual behavior. Cambridge, MA:
month after training has ended and also MIT Press, 1982:367-385.
transfer to other reaching tasks.
14. Hofsten C von, Fazel-Zandy S. Development
of visually guided hand orientation in reach-
ing. J Exp Child Psychol 1984;38:208-219.
References
15. Hofsten C von, Ronnqvist L. Preparation for
1. Twitchell T. Reflex mechanisms and the de- grasping an object: a developmental study. J
velopment of prehension. In: Connolly K, ed. Exp Psychol 1988;14:610-621.

Mechanisms of motor skill development. NY: 16. Kuypers HGJM. Corticospinal connections:
Academic Press, 1970. postnatal development in the rhesus monkey.
2. McDonnell PM. Patterns of eye -hand coor- Science 1962;138:678-680.
dination in the first year of life. Can J Psychol 1 7. Kuypers HGJM. The descending pathways to
1979;33:253-267. the spinal cord, their anatomy and flinction.
3. Piajet J. The origins of intelligence in chil- In: Eccles JC, ed. Organization of the spinal
dren. NY: WWNorton, 1954. cord. Amsterdam: Elsevier, 1964.
4. White BL, Casde P, Held R. Observations on 18. Lawrence DG, Hopkins DA. Developmental
the development of visually-directed reach- aspects of pyramidal motor control in the rhe-
ing. Child Dev 1964;35:349-364. sus monkey. Brain Res 1972;40:1 17-1 18.
Sy6 Section 1\ LPPHREXTREMIITCONIROL
19. Corberta O, Mounoiid P. Early dc\clopnicnt tude of movement. I E.xp Psvchol 1954;4~;
ot" grasping and manipulation. In; Ban! t\ 381-391.
Fleun' M, Hay L, cds. Development oteye- 32. Hay L, B.u-d C, Fleurv' M. X'isuo-manual co-
hand coordination across the lifespan. Co- ordination fi-om 6 to 10: specification, con-
lumbia, SC: L'nivcrsirv- of South Carolina troland evaluation of direction and ampli-
Tress 1W0:189-213. tude p.irameters of movement. In: Wade MG,
20. Hofsten C von, 1 indhagen K.. t^bser\atii>ns Whiting HTA, eds. Motor development in
on the dexelopment ot' moving
rc.iching for children: aspects of coordination and control.
objects. I K\p Child IVchol W~»^;28:l,=;8- Dordrecht: Martinus Nijhotl", 1986.
173. 33. Rev A. Le freinage volontaire du mouvement
21. Napier JR. Ihe prehensile movements of the graphique chez Tenfant. In: Epreuves d'in-
hum.m hand. 1 Bone loint Surg 1^56; telligence pratique et de psychomotricite.
3SB:902-913. Neuchatel: Delach.uix &; Niestle. 1968.
Forssberg H, F.liasson AC, Kinoshita H, lo 34. Shumvvay-Cix>k A, \Voollacott M. The
hanssi>n RS, Westling G. Development of hu- grow th of stabilitv-: postural control tixjm a
man precision grip I: basic ciKirdination of developmental perspective. Journal of Motor
force. In press. Beh,\vior 1985;I~:I3I-147.
25. Gibson H, Walker AS. Development o( 35. Kerr R. Movement control and maturation in
knowledge of visu.il-tactu,il atlbrdance of elementarv grade children. Percept Mot
substance. Child Dev 1984;55:453-160. Skills I9-5;41:151-I54.
Rurt" HA. Infants' manipul.uive exploration 3ti. Sugden DA. Movement speed in children,
of objects: of age and object charac-
eftccts loumal of Motor Behavior 19S0;12:125-
teristics. Dev Psychol r984;20:9-20. 132.
Connolly KJ. The development of compe- 37. Hay L. Accuracy of children on an open-loop
tence in motor skills. In: Nadeau CH, H.illi- pointing task. Percept Mot Skills 1978;
well \VR, Newell KM, Roberts GC, eds. Psy- 4~:10~9-I082.
chology'of motor behavior and sp<.>rt. 38. Schellekens IMH, KolverKx-r .\F, Scholten
Champaign, IL; Hum.in Kinetics, 1979:229- CA. The microstnicturc of tapping move-
250. ments in children. Ioum.iI of Motor Behavior
26. Baylcy N. Manual for the Bayley scales of in- I984;I6:21V39.
fant development. NY: Psychological Cor- 59. Dellen T \",m, Kalverboer .\F. Single move-
poration. 19t>9, ment contail .uid infomiation paxressing, a
Hein .\, Held R. Dissociation of the visual development.il study. Behav Brain Res 1984;
placing response into elicited and guided I2:23~-23S.
compiinents. Science 1967;15S:390-392. 40. Hay L. Spatial-temporal anah-sis of move-
Held R, Hein A. Movement- paxluced stim- ments in children: motor programs versus
ulation in the development of visually guided feedback in the development of reaching.
behavior. lounial of Comparative and Physi- Journal of Motor Behavior 19--9;1 1:189-
ological Psychology 1963;56:872-876. 200.
Hav L. Development.il changes in eye-hand 41. Weltord AT, Motor skills ,ind aging. Mor- In:
co«.irdination behaviors: Preprogroinming timer I.\, PiR-izzolo FT, Maletta GI, eds.The
versus feedKick contml. In; Barvi C, Fleurv- aging motor svsteni. NT; Praegcr, 1982:152-
M, H,iy L, eds.Development of eve-hand 18".'
coordination .icross the lifespan. Columbia, 42. Williams H. Aging and eye-hand coordina-
SC: Universin- of South Carolina Press tion. In: Bard C, Fleurv- M, Hay L, eds. De-
1990:217-244. velopment of eye-hand coordination across
30. Brown ]\, Sepehr MM, Ettlinger G, Skrc- the lifespan. Columbia, SC: University of
czekW. The accuracy of aimed movements South C,u\>lina Press 1990:32:^-357.
to visual targets during development: the a>le 43 \\'eltbal AT. Motor Performance. In: Birren
of visual information. 1 Exp Child Psvchol G, Schaie K, eds. HandKx->k of the psychol-
1986;41:443-160. ogv of aging. NY: \"an Ncxstrand Reinhold,
31. FinsPM. The infomiation capacity of the hu- 19— :3-20.'
man motor system in controlling the ampli- 44. GL Willianison, CI Leipcr, NH Mayer. Bea-
(."Iiaptcf Sixteen L'rri R l-\rRiMirv M.\Niiaii.MioN Sku is: C'nANia-s AtRoss l.itt; SiWN 397
\cr Côllciic Assessment ot'spccii and .iei.'urae\' uli of inca-asing complexity as a function of
Dtiuovenient in i>Kler adults using I'itts' tap ageing. Br ] Psychol iy77;6S:18y 201.
pinj; test, Neuix>s>.i Ahsti bJ'J3;iy:55t». 4^J. Clark I, 1 anphear A, Riddick C~. Ihe etVccts
-IS Rahhit r, Roi;ei's M. Age Mui choice between of \ideogan>e playing on the response selec-
I'cspiinses in a self paced repetitive task. Vv tion processing of elderlv adults. derontol 1
gononiics 1^'65;S:435—444. l^>8^;42:82-85.

46. Rabbit T, Binen ]\- . Age and respon.ses to .>ic 50. Wright jM von. A note on the role of "giud
qucnccs of repetitive and intcrrupti\e signals. ancc" in leainmg Br I Ps\clu>l b>5";48 133
1GenMituI iyft7;22: 143-150. 137,
47. Halduto 1 , Baron A. Age related etVects ot 51. Surberg PR. Aging and etVect of physical
practice and task complexity <.>n card sorting. mental practice upon acquisition and reten-
1 C'.eriMitol 19S6;4l:6S« 661. tion of a motor skill ».;eronti>i l'>76;31:64-
1
48 liMdan 1', Rabbitt P. Resp<.>nsc times to stun

Chapter 17
Abnormal Upper Extremity

Manipulation Control
Introduction Impairments Affecting Grasp and Release

Target Location Problems: Eye-Head Motor Problems
Coordination Problems Affecting Activation and
Visual Deficits Coordination of Agonist Muscles
Transport Problems Inappropriate Activation of Antagonist
Motor Dyscoordination Muscles
Musculoskeletal Constraints Sensory Problems
Pain and Edema Postural Problems
Weakness Problems with Adaptation
Spasticity Visual Deficits
Mass Patterns of Movement Somatosensory Deficits
Dyscoordination Anticipatory Aspects
Motor Impairments Affecting the Apraxia
Nonhemiparetic Limb Summary
Sensory Impairments
Optic Ataxia
Somatosensory Loss
tremit)' dysfunction requires an understand-

Introduction ing of the problems associated with specific
Normal upper extremity' function is the types of neurological impairment, and the way
basis for fine-motor manipulation skills im- inwhich these problems affect the key com-
portant to activities such as feeding, dressing, ponents of upper extremity fimction.
grooming, and handwriting. In addition, up- In previous chapters we suggested the
per extremity' function plays an important role key components of upper extremity function
in gross motor skills such as crawling, walking, included (a) locating a target, requiring the
the ability to recover balance, and to protect coordination of eye-head movements, (b)
the body from injury when recovery is not reaching, involving transportation of the arm
possible. and hand in space, {c) manipulation, includ-
Because upper extremit^' control is ining grip formation, grasp, and release, and (d)
tertwined with both fine and gross motor postural control. In addition, goal-directed
skills, recovery of upper extremity fiinction is upper extremity movements also involve
an important aspect of retraining the patient higher cognitive processes necessary for gen-
with impaired motor control and falls within erating and planning an intent to act, and
the purview of most areas of rehabilitation, maintaining that intent long enough to carry
including both occupational and physical out the action plan.
therapy. As we mentioned in the last two chap-
Abnormalities of upper extremity fimc- ters, tlie research on upper extremity dyscon-
tion in the patient with a neurological dis- trol contradicts traditional assumptions that
ability can result from a wide variety of sen- upper extremity function is due to a single
sory, motor, and cognitive impairments. controlling system that develops in a proximal
Thus, assessment and treatment of upper ex- to distal manner. This has a number of im-
399
400 Section W UPPER EXTREMITY CONTROL
portant implications for clinicians involved in vestibulo-ocular reflex control of eye move-
retraining the patient with upper extremity' ments in response to head movements; and
dysfiinction. It suggests that neural pathology' (c) inability' to adapt the vestibulo-ocular re-
can affect some aspects of upper extremity' flex to changes in task demands due to cere-
fiinction, while leaving others unaffected. bellar damage (2). All of these t\'pes of prob-
In addition, it suggests that therapy di- lems affect the patient's abilit)' to stabilize
rected at recover}' of hand hinction should oc- gaze on an object when moving the head.
cur simultaneously with retraining of shoulder However, in this chapter, we focus primarily
or more proximal components of the move- on problems related to visually driven eye
ment, rather than waiting to work on hand movements.
control until proximal control has been de-
veloped, which has been a traditional rehabil- Visual Deficits
itation approach.
This chapter focuses on abnormal upper Central lesions affecting the processing
extremity' function as it relates to manipula- of visual signals may also disrupt upper ex-
tor)' skills. We review problems related to the tremity' motor control in the patient with a
key components of upper extremitv' control, neurological deficit. Visual field deficits fol-
incorporating a discussion of senson,', motor, lowing a stroke, such as homonymous hemi-
and higher level problems that affect each as- anopia, restrict a patient's abilit\' to see objects
pect of upper extremity' control. We begin in one-half of the visual field ( 1 ).
with a review of problems related to locating Until recentiy, pathology causing le-
a target, requiring the coordination of eye- sions in the visual cortex of humans was
head movements. thought to cause total blindness, except for a
rudimentary abilit}' to detect changes in visual
Target Location illumination. However, recentiy it has been
Problems: Eye-Head shown that monkeys with lesions in the striate
cortex are able to reach toward objects
Coordination moving
still
across their visual field (3). It has

A critical aspect of manipulaton,' dys- been hypothesized that these reaching move-
function is the inabilin,' to locate a target and ments may be due to visual processing occur-
maintain one's gaze on that target preceding ring in subcortical structures, such as the su-
the reach. Remember from the last two chap- perior colliculus.
some target location tasks require eye
ters that Research with humans has now con-
movements alone, while others require a firmed these findings. When patients were not
combination of eye-head movement, and still asked whether they could "see" the object,
other tasks require a combination of eye- but simply to move their eyes toward where
head-trunk movements, depending on the ec- they thought the object might be, the direc-
centricit}' of the target in space. This has led tion and amplitude of their eve movements
researchers to suggest that eye-head coordi- were significantiy correlated with the position
nation is not controlled by a single mecha- of the targets (4).
nism, but rather emerges from an interaction More recent experiments have been per-
of several different neural mechanisms (1). formed on subjects with hemianopia due to a
What Di'pes of problems afreet the ability' hemispherectomy on one side (5). Patients
to stabilize gaze during dift'erent head move- with hemianopia were asked to point to the
ments toward and thus potentially
a target, target when it appeared in either their normal
affect the accuracy and precision of reaching visual field or their affected field. When it was
movements? Problems include (a) disruption in the affected field, they were asked to
of \'isually driven eye movements due to dam- "guess" where it was, and to point there,
age within the oculomotor system; (b) dam- since patients said that they couldn't see it.
age to the vestibular system, which disrupts Again, pointing positions in the hemianopic
Chapter Seventeen Abnormal Upper Extremity Manipulation Control 401
field were definitely correlated with the target impairments due to weakness, spasticity, or
positions. the presence of mass patterns. Musculoskele-
Although subjects were initially very tal impairments limit the ability to move the
poor at reaching for objects in this manner, arm freely in space. Shoulder subluxation is a
theirperformance improved with training. If frequent accompaniment of other primary
they were simply told that the target would motor problems in the flaccid stroke patient.
appear at a different location for each trial, Tightness in the chest muscles and ligaments
with practice, they showed a clear and rapid can develop in patients who habitually hold
improvement in their abilities (1, 6). the involved arm in a flexed posture, which
Patients with parietal lesions also show includes internal rotation of the shoulder and
problems with eye movements when these protraction of the scapula. Tight elbow, wrist,
movements are a part of exploratory visual and finger flexors limit the patient's ability to
searches or reaching behavior. They may have actively extend the hand (9-11).
problems breaking visual fixation (Balint's
syndrome) or in optic ataxia; they also may PAIN AND EDEMA
have slowed reaction time for saccades, with
Another complication that interferes
the saccades subdivided into staircase patterns
with the recover)' of upper extremity fijnction
(7,8).
following a stroke is pain and/or swelling in
Transport Problems the hemiplegic arm. In addition to shoulder

pain, the shoulder-hand syndrome is a rec-
Remember from the previous chapters ognized concomitant of stroke, which occurs
that research hasshown that the transport in approximately 15% of all stroke patients.
component of reaching varies, depending on The shoulder-hand syndrome encompasses
the goal of the task. Thus, the trajectory and pain with motion, and loss of range of motion
duration of the movement used during the in both the shoulder and the hand. In severe
transport phase varies, depending on whether cases, there is pain at rest. If shoulder-hand
the goal is to touch a target or grasp an object, syndrome is prolonged, it can lead to a "fro-
or to grasp and move the object in different zen" shoulder. There is no agreement con-
ways. We hypothesize that one consequence cerning the underlying cause of shoulder pain
of an upper motor neuron lesion is the loss of following stroke, nor is there agreement on
this task-specific flexibilit)' in how movements methods for treatment (12-15).
are organized.
WEAKNESS
Motor Dyscoordination
Neural lesions affecting the ability to
As we mentioned earlier, reaching is generate force are a major limitation in many
controlled by a dififerent neural mechanism
patients with a neurological impairment.
from that of grasping; hence, patients can Strength is defined as the ability to generate
have impaired reach but intact grasp, or vice- sufificient tension in a muscle for the purposes
versa. For many patients who are neurologi- of posture and movement (16). Strength re-
cally impaired, both reach and grasp are af- sults from both properties of the muscle itself
fected, reflecting dysfunction in the multiple
and the appropriate recruitment of motor
systems controlling upper extremity function. units, as well as the timing of their activation
This section reviews a variety of musculoskel-
(16, 17). Neural aspects of force production
etal and motor constraints that affect reach- reflect (a) the number of motor units re-
ing.
cruited, (b) the type of units recruited, and (c)
the discharge frequency (18).

MUSCULOSKELETAL CONSTRAINTS
Weakness, or the inability to generate
Many patients with upper motor neuron tension, major impairment of fianction in
is a
lesions develop secondary musculoskeletal many patients with upper motor neuron le-
sions. Stroke patients have been shown to Other studies examining reaching and
have abnormal and reduced firing rates of mo- other types of movements have found inap-
tor neurons (19). Thus, weakness, or the in- propriate "shortening reactions" that con-
ability to recruit motor neurons, major
is a strain upper extremity movements. A short-
constraint affecting all aspects of upper ex- ening reaction is the inappropriate acdvation
tremity function, including the ability to of the stretch reflex during shortening con-
transport, grasp, and release objects. tractions of a muscle, thus impairing a pa-
tient's ability to move the arm. Inappropriate
SPASTICITY shortening reactions have been reported in
patients following stroke (21 ) and in patients
As we mentioned in Chapter 9, the with Parkinson's disease (22).
range of muscle tone abnormalities found It is important to note that none of the
within patients who have UMN lesions is studies described denies the fact that spasticity
great. We defined normal muscle tone as the impairs motor control in the patient with neu-
muscle's resistance to being lengthened, or its These studies do, however,
rological deficits.
and that stiffness or tone is the result
stiffness, challenge the assumption that spasticity is the
of both non-neural and neural components. primary impairment to normal motor control
On the upper end of the tone spectrum is hy- (23,24).
pertonicity or spasticity, often defined as "a Another eftect that spasticit)' may have
motor disorder characterized by a velocity-de- on upper extremity motor control relates to
pendent increase in tonic stretch reflexes stiffness abnormalities during reaching. Re-
(muscle tone)" (20). However, since muscle member that in Chapter 16, we explained that
tone or stiffness is a result of both non-neural the location programming theory hypothe-
and neural components, it has also been em- sizes that when a person makes an arm move-
phasized that increased muscle stifthess in ment, the nervous system programs the rela-
spastic hypertonia may be due to changes in tive balance of tensions (or stiffness) of two
the intrinsic properties of the muscle fibers opposing (agonist and antagonist) muscle sets
themselves. in multiple joints in order to move the arm to

Many clinicians view spasticity or dis- a new position in space. Thus, in patients with
orders of tone to be the most significant im- stiffness control problems, such as spasticity,
pairment constraining function in the patient location programming would be ver)' difficult
with upper motor neuron disease (9-11). or impossible.

However, the extent to which spasticity
impairs upper extremity fiinction is still MASS PATTERNS OF MOVEMENT
unclear.
Results from a study examining the ex- The presence of mass patterns of move-
tent to which abnormal stretch reflexes in an- ment is another major limitation of upper ex-
tagonist muscles impair arm movements in tremity fiinction in many patients who have
stroke patients raises questions about the ex- suft'ered stroke or traumatic brain injury. Nor-
tent to which velocity-dependent spasticity mal upper extremity' fiinction requires the
limits upper extremity control (21). For ex- combine various t}'pes of move-
ability to
ample, it has been hypothesized that flexor ments, and to use the fingers independentiy.
spasticity in the biceps may prevent eftective The upper extremity flexion synergy usually
and extension of the
activation of the triceps involves abduction, extension, and external
arm. Results from studies examining upper rotation of the shoulder, elbow flexion, fore-
extremity reaching movements in patients arm supination, and flexionof the wrist and
with hemiplegia do not support the hypoth- fingers. The extensor pattern is forward flex-
esis that the primary constraint on upper ex- ion, adduction, and internal rotation of the
tremity reaching is spasticity of the biceps, but shoulder, extension of the elbow, forearm
rather weakness and inability to recruit mo- pronation, and extension of the fingers and
toneurons in the triceps (21). tiiumb. More recently, massed patterns of
Chapter Seventeen Abnormal Upper Extremity Manipulation Control 403
movement have been viewed as invariant co- ipsilateral to a lesion, researchers have also
ordinative structures (27). been able to understand the contribution of
a particular hemisphere to a reaching move-
DYSCOORDINATION ment without the confounding influence of

the severe sensory and motor loss that occurs
Studies examining the trajectories of the in the contralateral limb.
hemiplegic arm in patients with hemiparesis An initial study examining motor prob-
found thatmovement amplitudes were lems in the nonhemiparetic limb has sug-
smaller, and movement times were longer gested that weakness is a contributing factor
than in nondisabled subjects, with disruptions in reaching problems in this limb as well as
of interjoint coordination between the elbow the hemiparetic limb (35). Other studies have
and shoulder (28). found that problems in reaching in the non-
Studies have looked at the contributions hemiparetic arm following unilateral hemi-
of specific brain subsystems to these prob- spheric lesion may involve other factors as
lems. The cerebellum appears to be important well.
in the programming of ballistic movements In recent studies (36, 37), researchers
(also referred to as open-loop or nonfeed- examined reaching abilities in two groups of
back-controUed movements). This is because patients suffering from unilateral damage to
studies have shown that padents with cere- the left or right cerebral hemispheres. Sub-
bellar disorders show abnormal ballistic arm jects were required to reach quickly and ac-
movements. There is some evidence that slow curately to a small visual target using the arm
closed-loop movements may be less affected ipsilateral to the lesion. Both patient groups
by cerebellar dysfunction (29-33). were found to be less accurate than controls
One study examined the ability of 7- and required more time to complete the reach
year-old children with mild cerebral palsy (ei- after the target was illuminated. The research-
ther ataxic or athetoid) to reach for a moving ers found, however, a significant difference
object. It was found that the reaches of these between the performance of subjects with
movement-impaired children had longer While the
right vs. left hemisphere lesions.
transport phases compared with nondisabled right hemisphere lesioned group took longer
children and thus were less efficient (34). In- to initiate a reach, the movements themselves
terestingly, despite their motor impairments, were similar to those of the control group. In
the disabled children were able to reach for contrast, the left hemisphere lesioned group
and grasp even quickly moving targets. The did not have problems in the time required to
researchers found that the children aimed initiate the reach, but took much longer to
their reaches well ahead of the moving targets, execute the reach itself Thus, reach was im-
suggesting that the children were able to paired in both groups, but for apparently dif-
compensate for their movement deficits when ferent reasons. These authors suggest that a
planning a reaching movement. They aimed lesion in the right hemisphere affects the pa-
their movements enough ahead of the target tient's ability to quickly detect the spatial po-
so they could sustain accuracy when reaching, sition of the target (higher level visual pro-
despite their movement impairments (34). cessing). In contrast, a lesion in the left
hemisphere appears to affect the patient's

MOTOR IMPAIRMENTS AFFECTING ability to an appropriate program
select
THE NONHEMIPARETIC LIMB (higher level motor processing) to achieve the

target position, and/or to modify that pro-
Traditionally, researchers have main- gram as it is being executed (36, 37).
tained that unilateral cerebral lesions manifest
in the limb contralateral to the lesions. Now Sensory Impairments
researchers are also finding subtle deficits af-
fecting the ability to reach on the nonhemi- Both visual and somatosensory impair-
paretic side. By studying reaching in the arm ments may have significant effects on manip-
ulatory function. One common visual system

problem
reaching
affecting
is
the
optic ataxia.
transport phase of
^f^Z
OPTIC ATAXIA
Lesions on either side of the posterior

parietal area in humans can cause marked eye-
hand coordination impairment, or optic
ataxia. Optic ataxia is defined as the inabilit)'
to reach for objects in extrapersonal space, in

the absence of extensive motor, visual, or so-
matosensory deficits (1). Patients with optic
ataxia typically misreach for objects within
the visual field that is contralateral to their
lesion.
This disorder was first described by Bal-
int in 1909 (7) using the term "visual disori-
entation." He noted that the patient could
reach normally with his left hand, but when
asked to reach with his right hand, he made
mistakes in all directions, until he eventually
bumped into the object with his hand. He
fi)und that the problem was related to visual
control of that hand, because if he asked the
patient to first point to the object with his left
hand, then he could reach accurately with his Figure 17.1. Drawing of the grip patterns of a patient
right hand. On autopsy, it was found that the with optic ataxia. A, Normal hand. B, Affected hand, vi-
patient had a lesion in the posterior parietal sual feedback. C, Affected hand, no visual feedback.
(From Jeannerod M. The neural and behavioral organi-
areas, including the angular gyrus and the an-
zation of goal-directed movements. Oxford: Oxford Uni-
terior occipital lobe on both sides of the brain
versity Press, 1990:225.)
Patients with optic ataxia have now been

tested carefiiUy in the laboratory setting. It
has been found that in the absence of visual normal hand («), affected hand with visual
feedback concerning their hand movement, feedback (^), and (c) without vision. Note
patients with unilateral lesions misreach to- that even with visual feedback, the affected
ward the side of the lesion, when they use the hand didn't begin to close until the last mo-
hand contralateral to the lesioned hemi- ment and the terminal grip size was too big.
sphere. Thus, there is always a directional er- Without visual feedback, grasp formation did
ror in reaching. not occur.
There are specific motor disorganiza- Why would patients with optic ataxia
tion problems in these patients as well. It has show no grasp formation.' Is it simply a strat-
been hypothesized that their problems relate egy they use because they have so much error
toprogramming visually guided goal-directed in their reach.* It has been argued that this
movements (1). It has been shown that the isn't the case, because they even show this
deceleration phase of reaching much longer
is with visual feedback. Thus, it has been con-
than that in the normal hand, with many small cluded that it results from a specific problem
peaks. In addition, these patients have prob- with eye-hand coordination mechanisms re-
lems with grasp formation. Figure 17.1 shows sponsible for adjusting finger posture to the
a reach of a patient with optic ataxia with his shape of the object { ).
1
Chapter Seventeen Abnoral\l Upper Extremiti' Manipulation Control 405
SOMATOSENSORY LOSS perature sensation were impaired or totally

absent in both hands. Tests showed that in
As noted in earlier chapters, experi- spite of these problems, the patient could per-
ments by Sherrington in the late 1800s form many motor tasks, even without \ision.
showed that monkeys that were deafferented For example, the patient could tap, do fast
on one side of the spinal cord stopped using alternating flexion and extension movements,
the affected limb. He thus concluded that and draw figures in the air, using only the
senson' feedback was critical to movement wrist and fingers (I, 41 ).
control. In contrast, researchers who deaffer- It was also noted that EMG actiNity dur-
ented both limbs of animals showed that the ing flexion and extension of the thumb was
animals recovered motor fimction. Move- similar to that seen in normal subjects. The
ments were initially awkward, but impro\'ed subject could also learn new thumb positions
within as little as 2 weeks, as long as \isual with vision and then reproduce those posi-
feedback was available (38). tions without \'ision. Thus, motor learning
Interestingh', recoven,' starts \\'ith the was also possible. However, the patient's per-
animals only being able to sweep the object formance rapidly deteriorated when asked to
across the floor. Then, a coarse grasp with all repeat the mo\'ement many times with the
four fingers develops, and then a pincer grasp eyes closed.
reappears (39). It has been suggested that In a second study on patients with pe-
when unilateral deafferentation occurs, the ripheral sensor\' neuropathy, patients could
animals may learn not to use the deafferented perform repetitive flexion and extension
limb, or even develop inhibition of the deaf- movements of the wrist, with normal EMG
ferented arm (39). This "learned disuse" hy- activit}-, as long as the mo\ements were not
pothesis is supported by the fact that unilat- too fast. At a certain point, however, the in-
erally deafferented animals recover movement ten'als bet%veen the EMG bursts tended to
coordination as well as bilaterally deafferented disappear. Itwas hypothesized that this was
animals if their normal limb remains immo- due to the higher level of cocontraction of
bilized so that they have to use the deaffer- agonist and antagonist muscles seen in these
ented Umb (1). patients (42).
Also, remember from the chapter on Patients could also hold a steady posture
normal eye-hand coordination that experi- with their deafterented limb as long as they
ments on deafferented monkeys showed that, had visual feedback. However, without visual
when making single-joint movements, they feedback, large errors were made, and the
could reach targets with relative accuracy, limb drifted back to its initial position, as
even when they could not see the hand. Dis- shown in Figure 17.2 (42).
placing the forearm just prior to movement What does this information tell us about
onset during a reach also did not significantly the role of kinesthetic feedback in reaching?
disturb pointing accuracy in these deaffer- It appears that it is not required for movement
ented animals. was thus concluded that sin-

It initiation and execution. However, it is still
gle-joLnt movements depend on changes in important for accurate reaching involving
muscle activation levels that are programmed multiple joints. Researchers testing humans
prior to movement onset and that no feed- with peripheral sensor\' neuropathy found
back is required for reasonably accurate exe- that patients were able to make accurate
cution of these movements (40). mo\'ements only if they involved single joints.
In addition, recent experiments on hu- They showed great problems in performing
mans after pathological deafferentation have natural movements used in normal life (43).
confirmed the from experiments on
results
monkeys (41). One patient had suffered a se- IMPAIRMENTS AFFECTING GRASP
vere peripheral sensory neuropathy, so that
AND RELEASE
there was loss of sensation in both the arms The range of grips required for daily life
and the legs. Light touch, vibration, and tem- tasks is great, varying from grips requiring
406 Section rV UPPER EXTREMITY CONTROL
a.
Normal
Figure 17.2. Recordings of wrist position of a patient with peripheral sensory neuropathy. The patient was asked to
rotate the wrist to reach positions of 5°, 10°, and 20°, against an elastic load. A, With vision, the patient had no
problems. B,Without vision, the position drifted back in the direction of the load. C, A normal subject's performance
without vision. (Adapted from Sanes JN, Mauritz KH, Dalakas MC, Evarts EV. Motor control in humans with large-fiber
sensory neuropathy. Human Neurobiology 1985;4:1 10.)
great precision, but not much force, to those sent, delayed, or prolonged in patients with
requiring greater amounts of force, but not pyramidal tract lesions, affecting the timing
much precision. Precision grip involves con- and precision of hand movements (48).
trol of individual finger motions, and is largely Unfortunately, it appears that no other
carried out by the intrinsic hand muscles (44, area of the CNS (not even corticomotoneu-
45). In contrast, a power grip appears to in- rons in the opposite hemisphere) can substi-
volve a generalized coacdvation of all the tute for these neurons when they are injured.
digits, primarily uses the extrinsic hand mus- In addition, there do not appear to be any
cles, and does not require a fine degree of alternative tracts within the CNS that can sub-
control. The two grips appear to be controlled stitute for loss of descending corticospinal
by different cortical neurons (46, 47). tracts. This limits the recover)' of precision
grip in patients with neural lesions affecting
Motor Problems the primar)' motor cortex or its descending
PROBLEMS AFFECTING ACTIVATION tracts (49-51).
AND COORDINATION OF AGONIST Research on reaching behavior in mon-

MUSCLES keys whose corticospinal system had been le-
sioned at birth has shown that the distal com-

Corticomotoneurons play an essential ponent of the reach never matures. For
role in precision grip, and their loss due to example, when area 4 of the motor cortex is
neural injur\' results in an inability' to recruit lesioned in infancy, a precision grip never de-
distal muscles, particularly the intrinsic hand velops. Remember fi-om the chapter on de-
muscles. Muscle activation bursts can be ab- velopment of upper extremity and manipula-
Chapter Se%enteen .\bnoral\l Upper Extremity' AL\nipul.\tiox Control 407
ton' control that precision grip usually

develops in monkeys at about 8 months of
age, at the time that the pyramidal track ma-
tures (1, 52).
Research snidies have examined recov-
er\' of upper extremity' control in adult pa-
tients uho ha\e had a stroke, with primar>'
damage to the motor cortex areas and the py-
ramidal pathways. These studies have shown
that movement in proximal joints recovers
first, with normal force returning in 4 to 6
weeks. However, isolated finger mo\ements
were permanendy lost in these patients (53).
Remember that control of proximal joints in-
volves a different svstem fi-om that controlling
distal muscles.
Other studies examining recover)' of up-
per extremity- function in adult stroke patients
have found that the shoulder-elbow synerg>-
for transporting hand to the object
the
showed was passi\ely
recover\' if the shoulder
supported against gravit\', but finger move-
ments were always clumsy. The patients could
not shape the hand in anticipation of the Figure 17.3. Drawing from film records of the reaches
of a hemiplegic 23-month-old child reaching for a prong
grasp. Also, the grasp was made by using the
from a pegboard with the normal hand and the affected
palm of the whole hand, rather than by using
hand, with visual feedback. A, Normal hand. B, Affected
the pincer grasp with the fingertips (1, 54). hand. (From Jeannerod M. The neural and behavioral or-
Studies have examined the reaching be- ganization of goal-directed movements. Oxford: Oxford
ha\ior of developmentally disabled children, Universit>' Press, 1990:72.)
including those with hemiplegia or Down

svndrome (55-57). In some cases of mild im-
pairment, hemiplegia is not readily identified grasp formation. There was a \er\' sUght clos-
untilabout 40 weeks, when the infant first ing of the hand afi:er contact with the object,
begins to use the pincer grasp and manipulate giving a verv' clumsy grasp ( 1 ).
objects (1). In one child of 23 months, the In a second child of 5 years of age, the
hand with hemiplegia was used only when the hemiplegic hand showed more normal reach
normal hand was immobilized, and even then, and grasp movements. The authors suggest
it was with great difficult}' that the child that more normal movement patterns may be
grasped objects. Figure 17.3, adapted fi-om the result of many years of rehabilitation train-
film records, illustrates the child reaching for ing (1). Figure 17.4 depicts film records of
a prong fi-om a pegboard with the normal her reaching movements with her normal
hand and the atfected hand, with \isual feed- hand (^4) and her hemiplegic hand (B, C, D).
back. Note that the normal hand did not an- Note that reaching in the hemiplegic hand
ticipate the shape of the object well, but a fin- was only affi;cted in relation to the pattern of
ger extension/flexion pattern was used. Also, grip formation. Finger shaping was abnormal,
contact of the hand with the object caused the with the index finger extended in an exagger-
fingers to close around the object, giving an ated manner, and then fle.xing only slighdy, if
accurate grasp. However, the hemiplegic at all, before contacting the object. Due to
hand showed an exaggerated opening during these problems, the objects were sometimes
the entire movement, with no anticipatoni' dropped during the grasp ( I ).
Figure 17.4. Drawing from film records of the reaches of a hemiplegic 5-year-old child after many years of rehabil-
itation reaching for a prong from a pegboard with the normal hand and the affected hand, with visual feedback. A,
Normal hand. B, C, and D, Affected hand. (FromJeannerodM. The neural and behavioral organization of goal-directed
movements. Oxford: Oxford University Press, 1990:73.)
INAPPROPRIATE ACTIVATION OF vs. pyramidal systems controlling these two

ANTAGONIST MUSCLES types of movements.
Abnormalities of upper extremity' func- SENSORY PROBLEMS

from disturbances to the tim-
tion can result
ing and amplitude of contraction of antago- Open-loop control or ballistic reaching
nist muscles. In some patients, for example, movements are preprogrammed, and there-
those with athetoid cerebral palsy, the antag- fore do not require sensory feedback to con-
onist muscle is inappropriately active. Antag- trol the movement. Relatively normal ballistic
onist activation can occur prior to the agonist upper extremit)' movements have been found
muscle, causing movement in the wrong di- in patients with complete limb deafferenta-
rection. Alternatively, antagonist activation tion (59). In contrast, closed-loop move-
can occur simultaneously with the agonist, re- ments, such as precision hand movements, re-
sulting in decreased amplitude of movement quire continuous senson,' inputs, and are
(48). significantly impaired in patients with sensor)'
Surprisingly, in many patients who are loss (60).
neurologically impaired, abnormal coordina- Recent studies have suggested that the
tion of muscles is not found consistently in all control processes involved in precision grip
types of upper extremity' movements. For ex- and lift are better described as "discrete event
ample, when patients with severe dystonia driven control" rather than as a "continuous
were asked to wa\'e, muscle activation patterns closed-loop" control (61, 62). As a basis for
underlying the alternating wrist flexion and this hypothesis, research has shown that sen-
extension movements were normal. In con- sor}' information is used intermittently at crit-
trast, excessive and inappropriate coactivation ical times within a precision grip and lift task,
of agonist and antagonist muscles was present rather than continuously.
in these same patients during precision hand A precision grip and lift task is organized
movements such as when asked to write their into distinct phases which are linked together.
name (58). This may be due to nonpyramidal The pattern of muscle activity' used in the dif-
Chapter Seventeen ABNORMAL UPPER Extremity Manipulation Control 409
movement is determined
ferent phases of this while in the second reach, there was incom-
by combination of previous experience and
a plete grip formation (1).
afferent information (visual and somatosen- Patients with central lesions to the pa-
sory) generated during the performance of rietal lobe, particularly the post-central gyrus
the task (61, 62). and the supramarginal gyrus, show similar
Tactile input is necessary to determine patterns for reach and grasp as patients with
the appropriate grip force. If grip force is too peripheral sensory problems. In a detailed
tight, the object can't be manipulated; if it is study on the recovery of reach and grasp in a
too loose, the object will be dropped. In a patient with a parietal lobe lesion, researchers
precision grip, forces for gripping anci lifting found the patient did not use her right hand
are generated simultaneously and appear to be spontaneously immediately following her le-
very dependent on cutaneous input. When sion, but later used it in many actions, as long
the fingers of neurologically intact subjects are as she had visual feedback. Without visual
anesthetized, grip forces are often inappro- control, her movements were very awkward.
priate to the object being gripped (61). For example, she couldn't sustain repetitive
What happens to eye-hand coordination tapping movements unless she could see or
skills in the patient with a neurological im- hear her fingers moving (1).
pairment with loss of somatosensation? Ex- In contrast to patients with peripheral
periments have been performed in which the deafferentation, who could grip normally as
reaching of patients with lesions

skills in the long as visual feedback was present, grip for-
somatosensory pathways at brainstem levels mation was impaired in the patient with a pa-
and at parietal cortex levels were examined rietal lesion, even with visual feedback present
( 1 ). In the patient with the lesion at the brain- (1). Figure 17. 6A shows the grasp compo-
stem level, the hand ipsilateral to the lesion nent of a reach with the patient's normal
was affected. When vision was present, the hand, while Figure 17.65 and C show the
reach was normal, as shown in Figure 17.5^1, grasp of the affected hand both with and with-
except that was longer in duration than it
it out visual feedback. When she reached with
was in the normal hand. However, without the affected hand with vision, the patient
vision, the grasping movements were critically made grasps using the whole palm of the
changed (Fig. 17. 5B and C). Finger grip was hand. Without visual feedback, only the initial
either absent altogether, or incomplete. In the part of the transportation phase was normal.
first reach the patient made with no visual Then the hand seemed to "wander above the
feedback, there was no grip formation at all. object, without a grasp" (1, p 207).
Figure 17.5. Drawing from film records

of the grip patterns of a patient with lesion
of the somatosensory pathways at the
brainstem level. A, With vision grasp was
normal. B and C, Without vision, was it
absent or incomplete. (From Jeannerod M.

The neural and behavioral organization of
goal-directed movements. Oxford: Oxford
University Press, 1990:205.)
No visual feedback
Figure 1 7.6. Drawing of the grip patterns

of a patient with parietal lobe lesions. A,
Normal hand, no visual feedback. B, Af-
fected hand, visual feedback. C, Affected
hand, no visual feedback. (From Jeannerod
M. The neural and behavioral organization
of goal-directed movements. Oxford: Ox-
ford University Press, 1990: 208.)
c^^
Thus, loss of sensory information results normal upper extremity control. Sensory in-
in abnormal grip and lift forces, and problems formation is critical to adapting movements
in the control of small, fine movements of the and is used to correct errors during the exe-
hand (41, 61-63). cution of upper extremit}' movement, ensur-
ing accuracy during the final portions of the
Postural Problems movement.
The ability to move our arms purposc- Visual Deficits

fiilly requires good postural control to main-
tain an appropriate orientation and stability of The primary function of visual feedback
the body during the performance of upper ex- in reaching appears to be related to the at-
tremity tasks. In the chapters on normal and tainment of final accuracy. It has been hy-
abnormal postural control, we reviewed in de- pothesized that the constancy of thumb po-
tail the research on anticipatory postural ad- sition with relation to the wrist during
justments that are normally made in advance reaching may be part of a strategy of provid-
of potentially destabilizing reaches to prevent ing clear visual feedback information regard-
or minimize displacement of the body. We ing the endpoint of the limb (64).
also discussed abnormalities in the ability to
effectively preprogram these anticipatory pos- Somatosensory Deficits
tural in a variety of patient
adjustments seen
populations, thus resulting in slowed and in- Is somatosensory input essential for the
efficient upper extremit}' movements. Please production of reaching movements? Consid-
refer back to these chapters for more infor- erable research (1, 38) has shown that mon-
mation on problems with postural control as keys that were deafferented still were able to
it relates to upper extremir\' movements. perform adequate reaching and grasping
movements as soon as 2 weeks after the lesion
Problems with was made, as long as vision was available.
Adaptation They noted that movements were awkward at
first, with animals only sweeping objects along
The ability to adapt upper extremity the floor. They then developed a primitive
movements to changes in task and environ- grasp with four fingers together and no
mental demands is an essential component of thumb, and finally redeveloped a crude pincer
Chapter Seventeen AbnormaI- Upper Extremity' Manipulation Control 411
grasp a few months after the lesion was made ticipator\' action and both intermittent and
(1). continuous sensor\' information about the
Other such experiments have shown ongoing events. Anticipator\' activity is based
that deafferentedmonkeys can still make rea- on prior knowledge about the task itself, and
sonably accurate single joint pointing move- the movements that need to be made.
ments, even when vision of the arm is oc- An essential component of all reaching
cluded, when the pointing task was learned movements is proactive visual and somatosen-
before deafferentation (40). In this case, even soPi' control, which is responsible for the cor-
displacing arm before the movement
the rect initial directionof the limb toward the
didn't affect terminal accuracy, even though target and the initial coordination between
they couldn't see or feel their arm position! limb segments. In addition, visual informa-
Thus, it was concluded that the monkey is cation about the characteristics of the object to
pable of using a central motor program to be grasped is used proactively to preprogram
perform reaching movements and that kines- the forces used in precision grip.
thetic feedback isn't required for achieving It has been hypothesized that visual and
reasonable accuracy. However, the animal somatosensor\' information is used to update
could not adapt the reaching movement to a proprioceptive and visual body maps, which
change in shoulder position, which thus programming of reaching
allows the accurate
changed the initial coordinates of the arm in movements. To determine the influence of
space. updated maps of the body workspace on the
Experiments performed \\ith humans accuracy of a reaching movement, experi-
with severe peripheral sensor^' neuropathy in ments were performed to manipulate visual
all four limbs ha\e shown similar results (41 ). information regarding hand and target posi-
The patient was able to perform a wide variety tions prior to movement. When the subject
of hand movements, such as tapping move- could not see the hand prior to movement,
ments, and could draw figures in the air, even there were large errors. It was thus concluded
with the eyes closed. Howexer, when he was that a proprioceptive map of the hand, by it-
asked to repeat the moxement many times self,was not adequate to appropriately code
\\'ith the eyes closed, the performance deteri- the hand position in the reaching workspace.
orated quickly. Thus, apparently, somatosen- This means that somatosensor*' inputs must
sor}'information isn't required for arm move- be calibrated by vision in order for the pro-
ment initiation or execution as long as the prioceptive map and the visual map to be
movements are simple or nonrepetitive. matched (I). No experiments have yet been
However, if patients have to make complex performed to determine how often the pro-
movements requiring coordination of many priocepti\'e map needs to be updated by visual
joints, or repeat movements, then without vi- inputs to ensure accurate movements.
sual feedback, they are unable to update their Loss of motor control resulting in hm-
central representations of body space. Ac- ited ability to move may also aft'ect explor-
cordingly, they show considerable movement atorv' aspectsof motor control. This reduction
"drift" and problems with coordination (1). of exploratory' mo\'ements may contribute to
Although these experiments suggest that cer- impaired upper extremit\' control by affecting
tain movements may be carried out without anticipatoPi' aspects of the movement.
somatosenson,' feedback, considerable work
has also shown the important contributions of Apraxia
sensor)' feedback to the fine regulation of
movement. Up until now, our discussion of abnor-
mal upper extremit)' manipulatixe control has
Anticipatory Aspects
related to tlie examination of problems in
The abilitv' to move smoothly, grasp, each of the constituent components. How-
and pick up objects, is a combination of an- ever, the use of the upper extremity' in the
412 Section W UPPER EXTREiVim' CONTROL
performance of simple even'day tasks is more formance of common acti\ities of daily living,
than the simple summation of these compo- including buttering hot coftee, putting
nents. It requires the integration of these clotheson backwards or inside-out, drinking
components into an action plan. An action from an empt}' cup, skipping key steps during
plan specifies the conceptual content of the activities such as shaving, toothbrushing, or
action, along with its hierarchical and sequen- hairbrushing, using a fork to eat cereal, put-
tial organization (65). The left cerebral cortex ting toothpaste on a razor, scrubbing the up-
includes structures specialized for higher-or- per and chin with a toothbrush, eating
lip
der motor programming or the formation of toothpaste, and apphing arm deodorant over
action plans (66). a shirt (66). In a classic paper, Luria describes
One way researchers have studied the the behavior of a frontal apra.xia patient who
nature of these motor programs is by analyz- would light a candle and put it in his mouth
ing the t\'pes of errors made by patients with to perform the habitual movements of smok-
left hemisphere damage. Disorders that result ing a cigarette (67).
from dysfiinction of this specialized left hemi-
sphere have been termed apraxias. One t\pe
of apraxia that has been studied extensively is
ideational apraxia, also referred to as frontal Summary
apraxia (67), or frontal lobe executive disor-
der (68). This is a disorder of the execution 1. Understanding the cause of impairments in
eye-hand coordination may be difficult, due
of movement that cannot be attributed to
to the complexity of the interactions between
weakness, incoordination or sensory loss, or
neural substrates involved in reaching skills.
to poor language comprehension or inatten-
2. Lesions to the motor cortex areas and the py-
tion to commands. ramidal pathways following stroke show re-
To understand this disorder, it is helpftil covery of function in the proximal joints first,
to first appreciate what occurs when a normal with normal force returning in 4 to 6 weeks.
adult decides to perform a task. It is h\poth- However, recovery of isolated finger move-
esized that the first step involves formulating ments almost never occurs.
the intention to perform the task and then 3. Studies on patients with interhemispheric le-
formulating an action plan. The essential sions suggest that proximal arm movements
requirement of an action plan is that it spec- are controlled by a diffuse cortical and sub-
cortical uncrossed pathway, while hand
ifies the £ioal of the action along with the hi-
movements are controlled only by the contra-
erarchical and sequential organization of
lateral motor cortex.
nested actions that are required to achieve the
4. Patients with lesions in the visual striate cortex
ultimate goal. Intentions, as defined by acti-
are still able to reach toward objects moving
vated action plans, are an integral feature of across their visual field, even though they are
all purposeful behavior. It has been hypothe- considered totally blind, possibly due to sub-
sized that the core of the intentional disorder cortical visual processing in the superior col-
of frontal apra.xia is a weakening of the top- liculus.
down formulation of action plans, that is, an 5. Patients with peripheral sensory neuropathy
inabilit)' to sustain the intent to the comple- can make accurate single joint movements,
tion of the action plan (66). but show great problems in performing most
normal movements. With visual feedback,
As a result, irrelevant objects exert a
reaching is reasonably normal, but without vi-
strong influence on the action plan, and this
sion, finger grip is either absent or abnormal.
leads to numerous performance errors. Re-
6. Lesions on either side of the posterior parietal
searchers havebegun to develop a system for area can cause optic ataxia or the inability to
coding performance errors based on this con- reach for objects in extrapersonal space (in the
cept of hierarchically organized units of action absence of extensive motor, visual, or soma-
within an action plan. These studies have enu- tosensory deficits).
merated examples of errors during the per- 7. Damage to the left hemisphere may cause
Chapter Seventeen ABNORMAL Upper ExTREMm' Manipulation Control 413
apraxia, a disorder of the execution of move- DS. Shoulder-hand syndrome in a hemiplegic

ment that can be accounted for neither by population: a 5-\ear retrospecti\e study. Arch
weakness, incoordination or sensory loss, nor Phys Med Rehabil 1977;58:353-356.
by poor language comprehension or inatten- 13. Patridge CJ, Edwards SM, Mee R, Langen-
tion to commands. The core of this disorder berghe HVK \an. Hemiplegic shoulder pain:
may be a weakening of the top-down formu- a study of t\vo methods of physiotherapy
lation of action plans, that is, an inability to treatment. CUnical Rehabilitation 1990;
sustain the intent to the completion of the ac- 4:43^9.
tion plan. As a result, irrelevant objects exert 14. Roy CW. Shoulder pain in hemiplegia: a lit-
a strong influence on the action plan, leading erature re\iew. Clinical Rehabilitation 1988;
to performance errors. 2:35-44.
15. Cailliet R The shoulder in hemiplegia. Phil-
adelphia: FA Davis, 1980.
References 16. Smidt GL, Rogers MW. Factors contributing
to the regulation and clinical assessment of
1. Jeannerod M. The neural and beha\ioral or-
muscular strength. Phys Ther 1982;62:
ganization of goal-directed movements. Ox-
1283-1290.
ford: Oxford Universin.' Press, 1990:283.
2. Martin TA, Keating JG, Goodkin HP, Bas-
17. Buchner DM, DeLateur BJ. The importance
of skeletal muscle strength to physical func-
tian AJ, Thach \\T. Storage of multiple gaze-
tion in older adults. Annals of Behavioral
hand calibrations. Neuroscience Abstracts
xMedicine 1991;13:1-12.
1993;19:980.
3. Humphre)- NK, Weiskrantz L. Vision in 18. Rogers MM. Musculoskeletal considerations
monkeys after removal of the striate cortex. in production and control of movement. In;
Nature 1969;215:595-597. Montgomer)- P, Connolly BH, eds. Motor

control and physical therapy. Hixson, TX:
4. Peoppel E. Letter to the editor. Nature 1973;
243:231. Chattanooga Group, 1991:69-82.
5. Perenin MT, Jeannerod M. Msual fiinction 19. Duncan P, Badke MB. Stroke rehabilitation.
within the hemianoptic field following earh' Chicago: Year Book, 1987.
cerebral hemidecortication in man. 1. spatial 20. LanceJAV: Symposium synopsis. In: Feldman
localization. Neuropsychologia 1978;16:1- RG, Young RR, Koella WP, eds. Spasticit\-:
13. disordered motor control. Chicago: Year
6. Zihl Werth R. Contributions to the study
J,
Book Medical Publishers, 1980:485.
of "blindsight." II. The role of specific prac- 21. Sahrmarm SA, Norton BJ. The relationship
tice for saccadic localization in patients with of \oluntar\' movement to spasticit\' in the
postgeniculate visual field defects. Neuropsv- upper motor neuron syndrome. Arch Neiu-ol
chologja 1984;22:13-22. I977;2:460-^65.
7. Balint R. Seelenhamung des "Schauens," op- 22. Johnson RH . Disorders of stretch reflex mod-
tische Ataxic, raumlische Stoning des Auf- ulation during \olitional movements. Brain
mersamkeit. Monatshr Psychiatr Neurol 1991;! 14:443^60.
1909;25:51-81. 23. Katz R, Rymer Z. Spastic h\pertonia: mech-
8. Waters RL, Wilson DJ, Sa\inelli Rehabil- R anisms and measurement. Arch Phys Med
itation of the upper extremit>' follo\\ing Rehabil 1989;70:144-155.
stroke. In: Hunter J, Schneider LH, Mackin 24. Gordon J. Assumptions underlying physical
EJ, Bell JA, eds. Rehabilitation of the hand. therapy intenention: theoretical and histori-
St. Louis: CV Mosby, 1978;505-520. cal perspectives. In: Carr J, Shepherd R, Gor-
9. Carr JH, Shepherd RB. A motor relearning don J, Gentile AM, Held J, eds. Mo\ement
programme for stroke. Rockxille, MD: As- science foundations for physical therap)- in re-
pen, 1983. habilitation. Rock\ille; Aspen, 1987.
10. Davies P. Steps to follow. New York: 25. Brunnstrom S. Movement therapy in hemi-
Springer- Verlag, 1985. plegia: a neurophysiological approach. New
11. Bobath B. Adult hemiplegia: evaluation and York: Harper & Row 1970.
,
treatment. London: William Heinemann 26. Gowland C. Staging motor impairment after
Medical Books, 1978. stroke. Stroke 1990;21{suppl 11):I1-19-I1-
12. Dans SW, Petrillo CR, Eichberg RD, Chu 21.
414 Section R' UPPER EXTRENUTi' CONTROL
27. Kamm K, Thelen E, Jensen J. A dxTiamical 38. Taub E, Berman AJ. Movement and learning
s>'stems approach to motor development. In: in the absence of senson,' feedback. In: Freed-
Rothstein J, ed. Movement science. Alexan- man SJ, ed. The neurophysiology' of spatially
dria, \'A:American Physical Therapy Associ- oriented beha\ior. Homewood: Dorey Press,
ation, 1991:11-23. 1968:173-192.
28. Le\in MF, Horowitz M, Junius C, Lamothe 39. Taub E. Motor beha\ior following deaffer-
AG, Feldman AG. Trajector>- formation and entation in the developing and motoricallv
interjoint coordination of drawing move- immature monkey. Herman RM, Grillner
In:
ments in normal and hemiparetic subjects. S, Stein DG, Stuart DG, eds. Neural control
Neuroscience Abstracts 1993;19:990. of locomotion. NT: Plenum Press, 1976:
29. Flowers K. Msual "closed loop" and "open 675-705.
loop" characteristics of voluntary movement 40. Polit A, Bizzi E. Processes controlling arm
in patients with Parkinsonism and intention movements in monkeys. Science 1978;
tremor. Brain 1976;99:269-310. 201:1235-1237.
30. Hallet M, Marsden CD. Physiolog>- and 41. RotiiweU JC, Traub .\L\I, Day BL, Obeso JA,
pathophysiology of the ballistic movement Thomas PK, Marsden CD. Manual motor
pattern. In: Desmedt JE, ed. Progress in clin- performance in a deafferented man. Brain
ical motor unit t\pes, re-
neurophysiolog)-: 1982;105:515-542.
cruitment and plasticit)' in health and disease. 42. Sanes JN, Mauritz KH, Dalakas MC, Evarts
Basel: Karger, 1981:331-346. E\'. Motor control in humans with large -fiber
31. Hallet M, Shahani BT, Young RR. EMG senson.- neuropathy. Human Neurobiology'
analysis of patients with cerebellar deficit. J 1985;4:101-114.
Xeurol Xeurosurg Ps>xhiatr\' 1975;38: 43. leannerod M. The formation of finger grip
1163-1169. during prehension. A cortically mediated \i-
32. Thach \\T. Correlation of neural discharge suomotor paaem. Beha\' Brain Res 1986;
with pattern and force of muscular acti\it\-, 19:99-116.
joint position and direction of intended next 44. Muir RB, Lemon RM. Corticospinal neurons
movement in motor cortex and cerebellum. J with a special role in precision grip. Brain Res
Neurophysiol 1978;41:654-676. 1983;261:312-316.
33. Deecke L, Komhuber HH. Cerebral poten- 45. Muir RM. Small hand muscles in precision
tial and the initiation of voluntan- movement. grip. A corticospinal prerogative. E.xp Brain
In: Desmedt JE, ed. Progress in clinical neu- Res 1985;10:155-173.
rophysiolog\', vol 1: Attention, volimtaty 46. Hofftnan DS. Luschell ES. Precentral cortical
contraction and event-related cerebral poten- cells during a controlled jaw- bite task. J Neu-
tials. Karger: Basel, 1977:132-150. rophysiol r980;44:333-348.
34. Forsstron A, von Hoftten C . Msually direaed 47. Fahrer M. Surgical approaches to the nerves
reaching of children with motor impairments. of the upper Umb. In: Tubiana R, ed.
Devel Med Child Xeurol 1982;24:653-661. The hand. Philadelphia: \VB Saunden,
35. GiullianiC, Geneva PA, Purser KE, Light 1988:539-547.
KE. Limb trajectors' in non-disabled subjects 48. Hallett M. Analysis of abnormal voluntary-
under t\so conditions of external constraint and in\oluntan' movements with surface elec-
compared with the non-paretic limb of sub- tromyography. In: Desmedt JE, ed. Motor
jects with hemiparesis. Neuroscience Ab- control mechanisms in health and disease.
stracts 1993;19:990. New York: Raven Press, 1983:907-914.
36. Fisk JD, GoodaJe ALA The effects of unilat- 49. Lawrence DG, Hopkins DA. The develop-
eral brain damage on \isually guided reach- ment of motor control in the rhesus monkey:
ing: hemispheric differences in the nature of evidence concerning the role of corticomo-
the deficit. Exp Brain Res 1988;72:425-t35. toneuronal connections. Brain 1976;99:235-
37. Smutok AL\, Grafinan J, Salazar AM, Swee- 254.
ney JK, Jonas BS, DiRocco PJ. Effects of uni- 50. Wise SD, Evarts EV. The role of the cerebral
lateral brain damage on contralateral and cortex in movement. Trends Neurosci 1981;
ipsilatral upper extremit}- function in hemi- 4:297-300.
plegia. Phys Ther 1989;69:195-203. 51 . Passingham R. Perry H, Wilkinson F. Failure
Chapter Seventeen Abnoral\l UPPER Extremity' MANIPULATION Control 415
to develop a precision grip in monkeys with recruitment and health and dis-
plasticitv' in
unilateral neocortical lesions made in infancy. ease. Karger: Basel,1981:331^346.

Brain Res 1978;145:410-414. 60. Fromm C, Evarts E. Relation of motor cortex
52. Kuypers HG. Corticospinal connections: neurons to precisely controlled and ballistic
postnatal development in rhesus monkey. Sci- movements. Neurosci Lett 1977;5:259-265.
ence 1962;138:678-680. 61. Johansson RS, Westling G. Roles of glabrous
53. Hecaen H, de Ajuriaguerra J.Etude des trou- skin receptors and sensorimotor memory in
bles toniques, moteurs et vegetatifs et de leur automatic control of precision grip when lift-
recuperations apres ablation limitee du cortex ing rougher or more slippen' objects. Exp
moteur et premoteur. Congres des Medecins Brain Res 1984;56:550-564.
Alinenistes et Neurologistes 1948:269-274. 62. Johansson RS, Westling G. Signals in tactile
54. Lough S, Wing AM, Eraser C, Jenner JR. afterents from the fingers eliciting adaptive
Measurement of recover)' of fiinction in the motor responses during precision grip. Exp
hemiparetic upper limb following stroke: a Brain Res 1987;66:141-154.
preliminarv' report. Human Movement Sci- 63. Jeannerod M, Michel F, Prablanc C. The
ence 1984;3:247-256. control of hand movements in a case of hem-
55. JeannerodM. Mechanisms of visuomotor co- ianaesthesia fiUowing a parietal lesion. Brain
ordination: a study in normal and brain-dam- 1984;107:899-920.
aged subjects. Neuropsychologia 1986; 64. Wing AM, Frazer C. The contribution of the
24:41-78. thumb to reacJiing movements. Q J Exp Psy-
56. Eliasson AC, Gordon AM, Forssberg H. Ba- chol 1983;35A:297-309.
sic coordination of manipulative forces in 65. Mack L, Verfaellie M, Rothi LJG,
Poizner H,
children with cerebral palsy. Dev Med Child Heilman KM. Three-dimensional computer-
Neurol 1991;134:126-154. grapic analysis of apraxia. Brain 1990;
57. Cole KJ, Abbs JH Tuner GS. Deficits in the 113:85-I0L
production of grip forces in Dov\n's syn- 66. Schwartz MF, Reed ES, Montgomery M,
drome. Dev Med Child Neurol 1988; Palmer C, Mayer NH. The quantitative de-
30:752-758. scription of action disorganization after brain
58. Rothwell JC, Obeso JA, Day VL, Marsden demage: a case study. Cogniti\'e Neurops\'-
CD. Pathophysiolog\' of dystonias. In: Des- chology 1991;8:381^14.
medt JE, ed. Motor control mechanisms in 67. Luria AR. Higher cortical fijnctions in man.
health and disease. New York: Raven Press, NY: Basic Books, 1966.
1983:851-864. 68. Norman DA, Shallice T. Attention to action:
59. Hallet M, Marsden CD. Physiolog\' and willedand automatic control of behavior. In:
pathophysiology of the ballistic movement Davidson RJ, Schwartz GE, Shapiro D, eds.
pattern. In: Desmedt JE, ed. Progress in clin- Consciousness and self-regulation, vol 4.
ical neurophysiology, vol 9. Motor unit t^'pes. New York: Plenum, 1986.
Chapter 18

THE Patient with Upper
Extremity Manipulatory
Dyscontrol
Introduction Transition to Treatment
Assessment Short-Term Goals
Functional Assessment of Upper Extremity Long-Term Coals
Control Treatment
Tests of Activities of Daily Living Is Proximal Control a Prerequisite for
Standardized Tests of Manipulation and Retraining Hand Function?
Dexterity Treating at Impairment Level
Jebsen Hand Function Test Reducing Musculoskeletal Impairments
Purdue Pegboard Test Sensory Reeducation
Minnesota Rate of Manipulation Retraining Key Components of Upper
Test Extremity Control
Physical Capacity Evaluation Retraining Eye-Head Coordination
Assessing Key Components of Upper Retraining Transport Phase of Reach
Extremity Control Biofeedback and Functional Electrical
m Locating a Target —
Eye-FHead Stimulation
* Coordination Adapted Positioning
Eye-FHand Coordination Retraining Task-Dependent
Reach and Grasp Characteristics of Reach
Transport Phase of Reach Retraining Grasp
Grasp Retraining Task-Dependent Changes
In-Hand Manipulation Skills in Grasp
Postural Control Attending to Relevant Perceptual
Assessing Planning and Sequencing Cues
Activities of Daily Living Release
Assessing Impairments Affecting Upper In-Hand Manipulation Skills
Extremity Function Postural Control
Range of Motion Retraining Problems in Planning and
Strength Sequencing of Activities of Daily Living
Abnormal Synergies Skills
Sensation Learned Disuse
Edema and Pain , ,_ Summary , . , -. . .^ •>'-<^^^^^
Introduction confined to a fixed synergy, either flexion or

extension, she is now able to move outside the
Mrs. Poirot, our pataent from Chapter context of these fixed coordinative patterns.
15, has been referred for therapy to improve Arm fiinction in the hemiparetic arm has be-
her upper extremity manipulator}' control. gun to return, and she has developed the abil-
Remember that 6 weeks ago she had a stroke ity tomove her arm, though movements are
in the left hemisphere, and has some residual weak and dyscoordinated. She is also begin-
right hemiparesis, with mild spasticity. While ning to show some ability to coordinate finger
in the past, movements of her right arm were movement for grasp. She also appears to have
417
418 Section rV UPPER EXTREMITY CONTROL
subtle but distinct problems in coordinating levels, including (a) objective measurement of
reach and grasp in her nonhemiparetic arm. fijnctional upper extremity' skills, (b) evalua-
Mrs. Poirot has sensor>' and perceptual tion of the key components that form the ba-
problems as well. She has a moderate hemi- sis for functional skills, and (
c) quantification
anopsia, decreasing the peripheral visual field of the underlving sensory, motor, and cog-
on the right. She has decreased cutaneous and nitive impairments that constrain perfor-
proprioceptive sensory hinction in the hemi- mance of skills and 2. In addition,
at levels 1
paretic arm. Mrs. Poirot's stroke occurred in a thorough assessment must include a review
the left cerebral hemisphere; as a result, she of the patient's medical and social history', as
has mild aphasia, affecting primarily her ability well as a review of current symptoms and con-
to verbally express herself. In adciition, she is cerns.
showing ADL performance errors, which sug- Since no one test will measure all levels
gests that she has problems related to ide- of flmction, the clinician must assemble a bat-
ational apra.xia. tery of tests that best meets the needs of the
This combination of sensor)- and motor t\'pe of patient being evaluated, while exam-
impairments will likely aft'ect all aspects of up- ining performance in each of these areas (1-
per extremity' control, including eye-head co- 4). A three-level assessment allows the clini-
ordination for locating a target, reach, grasp, cian to answer the following questions:
release, manipulation, and the use of antici-
patory postural adjustments before upper-ex- 1 To what degree can the patient perform
tremity movements. fijnctional tasks.'
This chapter presents a task-oriented ap- 2. What strategies does the patient use to
proach to assessment and treatment of the pa- perform the tasks, and can they adapt
dent with upper extremity' manipulator^' con- changing task conditions?
strategies to
trol problems. As we noted in Chapter 5, a 3. What are the senson', motor, and cog-
task-oriented approach reflects a conceptual niti\'e impairments that constrain how
frame\\'ork for clinical pracuce that incorpo- they perform the task, and can these im-
rates four key elements: the clinical decision- pairments or strategies be changed
making process, hypothesis-oriented clinical through intervention, thereby improv-
practice, a model of disablement, and a sys- ing the level of functional performance,
tems theor}' of motor control. An important or is the patient performing optimally,
part of the task-oriented approach is the abil- given the current set of impairments?
ity' to generate multiple hypotheses about the
potendal causes of dysftmcdon in the padent, The information gained through assess-
and systematically test those hypotheses in or- ment used to develop a comprehensive list
is
der to refine one's understanding of the prob- of problems, establish short- and long-term
lems contributing to loss of function. goals, and formulate a plan of care for retrain-
It is important to remember that the de- ing upper extremit)' control.
velopment of clinical methods such as a task-

oriented approach based on a systems theor\' Functional Assessment of Upper
of motor control is just beginning. As sys- Extremity Control
tems-based research provides us with an in-
creased understanding of normal and abnor- A task-oriented approach to evaluating

mal upper extremit)' control, new methods for upper extremity' ftmction begins with a flmc-
assessing and treating problems affecdng up- tional assessment to determine how well a pa-
per extremit}' fijnction will emerge. tient can perform a variety of skills that de-
pend on control of the upper extremity'. A
Assessment fiinctional assessment can provide the clini-
cian with information on the patient's level of

A task-oriented assessment of upper ex- performance compared to standards estab-
tremitv control evaluates behavior at three lished with normal subjects. There are a num-
Chapter Eighteen ASSESSMENT AND TREATMENT 419
ber of approaches to assessing functional per- these tasks require other key components of
formance related to upper extremity control. upper extremit)' fimction, such as eye-head
Tests can be categorized into ADL scales, coordination, postural control, and arm trans-
physical capacity evaluation tests, and tests port. When evaluating patients with apparent
that examine hand dexterity and manipula- keep in mind that of-
unilateral involvement,
tion. body evidence problems
ten both sides of the
with upper extremity control. Hence, the
clinician should assess both sides. Some
TESTS OF ACTIVITIES OF DAILY
commonly used tests of hand dexterity and
LIVING
manipulation skills for adults are detailed
below.
Standardized Activities of Daily Living
(ADL) scales test bathing, dressing, groom-
Jebsen Hand Function Test
ing, toileting, feeding, mobility', and conti-
nence. Examples of ADL scales include the
The Jebsen Hand Function Test was
Katz Index (5), Functional Independence
one of the earliest objective standardized tests
Measure (FIM) (6), PULSES and
profile (7),
of hand function (13). It contains seven timed
the Barthel Index (8).
subtests: writing, card turning, picking up
Standardized Instrumental Activities of
small items, simulated feeding, stacking
Daily Living (lADL) scales also offer an ap-
checkers, picking up light cans and heavy cans
proach to assessing upper extremiU' function
(Fig. 18.1). The Jebsen-Taylor test requires
by examining skills that require environmen-
that both hands be tested, with the nondom-
tal interactions, such as telephone usage, trav-
inant hand tested first. The test is relatively
eling, shopping, preparing meals, housework,
quick to administer (10 to 15 minutes), and
and finances.
uses inexpensive and readily available materi-
als. In addition, it has established norms
STANDARDIZED TESTS OF which a
against patient's performance can be
MANIPULATION AND DEXTERITY compared (13, 14). The Jebsen-Taylor gen-
erally has excellent test-retest reliabilit)', with
One of the most common approaches to the exception of v\ riting and feeding subtests,
evaluating upper extremity fijnction focuses which tend to show practice effects (15).
on the assessment of manipulation and dex- The purpose of the Jebsen-Taylor test is
terity. Manipulation refers to general move- to assess hand functions that are common to
ment of an object in space, or with reference many ADL tests. Studies have found a mod-
to another object. It can also be used to refer erate correlation between scores on the Jeb-
to in-hand manipulation of an object, such as sen-Taylor test and ADL ability as measured
when one modifies or adjusts an object while by the Klein-Bell ADL Scale (16). However,
it remains in the hand (10). Manipulation re- the correlations have not been high enough
quires that hand and finger fiinction adapt ac- to warrant substitution of the Jebsen for an
cortiing to the physical and spatial properties ADL test (16).
of objects (II).
Manipulation typically involves several Purdue Pegboard Test
categories of skills, including tool-use skills
(pencils, pens, scissors), dressing (buttoning, The Purdue Pegboard test is another
t^'ping), eating skills (use of a knife, opening testof finger manipulation and hand dexterity
containers), and other skills such as money (17). This test is a time-based measure of dex-
handling (12). Performance of these skills re- terity, requiring the placement of pins into
quires a variety of hand movements with ref- holes, or the assembly of a group of pins,
erence to an object, including pushing, pull- washers, and collars. There are four subtests
ing, shaking, and throwing, transferring, and that examine prehension in the right, left, and
releasing. In addition to hand movements. both hands, and a bimanual assemblv task.
Figure 18.1. Jebsen-Taylor Hand Function Test. Two items from this seven-item test are shown, A, lifting light cans,
and B, stacking checkers.
Thus, the test scores dexterity in each hand trol. A Physical Capacity Evaluation involves
tested separately, as well as bimanual dexter- a comprehensive assessment batter)' of the pa-
it)'. The patient's performance is compared tient's functional and physical abilities, and is
against standardized normative data. Like all used to determine the patient's abilit)' to re-
the timed tests, the Purdue does not evaluate turn to work. A detailed discussion of PCE is
the cause of impaired prehension; it only doc- beyond the scope of this chapter, but a brief
uments that an impairment exists. review of some of the available tests that eval-
uate upper extremity control is presented.
Minnesota Rate of Manipulation Test A number of commercially available

standardized tests examine upper extremit)'
function related to physical work. The Valpar
The Minnesota Rate of Manipulation
Work Samples (Valpar Work Samples, Valpar
Test (MRMT) (American Guidance Sendee)
Corporation, Tucson) is a standardized test
is a standardized test of manual dexterit)', and
that uses 19 v\ork samples to evaluate reach-
contains five subtests: the placing test, turning
ing, handling, manipulating, and feeding
test, displacing test, one-hand turning and
placing test,and tvvo-hand turning and plac-
( 18). The BTW Work Simulator (Work Sinv
ulator, Baltimore Therapeutic Equipment
ing test (18). These are timed tests of dexter-
ity that require the subject to manipulate
Co., Baltimore, MD) upper ex-
also evaluates
tremit)' function through the use of a me-
blocks and place them into a series of holes.
chanical device with 18 different tool attach-
The placing and the turning tests are the most
ments. Thus, the patient's abilit)' to handle
commonly administered of the five subtests.
and manipulate tools is quantified and perfor-
The test is standardized, and norms are avail-
mance compared to normative data (19).
able.
The advantage of fiinctional perfor-
mance-based assessments, whether ADL,
PHYSICAL CAPACITY EVALUATION PCE, or hand dexteiit)', is their abilit\' to
quantiiy' functional performance and compare
Physical Capacity Evaluation (PCE), al- it to established norms. Two limitations of
ternatively referred to as a fijnctional capacity' this t)'pe of testing include (a) tests do not
or work skills evaluation, is another approach assess how a patient performs the task, that is,
to evaluating fimctional upper extremity' con- the qualit)' of movement used; and (b) the
Chapter Eighteen Assessment and Treatment 421
tests do not provide insight into why a patient problems in key components and suggest di-
is unable to perform a functional skill, that is, rections for clinical intervention.
into the underlying impairments that con-
strain fiinction. This latter point is particularly LOCATING A TARGET— EYE-HEAD
significant, since a major focus of treatment is COORDINATION
directed at tndng to resolve underlying im-
pairments and helping patients to recover Research has shown that when an object
strategies that will enable them to perform to be grasped appears in the peripheral visual
the key components of upper extremit)' field, there is normally a predictable sequence
control. of movements. The eyes reach the target first,

then the head, and finally the arm. Locating
Assessing Key Components of an object to be grasped requires various com-
Upper Extremity Control binations of eye, head, and trunk motion, de-
pending on the location of the object with
In previous chapters, we described the reference to the visual field (20). Apparently,
second level of a task-oriented assessment as these task-specific movements of the eye,
a strategy assessment. A strategy assessment head, and/or trunk are controlled by different
is a qualitative examination of the way in systems. Thus, evaluating eye-head coordi-
which a subject coordinates movements to ac- nation underlying the abilit)' to locate an ob-
complish specific tasks. It also includes strat- ject to be grasped requires the assessment of
egies for organizing sensory/perceptual in- three systems: eye movements to a target close
formation underlying a particular task. In this to the central visual field; eye and head move-
chapter, strateg)' assessment refers to the eval- ments to a target located in the periphen,'; and
uation of the key elements of upper extremit)' eye, head, and trunk movements to locate a
control that underlie a broad range of fimc- target, which is even more eccentric (20).
tional skills. Ideally, assessment of upper ex- A clinical approach to assessing eye-
tremity control would evaluate these key ele- head coordination has been proposed based
ments separately. Thus, standardized tests on this understanding of the separate systems
would allow the evaluation of the following controlling eye-head movements (21, 22).
key components in patients with upper ex- Eye movements used to locate targets pre-
tremity dysfunction: (a) eye-head coordina- sented within the central visual field are as-
tion, (b) postural control (weightbearing on sessed The patient is asked to keep the
first.
arms for support, as well as postural adjust- head still and move only the eyes. Both sac-
ments in the legs and trunk, which support cadic eye movements to fixed targets and
reaching movements), (c) arm and hand smooth pursuit eye movements used to track
transport, (d) grasp and release (fine pincer moving targets are tested. This is shown in
and power grasp), and (f) manipulation skills, Figure 18.2. The patient's abilit)' to locate the
including in-hand manipuladon as well as bi- target with the eyes and maintain a stable gaze
lateral manipulation skills. In addition, tests on the target for 15 seconds is assessed (22).
would assess the patient's ability to aciapt Eye movements are graded subjectively ac-
these components to changes in task and en- cording to a three-point scale: intact, im-
vironmental demands. For example, they paired, or unable. In addition, any subjective
would assess the abilit)' to adapt grasp to ob- complaints related to blurred or unstable vi-
jects of different size, shape, weight, and tex- sion reported by the patient are recorded. In
ture. assessing our patient, Mrs. Poirot, you might
There is a great need for standardized note that she has difficulty making accurate
assessment tests that isolate and evaluate key eye movements to targets that are presented
components of upper extremity manipulator)' in her right visual field. She also has difficulty
fi.mction. The development of such tests tracking moving targets.
would allow clinicians to isolate and quantify Next, the patient's abilit)' to locate and
422 Section IV UPPER EXTREMITY' CONTROL
Figure 18.3. Testing eye-head coordination. The pa-

tient's ability to make coordinated eye-head movements
to locate a target in the far peripheral visual field is
shown.
abilities, eye-head coorclination may be tested

in standing and during walking as well (22).
Figure 18.2. Testing the patient's ability' to make sac-

cadic eye movements to locate and maintain gaze on a EYE-HAND COORDINATION
target located within the near peripheral field.
Tests of visuomotor coordination eval-

uate the coordination between \'ision and arm
stabilize gaze on targets presented in the pe-
movements. With the growing use of personal
is assessed. This is shown
ripheral visual field
computers, many tests have been computer-
ized, using a joystick and computerized xddeo
in The coordination of com-
Figure 18.3.
graphics to examine the ability to track still or
bined eye-head movements is graded subjec-
tively using the three-point scale just pre-
moving visual targets presented on a com-
sented. Patients should be able to localize a puter screen (23). Computerized visual track-
ing tests allow the objective quantification of
target with the eyes and inaintain a stable gaze
on that target while the head is moving. We errors in eye-hand coordination. Unfortu-
note that Mrs. Poirot is able to make coor-
nately, they do not distinguish between errors
dinated eye-head moxements to targets lo-
from an inabilit\' to visually locate
that result
cated in her and maintain a stable gaze on the target and

left peripher}', but is unable to
coordinate eye-head movements to targets
movement errors in controlling the joystick.
presented on her right.

Finally, the patient's ability to make eye- REACH AND GRASP
head-trunk movements necessary to locate
targets oriented in the far peripher\' is as- Reach and grasp is a sensoPi'-motor skill
sessed. In the case of Mrs. Poirot, you may with higher le\'el adaptive components as
note that she evidences a similar inability' to well. Locating a target in space requires visual
make coordinated eye, head, and trunk move- pathways including areas from the visual cor-
ments to targets presented in her far right vi- tex to the parietal lobe, the so-called "where"
sual field. pathways see Chapter 3 for further informa-
(
Patients are tested initially in the seated tion on these pathways). This information is
position; however, depending on the patient's then used to form an action plan that allows
Chapter Eighteen .Assessment .\n"d Tre.\tment 423
coordinated movements of the eyes, head, pon phase of reach? First, one can look at the
and hand in space. accuracy' of the transport phase itself. In ad-

The motor plan is task-dependent. dition, one can look at the duration of trans-
^Mien the task invoh es pointing, the arm and port, determining the time it takes for the pa-
the hand are controlled as a unit. In contrast, tient to get to the target. Yet, the greatest
when the task involves gxasping an object, the insight comes from obser\ing the movement
arm that is in\olved in the transport part of profile itself, including the relative coordina-
the moxement is controlled separately from tion among the segments of the upper ex-
the hand, which is invohed in the grasp por- tremity making the mo\ement.
tion of the movement. While these t^vo com- What modifications would you expect
ponents are controUed separately, the\- are to see in Mrs. Poirot's transport phase of a
linked spatially and temporally. reaching movement w hen she performs vari-
Grasp formation begins during the ous t\pes of tasks.- Normally, when \ou grasp
transport phase of movement. We are not and lift an object, the movement speed will be
aware of an\- standardized tests that evaluate faster than if you simply point at an object,
and dociunent specific movement strateg\- since the momentum of the transport phase
problems related to the organization and co- can be used to assist the lifting of the object.
ordination of the different components of You should ask yoiu-self does the patient con-
movements involved in reach and grasp. The tinue the momentum diuing the entire reach-
following sections offer suggestions to help ing movement? Does the patient slow the
the clinician think about ways to assess move- mo\ ement unnecessarily in order to grasp the
ment strategies used in reach and grasp. We object prior to the lift? During a reach and
too await the development of standardized grasp task, invohing a stationan' target, does
systematic assessment tools for assessing reach the patient show clear acceleration and decel-
and grasp. eration of the movement profile, slowing the
movement appropriately in order to grasp the
Transport Phase of Reach object rather than knocking it over?
To assist in improxing the abilit\- to an-
Remember from pre\"ious chapters that alyze the different characteristics of the trans-
reaching istask-dependent. Thus, the char- movements may
port phase, \ideotaping the
acteristics of the mo\ement for example, how-
( be Mdeotaping allows the clinician to
helpfiil.
fast one moves or how straight the trajectory- \iew the movement over and o\"er, analyzing
is) depend on the nature of the task being per- its various components. In addition, the abil-
formed. Therefore, when assessing the trans- it\- to slow the \ideotape and stop the film as
port phase of reaching, the clinician will want necessar\- can facifitate the clinician's abilit\- to
to use a variety' of tasks. For example, tasks understand the movement profile. Use of \id-
could include pointing at a target, reaching eotaping can be helpful to understanding the
forand grasping a target, or reaching, grasp- movement, but it is time-consuming, and for
ing, and lifting a target. Though these tasks that reason is not often used on a routine basis
use a varier\' of grasp components, onlv the in the clinic.
transport phase will be observed at this point. Remember from Chapter 15 that there
In addition, one will want to locate the is between mo\ement speed,
a relationship
targets at different places relati\"e to the pa- cUstance, and accuracy; this is represented by
tient'sbody, for example, ipsUateral to the law (24). Thus, when subjects are asked
Fitts"
reaching arm vs. on the contralateral side of to move with greater accuracy, they slow
the body, and close to the patient vs. far away down. Clinicians who are assessing task-de-
(within arm's reach or requiring forward lean pendent movements involved in reaching
of the trunk). should examine the relationship between the
WTiat should the clinician be looking at patient's movement speed and accuracy'. For
in order to assess the efficiency' of the trans- example, one approach to examining this
trade-off benveen speed and accuracy would objects of different orientation, gripping ob-
be to have the patient hold a pencil and time jects of different sizes and shapes, gripping
how long it takes the patient to move the pen- objects of different weights, gripping and lift-
cil point back and forth 10 times between two ing objects, vs. gripping and throwing ob-
circles, drawn on a piece of paper. One could jects, vs. gripping and fitting objects into
tell the patient to move as fast as possible holes. Finally, the clinician might want to as-
while attempting to place the pencil point sess the patient's abilit)' to reach and grasp not
within the circle. Accuracy is determined by only stationan,' objects but those that are
the number of dots outside the two circles. moving at various speeds.
The difficult)' of the task can be \'aried by how What should the clinician watch for
large the circles are, and how far apart they when assessing grasp? It will be important to
are. This gives the clinician insight into the obser\'e the orientation of the hand and the
patient's ability to make accurate movements shape of the fingers relative to the thumb dur-
and how he or she improves performance ing the reach. One should ask: At what point
with to speed and accuracy over
respect in the movement trajectory' does the hand
time. We
might anticipate that Mrs. Poirot open maximally, and then begin to close in
will show impairments in movement time anticipation of a grasp? It is also important to
and accuracy in both her arms, with defi- obsene the position of the thumb during the
cits more pronounced in the hemiparetic course of the movement. Does it remain in a
arm. stable position during the course of the move-
Finally, during the process of assessing ment, thereby serving as a reference point for
reach, the clinician will want to look at the the reach and grasp?
patient's abilit)' to use visual information to Remember that grip formation takes
correct the accuracy of a reach. One approach place during the transportation phase and is
could use prism glasses to offset the visual anticipatory of the characteristics of the object
image that the subject sees. During the course to be grasped. Some problems you might ex-
of a reach, once the hand comes into view, pect to see in patients who have impaired
the person perceives the error between the grasp formation are (a) an absence of antici-
hand position and the perceived target posi- pator*' hand shaping during the transport
tion and uses visual cues to update the move- phase of the movement, (b) a hand that does
ment path to reach the target accurately. Us- not close appropriately in relation to the ob-
ing this tii'pe of test, we would expect that ject to be grasped, or alternatively closes too
Mrs. Poirot would be unable to use visual soon, or (
c) an inability' to alter the hand shape
feedback and therefore would be unable to to accommodate objects of different shapes
correct the movement path, thus making in- and sizes.
accurate reaches. Questions that should be asked con-

cerning the patient's performance are: What
Grasp t\'pe of grip does the patient use? Does the
patient van*' the number of fingers used dur-
The same t)'pes of tasks that were used ing grasp, depending on weight and size char-
to evaluate the transport portion of a reach acteristics of the object to be lifted? Does the
and grasp movement can be used to assess patient show errors in slip or in pushing the
grasp. Now the clinician will focus on evalu- object to be gripped into the surface? Our pa-
ating the characteristics of the grasp portion tient, Mrs. Poirot, makes frequent errors in
of the movement. Again, videotaping a pa- programming grip and lift forces, as evi-
tient's performance will assist the clinician in denced by the fact that objects frequendy slip
identifi'ing problems related to hand orien- from her grasp when she is trying to lift. Also,
tation and grasp formation. In addition to the she sometimes uses too much force when lift-
tasks mentioned earUer, the clinician might ing a paper cup, causing compression of the
want to structure tasks that include gripping sides of the cup and spilling of the liquid in-
side.This suggests errors in the programming example, lifting a light object slowly while
of grip and lift forces. standing is less destabilizing to the body's po-
sition in space than lifting a heavy object
IN-HAND MANIPULATION SKILLS quickly. For ftirther ideas, refer to the chapter
on clinical assessment of postural control.
In-hand manipulation is defined as the
process of adjusting objects within the hand ASSESSING PLANNING AND
after grasp. The In-Hand Manipulation Skills SEQUENCING OF ACTIVITIES OF
(TIME) test evaluates in-hand manipulation DAILY LIVING
skills based on Exner's Classification system
(10). This test examines the following cate- As you remember from Chapter 1 7, def-
gories of in-hand manipulation skills: (a) icits in executive control can impair a patient's
translation, that is, moving an object from the ability' to carr)' out fijnctional ADL activities.
fingers to thepalm and back, as when picking This causes the patient to produce frequent
up a coin and moving it to the palm; b) shift, ( errors resulting from misuse of objects, to
defined as adjusting the position of an object perform actions out of sequence, and to per-
held near the DIP joints of the fingers with severate on tasks. Tasks that require planning
the thumb opposed, for example, moving a and organization over time are most fre-
pen so it is held closer to the point for easier quently affected. These types of problems
writing; (c) rotation (simple or complex) in- have been referred to as frontal apraxia by Lu-
volving rotating an object, stabilizing, and ria (25-28).
then moving the object, for example, turning It has been suggested that to evaluate
a paper clip so it can be used. these types of performance errors, you need a
descriptive theory that will allow you to pick
POSTURAL CONTROL out units of action and define their properties
at different levels of organization. Such a de-
Examining the ability of the patient to scriptive theory of action has been suggested,
control the body's position in space during and is the basis for an action coding system
the execution of upper extremity movements that allows the identification and documen-
is an important part of assessing upper ex- tation of performance errors associated with
tremity manipulator)' control. There are sev- ADL (26).
eral ways one could assess the relationship be- This action coding system takes com-
tween postural control and manipulation monly performed ADL and breaks them
skills. One approach would be to keep the ma- down into their constituent parts, hierarchi-
nipulation task constant and var^' the postural cally organizing them and identifying those
conditions. For example, one could ask a pa- units that are critical to achieving the goal
tient to reach for a target while sitting in a (26).
chair with support, vs. a chair with no back At the lowest level of the action hierar-
support, vs. standing on a firm surface or a chy are what the authors refer to as Al level
moving surface. The clinician would observe and Al crux level acts. Multiple Al level acts
the patient's ability' to maintain stabilit)' and are then combined into A2
which level acts,
the patient's ability to maintain the efficiency are basic task subgoals. For example, when
and accuracy of the reaching task, as postural sugaring a cup of coffee to drink, opening the
conditions become more stringent. sugar is considered an Al level action, pour-
An alternative approach to examining ing the sugar into the coflTee is considered a
the relationship between posture and upper critical or Al crux level of action, and stirring
extremity control would be to maintain a con- the coffee is another Al action. The aggre-
stant postural condition, such as standing un- gation of all these level Al's form the basic A2
supported, and to vary the destabilizing char- task subgoal of putting the sugar into the cof-
acteristics of the upper extremity task. For fee (26).
426 Section W UPPER EXTREMITY" CONTROL
This t\pe of acrion coding system allows tion can be documented using goniometric
of the t\pe and frequena' of er-
identificarion recordings. Decreased motion is often an in-
rors when performing a task. For example, a direct effect of neurological lesion, associated
patient might attempt to pour sugar into his with paresis, mass flexion patterns, or spasti-
coffee prior to opening the sugar packet. Or, cit\'.
the patient might stir hot water with his spoon The American Societ\' for Surger\' of the
prior to putting in the coffee. Or, the patient Hand (29) recommends a composite measure
could reach for the sugar and get distracted to represent joint motion of the fingers and
by the butter, which is nearby, and does not thumb. Total motion T.\M is the sum
active ( )
satistx' the Al crux, or the goal of the task of active flexion measurements of the meta-
(26).' carpophalangeal, proximal, and distal inter-
\\Tien applying this t\pe of analysis to phalangeal joints, minus the active extension
Mrs. Poirot's ADL skills, we might expect to deficits of the same joints. Total passive mo-
see frequent errors in sequencing her perfor- tion (TPM) is calculated in the same way, us-
mance and to be frequently distracted and un- ing passive measurements. Total motion mea-
able to frilfill the goal of the task. sures provide a single value that represents the
This approach to e\aluating ADL per- motion of a digit. This approach
total active
formance errors is now under development. to measurement is discussed more fiiUv else-
It offers great potential for clinicians in iden- where (2).
tifiingand documenting higher level prob-
lems affecting the planning and execution of STRENGTH
upper extremity' tasks.
The abilit\' to generate force is ftinda-
Assessing Impairments Affecting mental to moving the arm for upper extremit\'
Upper Extremity Function fiinction. Generally, the ability' to generate
force is evaluated through manual muscle
The third level of assessment of upper testing, which examines strength during a
extremit)' control examines the sensor\', mo- concentric voluntar\' contraction. As de-
tor, and cognitive subs\'stems and processes scribed in previous chapters, strength testing
involved in the generation of task-specific in the patient with an upper motor neuron
movement. These include measurement of lesion is contro\ ersial (review Chapters 10
musculoskeletal and neuromuscular limita- and 14).
tions, including range of motion, strength, .\mong the concerns related to strength
spasticity', mass patterns, assessment of sen- testing in patients with neurological disorders
son,- impairments, and evaluation of cognitive is whether the abilit\' to generate force during
and perceptual limitations. Since assessment a manual muscle test accurately predicts the
of these underhing systems has been dis- muscle's capacit}' to fiinction properly during
cussed in detail in previous cUnical chapters, a task-dependent movement. While there is
only a brief review of strategies for assessing no universal agreement on many of the issues
impairments specific to upper extremity' con- related to strength testing, it continues to be
trol is presented. considered an essential part of an evaluation
of upper extremity- fijnction. It is particularly
RANGE OF MOTION important to evaluate strength of both the ex-
trinsic and intrinsic muscles of the hand.
Since mobility' in the hand and arm is Strength of pinch and grasp can be measured
necessar\' to upper extremit)' function, evalu- using manual muscle tests.
ating and documenting limitations in active An objective measurement of grip
and passive joint motion in upper extremin*' strength can be determined using a dyna-
joints areimportant for understanding per- mometer, shown in Figure 18.4 (Jamar Dy-
formance limitations. Limitation in joint mo- namometer, Asimow Engineering Co, Los
Angeles) (30-32). The use of a dynamometer

allows grip strength to be evaluated at various
hand widths. This is important because grip
strength varies according to the size of the
object being grasped (2, 33). Normative ta-
bles for grip fimction are available (34-35).
Dynamometer testing requires that the
patient have sufficient grip strength to grasp
the dynamometer. In cases of very weak grip,
a bulb dynamometer, or a blood pressure cuff
rolled to 5 cm and inflated to 5 Hg, can mm
be used to docuinent grip strength. Change
in the millimeters of mercuiy is recorded as
the power of grip (33).
Electronic pinch meters (Pinch Gauge,
Figure 1 8.4. A |amar dynamometer can be used to ob- B&L Engineering, Santa Fe Springs, CA) can
jectively measure grip strength.
be used to measure strength of pinch (33).
Usually, three types of pinch are assessed: (a)
tip-to-tip (thumb tip to index finger tip), (b)
thumb pulp to index and long finger pulp (3-

point chuck), and (c) thumb pulp to lateral
aspect of index finger (key or lateral grip).
These are shown in Figure 18.5. The mean of
three trials is compared to norms that are spe-
cific to the type of pinch meter used.
ABNORMAL SYNERGIES
In the patient who has suffered a stroke,
a major constraint on the ability' to move the
shoulder, elbow, and hand voluntarily is the
presence of abnormal synergies of moveiTient
(36, 37). Determining the degree to which a
patient is constrained by abnormal synergies
is still considered by many clinicians an im-
portant part of assessing upper extremity con-
trol in stroke patients (38).
The assessment scale shown in Table

18.1 was proposed by Signe Brunnstrom (36)
and is based on proposed stages of recovery
as indicated by the degree to which voluntary
control is constrained by abnormal synergistic
movement. This approach to assessment is the
basis for the more recent test, the Fugl Myer
Measurement of Physical Performance used
to quantif)' movement disorders in the patient
who has had a stroke (37).
SENSATION
Figure 18.5. Lisf ol ,i |)iin li nuit-i lu measure precision
grip strength. Three types of pinch are assessed: A, tip- Sensation testing is a critical part of eval-
to-tip, B, three-point chuck, and C, key or lateral grip. uating hand fimction because degree of sen-
428 Section IV UPPER EXTREMm' CONTROL
Table 18.1. Hemiplegia —Classification and Progress Record. Upper Limb—Test Sitting^
Name Age Date of onset Side affected_
Date
Passive motion sense, shoulder elbow
pron.-supin _wrist flex.-ext._
1. NO MOVEMENT INITIATED OR ELICITED.

2. SYNERGIES OR COMPONENTS FIRST APPEARING. Spasticity developing.
Flexor synergy
Extensor synergy.
3. SYNERGIES OR COMPONENTS INITIATED VOLUNTARILY. Spasticity marked.
FLEXOR SYNERGY Active Jo int Range Remarks
Shoulder girdle Elevation

Retraction
Hyperextension
Shoulder joint Abduction
Ext. rotation
EJbow Rexion
Forearm Supination
EXTENSOR SYNERGY
Shoulder Pectoralis major
Elbov^ Extension
Forearm Pronation
4. MOVEMENTS Hand to sacral
DEVIATING region
l-HUM bAblC Raise arm
SYNERGIES forw.-horiz.
^pasiicty
p,on..supin.
decreasmg îôw at 90'
5. RELATIVE IN- Raise arm

DEPENDENCE side-horiz.
OF BASIC Raise arm
SYNERGIES overhead
:»pasi.aiy
Pron.-suptn.
*'^"'"B elbow extended
6. MOVEMENT COORDINATION NEAR
NORMAL. Spasticity minimal
'From Brunnstrom S. motor testing procedures in hemiplegia: based on sequential recovery stage. Phys Ther. 1966; 46: 357-375.
sibilit}' has been shown to be a valid predictor A hierarchy of sensor)' testing in the up-
of recoven' of hand fiinction (39^1). Sen- per extremit)' has been suggested (2). The
son,' testing includes two-point discrimina- simplest level of sensor\' fiinction is the abiht>'
tion, cutaneous sensation (monofilament to discriminate a single point touch-pressure
test), stereognosis, vibration, proprioception srimulus. Touch-pressure sensibilit)' can be
(or position sense), thermal (head/cold), and assessed objectively by using calibrated nylon
pain (pin-prick). monofilaments (42^3). The use of gradu-
Chapter Eighteen ASSESSMENT and Treatment 429
proved if patients can learn to substitute vi-

sion for impaired tactile sensation (44).
The ability to discriminate textures is
proposed as the next level of sensory /percep-

tual hierarchy. This includes the ability to dis-
tinguish the roughest, smoothest, and most
irregular textures. Finally, at the top of the
proposed senson'/perceptual hierarchy is ob-
ject recognition, representing the most com-
plex of the sensors'/perceptual skills (2).
EDEMA AND PAIN

Upper extremity' edema is a common
problem in many types of patients and is at-
tributed to an inadequate pumping mecha-

nism acting on the venous and lymphatic sys-
tems; this reduces the fiinctional of abilities
thehand by restricting motion. Edema can be

measured using circumferential assessment
(measurement of the circumference of the
hand) or through volumetric assessment (44,
45).
Figure 18.6. The use of graduated nylon monofila-
Volumetric assessment measures the
ments to test touch/pressure sensibility. A, A standard-
ized series of graduated monofilaments is used to test B,
water displaced when a limb is immersed.
touch pressure. There are numerous commercially available
volumeters consisting of a plastic tank with a
dowel centered in the lower third of the con-
ated nylon monofilaments (Fig. 18.6) is container to control the depth of hand immer-
sidered one of the most reliable and valid tests sion. A spout at the top of the container al-
of sensor\' capacity and its relationship to lows the displaced water to collect in a 500-ml
functional abilities. A detailed description of graduated cylinder (46). There are significant
sensor)' testingand norms using graduated differences between dominant and nondom-
monofilaments may be found in a review by inant hands, suggesting that clinicians should
Levin and colleagues (41 ). not compare volume measurement of affected
The next level in the proposed sensory and unaffected arms. Instead, volume of the
hierarchy is tvvo-point discrimination. This impaired extremity should be compared to it-
test examines the patient's abilit)' to detect self over time (44).
two stimuli applied simultaneously to the up- Volumetric assessment of upper extrem-
per extremities. The abilit}' to detect still and ity edema can be carried out in sitting or in
moving stimuli is tested. Reliability can be in- standing, and shows high test-retest reliabil-
creased by using a commercial two-point ity. However, volumes are lower in sitting
discriminator testing instrument (2). Two- than in standing, suggesting the importance
point discrimination tests can be important of maintaining a consistent position during
predictors of recovery of hand function. Stud- testing. Best reliability in testing found
is
ies have shown that an inability to discrimi- when volumetric assessment is performed in
nate between two points applied simulta- the seated position (44).
neously to the fingertips 1 cm apart is Another complication that interferes
poor recovery of hand function
predictive for with the recovery of upper extremity fiinction
(44). Recovery of hand fijnction can be im- is pain. Assessment of pain usually involves
questioning the patient about the location control for reaching movements), (c) arm and
and extent of pain symptoms, determining hand transport, (d) grasp and release (fine
whether pain is constant or intermittent, and pincer and power grasp), (e) manipulation
whether it is present at rest or only when the skills, including in-hand manipulation as well
patient moves. Intensit}' of pain is determined as bilateral manipulation skills. In addition,
by asidng the patient to grade the intensity of short-term goals may be described in terms of
the pain on a subjective scale, for example, on interim steps to achieving independence in a
a scale of to 5 or to 10 (47-49). fimctional task.
In summary, a comprehensive assess-
ment of upper extremity fiinction requires a Long-Term Goals
batter)' of measures that examine performance
at many different levels. Further information As proposed in Chapter 5, long-term

on guidelines for assessment of hand function goals should be objective and measureable,
can be found through the American Societ)' and can be expressed in terms of recovery of
for Surgery of the Hand (29) and the Amer- fijnctional capacity of the upper extremity, re-
ican Society of Hand Therapists (50). lated to either ADL, work, or use of the upper
extremity in posture and mobility tasks.
Transition to Treatment
Treatment
Developing therapeutic strategies to re-
train upper extremity' control in the patient The goals of a task-oriented approach to
v\'ith neurological dysfiinction begins with the retraining the patient with upper extremity'
identification of a comprehensive list of pa- manipulator)' dyscontrol include (a) resolve
tient problems, including both the functional or prevent impairments; (b) develop strategies
limitations, or disabilities, as well as the spe- related to the recover)' of the key components
cific impairments that constrain fianction of upper extremity control, including eye-
(51). As mentioned in Chapter 5, when iden- head coordination, postural control, arm and
tifying impairments, it is important from a hand transport, grasp and release (fine pincer
therapeutic standpoint to distinguish perma- and power grasp), and manipulation skills (in-
nent impairments from those that are tem- hand as well as bilateral skills), and (d) retrain
porary and thus potentially amenable to treat- fimctional tasks, including the capacit)' to
ment. adapt strategies so that functional tasks can be
From a comprehensive list, the therapist performed in changing environmental con-
and patient problems that will
identify priority texts.
become the focus for initial intenention strat- The therapeutic techniques used to re-
egies (52). Thus, a list of short- and long-term train upper extremit)' control will vary, de-
treatment goals that are objective and mea- pending on the particular constellation of
surable are established and a specific problems facing each patient. For example,
treatment plan is formulated for each of the retraining ADL skills in a patient with hemi-
problems identified. plegia may require passive mobilization of
proximal (trunk, scapula, and shoulder com-
Short-Term Goals plex) and distal structures to remediate mus-
culoskeletal The presence of
impairments.
Short-term goals should be described in weakness and neuromuscular dyscontrol may
objective and measurable terms. They may be require clinical techniques to facilitate active
described in terms of resolving impairments movement necessar)' for transporting the arm
and recovery of key components of upper ex- and hand in space. Sensory reeducation may
tremity control, including (a) eye-head co- be used to improve sensibility. For those pa-
ordination, (b) postural control (weightbear- tients constrained by impaired eye-head co-
ing on arm for support, as well as trunk ordination, a program to improve visual lo-
Chapter Eighteen ASSESSMENT AND Treatment 431
cation and gaze stabilization may be control to emerge before working on hand
appropriate. Finally, strategies to retrain grasp fianction, since the two systems controlling
and release capabilit)' in the hand will likely be them are different.
needed.
Treating at the Impairment Level
Is Proximal Control a Prerequisite
The goal of treatments aimed at the im-
for Retraining Hand Function?
pairment level is to correct those impairments
that can be changed, and prevent the devel-
Current research examining the neural
opment of secondary impairments. Alleviat-
basis for reach and grasp has a number of im-
ing underlying impairments enables the pa-
portant implications for clinicians when re-
tient toresume using previously developed
training upper extremit)' control in the patient
upper extremit)' control. When
strategies for
with a neurological lesion. Research on nor-
permanent impairments make resumption of
mal reaching suggests that proximal functions
previously used strategies impossible, new
including posture, arm, and hand transport,
strategies will have to be developed.
are controlled bv different mechanisms from
Treatment strategies aimed at modif)'-
those controlling distal hmctions related to
ing sensor)' and motor impairments were pre-
grasp and release. In addition, studies have
shown that CNS lesions can have a selective
sented in detail in Chapters 10 and 14. A brief
discussion of some treatment suggestions of-
effect on transport versus manipulation as-
ten used in modif)'ing impairments in the up-
pects of upper extremit)' function. Because
per extremit)' are presented below (61-63).
these two aspects of upper extremity fianction
are controlled separately, they may recover at
REDUCING MUSCULOSKELETAL
different rates (53).
IMPAIRMENTS
In addition, the degree to which recov-
ery occurs is dependent on the extent to
An important part of retraining upper
which other areas of the CNS can substitute extremity control is reducing the musculo-
for those parts of the CNS that are injured. skeletal constraints that develop secondar)' to
Proximal functions involving the transport other impairments such as paresis or spasti-
phase and/or stabilit}', may be easily substi- cir\'. Passive and active exercises are used to
tuted by other neural mechanisms. In con- mobilize structures essential to upper extrem-
trast, lesions affecting precision movements of ity control, including both proximal stitic-
the hand may find no substitute in the CNS, tures such as the trunk, scapula, and shoulder
and thus recover)' may be limited. musculature, as well as distal structures in-
On a hopeffil note, research suggests volving the hand and wrist.
that training can help improve hand fianction Many sources describe in detail ap-
despite lesions to areas thought to be critical proaches to mobilizing the trunk, scapula,
to these movements. Cortically controlled and shoulder structures in the patient with a
hand movements require more attention and neurological impairment (37, 61-65). These
active participation of the subject than do au- techniques, however, have yet to be validated
tomatic movements (58, 59). In addition, through controlled research. For example,
cortically induced movements require long Figure 18.7 shows one approach to mobiliz-
periods of training and are very labor-inten- ing musculoskeletal structures in the trunk,
sive (53,60). arm, and hand (63). In this approach, the pa-
Thus, it does not appear that control tient is in the supine position, and rotates the
over proximal body segments is a necessar}' shoulders and hips in the opposite direction
precursor to working on hand function,
distal to lengthen the tmnk (63). This approach is
suggesting that the two can be worked on si- used to elongate trunk, arm, and hand mus-
multaneously, rather than sequentially (53). cles that have shortened because of paresis or
It may not be necessary to wait for proximal spasticity.
432 Section FV UPPER EXTREMITl' CONTROL
Figure 18.7. Counter-rotation between the shoulders

Figure 1 8.8. Passive extension of the wrist and hand to
and hips results in elongation of the trunk and is used to
reduce tightness in the upper extremity.
reduce muscle tightness in trunk muscles.
Elongating tight wrist and hand flexors ing programs, it is important to note that re-
is often considered a necessary part of retrain- search validating the effectiveness of these
ing upper extremit}' control in the patient techniques has not yet been done.
with a neurological Hemiplegic pa-
deficit.
tients who habitually hold the invoh'ed upper SENSORY REEDUCATION

extremity in mass flexion develop tightness of
these structures, which limits return of acti\'e A number of reports have been pub-
movement. Figure 18.8 gives an example of lished that recount approaches to retraining
how the wrist and hand can be brought into sensibilit)' in patients with peripheral and cen-
extension passively in order to lengthen tight tral neural injuries resulting in decreased sen-
muscles (61-63). sation (77-79). How much can fiinctional
Other approaches to remediating mus- sensation be impro\'ed following a peripheral
culoskeletal constraints include the use of or central lesion affecting sensibilitv? The an-
plaster casts, splints, and orthoses to increase swer is: we don't know.
range of motion and mobility of arm and Several authors have recommended that
hand structures (66-72). sensor)' reeducation programs focus on both
The use of splints, whether passive or discriminative and protective senson,' ftinc-
dynamic, will not be covered within the con- tions (39, 78-81). Early training focuses on
text of this book. The reader is referred to the detection and localization of moving and
other texts that discuss splints in detail (64, stationary' light touch stimuU. As patients
66, 72-76). The use of upper extremity' learn to perceive constant and moving touch,
weightbearing activities that stretch hand and sensor\' reeducation focuses on learning to
arm structures has also been suggested as an discriminate size and shape, object recogni-
approach to reducing musculoskeletal limitation, and two-point discrimination. A large
tions (37, 61-63). Upper extremit\' weight- part of the training makes use of higher cor-
bearing activities include having the patient tical fijnctions, including attention, learning,
practice maintaining body weight through an and memor\', to facilitate sensoPi' detection,
extended arm placed to the side (Fig. 18.9^), recognition, and localization (78). Tactile re-
to the back (Fig. 18.95), and to the front of training generally occurs both with and with-
the body (Fig. 18.9C). In addition, weight- out vision.
bearing through an extended arm can be prac- Since the patient with decreased sensi-
ticed in the standing position (Fig. 18.10). bility will not experience discomfort or pain,
While these activities are routinely sug- an important part of sensor\' reeducation is
gested as part of most upper extremity- retrain- teaching patients strategies to protect the
Figure 18.9. Upper extremity weightbearing activities. Retraining upper extremity weightbearing A, to the side; B, to
the rear; and C, forward.
limb with decreased sensation from noxious

and injurious stimuli (80). To protect the
hand and arm from injur)', a series of guide-
lines have been recommended and are sum-
marized in Table 18.2.

It is unclear whether sensory reeduca-
tion teaches patients

ing
it
sensibilit\'
how to use the remain-
to their advantage or
actually alters the physiological basis for
sensation (39). For example, since
whether
it is known
wn
that moving stimuli are more detectable than
stationar\' stimuli, the patient can be taught
to move the hand to achieve a moving stim-
ulus, and thus improve the chances for sen-
sory awareness. Alternatively, vision can be
used to compensate for deficits in tactile sen-
sation; thus, the patient can be taught to look

at the hand when reaching or grasping for an
object (39).
Investigators involved in retraining sen-
sory fiinction report that the capacit)' to adapt
to impaired sensibility is dependent on the pa-
tient's motivation as well as training. Those
patients who were willing to use the impaired
limb were better able to recover function. For
a more complete discussion of methods for
sensory reeducation, the reader is referred to
articlesby Callahan (79) and Bell-Krotoski Figure 18.10. Upper extremity weightbearing can also
and colleagues (39). be trained in the standing position.
Table 18.2. Protective Strategies for Patients nisms underlie the movements of eyes, head,
with Decreased Sensibility in the Upper
and trunk, these systems need to be trained
Extremity"
separately ancH in combination.
1 Avoid exposure to thermal extremes and sharp A progression of exercises for retraining
objects.
eye-head coordination and gaze stabilization
2. Do not use excessive force when gripping a tool or
in patients with vestibular ciysfLinction has
object.
3. Build up small handles in order to distribute force
been proposed by Susan Herdman, a physical
and avoid localized increase in pressure. therapistand David Zee, M.D., fi-om Johns
4. Avoid tasks that require the use of a uniform grip Hopkins University Medical School (21, 82).
over long periods of time. These exercises have been used successhilly to
5. Change tools frequently to alter grip and to rest
retrain eye-head coordination problems in pa-
tissues.
6. Observe skin for signs of stress.

tients with central neurological disorders
7. Treat blisters and lacerations quickly and with care (83).
to avoid infection. This approach is reviewed in Table 18.3
8. Maintain daily skin care including soaking, and oil
and begins with exercises to retrain saccadic
massage to maintain optimal skin condition.
and smooth-pursuit eye movements while the
'Adapted from Brand PW. Management of sensory loss in the
head is still (21, 82). Exercises are progres-
Omer E, Spinner M, eds. Management of periph-
extremities. In: sively given to retrain coordinated eye move-
eralnerve problems. Philadelphia: WB Saunders, 1 980:262-272.
ments in conjunction with head movements
to targets located in the peripheral visual field.
Also practiced are exercises to maintain a sta-
Retraining Key Components of ble gaze on an object moving in phase with
Upper Extremity Control the head. Finally, movements involving eye,
head, and trunk motions are practiced as pa-
A task-oriented approach to retraining tients learn to locate targets oriented in the
involves more than just the resolution of im- far periphery. Exercises are practiced in sit-
pairments when possible. Ideally, patients ting, standing, and walking.

should be guided to recover or develop sen- Research in the field of retraining visual
sory and motor strategies that are effective in perception in patients with central neural le-
performing the key components of upper ex- sions is just beginning. Traditionally, strate-
tremity control. Since research has shown that gies to assist patients with visual field deficits,
key components such as reaching and grasp- such as homonymous hemianopsia, involved
ing are driven by the nature of the task, re- teaching patients to consciously scan the
training these key components must be done space represented by the impaired visual field.
within the context of purposeful tasks. There- Until recently, it was thought that le-
fore, we are combining a discussion of the two sions to the visual cortex resulted in perma-
levels of training, the strategy and the adaptive nent impairments to the visual system. How-
task level, into one section. ever, as described in the previous chapter,
patients with lesions in central visual struc-
RETRAINING EYE-HEAD tures are able to make fairly accurate eye
COORDINATION movements and/or reaching movements to
targets when told to move toward where they
An important part of regaining upper thought the object might be (84-85). Ini-
extremity control is retraining eye-head co- tially, patients were very poor at reaching for
ordination, which is essential to locating and objects in this manner; however, their perfor-
stabilizing gaze on a target or object to be mance improved with practice (20).
grasped. Problems that affect the ability to lo- These studies raise many questions
cate objects and stabilize gaze potentially af- about the potential for retraining visual fijnc-
and precision of reaching
fect the accuracy tion in the patient with impaired visual per-
movements. Since different control mecha- ception due to centra] neural lesions. Re-
Table 18.3. Eye-Head Coordination Exercises for Gaze Stabilization^
Stage I. Eye Exercises

A. Exercises to improve visual following (smooth pursuit)
1 Sit in a comfortable position; do not move your head.
2. Hold a small target (about 2" x 2", like a matchbook cover) containing written material at arm's length in
front of you.
3. Keep your head still.
4. Move your arm slowly from side to side (about 45°). Try to keep the words in focus as you move.
5. Move your arm to the left, then right, then center. Rest for 3 seconds. Repeat 5 times.
6. Move your arm up and down about 30°. Move your arm up, then down, then center. Rest for 3 seconds.
Repeat 5 times.
B. Exercises to improve gaze redirection (saccade)
1 Sit in do not move your head.
a comfortable position;
2. Hold two small targets (2" x
one in each hand, about 2") 1
2" apart in front of you.
3. Move your eyes only from one target to the other.

4. Move right, move left. Stop and rest.
5. Repeat 5 times.
6. Hold the two targets in front of you vertically, above and below the midline. Keep your head still; move your
eyes only from one target to the other.
7. Move eyes up, eyes down. Stop and rest.
8. Repeat 5 times.
Stage Head Exercises
II.
A. Move head, object still

1 Side-to-side movements: Hold at arm's length a small target (like a matchbook). Try to keep the words in clear
focus; move your head slowly from side to side. Move head to the right, move head left, move head to the
center. Rest. Repeat 5 times.
2. Up and down movements: Repeat, but move your head up and down while keeping your eyes on the target
held in front of you. Move head up, move head down, come to the center. Stop and rest. Repeat 5 times.
3. To progress yourself, move your head at faster and faster speeds, until you can no longer read the words.
Repeat using a target that is attached to the wall, 6 feet away.
4. Practice both ( /) and (2) with your eyes closed. You should try to visualize in your mind the target, and focus
on it as though your eyes were open.
S(age Eye-Head Exercises
///.
A. Move eyes and head to stationary objects

1 Side-to-side movements: Hold two small targets (2" X 2"), one In each hand, about 36" apart in front of you.
Move your head and eyes to look at first, one target, then the other. Try to clearly focus on the words on each
target each time you move your head and eyes. Look left, look right, then rest. Repeat 5 times.
2. Up-and-down movements: Hold the two targets in front of you vertically, above and below the midline, about
36" apart. Move your head and eyes to look at first, one target, then the other. Try to clearly focus on the
words on each target each time you move your head and eyes. Look left, look right, then rest. Repeat 5 times.
3. To progress yourself, repeat /) and (2) moving your head at faster and faster speeds, until you can no longer
(
read the words. Repeat, using a target that is attached to the wall, 6 feet away.
B. Move eyes and head and object in phase together
1 Side-to-side movements: Hold a small x 2", like a matchbook cover) containing written
target (about 2"
material at arm's length in front of you. Move your arm and head together from side to side. Try to keep the
words in clear focus while you move your arm and head together slowly from side to side (about 45°). Move
left, move right, move center, and rest. Repeat 5 times.
2. Up-and-down movements: Hold a small target (about 2" x 2", like a matchbook cover) containing written
material at arm's length in front of you. Move your arm and head together up and down. Try to keep the
words in clear focus while you move your arm and head together slowly up and down (about 30°). Move up,
move down, move center, and rest. Repeat 5 times.
3. To progress yourself, repeat ( /) and (2) moving your head at faster and faster speeds, until you can no longer
read the words. Repeat, using a target which is attached to the wall, 6 feet away.
'From Zee DS. Vertigo. Current therapy in neurologic disease. 1985:1-13.

436 Section YV UPPER EXTREMITY CONTROL
search needed to develop new strategies for

is shoulder flexed and the elbow extended (Fig.
retraining visual impairments, and to test the 18.11^). This position minimizes the
efficacy of these strategies on recoven,' of vi- amount of force the patient must generate to
sual localization of targets in space. move the arm actively against gravitv'. In some
cases, graviu' can assist motion. For example,
RETRAINING THE TRANSPORT as shown in Figure 18.115, when the patient
PHASE OF REACH is asked to touch his hand to his nose (or
shoulder or head), gravit\' assists elbow flex-
Transport requires the abilit\' to move ion, while the patient eccentrically activ-
the arm in a coordinated way in all directions. ates the triceps to control the descent of the
It includes transporting the hand to an object hand.
to be grasped, as well as transporting the Other seated activities involve exercises
grasped object to a new location. to regain concentric and eccentric contraction

When a neurological lesion results in pa- of shoulder flexors. The patient is asked to sit
resis and the inabihty to recruit motor neu- with the arm supported on a table (Fig.
rons for active movement, retraining upper 18.12) and lift the arm, then drop the arm
extremity movement control often begins back to the surface. Shoulder horizontal ab-
with therapeutic strategies used to facilitate duction can be practiced by asking the patient
active motion by the patient. Several authors to reach for the opposite shoulder with sup-
have laid out a progression of activities for report given under the elbow as needed.
training arm fiinction primarily in stroke pa- These exercises are based on the as-
tients, which include retraining control of arm sumption that practicing activation of isolated
movements underlying the transport phase of muscles will carr>' over to hanctional tasks, in-
upper extremitTi' fijnction (37, 61-63, 65). cluding transport skills involving the arm and
Most of these suggestions relate to prac- hand. As is true for most clinical inter\'ention
ticing control of isolated joint movements in techniques, therapeutic strategies for retrain-
supine, sitting, and standing. For example, re- ing arm fiinction are based on clinical obser-
training active control of arm movements is vations, and have yet to be validated through
often begun in the supine position with the controlled studies.
Figure 18.11. Early retraining of active arm movement. Use of the supine when retraining active control of
position
the upper extremity can eliminate A, the effects of gravity; or B, alternatively make use of gravity to assist movement.
Chapter Eighteen ASSESSMENT and Treatment 437
improvements later than did the targeted

training group (87).
Adaptive Positioning
Adaptive positioning, particularly mod-

if\'ing seat posture, has traditionally been con-
sidered an important approach to improving
upper extremitii' control in the patient with a
neurological impairment. Adaptive seating

programs are based on three assumptions: {a)
adaptive seating will reduce abnormal muscle
tone; (b) improved muscle tone will improve
the ability to stabilize posture; and (c) in-
creased postural stability will increase the abil-
ity' to control the upper extremity (88-91).
Several studies have examined the effect
of altered seat angles on arm movements in
children with and without cerebral palsy.
While one study reported faster arm move-
ments in cerebral palsy children with a back-
Figure 18.12. Practicing isolated shoulder motion in
rest of 90° (92), most studies have not found
the sitting position.
that seating posture made a difference on im-
mediate reaching movements, as measured
Biofeedback and Functional Electrical through kinematic analysis of upper extremit\'
Stimulation moveinents (89, 93). These results do not
rule out a long-term effect of altered seating
Biofeedback and fiinctional electrical posture on reaching.
stimulation (FES) have been used to facilitate Quite possibly, the effects of positioning
motion in paretic limbs (86, 87). In one are specific to the t\'pe of upper extremity task
study, experimenters compared two different being performed. One study compared the ef-
biofeedback approaches to retraining hjnc- fects of seating and prone standing in subjects
tional control of shoulder, elbow, and the dis- with cerebral palsy on the Jebsen-Taylor
tal upper extremity' in 20 chronic stroke and Hand Function Test (94). This study exam-
six head-injured patients (1 to 7 years postin- ined the effects of positioning on the time re-
jur\') (87). All patients had some abilit)' to in- quired to complete eight simulated fianctional
itiate voluntar\' wrist and finger extension and tasks on the Jebsen-Taylor test, and found
thumb abduction moveinents. Ten patients that some subtests were performed faster in
used a "motor copy" approach, in which pa- the seated position (small objects subtest),
tients attempted to activate muscles in the in- while other subtests were performed faster in
volved side using, as a reference, output from the prone standing position (simulated feed-
the noninvolved extremity. Ten patients used ing). The authors report the most atypical
"targeted training," which required the pa- grasping patterns occurred during the simu-
tient to reduce activity' in a spastic muscle and lated feeding subtest. These results suggest
recruit activity in the antagonist muscle. that the effects of positioning may be task-
The study found that both approaches specific (94).
were equally effective in making changes in
Retraining Task- Dependent Characterisdcs
upper extremity fiinction in patients with a
of Reach
chronic neurological lesion; however, the
timing of those changes was different, since Since the characteristics of the transport
the motor copy group tended to show their phase vary according to the task to be per-
438 Section IV UPPER EXTREMITY' CONTROL
formed, it is important to structure retraining modulation of stiffness in the agonist

so that the patient learns to modify the move- and antagonist muscles around the
ments used to transport the arm and hand in joints (refer back to the discussion of
space in a task-dependent way. The following location programming in Chapter 15).
list oflfers various possible ways to retrain By giving patients tasks requiring loca-
reaching based on research examining the tionprogramming, the clinician can as-
characteristics of transport movements during sist the patient in learning to modulate
upper extremity tasks. It is important to re- levels of stiffness in the upper extremit)'.
member that these suggestions, like other One approach might be to place the pa-
suggestions made throughout this chapter, where he/she could lo-
tient at a table
have yet to be validated through experimental cate the target visually but not be al-
testing. lowed to see his/her hands. The
clinician would determine if they could
1 Since the transport phase of movements still be accurate in locating the target in
such as pointing, reaching, and grasp- space, based on programming stiffness
ing, and reach, grasp, and moving an of the agonist and antagonist muscles.
object have very different movement
characteristics, one cannot train a pa- RETRAINING GRASP
tient in one task and expect that the per-
formance skills will automatically carry Every day we are called upon to handle
over to the transport phase of the other a great variety of objects that varj' in size,
reaching tasks. Therefore, we suggest shape, weight, and texture. Hand fianction re-
that training needs to be specific to each quires the abilit)' to grasp, release, and manip-
of these task t>'pes. ulate objects, as well as the capacit)' to adapt
2. It has been shown that visual feedback how we grasp in response to characteristics of
is important when making corrections the object to be grasped.
during a movement for increased ac- Often, retraining grasp function in the
curacy. Thus, training patients to be- patient with paresis and dyscontrol begins
come proficient in using visual infor- with retraining a power grasp, then moves to
mation to correct ongoing movements progressively more precise grips (95). A
is essential to retraining upper extremit\' power grasp utilizes a symmetrical grasp pat-
control. To do this, the clinician should tern and allows for cylindrical hold on objects.
have the patient practice slower move- When retraining power grasp, patients
ments, drawing the patient's attention are often assisted in molding the hand to the
to visual cues relating hand movement, shape of variously sized cylindrical objects
particularly thumb position to target lo- with a symmetrical finger flexion pattern, with
cation. thumb opposed. Pov^'er grasp is practiced in
3. By asking the patient to move quickly both the vertical and horizontal planes. In ad-
in one motion to targets placed at var- dition, a power grip is critical to holding as-
ious distances, the clinician can assist sistive mobility' devices. This is shown in Fig-
the patient in learning to modulate the ure 18.13.
initial forces needed to move the arm It has been recommended that grasp re-
towards a target. In this way, the patient training progress to teaching patients a suc-
learns to program forces appropriately cession of more precise grips. For example,
for quick and accurate movements. patients are taught to grip using a three-jaw
4. Research also suggests that the ability' to chuck pattern involving the thumb opposed
move to a new position in space without to t\^'o fingers (95). Finally, a pincer grip (ei-
the use of visual feedback is important ther tip-to-tip or lateral), which involves index
when making reaching movements. fingerand thumb opposition, is taught (refer
This can be accomplished through back to Fig. 18.5 to review these grips).
Chapter Eighteen ASSESSMENT AND Treatment 439
Figure 18.13. Retraining power grasp. Power grasp used to pick up object oriented A, vertically; B, horizontally; C,
in power grip on an assistive device for locomotion.
Retraining Task- Dependent Changes in hand. Place a cup in front of you, as shown in
Grasp Figure 18.14A. Now reach for the cup and grasp
It In preparation for pouring water into it. Notice
many of the el-

Research has shown that the orientation of the hand and the movement you
ements of grasp, including how we orient and use to pick up the glass for this task. Now turn the
glass over, as shown in Figure 18.146. Your task
shape our hand and the amount of force we
is identical, to pour water into the glass without
use to grip is preprogrammed, that is, deter-
setting down. Notice how you modify the ori-
it
mined before we even touch the object to be entation of your hand and the movement you use
grasped. Hand orientation, shape, and force to accomplish the task.
characteristics are determined based on our
previous experience with grasping objects, in
Thus, an important part of regaining
conjunction with our ability' to perceive rele-
tiinctional recover^' of upper extremity' con-
vant cues about the object to be grasped.
trol requires learning to modif.' grasp strate-
These two factors are used to program hand
gies for changing task demands. Retraining
shape and force characteristics of grip (96-
the ability' to adapt grasp should address both
98).
motor and perceptual aspects of the task. This
is because recovery of effective grip requires
control over extrinsic and intrinsic muscles of
ACTIVE LEARNING MODULE the hand, as well as the ability' to discriminate
You can see this for yourself. Your perceptual cues critical to preprogramming
task is to pick up a glass and pour wa- hand shape and force.
ter in it from a pitcher held in the other Errors in grasp, including gripping too
440 Section rV' UPPER EXTREMIT\' CONTROL
characteristics that are important in correcdv

programming hand shape and force. Prior to
actually grasping and lifting an object, asking
patients questions about their perceptions re-
garding essential characteristics of an object is
another wa\' to help patients attend to rele-

vant perceptual cues related to the task. For
example, the clinician could ask: "Do you
think that object is hea\y or light.- Is it slipper\'
or not.-"
Though research suggests that retrain-
ing perceptual aspects of grip is important to
the recover)' of control, strategies for such re-
training are just emerging and have yet to be
tested experimentally.
RELEASE
To achieve a fiinctional grasp, patients

not only must be able to grasp, but release
objects. For many patients with neurological
lesions, a power grasp is accomplished by us-
Figure 18.14. Modihing reach to changing task de-
ing a mass pattern of flexion. While this pat-
mands. A, Reaching for a cup when it is oriented right-
tern is successfiil in creating grasp, the patient
side up. B, Changes in orientation of the hand when
reaching tor a cup that is inverted. is unable to actively extend the fingers and
release the grasp without the assistance of the
other hand. Alternatively, a patient may use
loosely and letting objects slip, or gripping wrist flexion to passively extend the fingers
too tightly and crushing objects, result from and thus accomplish release (64, 95).
inappropriate force control. However, it is not Rhoda Erhardt, an occupational thera-

always easy to determine if errors are the result pist, has published an extensive assessment
of poor control over muscles, or alternatively, form that describes a developmental sequence
from errors in perceiving characteristics of the for releasing objects (95). This sequence has
object to be gripped and thus programming been used as the basis for a program to retrain
force incorrectly. release in the adult patient with a neurological

impairment (64). The suggested sequence be-
Attending to Relevant Perceptual Cues gins with learning to release an object that is
externally stabilized. This approach is based

Modifring the shape of the hand and on the obsenation that children learn to re-
the force used to grasp are done in anticipa- lease an object externally stabilized on a sup-
tion of the grasp, based on previous experi- porting surface prior to learning to release ob-
ence and relevant perceptual cues. Thus, an jects in space (95).
essential part of retraining hand fiioction is Thus, patients are taught to release ob-
helping patients relearn the capacity' to dis- jects that are stabilized either by the patient's
criminate perceptual cues, which are critical to other hand, by the therapist's hand, or by a
programming hand shape, and force charac- supporting surface. This is followed by learn-
teristics of grasp. ing to release objects that are not supported.
This could be accomplished b\' helping Patients practice releasing an object using a
patients explore objects both \-isually and with pattern of finger extension with the wrist in
their hands, drawing their attention to object neutral, as opposed to release resulting from
Chapter Eighteen Assessment AND TREATMENT 441
wrist flexion, producing mechanical extension so much related to specific movement coor-
of the fingers. dination and sensory disorders, as it is to
problems in the planning and sequencing of
IN-HAND MANIPULATION everyday acts. The treatment of these types of
An important part of recovery of hand problems is a complex issue and beyond the
fianction is helping patients regain the ability scope of this chapter.
to manipulate objects within the hand itself,
down
Learned Disuse
without setting the object or transfer-
ring it to the opposite hand. A
major constraint on recovery of arm
function may be the unwillingness of patients
to use an impaired upper extremity when the
ACTIVE LEARNING MODULE nonimpaired extremity is available. Thus,
You can see this for yourself. Pick "learned disuse" in both primates and hu-
up a pencil and begin to write. Now, mans often follows sensor)' or motor loss af-
you
alter the position of the pencil so fecting one side of the body (101-103). Re-
can erase. In-hand manipulation skills enable you searchers have known for many years that
to change the position of the pencil in your hand restraining the intact limb in monkeys will
without using the other hand, or setting the pencil force the animal to use the impaired arm
down to alter your grip.
(104, 105). This knowledge has led to the
development of "forced-use" paradigms to
Padents learning to regain in-hand ma- encourage the use of a chronically impaired
nipulation skills practice "intrinsic move- upper extremity (106, 107).
ments," which allow objects of various sizes In these research studies, hemiplegic pa-
and shapes to be moved within the hand itself tients (1 to 5 years poststroke or head-in-
Movements practiced inclucie (a) moving an jured) were required to wear slings restraining
object from the fingers to the palm and back, the nonimpaired arm during waking hours for
called translation, [b) rotating an object end 14 days. In each case, patients spent 6 to 7
over end, referred to as rotation, and (c) ad- hours a day practicing tasks that required up-
justing the position of an object held near the per extremit)' flinction (eating, throwing a
distal joints of the fingers with the thumb op- ball, writing, pushing a broom, manipulating
posed, referred to as sliifl: (99, 100). checkers and pegs).
motor ability was
Results suggested that
POSTURAL CONTROL improved following restraint of
significantly
As noted earlier, postural control is an the noninvolved arm in chronic stroke pa-
important aspect of upper extremity control. tients. Patients involved in the forced-use par-
Postural dyscontrol can be a contributing fac- adigm significantly improved ADL abilities
tor to problems related to upper extremity compared to a control group, and improve-
dysfiinction. Therefore, treatment of postural ments were sustained 1 to 2 years postinter-
disorders is considered a key component of vention(106, 107).
retraining upper extremity control. This topic Forced-use paradigms are still at a re-
was covered in detail in Chapter 10, and the search stage in development. It is not clear if
reader is urged to review concepts related to and how they might be used in retraining up-
retraining postural activity associated with upper extremity control in the patient with an
per extremity movement. upper motor neuron lesion. Criteria need to
be established for patients for whom this ap-
RETRAINING PROBLEMS IN Many
proach might be appropriate. patients
PLANNING AND SEQUENCING are excluded from this type of treatment due
ACTTVITIES OF DAILY LIVING SKILLS
to imbalance, and the need to use the unim-
For many patients, limitation in the re- paired upper extremity for balance control
covery of independence in ADL skills is not and to prevent falls (106).
One important and encouraging aspect separately, and therefore can be retrained si-
of these forced-use studies is the awareness multaneously, rather than sequentially.

that motor improvements are possible even in 6. Hand function requires the ability to grasp, re-
chronically impaired patients who lease, and manipulate objects, as well as the
are 1 to 5
capacity to adapt how we grasp in response
years post-onset.
to characteristics of the object to be grasped.
Many elements of grasp, including hand
Summary shape and force characteristics, are prepro-

grammed using previous experience and rel-
1 Retraining control of the upper extremity is evant perceptual cues about the object to be
important to most areas of rehabilitation in- grasped. Thus, retraining hand function re-
cluding physical and occupational therapy. quires attention to both motor and perceptual
While both areas of therapy retrain upper ex- aspects of the task.
tremity control, physical therapists tend to fo- 7. Sensory reeducation programs focus on sev-
cus on postural and mobility aspects of upper eral aspects of sensory function including dis-
extremity function, while occupational ther- crimination and protective sensory functions.
apists tend to function on ADL aspects, in- It is unclear whether sensory reeducation
cluding the recovery of fine motor hand skills. teaches the patient how to use the remaining
2. A task-oriented approach to assessment of up- sensibility to their advantage, or whether it ac-
per extremity function requires a battery of tually alters the physiological basis for sensa-
tests that measure (a) functional performance, tion. It is clear, however, that the capacity to
either ADL or work-related; (b) key compo- adapt to impaired sensibility is dependent on
nents of control, including eye-head coordi- the patient's motivation as well as training.
nation, posture, transport, grasp and release, Studies have shown that those patients who
and manipulation; and (c) underlying sensory, were willing to use the impaired limb were
motor, and cognitive impairments, including better able to recover function.
range of motion, strength, sensation, volume, 8. A major constraint on recovery of arm func-
and coordination. tion may be the unwillingness of patients to
3. Preparing treatment plans to retrain upper ex- use an impaired upper extremity when the
tremity control requires the identification of a nonimpaired extremity is available. Results
comprehensive list of patient problems, in- from studies examining "forced-use" para-
cluding the functional limitations, or disabili- digms suggest that motor ability can be sig-
ties, as well as the specific impairments that nificantly improved by limiting the hemiple-
constrain function. From this list, short- and gic patient's use of the noninvolved arm. One
long-term treatment goals are established, and important and encouraging aspect of these
therapeutic strategies are developed to meet forced-use studies is the awareness that motor
those goals. improvements are possible even in chroni-
4. A task-oriented approach to retraining upper cally impaired patients who are 1 to 5 years
extremity control seeks to minimize impair- post-onset.
ments while maximizing the patient's capac-
ity for function. Retraining involves the de-
velopment of therapeutic strategies to (a)
References
remediate as many sensory, motor, and cog-
nitive impairments as possible, (fa) generate 1. Smith HD. Assessment and evaluation: an
strategies to achieve the key components of overview. In: Hopkins HL, Smith HD, eds.
upper extremity control, and (c) develop the Willard and Spackman's occupational ther-
capacity to perform functional tasks under a apy, 8th ed. Philadelphia: JB Lippincott,
variety of environmental contexts. This re- 1993:158-165.
quires the development of adaptive capacity. 2. Fess EE. Documentation: essential elements
5. Research suggests that the development of of an upper extremity' assessment batter}'.
control over proximal body segments is not a In: Hunter JM, ed. Rehabilitation of the
necessary precursor to the emergence of distal hand, 2nd ed. St. Louis: CV Mosby,
hand function. Proximal and distal segments 1990:53-81.
of the upper extremity appear to be controlled 3. Bear Lehman J, Abreii BC. Assessing the
hand: issues in reliabilit\' and validity'. Phys of daily living. Occup Ther
activities J Res
Ther 1989;12:1025-lo'33. 1989;5:316-318.
4. Caldwell CB, Wilson DJ, Braun RM, Eval- 17. Tiffin J. Purdue pegboard examiner manual.
uation and treatment of upper extremity' in Chicago: Science Research Associates,
the hemiplegic patient. Clin Orthop Rel Res 1968.
1969;63:69-93. 18. Baxter- Petralia P, Bruening LA, Blackmore
5. Katz S, Downs TD, Cash HR, Grotz RC. SM, McEntee PM. Physical capacity' evalu-
Progress in development of die index of ation. In: Hunter JM, et al., eds. Rehabili-
ADL. Gerontologist 1970;1:20-30. tation of the hand. St Louis: CV Mosby,
6. Keidi RA, Granger CV, Hamilton BB, Sher- 1990:93-108.
win FS. The functional independence mea- 19. Curtis RM, Engalitcheff J Jr. A work simu-
sure: a new tool for rehabilitation. In: Ei- lator for rehabilitating the upper extrem-
sentberg MG, Grzesiak RC, eds. Advances ity'
—preliminary' report. J Hand Surg I98I;
in clinical Rehabilitation, vol 1 . New York: 6^499-510.
Springer- Verlag, 1987:6-18. 20. Jeannerod M. The neural and behavioral or-
7. Granger CV, Albrecht GL, Hamilton BB. ganization of goal-directed movements.
Outcome of comprehensive medical reha- Clarendon Oxford, 1990.
Press:
bilitation: measurement of PULSES profile 21. Herdman S. Assessment and treatment of
and the Barthel index. Arch Phys Med balance disorders in the vestibular-deficient
Rehabil 1979;60:145-154. patient. In: Duncan P, ed. Balance: pro-
8. Mahoney Rl, Barthel DW. Functional eval- ceedings of the APTA Forum. Alexandria,
uation: the Barthel index. Md Med J 1965; VA: APTA, 1989.
14:61-65. 22. Shumway-Cook A, Horak F. Rehabilitation
9. Lawton MP. The fiinctionai assessment of of the patient yvith vestibular deficits. Neu-
elderly people. J Am Geriatr Soc 1971; rol Clin 1990;8:441^57.
19:465-481. 23. Behbehani K, Kondraske G, Richmond JR.
10. Exner CE. Development of hand functions. Investigation of upper extremity' visuomo-
In: Pratt PN, Allen AS, eds. Occupational tor control pertbrmance measures. IEEE
therapv for children. St Louis: CV Mosby, Trans Biomed Eng 1988;7:518-525.
1989. 24. Fitts PM. The information capacity of the
11. Corbetta D, Mounoud P. Early develop- himian motor .system in controlling the am-
ment of grasping and manipulation. In: movement.
plitude of J Exp Psychol 1954;
Bard C, Fleun,' M, Hay L, eds. Development 47:381-391.
of eye-hand coordination across the lifespan. 25. Poizner H, Mack L, Veriaellie M, Rodii
Columbia, SC: University of South Carolina LJG, Heilman KM. Three-dimensional
Press, 1990:188-216. computergrapic analysis of apraxia. Brain
12. Swanson AB, Goran-Hagert C, Swanson, 1990;113:85-101.
GD. Evaluation of impairment of hand 26. Schwartz MF, Reed ES, Montgomery' M,
fianction. In: Hunter JM, Schneider LH, Palmer C, Mayer NH. The quantitative de-
Macidn EJ, Bell JA, eds. Rehabilitation of scription of action disorganization after
the hand. St. Louis: CV Mosbv, 1978:31- brain damage: a case study. Cognitive Neu-
69. ropsychology 1991;8:381-414,
13. Jebsen RH, Taylor N, Trieschmann RB, n . Luria AR. Higher cortical ftmctions in man.
Trotter M], Howard L. An objective and NY: Basic Books, 1966.
standard test of hand function. Arch Phvs 28. Norman DA, Shallice T. Attention to ac-
Med 1969;50:311-319. tion: willed and automatic control of be-
14. Agnew PJ, Maas F. Hand fijnction related havior. In: Davidson RJ, Schwartz GE,
to age and sex. Arch Phys Med Rehabil Shapiro D, eds. Consciousness and self-
1982;63:269-271. regulation, vol 4. New York: Plenum,

15. Stern EB. Stability' of the Jebsen-Taylor 1986.
hand ftmction test across three test sessions. 29. American Society' for Surgery' of the Hand.
Am J Occup Ther 1992;7:647-649. The hand: examination and diagnosis. New
16. Lynch KB, Bridle MJ. Validity' of the Jeb- York: Churchill Livingstone, 1983.
sen-Taylor hand function test in predicting 30. Bechtol CO. Grip test use of dynamometer
444 Section IV UPPER EXTREMm' CONTROL
with adjustable hand spacing. JAMA 1954; stroke. In: Hunter JM, Schneider LH,
36:820-824. Mackin EJ, Bell JA, eds. Rehabilitation of
31. Schmidt RT, Toews J. Grip strength as mea- the hand. St. Louis: CV Mosbv, 1978:505-
sured by the Jamar dynamometer. Arch 520.
Phys Med Rehabil 1970;5:321-327. 45. Stern EB. Volumetric comparison of seated
32. Fike ML,
Rousseau E. Measurement of and standing test postures. Am J Occup
adult hand strength: a comparison of two Ther 1991 ;80 1-805.
instruments. Occupational Therapy Journal 46. Waylen J, Seibly D. A study to determine
of Research 1982;2:43^9. the average deviation accuracy of a com-
33. Fess EE, Harmon KS, Strickland JW, Stei- mercially available volumeter. J Hand Surg
chen JB. Evaluation of the hand bv objective 1981;6:300-313.
measurement. In: Hunter JM, Schneider 47. Partridge CJ, Edwards SM, Mee R, Langen-
LH, Mackin EJ, Bell JA, eds. Rehabilitation berghe HVX van. Hemiplegic shoulder
of the hand. St. Louis: CV Mosbv, pain: a study of two methods of physiother-
1978:70-93. apy treatment. Clinical Rehabilitation 1990;
34. Kellor M, Frost J, Silverberg N, et al. Hand 4:43^9.
strength and dexteritii': norms for clinical us- 48. Roy CW. Shoulder pain in hemiplegia: a lit-
age. Am J Occup Ther 1971;25:77-83. erature review. Clinical Rehabilitation 1988;

35. Mathiowetz v, Kashaman N, Valland G et 2:35^4.
al. Grip and pinch strength: normative data 49. Cailliet R. The shoulder in hemiplegia. Phil-
for adults. Arch Phys Med Rehabil 1985; adelphia: FA Daws, 1980.
66:69-74. 50. American Society' of Hand Therapists. Clin-
36. Brunnstrom S.Motor testing procedures in ical assessment recommendations. Garner,
hemiplegia: based on sequential recover\' NC: American Society of Hand Therapists,
stages. Phys Ther 1966;46:357-375. 1981.
37. Duncan P, Badke MB. Stroke rehabilitation: 51. Internationa] classification of impairment,
the recovery of motor control. Chicago: disabilities and handicaps: a manual of clas-
Year Book, 1987. sification relating to the consequences of
38. Gowland C. Staging motor impairment af- disease. Geneva: World Health Organiza-
ter stroke. Stroke 1990;21(suppl II):I1-19- tion, 1980.
11-21. 52. O'Sullivan, S. Clinical decision making:
39. Bell-Krotoski J. Weinstein S, Weinstein C. planning effective treatments. In: O'SuUivan
Testing sensibilit\', including touch-pres- S, Schmitz T. Physical rehabilitation: assess-
sure, two-point discrimination, point local- ment and treatment. 2nd ed. Philadelphia:
ization,and vibration. J Hand Therapy FA Davis, 1988:1-8.
1993;2:114-123. 53. Pehoski C. Central ner\ous system control
40. Bell JA. Sensibility' evaluation. In: Hunter of precision movements of the hand. In:
JM, Schneider LH, Mackin EJ, Bell JA, eds. Case-Smith J, Pehoski C, eds. Development
Rehabilitation of the hand. St. Louis: CV of hand skills in children. RockA-ille, MD:
Mosby, 1978:273-291. American Occupational Therapy Associa-
41. Levin S, Pearsall G, Ruderman R. Von tion, 1992:1-11.
Frey's method of measuring pressure sensi- 54. Growden JH, Chambers WW, Liu CN. An
bilit\' in the hand: an engineering anah'sis of experimental study of cerebellar dyskinesia
the Weinstein-Semmes pressure aesthesiom- in the rhesus monkey. Brain 1967;90:603-
eter.Hand Surg 1978;3:211.
J 630.
42. Semmes J, Weinstein S, Ghent L, Teaber 55. Lawrence DG, Hopkins DA. The develop-
HL. Somatosenson' changes after penetrat- ment of motor control in the rhesus mon-
ing brain wounds in man. Cambridge; Har- key: evidence concerning the role of corti-
vard Universit\' Press, 1960. comotoneuronal connections. Brain 1976;
43. Werner JL, Omer GE. Evaluation cutaneous 99:235-254.
pressure sensation of the hand. Am J Occup 56. Schwartzman RJ. A behavioral analysis of
Ther 1070;24:347-356. complete unilateral section of the pyramidal
44. Waters RL, Wilson DJ, Savinelli R. Reha- tract at the medullary le\el in macaca mu-
bilitation of the upper extremit)' following latta. .\nn Neurol 1978;4:234-244.
57. Chapman C, Wiesendanger M. Recoven' of and upper extremit\' inhibiti\c casting for
fianction tollo\\ing unilateral lesions of the children with CP. Dev Med Child Neurol
bulbar pyramid in the monkey. Electroen- 1991;33:379-387.
cephalogr Clin Neurophysiol 1982;53:374- 72. Cannon N. Manual of hand splinting. New
387. York: Churchill Livingstone, 1985.
58. Evarts EV. Role of motor cortex in volun- 73. Lindholm L. Weight-bearing splint: a
tary movements in primates. In: Brookhan method for managing upper extremin' spas-
JM, Mountcastle \T?, eds. Handbook of ticity'. Physical Therapy Forimi 1985;5:3.
physiolog)', section I. Volume II, Motor 74. Malick M. Manual on static hand splinting.
control. Bethesda, MD: American Physio- Pittsburgh: Harmamlle Rehab Center,

logical Societ>', 1981. 1980.
59. Wise SD, Evarts EV. The role of the cerebral 75. Neuhaus BE, Ascher B, Coullon M, et al. A
cortex on movement. Trends Neurosci survey of rationales for and against hand
1981;4:297-300. splinting in hemiplegia. Am J Occup Ther
60. Asanuma H, and indirect
Arissian S. Direct 1981;35:83-95.
sensoPi' input pathways to the motor cortex: 76. Zizlis J. Splinting of the hand in a spastic
its structure and fiinction in relating to hemiplegic patients. Arch Phvs Med Rehabil
learning motor skills. J Physiol 1984;39:1- 1964;1:41-43.
19. 77 Ma\nard CJ. Sensor\' reeducation following
61. Bobath B. Adult hemiplegia: evaluation and peripheral ner\'e injur)'. In: Hunter IM,
treatment. London: Wm Heinemann Med- Schneider LH, Mackin EJ, Bell JA, eds. Re-
ical Books, 1970. habilitation of the hand. St. Louis: CV
62. Carr Shepard R. Motor relearning pro-
J, Mosby, 1978:318-323.
gramme for stroke. Rock\'ille, MD: Aspen, 78. Dellon AL, Curtis RM, Edgerton MT. Re-
1983. education of sensation in the hand following
63. Davies P. Steps to follow. New York: ner\'e injur\'. Plast Reconstr Surg 1974;
Springer- Verlag, 1985. 53:297-305.
64. Boehme R. Impro\ing upper body control. 79. Callahan AD. Sensibility testing: clinical
Tucson, AZ: Therapy Skill Builders, 1988. methods. In: Hunter JM, Schneider LH,
65. Voss D, lonta M, Myers B. Proprioceptive Mackin EJ, Callahan AD, eds. Rehabilita-
neuromuscular facilitation: patterns and CV Mosby,
tion of the hand. St. Louis:
techniques. 3rd ed. Philadelphia, Harper & 1990:600-602.
Row, 1985. 80. Brand PW. Management of sensory loss in
66. Fess EK, Gettle D, Strickland J. Hand Omer E, Spinner M, eds.
the extremities In:
splinting: principles and methods. St. Louis: Management of peripheral nene problems.
CVMosby, 1981. Philadelphia: \VB Saunders, 1980:862-872.
67. Cusick B, Sussman M. Short leg casts: their 81. Vinogrand A, Taylor E, Grossmand S. Sen-
role in the management of cerebral palsv. sor\' retraining of the hemiplegic hand. Am
Phys Occup Ther Ped 1982;3/4:93-n0.' J Occup Ther 1962;5:246-256.
68. Yasukawa A. Upperextremit}' casting: ad- 82. Zee DS. Vertigo. Current therapy in neu-
junct treatment for the child with cerebral rological disease. Philadelphia: BC Decker,
palsy. In:Case-Smith J, Pehoski C, eds. De- 1985.
velopment of hand skills in children. Rock- 83. Shumway-Cook, unpublished obser\ations.
ville, MD: American Occupational Therapv 84. Peoppel E. Letter to the editor. Nature
Association, 1992:111-123. 1973;243:231.
69. Smith LH, Harris SR. L^pper extremity' in- 85. Perenin MT, Jeannerod M. Visual function
hibitive casting for a child with cerebral within the hemianoptic field following early
palsy. Phys Occup Ther Ped 1985;5:71-79. cerebral hemidecortication in man. I. spatial
70. Cruickshank DA, O'Neill DL. Upper ex- localization. Neuropsychologia 1978;16:1-
tremity inhibitive casting in a boy with spas- 13.
tic quadriplegia. Am J Occup Ther 1990; 86. Kraft GH, Fitts S, Hammond MC. Tech-
6:552-555. niques to inipro\e fiinction of the arm and
71. Law M, Cadman D, Rosenbaum P, Walter hand in chronic hemiplegia. Arch Phys Med
S, Russell D, DeMatteo C. NDT therapy Rehabil 1992;73:220-227.
87. WolfSL, LeCraw DE, Barton LA. Compar- SC: University of South Carolina Press,
ison of motor copy and targeted bioteed- 1990:75-98.
hack training techniques for restitution of 97. Forssberg H, Eliasson AC, Kinoshita H, Jo-
upper extremity fianction among patients hansson RS, Westling G. Development of
with neurologic disorders. Phys Ther 1989; human precision grip: basic coordination of
69:719-735. Res 199I;85:451^57.
force. Brain
88. McPherson J. Schiid R, Spaulding SJ, Bar- 98. Wesding G, Johansson RS. Factors influ-
samian, Transon C, White SC. Analysis of encing the force control during precision
upper extremity movement in four sitting grip. Exp Brain Res 1984;53:277-284.
comparison of persons with and
positions: a 99. Elliot J, Connolly K. A classification of ma-
without cerebral palsy. Am J Occup Ther nipulative hand movements. Dev Med
1991;2:123-129. Child Neurol 1984;26:283-296.
89. Shellenkens JM, Scholten CA, Kalverboer 100. Exner CE. In-hand manipulation skills. In:
AF. Visually guided hand movements in Case-Smith J, Pehoski C, eds. Development

children with minor neurological dvshmc- of hand skills in the child. Rock\ille, MD:
tion: response time and mo\'ement organi- American Occupational Therapy Associa-
zation. J Child Psych Psychiatn' 1983; tion, 1992:35^5.
24:89-102. 101. Seligman ME, Maier S. Failure to escape
90. Kluzik J, Fetters L, Coryell J. Quantification traumatic shock. J Exp Psvchol I967;74:I-
of control: a preliminary study of effects of 9.
neurodevelopmental treatment on reaching 102. Taub E. Motor beha\ior follo\\ing deaffer-
in children with spastic cerebral palsy. Phys entation in die developing and motorically
Ther 1990;2:65-78. mature monkey. In: Herman R, Grillner S,
91 Waksvik K, Lew R. An approach to seating Ralston JH, Stein PSG, Stuart D, eds. Neu-
for the cerebral palsied. Can J Occup Ther ral control of locomotion. New York: Ple-
1979;46:147-152. num, 1976.
92. Nwaobi OM, Brubaker CE. Cusick B, Suss- 103. Rothwell JC, Traub MM, Day BL, Obeso
man M. Electromyographic investigation of JA, Thomas PK, Marsden CS. Manual mo-
extensor activity in cerebral palsied children tor performance in a deafferented man.
in different seating positions. Dev Med Brain 1982;105:515-542.
Child Neurol 1983;25:175-183. 104. Knapp HD, Taub E, Berman J. Movements
93. Seeger BR, Caudrey DJ, O'Mara NA. Hand of monkeys with deafferented forelimbs.
fiinction in cerebral palsy: the effect of hip Exp Neurol 1963;7:305-315.
flexion angle. Dev Med Child Neurol 1984; 105. Taub E, Berman AJ. Avoidance condition-
26:601-606. ing in the absence of relevant proprioceptive
94. Noronha Bundy A, Groll J. The effect of
J, and exterocepti\e feedback. J Comp Phvsiol
positioning on the hand function of bo\'s Psychol 1963;56:1012-1016.
with cerebral palsy. Am J Occup Ther 1989; 106. WolfSL, Lccraw DE, Barton LA, Jann BB.
43:507-512. Forced use of hemiplegic upper extremities
95. Erhardt RP. Developmental hand dysfiinc- to reverse the effect of learned nonuse
tion: theory, assessment and treatment. among chronic stroke and head injured pa-
Tucson, AZ: Therapy Skill Builders, 1982. tients.Exp Neurol 1989;104:125-132.
96. Fisk ID. Sensory and motor integration in 107. Taub NE, Novack TA, et al. Tech-
E, Miller
the control of reaching. In: Bard C, Fleury nique to improx'e chronic motor deficit after
M, Hay L, eds. Developmental of eye-hand stroke.Arch Phys Med Rehabil 1993;
coordination across the lifespan. Columbia, 74:347-354.
Appendix A. Postural Control Assessment Form
Patient: Sex
Telephone: ^ Date: / —
Referring physician: Therapist:
I. HISTORY
A. Social History
Living Situation:
Home Retirement Center Nursing Home

Lives With:
Alone Spouse Friend Paid assistant
B. Medical History
Date of Onset of Condition:
Diagnosis;
Co-morbidities: (number)
List:
C. Fail/Imbalance History
How many falls?
No history of falls
1 Has fallen 1-2 times in last year
2 Has fallen 1-2 times in six months
3 Has fallen 1-2 times in the last six weeks
When was your most recent fall?
Did the fall occur inside or outside?
How did fall occur?
Injuries resulting from fall?
Dizziness during fall?
How often do you lose your balance, ie. trip slip or stumble?
No history of imbalance
1 Has imbalance monthly
2 Has imbalance weekly
3 Has imbalance daily
D. CURRENT MEDICATIONS
No of meds: Types:
E. BLOOD PRESSURE
Take patient's blood pressure in the supine position, again after moving into the seated
position, and again after standing.
B/P Supine: B/P Sitting: B/P Standing:

447
448 Appendix A POSTURAL CONTROL ASSESSMENT FORM
II. PERFORMANCE BASED FUNCTIONAL MEASURES OF BALANCE

A. Functional Balance Scale
(Reprinted with permission: Berg K, Measuring balance in the elderly: Validation of an instrument
[Dissertation]. Montreal, Canada: McGill University, 1993.)
1. Sitting to standing
Instruction: Please stand up. Try not to use your hands for support.
Grading: Please mark the lowest category which applies.
(4) able to stand, no hands and stabilize independently
(3) able to stand independently using hands
(2) able to stand using hands after several tries
(1) needs minimal assist to stand or to stabilize
(0) needs moderate or maximal assist to stand
2. Standing unsupported
Instruction: Stand for two minutes without holding.
(4) able to stand safely 2 min
(3) able to stand 2 min with supervision
(2) able to stand 30 sec unsupported
(1) needs several tries to stand 30 sec unsupported
(0) unable to stand 30 sec unassisted
IF SUBJECT ABLE TO STAND 2 MIN SAFELY, SCORE FULL MARKS FOR SITTING UN-
SUPPORTED. PROCEED TO POSITION CHANGE STANDING TO SITTING.
3. Sitting unsupported feet on floor
Instruction: Sit with arms folded for two minutes.
(4) able to sit safely and securely 2 min
(3) able to sit 2 min under supervison
(2) able to sit 30 sec
(1) able to sit 10 sec
(0) unable to sit without support 10 sec
4. Standing to sitting
Instruction: Please sit down.

Grading: Please mark the lowest categroy which applies.
(4) sits safely with minimal use of hands
(3) controls descent by using hands
(2) uses back of legs against chair to control descent
(1) sits independently but has uncontrolled descent
(0) needs assistance to sit

Appendix A POSTURAL CONTROL ASSESSMENT FORM 449
5. Transfers
Instruction: Please move from chair to bed and back again. One way toward a seat with
armrests and one way toward a seat without armrests.
(4) able to transfer safely with only minor use of hands
(3) able to transfer safely with definite need of hands
(2) able to transfer with verbal cueing and/or supervision
(1 needs one person to assist
(0) needs two people to assist or supervise to be safe
6. Standing unsupported with eyes closed
Instruction: Close your eyes and stand still for 10 sec.
(4) able to stand 10 sec safely
(3) able to stand 10 sec with supervision
(2) able to stand 3 sec
(1) unable to keep eyes closed 3 sec but stays steady
(0) needs help to keep from falling
7. Standing unsupported with feet together.
Instruction: Place your feet together and stand without holding.
(4) able to place feet together indep and stand 1 min safely
(3) able to place feet together indep and for 1 min with supervision
(2) able to place feet together indep but unable to hold for 30 sec
(1 needs help to attain position but able to stand 1 5 sec feet together
(0) needs help to attain position and unable to hold for 1 5 sec
THE FOLLOWING ITEMS ARE TO BE PERFORMED WHILE STANDING UNSUPPORTED

8. Reaching forward with outstretched arm
Instruction: Lift arm to 90 degrees. Stretch out your fingers and feach forward as far as you
can. (Examiner places a ruler at end of fingertips when arm is at 90 degrees. Fingers should
not touch the ruler while reaching forward. The recorded measure is the distance forward
that the fingers reach while the subject is in the most forward lean position.)
(4) can reach forward confidently >10 inches
(3) can reach forward >5 inches safely
(2) can reach forward >2 inches safely
(1) reaches forward but needs supervision
(0) needs help to keep from falling

9. Pick up object from the floor
Instruction: Pick up the shoe/slipper which is placed in front of your feet
(4) able to pick up slipper safely and easily
(3) able to pick up slipper but need supervision
(2) unable to pick up but reaches 1-2 inches from slipper and keeps balance indep
(1) unable to pick up and needs supervison while trying
(0) unable to try/needs assist to keep from falling
10. Turning to look behind/over left and right shoulders.
Instruction: Turn to look behind you over/toward left shoulder. Repeat to the right.
(4) looks behind from both sides and weight shifts well
(3) looks behind one side only; other side shows less weight shift
(2) turns sideways only but maintains balance
(1) needs supervision when turning
(0) needs assist to keep from falling
11. Turn 360 degrees
Instruction: Turn completely around in a full circle. Pause. Then turn a full circle in the
other direction.
(4) able to turn 360 safely in <4 sec each side
(3) able to turn 360 safely one side only in < 4 sec
(2) able to turn 360 safely but slowly
(1) needs close supervision or verbal cueing
(0) needs assistance while turning
DYNAMIC WEIGHT SHIFTING WHILE STANDING UNSUPPORTED
12. Stool touch
Instruction: Place each foot alternately on the stool. Continue until each foot has touched
the stool four times.
(4) able to stand indep and safely and complete 8 step in 20 sec
(3) able to stand indep and complete 8 steps in >20 sec
(2) able to complete 4 steps without aid with supervision
(1) able to complete >2 steps needs minimal assist
(0) needs assistance to keep from falling/unable to try

13. Standing unsupported, one foot in front
Instruction: (Demonstrate to subject) Place one foot directly in front of the other. If you
feel that you cannot place your foot directly in front, try to step far enough ahead that the
heel of your forward foot is ahead of the toes of the other foot.
(4) able to place foot tandem indep and hold 30 sec
(3) able to place foot ahead of other indep and hold 30 sec
(2) able to take small step Indep and hold 30 sec
(1 needs help to step but can hold 1 5 sec
(0) loses balance while stepping or standing
14. Standing on one leg
Instruction: Stand on one leg as long as you can without holding.
(4) able to lift leg indep and hold >10 sec
(3) able to lift leg indep and hold 5-10 sec
(2) able to lift leg indep and hold = or > 3 sec
(1) tries to lift leg; unable to hold 3 sec but remains standing indep
(0) unable to try or needs assist to prevent fall
TOTAL SCORE / 56
III. STRATEGY ASSESSMENT

A. Seated postural control
1 Alignment: eo ec
Ask patient to sit up as straight as they can.
2 = Normal alignment of body segments
1 = Partial correction towards normal alignment
= Abnormal alignment of body segments, ie lateral asymmetry, excessive rotation
of the pelvis, kyphosis, or forward flexion of the head, or inabilty to sustain
vertical.
2. Active Weight Shifts: eo ec
Ask patient to shift weight to one side, as far as they can without losing their balance.
Perform one side then the other, first with eyes open, then with eyes closed.
2 = Normal is defined as the patient's ability to shift weight symmetrically, elongate
trunk on weight bearing side, re-establish vertical with and without vision, not
dizzy.
1 = Able to partially complete
= Abnormal, inability to shift weight, asymmetrical weight shift, inability to re-
establish vertical.
B. Stance postural control
1 Alignment: Eyes open Eyes Closed Base of Support
Ask patient to stand up as straight as they can, measure base of support at mid foot.
2 = Normal alignment of body segments, vertical line of gravity at tragus, shoulder,
hip, knee and just ant to malleoli, even between both feet.
1 = Partially able to assume normal alignment
= Abnormal alignment, ie, center of mass laterally displaced, or displaced forward
or backward, or excessive rotation of the pelvis, thoracic kyphosis, or forward
flexion of the head, cannot sustain a vertical position.
2. Movement Strategies:
a. Self Initiated Sway Strategy:
Ask patient to sway forward and backward, but not take a step.
2 = Normal, ankle centered sway, inverted pendulum movement of body

with good range forward and backwards.
1 = Partial ankle strategy, reduced range
= Abnormal is inability to sway about the ankles, controlling the knees and
hips in a neutral position.
b. Reactive Balance Strategy
Within base of support
Holding patient at the hips, therapist displaces patient small distance by push-
ing/pulling at hips. Instruction: "Let me move you, try not to take a step, but
keep your balance."
2 = Normal, ankle centered sway, inverted pendulum movement of body
with good range forward and backwards.
1 = Partial ankle strategy, reduced range
= Abnormal is inability to sway about the ankles controlling the knees and
hips in a neutral position.
Outside base of support
Therapist displaces patient's center of mass outside base of support. Instruc-
tion: "Let me move you; you might have to take a step. It's OK."
2 = Ability to take a step with either foot, normal range
1 = Step with one foot only, or altered range
= Abnormal is inability to take a step to keep from falling.
3. Sensory Strategies
Clinical Test of Sensor^' Interaction in Balance:
Time —30 Sec. Sway 1 = normal sway, = Abnormal (asymmetric or excessive
sway)
Trial 1 Trial 2
Time Sway Time Sway
Eyes open, firm surface
Eyes closed, firm surface —

Visual dome, firm surface —
Eyes open, foam surface
Eyes closed, foam surface —

Visual dome, foam surface —
IV. SYSTEMS ANALYSIS
A. Mental Status
MINI MENTAL TEST (Mental Status)
1 . What is the date today?
2. What day of the week is it?
3. What is the name of this place?
4. What is your telephone number?
or What is your address?
5. How old are you?
6. When were you born?
7. Who is the President of the US now?

8. Who was the President before him? _
9. What was your Mother's maiden name?
10. Subtract 3 from 20 and keep subtracting 3 from each new number, all the way down
(20, 17, 14, n, 8, 5, 2).
Total Number of Errors
Oriented at all times (0-2 errors on MM test)
1 Mild intellectual impairment (3-4 errors)
2 Moderate intellectual impairment (5-7 errors)
3 Severe intellectual impairment (8-10 errors)
B. MUSCULOSKELETAL SYSTEM
1. Strength Right Left
Gastroc/soleus
TA
Quads
Hamstrings
Hip Flexor
Hip Extensors
Abductors
Adductors
Trunk (partial sit up)

B. MUSCULOSKELETAL SYSJiM— continued

2. Range of motion Right Left
Hip
Knee
Ankle
Cervical
Trunk
Scoring:
= ankylosed
1 = moderate hypomobility
2 = mild hypomobility
3 = normal
4 = mild hypermobility
5 = moderate hypermobility
6 = severe hypermobility
3. Muscle Tonus:
= No increase in muscle tone
1 = Slight increase in muscle tone, manifested by a slight catch and release or by
minimal resistance at the end of the range of motion u'hen the affected part(s)
is moved in flexion or extension
1+ = Slight increase in muscle tone, manifested by a catch, followed by minimal
resistance throughout the remainder (less than half) of the ROM

2 = More makred increase in muscle tone through most of the ROM, but affected
part(s) easily moved

3 = Considerable increase in muscle tone, passive movement difficult
4 = Affected part(s) rigid in flexion or extension.
4. Pain:
5. Cerebellar Coordination:
Finger to nose:
Pronation/supination:
Heel to shin:
Tremor:
Scoring: 5 = normal, 4 = minimal impairment, 3 = moderate impairment, 2 = severe
impairment, 1 = cannot perform.
3. Peripheral Sensibility
Test the follovk'ing senses. Score N if intact, A if abnormal
Proprioception:
(Great Toe, Ankle)

Appendix A POSTUR.\L CONTROL ASSESSMENT FORM 455
The following is by subject report:
Central visual acuit>'
Peripheral visual acuity
Depth Perception
EVALUATION SUMAAARY:
Problems:
Functional Level of Performance:
Strategies For Postural Control:
Alignment
Movement
Sensory'
Impairments:
Cognitive
Musculoskeletal
Neuromuscular
Senson.
PLAN:
Short Term Coals: (Expressed in temrs of underlying impairments, or interim steps towards
a long term goal)
Long Term Goals: (Expressed in terms of functional skills).
Frequencv of Treatment: Duration:
Treatment Plan:
GLOSS.\RY
Action potential- ~3ijc jump in vohage learning. An weak stimulus I9he condi-
initially
fv ihat is obeefved when tioned stimulus!!becomes hi^ly effective in
producing a response wfien it becomes associ-
Acd^:^. .^v : ~irn* — ^>ne ot the three maior ated with another stioriger stimulus (the uiKon-
-f - - - locomotion lelflect- ditikxied stimulusl. After repeated pairing of the
''£:-t;~ ^ ;: to meet the goals ot conditioned and the urKxmditioned stimulus,
the animai ana tne Demands of the environ- one beginsto see a conditioned response (CR)
ment. totheCS.
Adaptive postural confrof—modifying sensoiy CSnical dedsiou-makhig process a procedure —
and motor systems in response to changif^ task for gathering information essential to develop-
and environmental demands. ing a plan of care consistent with the problems
Agnosia —the inability- to recognize. Lesions in the and needs of the patient.
parietal lobe often cause agnosia or n^lect ol Cfaaal iocppfooess—motor control processing in
the conlralatefal side of the body, obfects. and which sensory feedback is used for the ongoirig
drawings. prorkidion of skilled movement
—
Alpha-motor neurons motor neurons within the CoffmUm processes—in this book, we define cog-
i
spinal cord that innervate skeletal muscle fibers. nitive processes broadly, to include higher level
A/T(idpala»y postmal confral—pretuning sensory neural processes such as planning, attention,
and motor systems in oqieclation of postural de- motivation, and emotional aspects of motor
mands based on previous experience and learn- control that underlie the establishment of intent
ing- or goals. It is difficult to make a distinction be-
AxM Mt it—the systematic acquisition of

l inibr- tween hiîer-level perceplualAnator processing
rnation that is relevant and meaningful in pro- and cognitive processing, since there is a grad-
.iding the clinician with a comprehensive pic- ual transition and overlap between the process-
~jre of the patient's abilities and probtems. irig levels.
—
Assodatire siage in the Rtts-Posner description —
dognHhie iUgn' in the Fitts-Posner description of
of motor learning, this is the serxmd stê. By motor learning, this is the first stage in the pro-
this time, the person has selected the best strat- cess. In it the learner is cortcemed with under-
egy for the task and b^ns to refine the skill. standing the nature of the task, developir^ strat-
Asywmitnc tonic neck rcAec—produces a be used to cany out the task,
egies that could
change in the position of the amis in response and determining how the task should be eval-
to change in head position. Turning the head uated.
produces extension in the Oce arm. and flexion —
CamperBaition behavioial substitution, that is,
in the skull arm. ahemative behavioral strategies are adopted to
Autonomous stage —in the Fitts-Pdsner descrip- complete a tafV
tion of motor learning, thb is the third stage. In CJorrceptuai fraru'eM ori —a logical structure that
ihis stage, there is automaticity in the sUM. and helps the : - tsnize dinical practices
a low degree of attention is required for its per- related to . ^d tieatnnent into a co-
formance- hesive anc - .?nlan.
Bod't -on-bod^ rishtirrg reaction — keeps the body CDordfaatfie -:rLc:^-r — commands that
r"e: 'r T- f-rl to the ground regardless aretempc-; : ?is;nals are sent
to musclef ~-is reduces
Bod\ -on-head rishtins reactlO'n — orients the head the degree- -ofled by
t.T .response to proprioceptive and tactile signals the nervoL - •> of
from the body in contact with a supporting sur- muscles t; -ler-
face. ent units

C^dlenop—the number of steps per unit of time, synonym).
c..,ii, ^.o^y^-*-^ ,c steps per minute.
Classical conditioning —a form of association ye spinal
457
458 GLOSSARY
cord, brainstem, and cerebellum intact. An area strained) to facilitate the return of function in
in the brainstem, called the mesencephalic lo- that arm.
comotor region, appears to be important in the —
Frozen gait pattern a gait pattern of patients with
descending control of locomotion. Decerebrate Parkinsons disease, characterized by an inabil-
cats will not normally walk on a treadmill, but ity to generate sufficient momentum so that for-
will begin to walk normally when tonic electri- ward progression is arrested.
cal stimulation is applied to the mesencephalic Gamma-motor neurons — motor neurons from the
locomotor region. spinal cord that innervate themuscle spindle
Declarative learning —The process of learning muscle fibers.
knowledge that can be consciously recalled and General static reactions (called attitudinal re-
thus requires processes such as awareness, at- flexes) —
involve changes in position of the
tention,and reflection. whole body in response to changes in head po-
Decorticate locomotor preparation — animal ex- sition.
perimental preparation with only the cerebral Glabrous skin — hairless skin.
cortex removed. In this preparation, an external Habituation a — decrease in responsiveness
stimulus is not required to produce locomotor that occurs as a result of repeated exposure to a
behavior, and the behavior is reasonably nor- nonpainful stimulus, (see synaptic detaclllta-
mal goal-directed behavior. tion.)
Degrees of freedom problem a motor control — —

muscle fibers Specialized muscle fi-
Intrafusal
problem involving how to control the many dif- bers foundin muscle spindles (extrafusal fibers
ferent joints and muscles of the body. are normal skeletal muscle fibers).
—
Denervation supersensitivity occurs when neu- —
HAT head, arm, neck, and trunk segments that
rons show a loss of input from another brain comprise the unit that must be balanced above
The postsynaptic membrane of a neuron
region. the legs during locomotion.
becomes hyperactive to a released transmitter Hierarchical processing — a system of neural pro-
substance. cessing in which higher levels of the brain are
Distributed practice —a training session in which concerned with issues of abstraction of infor-
the amount of rest beUveen trials equals or is mation. For example, higher brain centers inte-
greater than the amount of time for a trial. grate inputs from many senses, and interpret in-
Excitatory postsynaptic potential (EPSP)— the coming sensory information.
change in membrane potential in the postsyn- Hypothesis — a hypothetical explanation about the
aptic cell (typically depolarizing) made by the cause or causes of a problem.
excitatory transmitter substance released from Hypothesis-oriented clinical practice — a process
the presynaptic neuron. used to systematically test assumptions about
Excitatory summation —occurs when a series of the nature and cause of a patient's problems.
excitatory postsynaptic potentials (EPSPs) con- Inertia —the tendency to remain at rest; the in-
tinue to build up depolarization to the threshold ability to move spontaneously.

voltage for the action potential in the next neu- Intrinsic feedback —feedback that comes to the
ron. individual through the various sensory systems
Extrinsic feedback — information that supplements as a result of the normal production of the move-
intrinsic feedback (e.g., when you tell a patient ment (e.g., visual information concerning
that he/she needs to pick up his/her foot higher whether a movement was accurate, somatosen-
to clear an object while walking). sory information concerning the position of the
Flexor —
withdrawal reflex a cutaneous reflex limbs as one was moving).
caused by a sharp focal stimulus, producing —
Joint-based planning one possible way the CNS
withdrawal, or flexion, and causing protection could control movements toward a target, by
from injury. The typical pattern of response is using joint angle coordinates to program move-
ipsi lateral flexion and contralateral extension, ments.
which allow the support of body weight on the Knowledge of performance (KP) — feedback relat-
opposite limb. The reflex is mediated by group ing to the movement pattern that the performer
III and IV afferents. has made.
—
Forced-use paradigm a therapeutic approach in Knowledge of results (KR) —a form of extrinsic
which hemiplegic patients are forced to use feedback. It has been defined as verbal (or its
their hemiplegic arm (the intact side is re- equivalent) terminal feedback about the out-
Glossary 459
come of the movement, in terms of the move- ity for producing skilled action. It emerges from
ment's goal. a complex of perception-cognition-action pro-
Labyrinthine righting reaction — orients the head cesses. Involves the search for a task solution,
to an upright vertical position in response to ves- which emerges from an interaction of the indi-
tibular signals. vidual with the task and the environment.
Landau reaction —combines the effect of the lab- —
Motor program the term may be used to identify
yrinthine, optical, and body-on-head righting a central pattern generator (CPG), that is, a spe-
reactions. cific neural circuit like that for generating walk-
Learning —the process of acquiring knowledge ing in the cat. In this case, the term represents
about the world. neural connections that are stereotyped and
Local static reactions — stiffen the animal's limb hardwired. The term is also used to describe
for support of body weight against gravity. higher-level hierarchically organized neural
Long-term memory continuum — of processes in- processes that store the rules for generating
volving information storage. Initial stages reflect movements so that tasks can be performed with
functional changes in the efficiency of synapses. a variety of effector systems.
Later stages reflect structural changes in synap- Muscle tone — the force with which a muscle re-
tic connections. These memories are less sub- sists being lengthened.
ject to disruption. Neck-on-body righting reaction orients the —
Long-term potentiation (LTP) — similar to sensiti- body in response to cervical afferents, which re-
zation. In the hippocampus, LTP occurs when a port changes in the position of the head and
weak and an excitatory input arrive at the same neck.
region of a neuron's dendrite. The weak input is Neuronal shock (diaschisis) — the short-term loss
enhanced if it is activated in association with the of function in neuronal pathways at a distance
strong one. LTP appears to require the simulta- from the lesion itself.
neous firing of both pre- and postsynaptic cells. Operant conditioning —the process of learning to
After this occurs, LTP is maintained through an associate a certain response, from among many
increase in presynaptic transmitter release. thathave been made, with a consequence. Be-
Massed practice — a session in which the amount haviors that are rewarded tend to be repeated,
of practice time in a trial is greater than the while behaviors followed by aversive stimuli are
amount of rest between trials. reduced in number.
Memory trace —within Adam's closed-loop the- Optical righting reaction — contributes the to re-
ory of motor control, the memory trace is used flex orientation of the head using visual inputs.
to select and initiate a movement. Parachute or protective responses— protect the
Model of brain function —model of brain func- body from injury during a fall.
motor control, is a simplified rep-

tion, related to Parallel distributed processing — neural process-
resentation of the structure and function of the ing in which the same signal is processed si-
brain as it relates to the coordination of move- multaneously among many different brain struc-
ment. tures, though for different purposes.
Model of disablement — an approach to ordering Perceptual trace— within Adam's closed-loop
the effects of disease, enabling the clinician to theory of motor control, the perceptual trace is
develop a hierarchical list of problems towards considered an internal reference of correctness

which treatment can be directed. built up over a period of practice.
Monosynaptic reflex the simplest — reflex path- Performance-based functional measures assess- —
way, consisting of a sensory neuron, the la af- ment tools that focus on measuring performance
ferent neuron from the muscle spindle, an in- on functional tasks.
terneuron, the la inhibitory interneuron, and a Plasticity —the ability to show modification or
motor neuron, the a-motor neuron to the same change. Short-term functional plasticity refers to
muscle. The muscle contracts in response to changes in the efficiency or "strength" of syn-
stretch of the muscle spindle and activation of aptic connections. Structural plasticity refers to
the la afferent neuron. changes in the organization and numbers of
Motor learning — the study of the acquisition and/ synaptic connections.
or modification of movement; a set of processes Postural control — regulating the body's position
associated with practice or experience leading in space for the dual purposes of stability and
to relatively permanent changes in the capabil- orientation.
460 GLOSSARY
Postural fixation reactions — used to recover from definitions include the ability to achieve task
perturbations other than to the supporting sur- goals using effective and efficient means, but
face. not necessarily those used preinjury.
Posturalmotor strategies —the organization of —
Recovery of function the reacquisition of move-
movements appropriate for controlling the ment skills lost through injury.
body's position in space. —
Recurvatum hyperextension, which occurs
Postural orientation — the ability to maintain an when the knee has sufficient mobility to move
appropriate relationship between the body seg- posteriorly past neutral.
ments, and between the body and the environ- Reflex —a stereotyped muscle response to a sen-
ment for a task. sory stimulus. The simplest reflex pathway is the
Postural stability — the ability to maintain the po- monosynaptic stretch reflex pathway, consisting
sition of the body, and specifically, the center of a sensory neuron, the la afferent neuron from
of body mass (COM), within specific boundaries the muscle spindle, an interneuron, the la inhib-
of space, referred to as stability limits. itory interneuron, and a motor neuron, the a-
Postural tone — increased level of activity in anti- motor neuron to the same muscle. The muscle
gravity muscles that helps maintain the body contracts in response to stretch of the muscle
vertically against the force of gravity. spindle and activation of the la afferent neuron.
Procedural learning — the process of learning tasks — process of sprout-
Regenerative synaptogenesis
that can be performed automatically without ating of injured axons.
tention or conscious thought, like a habit. Response-produced feedback— the sensory all in-
—
Progression requirement one of the three major formation that is available as the result of a
requirements for successful locomotion, reflect- movement that a person has produced.
ing the need for a basic locomotor pattern that Resting potential —the neuron, when it is at rest,
can move the body in the desired direction. always has a negative electrical charge or po-
Propulsive gait pattern — gait pattern of patients tential on the inside of the cell, with respect to
with Parkinson's disease, characterized by an the outside. This is called the resting potential.
inability to restrain momentum, leading to un- Righting reactions — allow the animal to assume
controlled progression. or resume a species specific orientation of the
Reactive synaptogenesis (collateral sprouting) — body with respect to its environment.
process in which neighboring normal axons —
Schema an abstract representation stored in
sprout to innervate synaptic sites that were pre- memory following multiple presentations of a
viously activated by the injured axon. class of objects.
Recall schema —within Schmidt's Schema theory, Segmental static reactions involve more than —
when initiating a movement, it is used for the one body segment, and include the flexor with-
selection of a specific response. Inputs to this drawal reflex, and the crossed extensor reflex.
schema include the initial conditions, desired —
Self-organizing system a system that can spon-
goal of the movement, and the abstract memory taneousl)' form movement patterns that arise
of previous response specifications in similar simply from the interaction of the different parts
tasks. of the system.
Receptive field —the specific area of skin, retina, Sensitization — an increased responsiveness fol-
etc., to which a cell is sensitive when the skin lowing a threatening or noxious stimulus.
or retina is stimulated. The receptive field can —
Sensorimotor strategies reflect the rules for co-
be either excitatory or inhibitory. ordinating sensory and motor aspects of pos-
Recognition schema — within Schmidt's Schema tural control.
theory, this is used for the evaluation of a re- —

Sensory strategies organize sensory information
sponse. The sensory consequences and out- from visual, somatosensory, and vestibular sys-
comes of previous movements are combined tems for postural control.
with the current initial conditions to create a Short-term memory— "working" memory, which
representation of the expected sensory conse- has a limited capacity for information storage
quences. and lasts for a few moments only. This reflects
Recovery —stringent definition requires achieving momentary attentional processes.

the functional goal in the same way was
it per- Spared function — used to describe a function that
formed preinjury, that is, using the same pro- is not lost following injury.
cesses utilized prior to the injury. Less stringent Spasticity — a motor disorder characterized by a
Glossary 461
velocity-dependent increase in tonic stretch re- while flexor activity predominates in the lower
flexes (muscle tone) with exaggerated tendon extremities. Flexion of the head reverses this;
jertcs, resulting from hyperexcitability of the thus, there is an increase in flexion in the upper
stretch reflex (it is one component of the upper extremities and extensor activity in the lower ex-
motor neuron syndrome). tremities.
Spatial summation summation — that produces de- Synaptic defacilitation or habituation —when a
polarization because of the simultaneous action neuron been activated over a period of
that has
potentials of multiple cells synapsing on the time releases less transmitter, often due to trans-
same postsynaptic neuron. mitter depletion, and is less effective in influ-
Spinal locomotor preparation — animal experiencing the postsynaptic neuron.
mental preparation in which lesions are made —
Synaptic facilitation when a neuron that is acti-
at the low spinal level, to allow the observation vated over a short period of time begins to re-
of only the hind limbs, or at the high spinal lease more transmitter with each action poten-
level, to allow the observation of all four limbs tial and therefore more easily depolarizes the
as part of the preparation. For this preparation, next cell.
one needs an external stimulus, for example, an Synaptic transmission — in chemical synaptic
electrical or pharmacological stimulus, to pro- transmission, each action potential in a neuron
duce locomotor behavior. releases a small amount of transmitter sub-
Stability limits — boundaries of an area of space in stance. It and attaches
diffuses across the cleft
which the body can maintain its position with- to receptors on the next cell, which open up
out changing the base of support. channels in the membrane and depolarize the
Stability requirement —one of the three major renew cell. If the depolarization is sufficient, an
quirements for successful locomotion, reflecting action potential will be activated.
the ability to maintain stability, including the —
Synergy functional coupling of groups of mus-
support of the body against gravity. cles such that they are constrained to act to-
Step length — the distance from the foot-strike of gether as a unit (synonym: coordinative struc-
one foot to the foot-strike of the other foot. For ture).
example, the right step length is the distance Task-oriented approach — a therapeutic approach
from the left heel to the right heel when both to retraining the patient with movement disor-
feet are in contact with the ground. ders, based on a systems theory of motor con-
Strategy — a plan for action; an approach to orga- trol.
nizing individual elements within a system into —

Temporal summation summation that results in
a collective structure. depolarization because of synaptic potentials
Stride length —the distance covered by the same from a presynaptic neuron that occur close to-
foot from one heel-strike to the next heel-strike. gether in time.
Support moment— the algebraic sum of the joint Theory of motor control a group of abstract —
moments at the hip, knee, and ankle, during the ideas about the nature and cause of movement.
stance phase of the step cycle. The support mo- Theories are often, but not always, based on
ment is an extensor torque. This net extensor models of brain function.
torque keeps the limb from collapsing while Tilting reactions — used for controlling the center
bearing weight, allowing stabilization of the of gravity in response to a tilting surface.
body and thus accomplishing one of the re- Tonic labyrinthine reflex — produces a change in
quirements of locomotion. body posture in response to vestibular inputs,

Symmetric tonic neck reflex —changes the posi- signaling head position with respect to gravity.
tion of the limbs in response to a change in head When the body is in supine position, extensor
position. When the head is extended, extensor muscles are facilitated; conversely, the prone
activity predominates in the upper extremities, position results in facilitation of flexor muscles.
INDEX
Page numbers in italics denote figures; those followed by "t" denote tables.
Achromatopsia, 67 Ambulation. See Gait; Locomotion

Action, 3 Analysis
Action potential, 50, 447 kinematic, 124
Action systems, 20 kinetic, 125
Activities of Daily Living, 108, 317, 419 observational gait, 324-330
problems in planning and sequencing of task,36-37
assessment of, 425—126 Ankle movements during gait, 248
retraining for, 44 plantarflcxion contractures, 299-301
Adams, J.A., 28 retraining for problems with, 336-337
Adaptation valgus foot, 301
gait abnormalities and, 299 varus foot, 300-301
mobility retraining goals related to, 334 Ankle strategy, 127-129, 129, 187
motor adaptation problems, 196, 196-197, 198t retraining use of, 227, 227-228
for postural control, 121, 447 Antagonistic restraint, 191
motor strategies for, 130, 130-131 Anterolateral system, 57-59
in older adults, 176-178, 177-178 Approximation technique, 223
sensory mechanisms for, 134—137, 135-136 Apraxia, 411^13
of reach and grasp, 364—365, 365 constructional, 62
requirement for locomotion, 240, 447 Assessment, 101, 107-112, 114,447
sensorimotor, 202-203 abilities, 108-110
of ftinctional
sensory Gentile'staxonomy of movement tasks, 109, 109t
gaitand, 251-253, 278 of ftinctional tests, 109-1 10
limitations
for stance control, 163-164, 164 performance-based functional measures, 108-109,
upper extremity dyscontrol due to problems with, 449
410-tll of impairments. 111
Adductors integrating nypothesis testing with, 1 1 1-112
overuse of, 306 interpreting data from, 112
spasticity of, 304 of mobility, 209, 210t, 315-330
Aging of postural control, 208-221
alignment of body and, 192 of strategies, 110-111,211-216
balance retraining and, 180-181 three levels of, 107-108, lOS, 208, 418
heterogeneity of, 171 transition to treatment from, 112-114
mobility and, 269, 283-292 of upper extremity manipulatory dyscontrol, 418^30
models of, 169-170, 170 Assistive devices, 340-341
musculoskeletal system and, 172, 172-173 Association areas, 61, 61-62, 75
neuromuscular system and, 173-175, 174—175 Ataxia, 198
postural control and, 169-182 optic, 404, 404
postural responses and, 161, 162 Ataxia Test Battery, 108
primar\' and secondary factors related to, 170-171 Attentional capacity, 180, 341
recovery and, 39
sensory systems and, 175-178 Back propagation, 17
adaptations for postural control, 176-178, 177-1 Balance control, 3. See also Postural control; Vestibular
multisensory 76
deficit, 1 system
somatosensory, 175-176 assessment of, 209, 210t
vestibular, 176 reactions for, 149-150, 150-152
visual, 176 retraining of, 180-181
upper extremity manipulation skills and, 389-394 during walking, 247
Agnosia, 67, 447 Barthel Index, 108,421
Akinesia, 79 Basal ganglia, 49, 78-79, 80
Alignment, 123, 123-125 Behavioral interpretation, 5-6
abnormal, 191-192 Bernstein, Nicolai, 13-14
assessment of, 211, 211-212 "Binding problem," 67
in elderly persons, 192 Biofeedback, 228, 437
ideal, 126 Bipolar cells of retina, 64
retraining of, 224—226, 225-226 Bobath approach, 104
Amacrine cells of retina, 64 Body schema, 199
463
464 Index
Bradykinesia, 79, 197-198 Constructional apraxia, 62

Brain function, 46-50, 48-49 Contextual interference, 36
Brain ftinction models, 4—5, 449 Contractures
Brain 100, 192, 195, 202
injur\-, hip flexion, 303, 304
Brainstem, 47, 48 knee extension, 305-306
Brunnstrom approach, 104 knee flexion, 302, 307
plantarflexion, 299-301, 306, 336
Cadence, 242, 447 Contrast sensitivit\', 59, 64
Canes, 226, 340-541 Coordination problems, 191-198
Caudate nucleus, 78, 79, 80 assessment for, 219
Center of mass, 122, 126 due to alignment abnormalities, 191-192
abnormal alignment and, 191-192 due to cerebellar disorders, 198
excessive oscillation of, 196 due motor adaptation problems, 196, 196-197, 198t
to
diuing gait c\'cle, 243-244 due 197-198
to Parkinson's disease,
strategies to restore to position of stabilit\', 126-130, due to scaling problems, 195-196
128 due to stroke and hemiplegia, 197
ankle, 127-129, 129 due to timing problems, 192-195, 193-195
hip, 129, 129-130 eye-head, 400-401
in older adults, 175, 175 upper extremitii' dyscontrol due to, 401^03
stepping, 130 Coordinative structure, 447. See also Synergies
Center-surround inhibition, 65 Coronal plane foot de\iations, 300-301
Central pattern generators, 11-13 retraining for, 337
Cerebellar disorders, 76, 198, 229, 403 Corticobulbar svstem, 49
Cerebellum, 47, 48, 76-78 Corticospinal tract, 49, 71, 73, 78, 406-407
anatomy of, 77-78 CrawUng, 280-282, 281-282
input and output connections of, 76, 7(>-77 Crouch pattern gait, 309-310
role in procedural learning, 91-93, 92 Crutches, 340
role in step cycle, 250 Cutaneous receptors, 55
Cerebral cortex, 47^9, 48
Cerebral hemispheres, 47—19, 48 Deafiiess, 202
Cerebral palsy, 100-101 Degrees of freedom, 13-14, 448
abnormal movement patterns in, 192, 193 Dener\ ation supersensiti\'ity, 94-95, 448
activation of antagonist muscles in,408 Dentate nucleus, 77
gait abnormalities due 309-310
to, Depolarization, 50-51
musculoskeletal impairments in, 187-188 Development
sensorv' selection problems and, 202 of locomotion, 269-280
upper extremity' dyscontrol in, 403 motor, 143-145, 144
Cerebrovascular accident. See Stroke neiu-omaturational theon- of, 10, 10
Classical conditioning, 26, 26, 41, 89, 447 of postural control, 143-166
Clinical decision-making process, 100-101, 114,447 role of reflexes in, 150-151
Clinical practice of upper extremin' manipulation skills, 377-389
h\pothesis-oriented, 7, 100-102 Dexteriu- tests, 419-120, 420
models of disablement and, 100, 102, 102-103,449 Diaschisis, 94, 449
reflex-based neurofacilitation approaches to, 104—106 Diencephalon, 47, 48
relation of motor control theories to, 5-7, 99-100, Distance theories of reaching, 369-371
103-104 Distributed model of motor control, 13
systems-based task-oriented approach to, 106-107 Disuse atrophy, 187
task-oriented conceptual framework for, 107-1 14 Dizziness, 176
assessment, 107-112 Dorsal columns, 56, 57, 96
simimary of 114 Down syndrome, 189
transition from assessment to treatment, 112-114 dyssynergia in, 192, 195
Clinical Test for Senson' Interaction in Balance, sensor\' selectionproblems in, 202
212-214 upper extremit\' dvscontrol in, 407
Closed-loop theor)' of motor learning, 28, 447 Duke Mobility- Skills Profile, 319, 320t-322t
Cochlea, 67 Dynamic Gait Index, 322, 323t-324t
Cognitive processes, 4, 447 Dynamical action theor\' of motor control, 14—16, 15
assessment of, 2 1 6-2 1 7, 2 1 7t clinical implications of, 15-16
and postural control in older adults, 179-180 limitations of, 1
role in gait changes in elderly, 288 Dynamical systems model of motor control, 15
treating impairments of, 222, 222t Dynamometn-, 426-427, 427
Collateral sprouting, 95, 450 Dyskinesia, 79
Compass gait, 243 Dyss\nergia, 193-195, 193-195
Compensation, 38, 113-1 14, 447
Conceptual framework, 447 Ecological theorj' of motor control, 18-19, 19
task-oriented, 107-114 clinical implications of, 19
Cones, 62, 63 limitations of, 19
Index 465
Edema, upper extremity, 401, 429 recovery of, 23-24, 38-41, 450. See also Recovery of
Elderly persons. See Aging fiinction
Electromyography, 124 spared, 38, 39, 451
during gait cycle, 245 Functional Balance Scale, 209
showing age differences in postural responses, 161, Functional electrical stimulation, 228, 229, 437
162 Functional Independence Measure, 108, 317, 319, 322,
during transition to independent stance, 158, 419
158-159 Functional Reach Test, 208, 209, 209t
Embn,'onic movements, 270
Endolymph, 67-69 Gait,239-257. See also Locomotion; Mobility
Emironmental factors, 4, 4, 12 assessment of, 210t, 315-330. See also Mobility
120
affecting postural control, 120, assessment
ecological theory and, 18-19, 19 observational gait analysis, 324—330
Equilibrium, 120 characteristics of mature gait, 276, 276t
Erhardt Test of Manipulatory Skills, 108 control mechanisms for, 248-254
Excitatory postsynaptic potentials, 50, 87, 88, 88, descending influences on, 249-251, 250
448 goals to be met during, 240
Exercise programs, 180-181 head stabilization during, 278-280
Extrafijsalmuscle fibers, 52 initiation of, 254-255, 255
Eye. See Visual system joint kinetics and, 246-248, 247
Eye-hand coordination, 379-385, 405 stance phase, 246-247
assessment of, 422 swing phase, 247-248
development of pincer grasp, 383 kinematic description of, 243-244, 244
emergence of hand orientation, 381-382, 382 muscle activation patterns and, 244-246, 245
emergence of object exploration, 383-384 non-neural contributions to, 253-254
learning to grasp moving objects, 382 pattern generators for, 248-249
in older adults,391, 393-394 phases of, 240-242, 241
roleof experience in, 384-385, 384-385 sensory feedback and adaptation of, 251-253, 278
visually triggered vs. visually guided reaching, proactive strategies, 253, 278
381 278, 279
reactive strategies,
Eye-head coordination, 358-359 somatosensory systems, 251-252
assessment of, 421^22, 422 vestibular system, 252-253
problems with, 400^01 visual system, 252
retraining of, 434-436, 435t stair-walking, 255-257
Eye-head-hand coordination, 385 summary of, 263
temporal/distance factors and, 242
Facilitation, 104 cadence, 242
neurofacilitation approaches, 104—106 in elderiy, 284-285
proprioceptive neuromuscular, 104 step length, 242
synaptic, 51, 451 stride length, 242
Fall's, 100, 111, 122,287 velocity of, 255
fear of, 180,288 Gait abnormalities, 295-310
gait changes related to, 288 due to cerebral palsy, 309-310
among older adults, 171-172, 178, 283 due to musculoskeletal limitations, 187, 296
protective responses during, 149, 152 due to neuromuscular impairments, 296-298
stepping for prevention of, 229-230 control problems, 297-298
Fast movements, 8 muscle tone and changes in stiffness, 297
Fastigial nucleus, 77 weakness, 296-297
Feedback, 33-34, 228-229 due to pain, 299
definition of, 34 due to Parkinson's disease, 308-309
extrinsic, 34, 448 fi-ozen gait pattern, 309, 448
conciurent vs. terminal, 34 propulsive gait pattern, 309, 450
knowledge of results, 34—35 due to sensory disorders, 298-299
intrinsic, 34, 448 adaptation problems, 299
response-produced, 34, 450 misrepresentation of stability limits, 298-299
Fitts' law,366-367, 386, 387, 423 somatosensory deficits, 298
Flexibility vestibular deficits, 298
assessment of, 217-218 visual deficits, 298
spinal, 172, 173, 187, 189 due to stroke, 308
Flocculonodular lobe, 77 effects of impairments on stance phase, 299-305
Forced-use paradigm, 249, 441^42, 448 foot contact/loading, 299-301
Fractured somatotopy, 77 coronal plane deviations, 300-301
Frozen gait pattern, 311, 448 impaired heel-strike, 299-300, 300
Fugl-Meyer Test, 108 mid-stance, 301-304
Function backward lean of trunk, 303
definition of, 38 drop in pelvis, 303-304, 304
466 Index
Gait abnormalities continued Greene, Peter, 17-18

302-303, 303
excessi\e hip flexion, Grip. See Upper extremit\' manipulation skills
301-302
excessive knee extension, 301,
lateral lean of trunk, 303 Habituation, 25, 40, 41, 51
persisting knee flexion, 302 definition of, 87, 448
scissors gait, 304 physiological basis of, 87-88, 88
terminal stance, 304—305 Hair cells, 69-70
inadequate toe-off, 304 Hamstring overacri\ity, 302, 307
lack of hip h\perextension, 304 Hand orientation, 381-382, 382
pehic retraction, 304-305, 305 HAT (head, arms, and trunk segments), 247, 297,
effects of impairments on swing phase, 305-307 448
initial s«ing, 305 Head control, 18, 67, 154-155
inadequate hip flexion, 305, 306 described by reflex and systems theories, 155
mid-swing, 305-306 eye-head coordination, 358-359
excessive adduction, 306 motor coordination and, 154
inadequate knee flexion, 305-306, 306 senson- contributions to, 154—155
terminal swing, 307 spontaneous, in neonates, 1 54
inadequate knee extension, 307, 307 stabilization during gait, 278-280
research on, 296 Head motion, 67, 69, 70
during stair-walking, 307 Heel-strike abnormalities, 299-300, 300
summar\' of, 312 retraining for, 336-337, 337
treatment of See Gait training Held, Jean, 38-40
Gait Assessment Rating Scale, 330, 331t-332t Hemiplegia, 111
Gait changes with aging, 283-290, 293 dyssynergia and, 193-194
adaptive control and, 286-288 on stability limits, 199, 199
effect
288
in fallers vs. nonfallers, motor adaptation problems due to, 197
kinematic anah'sis of, 285, 285 motor impairments affecting nonhemiparetic limb,
kinetic analysis of, 286 403
muscle activation patterns and, 285-286 upper extremit>' dyscontrol due to, 407, 407-408
regression hypothesis of, 291-292 Hierarchical theor\' of motor control, 9-10, 9-11, 46,
role of patholog\' in, 288-290 448
stair-walking, 290 clinical implications of, 1
temporal/distance factors, 284-285 limitations of, 10-1
Gait training, 334-343 neurofacilitarion approaches associated with, 105
assistive devices for, 340-341 Hip abductors
at functional level, 341-342 spasticity of, 304
at impairment level, 334—335 weakness of 303-304
limitations of, 335 Hip extensor weakness, 303
pre-ambulation skill training, 334-335 Hip flexors
stair- walking, 343, 344-345 contracture of, 303
335-340
at strategy' level, spasticit)' of, 302-303
diuing stance phase, 336-339 weakness of, 305
heel-strike, 336-337, 337 Hip movements during gait, 247, 248, 299
mid-stance, 337-338, 338 effect of pain on, 299^303
terminal stance, 339, 339 excessive flexion, 302-303, 303
during swing phase, 339-340 flexion contractures, 304
mid-swing, 339, 340 inadequate flexion, 305, 306
pre-swing, 339 Hip strategy, 129, 129-130
terminal-swing, 339-340 retraining use of, 229
summan,' of, 352 use by older adults, 175
task-oriented, 342, 343 Hippocampal neurons, 90-91
Galloping, 276-278 Hoppmg, 276-278
Ganglion cells of retina, 64—65 Horizontal cells of retina, 64
Gastrocnemius/soleus spasticity, 300, 304 Huntington's disease, 79
Gaze, 298, 400 H\permetric responses, 196
General static reactions, 448. See also Reflex(es), Hypenonicity, 189-191, 190, 218, 404. See also
attitudinal Spasticity
Gentile's taxonomy of movement tasks, 109, 109t Hypothesis, 101,448
Genu recur^•atum gait pattern, 310 Hypothesis-oriented clinical practice, 7, 100-102, 114,
Gesell, Arnold, 145, 153, 165,280 448
Get-up and Go test, 208 clean result of hypothesis testing, 102, 111-112
Gibson, James, 18-19 integrating h\pothesis testing Into assessment,
Glabrous skin, 448 111-112
Globus pallidus, 78, 79, 80 Hypotonia, 77, 198
Golgi tendon organs, 53, 54, 251
Grasping. See Upper extremit)' manipulation skills Icing, 228, 228
Gravit)', 253 Immobilization, 187
Index 467
Impairment assessment. 111 Locomotion. See also Gait; Mobility

Impulse variabilit\' model of reaching, 370 assessment of, 315-330
Inertia, 448 development of, 269-280
In-hand manipulation, 425 early stepping behavior, 270—274, 271
retraining of, 441 gait adaptation strategies,278, 279
Inhibitor!,' techniques, 104 maturation of independent walking, 274—276, 275,
Instrumental Activities of Daily Living, 108, 316, 419 276t, 277
International Classification of Impairments, Disabilities prenatal, 270
and Handicaps, 102 run, skip, hop, and gallop, 276-278
Interposed nucleus, 77 summarv' of, 292
Intrafusal muscle fibers, 52, 53, 448 gait, 239-257
Inverse dynamics, 254 minimum standards for independent community
ambulator, 316-317
Jebsen Hand Function Test, 419, 420 in older adult, 269, 283-292
Joint kinetics, 246-248, 24~ requirements for, 240, 292
Joint receptors, 54—55, 251 adaptation, 240, 447
Joint-based planning, 448 progression, 240, 450
stabilit)', 240, 451
Katz Index, 108,421 Long-term potentiation, 90-91, 91. 449

Kinematic analysis, 124
ofgait, 243-244, 244 Ml responses, 129
285
age-related changes, 285, M2 responses, 129
273
infant vs. adult, 273, M3 responses, 128-129
of reaching movements, 387-389, 389-390 Manipulation tests, 419^20, 420
Kinetic analysis, 125 Mass panerns of movement, 402^03
of age-related gait changes, 286 Mechanoreceptors, 55
inverse dynamics, 254, 286 Medial lemniscal pathway, 57
Knee movements during gait, 248 Medial vestibulospinal tract, 70
effectof pain on, 299, 302 Membranous labyrinth, 67
excessive flexion, 302, 337-338 Memon', 88
extension contractures, 305-306 definition of, 86
flexion contractures, 307 learning and, 86
flexor svnerg\-, 307, 307 localization of, 86-87
h>perextension, 186, 186. 301, 301-303, 337 long-term, 86, 89, 449
inadequate extension, 300, 307 sensor\', 93
inadequate flexion, 305-306 short-term, 86, 89, 450
retraining for problems with, 337-338, 338 spatial, 90
Knowledge of performance, 34, 448 Memon- trace, 28, 449
Knowledge of results, 29, 34-35, 448^49 Mental practice, 37
Mental status assessment, 216-217
Landau reaction, 148, 149, 155, 449 Mesencephalic locomotor region, 249
Lateral geniculate nucleus, 65 Mini-Mental State Exam, 217
Lateral inhibition, 59 Minnesota Rate of Manipulation Test, 420
Law of developmental direction, 145 Mirrors, 225, 225
Law of effect, 27 Mobility'
Learned disuse, 405, 441-442 control of, 239-264
Learning, 24—28, 40—il See also Motor learning
. locomotion. See also Gait; Locomotion
associative, 25-28, 41 abnormal, 295-313
neuralplasticity' and, 89-93 development of, 269-280
classical conditioning, 26, 26, 41, 89, 447 gait 239-257
c-\'cle,
as continuum of short- to long-term changes, requirements for, 240

86, S6 of older adult, 269, 283-292
declarative, 27-28, 41, 90-91, 448 age-related gait changes, 283-290, 287t
definition of, 24, 85, 86, 449 compared with 291-292
infant,
on structure and
effect fijnction of brain neurons, rising-lrom-bed movement, 290-291, 291
87 sit-to-stand movement, 290
86-87
localization of, other than gait, 257-263
memor)' and, 86 abnormalities of, 311-312
nonassociative, 25, 41 bed mobilit)' skills, 312
physiological basis of, 87-89, 88 sit-to-stand, 311-312
operant conditioning, 26-27, 41, 89-90, 449 developing skills for, 280-283
perceptual, 93 prone progression, 280—282, 281-282
procedural, 27-28, 41, 91-93, 450 rolling, 280
sensorv', 25 supine-to-stand, 282-283, 283
Learning disabilities, 202 rising-trom-bed, 261, 261-262
Level of consciousness, 216, 217t rolling, 262, 262-263
Location programming, 369, 371-372, 371-373 sit-to-stand, 257-259, 258
468 Index
Mobility continued 19-20

integration of,
supine-to-stand, 259-261, 260 motor programming theories, 11-13, 12
transfers, 257 parallel distributed processing theory, 16, 16-17, 46,
summan' of,263-264 449
Mobility assessment, 209, 210t, 315-330 physiology' and, 45-46, 46
at fiinctional level, 316-319 reflex theoty, 7-9, 8
quantifying temporal/distance factors, 318-319, relationship of practice to, 5-7, 99-100, 103-104
319 dynamic, evolving ideas, 7
three-minute walk test, 317-318, 318 fi-ameyvork for interpreting behavior, 5-6, 6
at impairment level, 330 guide for clinical action, 6-7
levels of, 315-316 parallel dey elopment of practice and scientific
scales for, 319-322 theoty, 104
Duke Mobility Skills Profile, 319, 320t-322t yvorking hy-pothesis for assessment and treatment, 7
Dynamic Gait Index, 322, 323t-324t summaty' of, 20
Functional Independence Measure, 319, 322 systems theoty, 13-14
limitations of, 322 task-oriented theories, 16-18
at strateg)- level, 322-330 Motor cortex, 71-75, 72-74
obser\'ational gait analysis, 324—330 areas of, 7
Mobility Assessment Form, 328 contributions to corticospinal tract, 71
Mobility retraining, 330-351 fiinction of, 73-75
gait training, 334-343 sources of inputs to, 71
340-34
assistive de\ices for, Motor development, 143-145, 144
341-342
at fiinctional level, Motor learning, 23-37. Sre also Learning
at impairment 334—335
level, Adams' closed-loop theory of, 28, 447
stair-walking, 343, 344-345 definition of, 23-24, 40, 85, 449
at strategy- level, 335-340 factors contributing to, 33-37
goals of, 316, 330-334 feedback, 34-35
long-term, 330, 332 practice conditions, 35-37
related to adaptation, 334 Fitts-Posner stages of, 31, 31-32
related to progression, 333 associative, 3 1447,
related to stability, 333-334 autonomous, 31, 447

short-term, 332-333 cognitive, 31, 447
of other skills, 343-351 Newell's exploration theon' of, 32, 32—33
rising-from-bed, 349-350, 351 recovety of function and, 23-24, 40-41
rolling, 351 relation to performance, 24—25
sit-to-stand position, 345-351, 346-349 role of cerebellum in, 78
summary' of, 351-352 Schmidt's schema theoty of, 28-31, 30
Models of disablement, 100, 102. 102-103, 114, 449 summan' of, 4
clinical implications of, 103 Motor milestones, 144, 145
Nagi model, 103 Motor neurons, 52-54
Schenkman model, 103 alpha, 52, 447
World Health Organization model, 102-103 gamma, 52, 448
Modified Ashworth Scale for Grading Spasticity', 2 1 9t Motor problems
Moro response, 144, 145, 154 406—107, 407-408
affecting agonist muscles,
Motion perception, 220-221 408
affecting antagonist muscles,
Motion sense, 64, 67 401—103
dyscoordination,
Motivation, 217 Motor program, 449
Motor adaptation problems, 196, 196-197, 198t Motor programming theories, 11-13, 12
Motor Assessment in Hemiplegia, 108 clinical implications of, 12-13
Motor Assessment Scale for Stroke Patients, 108 limitations of, 12
Motor control, 3^ Motor strategy' assessment, 211-212, 211-214
definition of, 3 Movement strategies, 240. See also Mobility
as interaction of individual, task, and emironment, 4, assessment of, 21 1-212, 213-214
4, 120, 120 retraining of, 226-230
physiology' of, 45-81 Multiple corrections theoty of reaching, 369-370
reasons for study of, 4 Muscle control problems, 297-298
in relation to action, 3 Muscle reeducation approach, 104
in relation to cognition, 4 Muscle response latencies, 1 34
in relation to perception, 3^ Muscle spindles, 52-54, 53
Motor control theories, 4-20, 145-146 intrafusal fibers of, 52, 53, 448
brain ftincrion models and, 4—5 innenation of, 52
definition of, 4, 451 nuclear bag fibers, 52
differences betyveen, 5 nuclear chain fibers, 52
dynamical action theoty, 14—16, 15 sending information via groups la and II afterents,
ecological theon', 18-19, 19 52-53

hierarchical theory, 9-10, 9-11, 46, 448 utilization of information from, 53-54
Index 469
Muscle stiffness, 54, 190, 402 Mobility Assessment Form, 328

abnormal gait due to changes in, 297 Rancho Los Amigos Gait Analysis Form, 328, 329
Muscle strength 330
limitations of,
assessment of, 218 problem-oriented, 325t-326t
for upper extremit\" manipulation, 426^27, 425 during stance phase, 326-327
of older adults, 170,172-173 during swing phase, 327-328
role in transition to independent standing, 158 Occipital lobe, 66
Muscle strength training, 181, 222-223 Ocular dominance columns, 66
Muscle tone,"l25, 402,^449 Operant conditioning, 26-27, 41, 89-93, 449
abnormal gait due to problems with, 297 Optic ataxia, 404, 404
abnormalities of, 189-191, i 90 Optic nerve, 65
assessment of, 21 Optic tract, 65-66
treating impairments of, 223 Optimized initial impulse model of reaching, 370-371
Muscles Otoconia, 69-70
activation patterns in older adults, 285-286 Otoliths, 69, 133
agonist, 406-i07, 407-40S
antagonist, 408 Pain, 299, 302, 303
tonicallv active during control of quiet stance, 123, upper extremit)', 401, 429^30
126 Parachute responses, 149, 152, 449
Musculoskeletal impairments, 187-188, 18S-189 Parallel distributed processing theor\' of motor control,
abnormal gait due to, 298 16-17, 46, 449
i6,
assessment for, 2 1 7-2 1 17
cfinical implications of,
treatment of, 222 limitations of, 17
upper extremitT,' dyscontrol due to, 401 Parkinson's disease, 17, 78, 79, 100, 197-198
treatment of, 431^32, 432-433 assessing symptom severity in, 108-109
dyssynergia in, 195
Nagi model of disablement, 102, 103 gait abnormalities 308-309
due to,
Neonatal stepping patterns, 270-274, 271, 273 frozen gait pattern, 309, 448
Neural plasticit\', 86-97 propulsive gait pattern, 309, 450
associative learning and, 89-93 motor adaptation problems in, 196, 196
definition of, 86, 449 Pehic drop, contralateral, 303-304, 304
effect of learning on, 87 retraining for, 338
global aspects of, 95-97 Pelvic retraction, 304-305, 305
recoven' of fiinction and, 93-97 Perception, 3^, 45, 46
short-term functional, 86, 449 assessment of, 220-221
of somatosensory' cortex, 95-96, 96 treating impairments in, 224
structural, 86, 449 Perceptual trace, 28, 449
Neurofacilitation approaches, 104-106 Perceptual-motor workspace, 32, 32-33
changing practices and, 106 Performance
clinical implications of, 105-106 definition of, 24
limitations of, 106 knowledge of, 34, 448
underlying assumptions of, 105 relation to learning, 24-25
abnormal motor control, 105 Performance -based fiinctional measures, 108-110, 449
normal motor control, 105 limitations of, 109-110
recover}' of function and reacquisition of skill, 105 Perilymph, 67
Neuromaturational tlieor\' of development, 10, 10 Peripheral sensory neuropathy, 405, 406
Neuromuscular impairments, 188-198 Photoreceptors, 63-64
abnormal gait due to, 296-298 Physical Capacity Evaluation, 420-421
assessment for, 218-219 Physiology of motor control, 45-81
treatment of, 222-223 action systems, 70-79
Neuronal response to injury, 93-95, 94-95 association areas of prefi-ontal regions, 75
Neuronal shock, 94, 449 basal ganglia, 78-79
Neurons, 50, 51 cerebellum, 76, 76-78
Newell, Karl, 32, 32-33 motor cortex, 71-75, 72-74
Nociceptors, 55 brain ftmction, 46-50, 48^9
Nonlinear behavior, 14-15, 15 motor control theories and, 45—i6, 46
Novel movements, 8-9 neurons, 50, 51
Nuclear bag fibers, 52, 53, 54 sensory /perceptual systems, 51-70
Nuclear chain fibers, 52-54, 53 somatosensory system, 52-62
Nudge Test, 111 vestibular system, 67-70
Nystagmus, 70 \'isual system, 62-67
summan' of, 81
Object identification, 64 Physiology of motor learning and recovery of function,
Obsen'ational gait analysis, 324—330, 352
'
85-97
forms for, 328-330 effects of learning on structure and function of brain
Gait Assessment Rating Scale, 330, 331t-332t neurons, 87
470 Index
Physiology' of motor learning continued Postural control assessment, 208-22

learning and memory, 86, 86 fiinctional, 208-211
localization of,86-87 balance and mobility scale, 209, 210t
neural plasticity, 86-97 Functional Balance Scale, 209
associative learning and, 89-93 Functional Reach Test, 208, 209, 209t
definition of, 86 limitations of,209-21
recovery of function and, 93-97 Up and Go 208
test,
nonassociative forms of learning, 87-89, 88 interpretation of, 22
summary of, 97 patient safety during, 208
Placing reaction, 126 perception, 220-221
Plantarflexors motion perception, 220-221
contractures of, 299-301, 306 stability limits, 220
inadequate activation 302 of, strategy assessment, 211-216
retraining for problems of, 336 motor strategies, 211-212, 211-214
spasticit)' of, 301-302, 306 sensory strategies, 212-216, 214-216
weakness of, 304—305 summary of, 220t, 221, 232
Plasticity. See Neural plasticity systems assessment, 216-220
Polio, 104 cognitive systems, 216-217, 217t
Positioning of patient, 223 musculoskeletal system, 217-218, 218t
Postural control, 3, 18, HI, 119-140 neuromuscular system, 218-219, 219t
adaptive, 121, 447 sensory systems, 219-220
137-138, 447
anticipator^', for upper extremity manipulation, 425
development of, 164-165 Postural control development, 143-166
development of strategies for, 227-228 anticipatory postural actions, 164-165
in elderly persons, 178-179 emergence of independent sitting, 155-157
loss of, 197 emerging head control, 154-155
compensator)' strategies for, 185-186, 186 motor milestones and, 144, 145
defining systems for, 120-121, 121 new models of, 151-153
defining task of, 120 refinement of stance control, 160-164
definition of, 366,449 summary of, 165-166
motor mechanisms for, 122-131 theories of, 119-120, 145-151, 146-152, 153t
adaptation motor strategies, 130, 130-131 transition to independent stance, 157-160
control of quiet stance, 123-126 Postural dyscontrol, 185-203
alignment, 123, 123-125 compensatory strategies for, 185-186, 186
muscle tone, 125 due to musculoskeletal impairments, 187-188,
postural tone, 126 188-189
during perturbed stance, 126-130, 128 due to neuromuscular impairments, 188-198
ankle strategy, 127-129, 129 abnormalities of muscle tone, 189-191, 190
feedback and feedforward, 127 alignment, 191-192
laboratory studies of, 126, 127 cerebellar disorders, 198
muscle synergies, 126-127 of anticipatory postural control, 197
loss
requirements varying with fijnctional tasks, motor adaptation problems, 196, 196
121-122 motor problems by diagnosis, 197, 198t
seated, 138-139, 139 Parkinson's disease, 197-198
sensory mechanisms for, 131-137 scaling problems, 195-196
adaptations of, 134-137, 135-136 and hemiplegia, 197
stroke
to rotational support surface perturbations, timing problems, 192-195, 193-195
136 weakness, 188-189
sensorimotor, 136-137 due to positive and negative effects of CNS lesions,
when learning new tasks, 136 185
during perturbed stance, 133-134 due to sensory disorders, 198-203
during quiet stance, 133 inability to adapt senses, 200-202, 201-202
somatosensory inputs, 132 misrepresentation of stability limits, 199,
vestibular inputs, 132-133 199-200
visual inputs, 131-132 sensorimotor adaptation, 202-203
summary of, 139-140 summary of, 203
upper extremit)' manipulation skills and, 366 systems perspective of, 186-187, 187
Postural control and aging, 169-182 upper extremit)' ftinction and, 410
anticipatory postural abihties, 178-179 Postural dyscontrol treatment, 221-232
balance retraining, 180-181 at fiinctional task level, 231-232
behavioral indicators of instability, 171-172 at impairment level, 222-224
cognitive abilities, 179-180 cognitive impairments, 222, 222t
musculoskeletal system, 172, 172-173 musculoskeletal impairments, 222
neuromuscular system, 173-175, 174-175 neuromuscular impairments, 222-223
sensory systems, 175-178 perceptual impairments, 224
summary of, 181-182 sensory impairments, 223-224
Index 471
at strategy level, 224-231 Reaction-time reaching tasks, 385-386

alignment, 224-226, 225-226 Recall schema, 29, 450
developing coordinated ankle strategy, 227, Receptive fields, 64-65, 450
227-228 changes induced by lesions of, 96
developing coordinated hip strategy, 229 Recognition schema, 29, 450
developing coordinated step strategy, 229-230, Recovery, 450
230 Recovery of function, 38-41
movement strategies, 226-230 assumptions of neurofacilitation approaches
scaling problems, 228-229 regarding, 105
sensory strategies, 230-231, 231 assumptions of systems-based task-oriented approach
timing problems, 228, 228-229 regarding, 105
summary of, 232 vs. compensation, 38, 113-114
for upper extremity manipulation, 441 definition of, 23, 38, 450
Postural fixation reactions, 149, 151, 450 factors affecting, 38-40
Postural orientation, 120, 450 age, 39
Postural stability, 120, 450 experience, 39—40
Postural tone, 126, 450 qualit}' of lesion, 39
Practice and training, 35-37 training, 40
balance retraining for older adults, 180-181 motor learning and, 23-24, 40-41
contextual interference and, 36 neuronal response to injury, 93-95, 94—95
distributed, 36, 448 spontaneous vs. forced, 38, 41
effect on recovery, 40, 41 stages of, 38
fiinctional reorganization of somatosensory cortex Reflex arc, 7
due to, 96, 96 Reflex theory of motor control, 7-9, 8
generalizability of, 29, 36 cfinical impfications of, 9
guidance during, 37 limitations of, 8-9
massed, 36, 449 neurofacilitation approaches associated with,
mental, 37 104-106
mobility retraining, 330-351 Reflex(es), 5, 55
postural training, 221-232 attitudinal, 146, 147, 147, 450
transfer of, 37 chaining of, 8
treatment planning for, 101, 113-114 cutaneous, 252
variable, 36 definition of, 450
whole vs. part, 36-37 flexor withdrawal, 55, 448
Precision grip, 367-368, 406 hierarchical organization of, 9-11
Prefi-ontal cortex, 75 modification of, 1
Premotor cortex, 71, 72, 74-75 modulation during locomotion, 251-252

Pretectal region, 66 monosvnaptic, 449
Primary motor cortex, 71, 72, 73-74 Moro,'l44, 145, 154
Progression newborn stepping, 270-274
effect of musculoskeletal hmitations on, 296 overriding of, 8
mobilit)' retraining goals related to, 333 postural, 146-152, 146-152, 153t
prone, 280-282, 2S1-282 role in development, 150-151
requirement for locomotion, 240, 450 role in development of reaching beha\iors, 377-378
Proprioceptive Neuromuscular Facilitation, 104 spinal, 252
Propulsive gait pattern, 309, 450 stretch, 54, 125, 129, 251-252
Protective responses, 149, 152, 449 changes in threshold for activation of, 190-191
PULSES Profile, 419 tonic labyrinthine, 147, 147, 451
Purdue Pegboard test, 419^20 tonic neck, 126
Purkinje cells, 78,91-92 asymmetric, 146-147, 147, 147, 447, 451
Putamen, 78, 79, 80 vestibulocoUic, 126
Pyramidal tract. See Corticospinal tract vestibulo-oculomotor, 70, 78, 92-93
vestibulospinal, 126
Quadriceps Reflex-hierarchical theory of postural control, 119,
compensation for weakness of, 302, 303 146-151, 146-152, 153t, 166
spasticity of, 302, 306 emergence of independent sitting, 157
emerging head control, 155
Rancho Los Amigos Gait Analysis Form, 328, postural reflexes in human development, 146-150
329 attitudinal reflexes, 147, 147
Rancho Los Amigos Scale, 216, 217t balance and protective reactions, 149-150,
Range fractionation, 55 150-152
Range of motion righting reactions, 147-149, 148-149
assessment of, 217, 218t role of reflexes in development, 150-151
in older adults, 173 transition to independent stance, 160
for upper extremity manipulation, 426 Resting potential, 50, 51, 450
Reaching. See Upper extremity manipulation skills Retina. Sec Visual system
472 Index
Retraining strategies, 113-114 emergence of independent sitting, 155-157

Righting reactions, 146-149, 14S-149, 450 described by reflex and systems theories, 157
bodv-on-bodv, 149, 150, 157, 447 motor coordination for, 155-156, 156
body-on-hcad, 149,447 156-157
sensor\' contributions to,
labyrinthine, 147, 449 postural control for, 138-139, 139
Landau, 149 Sit-to-stand movement, 257-259, 258
neck-on-bodv, 149, 150, 157,449 abnormalities of, 311-312
optical, 149, 449 in older adults, 290
in Parkinson's disease, 195 retraining of, 345-351, 346-349
197-198
Rigidit)', force-control strategy, 345-347
Risingfrom-bed movement, 261, 261-262 momentum strategy, 347-350
abnormal, 312 Skipping, 276-278
in older adults, 290-291, iW Somatosensory system, 52-62
retraining of, 349-350, 351 ascending pathways of, 56, 57-59, 58
Rods, 62, 63 anterolateral system, 57-59
262-263
Rolling, 262, dorsal column-medial lemniscal system, 57
abnormal, 312 association cortices, 61, 61-62
development of, 280 influence on postural control, 132, 186
retraining of, 351 influence on 126
postural tone,
Rood approach, 104 in older adults, 175-176
Running, 276 peripheral receptors of, 52-55
cutaneous receptors, 55
Saccule, 69-70 Golgi tendon organs, 53, 54
Scale for Instrumental Activities of Daily Living, 108, joint receptors, 54-55
419 muscle spindle, 52-54, 53
Scaling problems, 195-196 role at spinal cord level, 55-57
treatment of, 228-229 role in modificationof gait, 251-252, 298
Scarpa's ganglion, 70 role in reach and grasp fiinctions, 363-364
Schema theory of motor learning, 28-31, 30, 450 somatosensory cortex, 59, 60
Schenkman model of disablement, 102, 103 fijnctional reorganization after injur\', 95-96,
Schmidt, Richard, 28-29, 30 96
Schwab Classification of Parkinson Progression, thalamus, 59
109 upper extremity dyscontrol due to problems of, 405,
Scissors gait, 304 406,410-411
Self-organizing system, 14, 450 5-7, 189-191, 190, 218
Spasticit\',
Semicircular canals, 68, 69, 70, 132 abnormal gait due to, 190, 297
Sensitization, 25, 41 assessment for, 218, 219t
definition of, 87, 450 definition of, 402, 451
physiological basis of, 88, 88-89 of gastrocnemius/soleus, 300, 304
short- and long-term, 89 of hip abductors, 304
Senson' assessment, 219-220 ofhip flexors, 302-303
of upper extremit\', 427^29, 429 of plantarflexors, 301-302, 306
Sensory differentiation, 62-64 retraining for, 336
Sensory disorders, 198-203 of quadriceps, 302, 306
gait abnormalities due298-299
to, treatment of, 223
gait changes in elderly due to, 288-289 upper extremity' dyscontrol due to, 402
inability to adapt senses, 200-202, 201-202 Speech sparing, 39
misrepresentation of stabilit)' limits, 199, 199-200 Spinal cord, 47, 48
sensorimotor adaptation, 202-203 Spinal locomotor preparation, 249-251, 451
treatment of, 223-224, 432^33, 434t decerebrate, 249, 447-4-48
upper extremity dyscontrol due to, 403^05, decorticate, 250-251, 448
408-410, 409-410, 432--i33 Spinocerebellar tracts, 77
Sensory Integration Therapy, 104 Splints, 432
Sensory selection problems, 202 StabUit\-, 120
Sensory stimulation techniques, 223 behavioral indicators of instability', 171-172
Sensory strategies measurement of, 318
assessment of, 212-216, 214-216 mobilit}' retraining goals related to, 333-334
for modifying gait, 251-253 requirement for locomotion, 240, 451
retraining of, 230-231, 231 120, 122, 199. 451
Stabilit>' limits,
Serial processing, 16 assessment of, 220

Sherrington, Sir Charles, 7-9 misrepresentation of, 199, 199-200
Short Portable Mental Status Questionnaire, 217 abnormal gait and, 298-299
Shortening reactions, 402 retraining of, 21
Shuflfling gait pattern, 308-309 Stair-walking, 255-257, 256
Sitting abnormal, 307
atypical postures for, 187, 188 descent, 256
Index 473
in older adults, 290 Summation, 50-51

retraining for, 343, 344-345 excitator\', 448
Stance, 122-138. Sec also Ciait spatial, 50-51,451
aging and, 173-175 temporal, 50, 451
adapting movements to changing tasks and Superior colliculus, 66
en\ironments, 175, 175 Supine-to-stand movement, 259-261, 260
changes in motor strategies during perturbed development of, 282-283, 283
stance, 173-175, 174 Supplementary motor area, 7 1 72, 74—75 ,
changes in quiet stance, 173 Support moment, 451

posture in, 122-138. See aim Postural control Surface-dependent pattern, 202, 214—215, 216
refinement of stance control, 160-164 Sway control. See Postural control
compensatory, 161-162, 162 Synaptic defacilitation, 51, 451
development of sensory adaptation for, 163-164, Synaptic effectiveness, 94, 95
164 Synaptic facilitation, 51, 451
motor coordination for, 160-162 Synaptic transmission, 50, 451
quiet stance, 160-161 Synaptogenesis, 95
sensor)' contributions to, 162-163 reactive, 95, 450
transition to independent stance, 157-160 regenerative, 95, 450
described by reflex and systems theories, 160 Svnergies, 13, 14, 126-127,451
development of adaptive capability for, 160 abnormal, 193, 193
de\elopnient of muscle svnergies for, 75cS', aflecting upper extremirs' manipulation,
158-159 427, 428t
motor coordination for, 157-159 associated with ankle strateg)', 127-129, 129
roleof strength in, 158 associated with hip strategy, 129, 129-130
sensory contributions to, 159-160 development of, 158, 158-159
Static reactions, 146 for upper extremit)' function, 402
general, 146 Systems approach, 20
local, 146, 449 Systems theory of motor control, 13-14, 20, 358
segmental, 146, 450 clinical implications of, 14
Step length, 242, 451 limitations of, 14
Stepping patterns, neonatal, 270-274, 271, 273 Systems theor)' of postural control, 119-121, 165, 166
Stepping strateg\', 130 emergence of head control, 155
retraining use of, 229-230, 230 emergence of independent sitting, 157
Strategy, 45 transition to independent stance, 160
sensorimotor, 450 Systems-based task-oriented approach, 106-107
sensor)', 230-231,257, 450 conceptual framework for clinical practice based on,
Strateg)' assessment, 110-111 107-114
of mobilitv', 322-330 underlying assumptions of, 106-107
of postural control, 211-216 abnormal motor control, 107
motor strategies, 211-212, 211-214 normal motor control, 106
sensory strategies, 212-216, 214-216 recovery of function and reacquisirion of skill, 107
of upper extremit}' manipulatory dyscontrol, 421^26
Strategy level treatment, 224-231 Tactile defensiveness, 25
alignment, 224-226, 225-226 Target location, 358-359
for mobilitN' retraining, 335-340 assessment of, 421—i22, 422
movement strategies, 226-230 problems with, 400—iOl
developing coordinated ankle strategy, 227, Task analysis, 36-37
227-228 Task-oriented conceptual fi'amework, 99-1 14, 207,
developing coordinated hip strategy, 229 451
developing coordinated step strategy, 229-230, assessment, 107-112
230 transitionfrom assessment to treatment, 112-1 14
scaling problems, 228-229 Task-oriented theories of motor control, 16-18, 106
timing problems, 228, 228-229 clinical implications of, 18
sensor)' strategies, 230-231, 231 limitations of, 1
perceived limits of stabilit)', 231 Tectopontine tract, 66

for upper extremit)' manipulator\' dyscontrol, Tectospinal tract, 66
434-^41 Temporal lobe 90
lesions,
Stride length, 242, 451 Temporal/distance factors, 242
Stroke, 5, 113 in elderly,284-285
dyssynergia and, 193-194, 193-195 quantification of, 318-319
gait abnormalities due to, 308 Thalamus, 59, 65, 80
motor adaptation problems due to, 197 Theories
sensor)' selection problems due to, 202 of motor control, 4—20, 45
weakness due to, 189 of motor learning, 28-33
Substantia nigra, 78, SO of postural control development, 1 19-120, 145-153
Subthalamic nucleus, 78, SO Thermoreceptors, 55
474 Index
Three-minute walk 317-318, 318

test, summaries of, 373, 394-395
energ)' efficicnq-, 318 summary of, 394-395
measuring stability', 318 theorieson control of reaching, 369-373
self-paced velocity, 317-318 distance theories, 369-371
Tibialis anterior inactivit)-, 300, 324 multiple corrections theory, 369-370
retraining for, 336-337, 337 optimized initial impulse model, 370-371
Tilting reactions, 149, 150, 451 Schmidt's impulse variability model, 370
Timing problems, 192-195, 193-195 distance vs. location programming, 369
treatment of, 228, 228-229 location programming, 371-372, 371-37S
Tinetti Test of Balance and Mobilit\', 108, 209, 210t Upper extremity' manipulatory dyscontrol,
Toe-drag, 306, 306 399^13
Training. See Practice and training adaptation problems, 410^1
Transfer of training, 37 anticipator)- aspects, 411
Transfer skills, 257 somatosensor)- deficits, 4 10^ 11
abnormal, 312-313 visual deficits, 410
Transpon problems, 401—i05 apraxia, 41 1^12
Treatment goals, 1 1 2- 1 1 3, 2 1 7, 3 1 6, 330-334, 430 assessmentof, 418—i30
Treatment planning, 101, 113-114 impairments affecting grasp and release, 405—110
Tremor, action (intention), 198 motor problems, 406—108
Trunk affecting agonist muscles, 406-407, 407-408
backward lean of, 303 affecting antagonist muscles, 408
for\\ard- flexed, 338 sensor)'problems, 408-110, 409^10
lateral lean of, 303 postural problems, 410
Two-point discrimination tests, 429 summan' of, 412^13
target location problems: eye-head coordination,
Unified Rating Scale for Parkinsonism, 109 400-101
Up and Go test, 208 400-401
visual deficits,
Upper extremity- manipulation skills, 357-373 401—105
transport problems,
basic characteristics of reaching tasks, 366-369 motor dyscoordinadon, 401^03
complex reaching and bimanual tasks, 367-369, 368 affecting nonhemiparetic limb, 403
development of, 377-389 dyscoordinarion, 403
eye-hand coordination, 379-385 mass patterns of movement, 402^03
development of pincer grasp, 383 musculoskeletal restraints, 401
emergence of hand orientation, 381-382, 382 pain and edema, 401
emergence of object exploration, 383-384 spasticit)', 402
learning to grasp moWng objects, 382 weakness, 401-102
role of experience in, 384-385, 384-385 sensor)' impairments, 403—105
visually triggered vs. \-isuallv guided reaching, optic ataxia, 404, 404
381 peripheral senson' neuropathy, 405, 406
eve-head-hand coordination, 385 Upper extremitv' manipulator\' dyscontrol assessment,
Fitts' law, 366-367, 386, 387 418^30
in -hand, 425,441 functional assessment, 418^21
innate vs. learned behavior, 378, 379-380 Physical Capacity Evaluation, 420-421
key elements of, 357-358, 399 of activities of daily living, 419
tests
kinematics of reaching movements, 387-389, of manipulation and dexterit)', 419-420
tests
389-390 Jebsen Hand Function Test, 419, 420
locating target: eye-head coordination, 358-359 Minnesota Rate of Manipulation Test, 420
movement accuracy, 386-387, 388 Purdue Pegboard test, 419—120
in older adults, 389-394 impairment assessment, 426-430
changes in reaching movement time, 391-393, abnormal synergies, 427, 428t
392, 392t edema and pain, 429^30
compensation and reversibility of decrements in range of motion, 426
reaching performance, 393-394 sensation,427^29, 429
summar)' of, 395 strength,426-427, 427
postural control and, 366 42 1^26
strategy assessment,
reach and grasp, 359-366 eye-hand coordination, 422
adaptation of, 364—365, 365 eye-head coordination, AlX^ll, 422
motor components of, 359-362, 360-361 in-hand manipulation skills, 425
precision grip, 365-366 planning and sequencing of acriWties of daily living,
role of senses in, 362-364 425-426
anticipator)' aspects of reaching, 364 postural control, 425
somatosensory' contributions, 363-364 reach and grasp, 422—425
visual guidance in reaching, 362-363 summar)' of, 442
visually controlled reaches across midline, 363 Upper extremir\' manipulator\' dvscontrol treatment,
reaction-time reaching tasks, 385-386 430-142
role of reflexes, 377-378 goals of, 430
Index 475
at impairment level, 431^33 in older adults, 176

reducing musculoskeletal impairments, 431^32, peripheral receptors of, 67-70
432-433 semicircular canals, 69
sensorv' reeducation, 432-433, 434t utricle and saccule, 69-70
learned disuse, 441^42 role in modification of gait, 252-253, 298
proximal control as prerequisite for, 431 static and dynamic fiinctions of, 70
at strateg)' level, 43 1 41 1 Visual dependence pattern, 201-202, 214, 216
in-hand manipulation, 441 Visual disorientation, 404
postural control, 441 Visual field, 65
release, 440-441 Visual proprioception, 62
retraining eye-head coordination, 434-^36, 435t Visual system, 62-67, 63
retraining grasp, 438^40, 439 central pathways, 65-66
attending to relevant perceptual cues, 440 lateral geniculate nucleus, 65
task-dependent changes in grasp, 439^40 pretectal region, 66
retraining problems in planning and sequencing superior colliculus, 66
activities of daily liWng, 441 higher order visual cortex, 66-67
retraining transport phase of reach, 436-437, influence on postural control, 131-132
436-^38 emergence of head control, 154—155
adaptive positioning, 437 emergence of independent sitting, 156-157
biofeedback and functional electrical stimulation, refinement of stance control, 162-163
437 transition to independent stance, 159, 159-160
task-dependent characteristics of reach, 437-^38 influence on postural tone, 126
summary' of, 442 in older adults, 176
transition to, 430 peripheral, 62-65
Upper motor neuron 185-186
disease, horizontal cells, 64—65
abnormal mobilirv' due 295-313 to, photoreceptors, 62-64
alignment abnormalities due to, 191-192 vertical cells, 64
motor adaptation problems due to, 196, 196-197, primary' visual cortex, 66
198t role in modificationof gait, 252, 253, 298
muscle tone abnormalities in, 189-191, 190 and grasp fijnctions, 362-363
role in reach
neurofacilitation approaches to, 104-106 upper extremity dvscontrol due to deficits of,
positive and negative signs of. 111, 185, 203 400^01,410
scaling problems due to, 195-196 optic ataxia, 404, 404
sensor\' disorders and, 199-203
timing problems due to, 192-195, 193-195 Walkers, 226, 340-341
weakness due to, 188-189 Walking. See Gait; Locomotion
Utricle, 69-70 Weakness, 188-189
gait abnormalities due to, 296-297
Valgus foot, 301 gait changes in elderly due to, 287-298
retraining for, 337 of hip abductors, 303-304, 304
Varus foot, 300-301 of hip extensors, 303
retraining for, 337 of hip flexors, 305
Velocity, 16,255,284, 316 of plantarflexors, 304-305
self-paced, 317-318 of quadriceps, 302, 303
Vermis, 77 upper extremity dyscontrol due to, 401-^02
Vertical cells of retina, 64 Weight training program, 181
Vestibular loss pattern, 216, 216 World Health Organization model of disablement, 102,
Vestibular system, 67-70, 68 102-103
central connections of, 70
influence on postural control, 132-134, 188 Zero condition, 146
influence on postural tone, 126
PHYSICAL THERAPY / OCCUPATIONAL THERAPY
The ONLY Bridge Between Theory and

Clinical Practice
Xhe virtual explosion of new research in neuroscience and

motor control has made it difficult to keep pace with the latest
fmdings. Now, Mdtor Control: Theory mndPructUtUApplicmtiom
brings you all the relevant information... in an easily digestible
format. The text focuses on scientific and experimental prin-
ciples and featiu*es:
to clinical practice;
I sFTTT^TTSTtTn iVTIS r
model tor reliabilitatioii
And to make it even easier for you to understand, each

chapter includes:
Motor Control: Theory mnd Practical Applicatione—^c ONE

gap between theorj' and cUnical
text that effectively bridges the
practice.
^.
no 7/97
'^'EWBOOK
ISBN D-bfl3-D77S7-D
45.50 90000

Anne Shumway-Cook, Motor Control Theory and Practical Applications Williams &amp; Wilkins (LK)

Uploaded by

Document Information

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

Anne Shumway-Cook, Motor Control Theory and Practical Applications Williams &amp; Wilkins (LK)

Uploaded by

Copyright:

Available Formats

Motor Control

Theory AND Practical Applications

Anne Shumway-Cookj Ph.D.

Marjorie H. Woollacott^ Ph.D.

Williams & Wilkins

book is protected by copyright. No part of this book may be

Printed in the United States of America

Library of Congress Cataloging in Publication Data

Shumway-Cook, Anne, 1947-

to do and provided us steadfast wisdom and support throufjhout its

Photographs by Da\id Trees, Education and Training Department, Northwest Hospital.

Section I THEORETICAL FRAMEWORK

Section III MOBILITY FUNCTIONS

Section IV UPPER EXTREMITY CONTROL

18. Assessment and Treatment of the Patient

Theories of Motor Control

Introduction relation to specific actions or acti\ities. For ex-

or stand still.- Researchers often study move-

STUDY OF ACTION and motor processes. Perception is essential

and the enxironment, and are clearly integral

While each of these aspects of motor Understanding motor control is easier

Why Should Clinicians Study ways, based on models of brain function.

tation of something, usually a simplified ver-

some extent simplify complex concepts. A • new ideas; and

to the coordination of movement. Theories of FRAMEWORK FOR INTERPRETING

arm and drawn close to her body.

is assumed to be a major determinant of ab-

NEW IDEAS: DYNAMIC AND

flect greater knowledge relating to the theor\'.

function, specifically the inability to acti\'el\-

is abstract. Rather, theories generate hypoth-

control are based on an understanding of the

upon which much of clinical practice is based. Response Response

fluence thinking about motor control today.

nervous system (CNS) related to reflexes, he CLINICAL IMPLICATIONS

sic unit of behavior if both spontaneous and

must be activated by an outside agent. low therapists to predict function. In addi-

absence of sensory input (7). a reflex theor)' to interpreting motor dyscon-

chaining model fails to explain the fact that a practices.

single stimulus can result in varying responses

novel or new situation. within different parts of the nerx'ous system.

control to explain the behaviors they saw in

has been referred to as a nciiromaturatioiial

Neuroanatomical Postural reflex Motor

stimuli or by central processes. of effector systems.

erator. Even when the sensor}' nerves were

an important flmction modulating action.

a locomotor rhxthm without either sensor\'

Synergy Sy-nergy Synergy

Right nana Right arm Left hand

ing movement control. in isolation.

CLINICAL LMPLICATIONS Systems Theory

The dynamical action approach does

emerges from the interaction of multiple el- Parallel Distributed Processing

Figure 1.9. Parallel distributed processing

equivalent to sensor\' neurons, interneurons, PDF is somewhat unique in its emphasis

glia. Clinical symptoms may not be apparent An adaptation of Greene's theor\' of

In the last 50 years, a tremendous LIMITATIONS

sential goal of the postural system. However,

The most significant implication of a

upon to retrain. This requires more than an

LIMITATIONS best theon,' of motor control is one that com-

ical methods for assessing and treating motor 1990.

the task, and the environment. 15. McGraw M. Neuromuscular maturation of

Anne Shumway-Cook, Motor Control Theory and Practical Applications Williams & Wilkins (LK)

Anne Shumway-Cook, Motor Control Theory and Practical Applications Williams & Wilkins (LK)