K

Kernel Methods
Learning via Linear Operators

Kindness

Knowledge Acquisition:
Constructing Meaning from
Multiple Information Sources
IVAR BRATEN, HELGE I. STRMS
Department of Educational Research, University of
Oslo, Oslo, Norway

Altruism and Health

Synonyms

Kinesthetic Communication
Dancing: A Nonverbal Language for Imagining and
Learning

Know-How
Tacit Knowledge

Knowledge
Affective and Cognitive Learning in the Online
Classroom
DICK Continuum in Organizational Learning
Framework
Discourse and the Production of Knowledge

Knowledge Acquisition
Complex Declarative Learning

Learning to construct and integrate; Multipledocuments literacy; Multiple-texts comprehension;

New literacy

Definition
Constructing meaning from multiple information
sources denotes a new form of literacy, which involves
locating, evaluating, and using diverse sources of information, digital as well as printed, for the purpose of
constructing an integrated, meaningful mental representation of a particular issue, topic, or situation.

Theoretical Background
The most influential conceptualization of how individuals construct meaning from multiple information
sources is the documents model proposed by Perfetti
et al. (1999). This model builds on and extends cognitive-processing models describing how readers comprehend a single text or document. Recently, the
documents model has been further elaborated by
Goldman (2004) and Rouet (2006).
In essence, the documents model explains how
individuals who want to understand the contents of
multiple information sources dealing with the same
issue from different perspectives must not only construct an integrated mental representation of the situation described across the sources but also keep the
different sources apart. Thus, both integration and separation processes are required. According to the

N. Seel (ed.), Encyclopedia of the Sciences of Learning, DOI 10.1007/978-1-4419-1428-6,

# Springer Science+Business Media, LLC 2012

1678

Knowledge Acquisition: Constructing Meaning from Multiple Information Sources

documents model, noting and remembering the

sources of the different perspectives or arguments
(i.e., who said what) are crucial because it allows
readers to achieve overall coherence in their mental
representation of the issue even when contradictory
perspectives or arguments are presented. A person
may, for example, read about the positive effects of
using sunbeds on the website of an organization
representing the industry. However, the same person
may also read a strict warning against sunbeds on
a noncommercial, government-sponsored health portal site. In such instances, noting and remembering
differences between the sources themselves, for example, with respect to authors and document types, probably help the person to accommodate both perspectives
in his or her global understanding of the issue with less
difficulty and confusion.
In addition to author and document type, individuals working with multiple information sources may
note and remember such source characteristics as rhetorical goal (intent, audience) and setting (place, date,
cultural setting). These source characteristics are presumably used when individuals assign weight and position to different pieces of information in their overall
representation of the issue. For example, a person may
think that information coming from a particular
source is less trustworthy because it is produced by
a layperson rather than a professional, printed in
a newspaper rather than an encyclopedia, old rather
than new, and intent on selling something rather than
inform. As a consequence, this piece of information is
given less prominence in the overall mental representation of the issue that the person constructs when
reading the sources than information coming from
sources that the person deems more trustworthy
based on their characteristics. According to Rouet
(2006), source characteristics cannot be ignored when
trying to construct meaning from multiple information sources because source information allows the
reader to differentiate documents, and to evaluate the
respective contribution of each document to a global
representation of the situation (p. 68).
Finally, the documents model holds that comprehending multiple information sources involves
constructing links that specify similarities and differences between sources. Such links are supposed to have
the form of predicates which, for example, state that
information in a source complements, supports, or

contradicts information in another source. According

to Perfetti et al. (1999), when a set of documents deals
with a controversial issue, such as in normal scientific
and argumentative discourse, predicates specifying
relationships between sources are often dominated by
a solidarity dimension, that is, indicate whether documents agree/disagree or support/oppose each other.
Whereas sourcesource predicates may sometimes be
explicitly expressed by authors through their use of
references, learners may also have to infer such relationships based on their prior knowledge of contents
and sources.

Important Scientific Research and

Open Questions
Systematic research on the construction of meaning
from multiple information sources started in the
1990s. In a landmark study, Wineburg (1991) found
that expert historians who worked through multiple
documents about a particular historical event tried to
piece together a coherent interpretation of the event
described in the documents, at the same time paying
close attention to the different sources on which this
interpretation was based. In this endeavor, the historians used a strategic approach composed of three
different heuristics. First, they relied on a corroboration
heuristic that involved the systematic comparison of
content across documents to examine potential contradictions or discrepancies among them. Second, they
employed a contextualization heuristic that involved the
use of prior domain knowledge to situate document
content in a broad spatialtemporal context. Third, the
historians relied on a sourcing heuristic, involving the
identification and consideration of source characteristics such as authors, document types, and place and
date of document creation. This information was used
to determine the evidentiary value of each document,
as well as to interpret the documents content. In contrast, high school students participating in Wineburgs
study seldom used these three heuristics when working
with the same documents, for example, paying much
less attention to source characteristics and having
difficulty resolving and even noticing discrepancies
among the sources.
Subsequent work, not only in history but also in
other domains (e.g., science and law), has confirmed
the importance of prior knowledge and strategic
processing to the comprehension and integration of

Knowledge Acquisition: Constructing Meaning from Multiple Information Sources

multiple documents about a particular issue (Goldman

2004; Rouet 2006). Prior domain knowledge seems
fundamental to drawing inferences needed to link
information and viewpoints across diverse sources,
and also to understanding important sources and
source characteristics in a domain, for example,
which publications or author qualifications may give
credence to a claim. Moreover, active strategic processing that involves cross-document elaboration to
compare, contrast, and integrate contents across documents, as well as the evaluation of sources to decide
which to trust and which to mistrust, seems important
for success in the complex task context of multiple
information sources.
Recently, individuals epistemic beliefs, that is, their
beliefs about knowledge and knowing in a domain,
have also been shown to play an important role in
their comprehension of multiple information sources.
Specifically, believing that knowledge in the domain is
tentative or complex or that knowledge claims need to
be justified by cross-checking of sources is more adaptive that believing that knowledge in the domain is
certain and simple and that there is no need or way to
justify knowledge claims (Braten et al. 2011). For example, individuals believing knowledge to be complex
may be more inclined to produce cross-document elaborations while working with multiple information
sources, whereas individuals believing knowledge to
consist of an accumulation of simple facts may be
more inclined to rehearse and paraphrase pieces of
factual information from single sources.
Not only individual but also document and task
variables have been shown to play a role in the comprehension of multiple information sources. For example, integration across documents is easier when the
contents of documents largely overlap, and crossdocument elaboration is facilitated by documents that
explicitly refer to and support or contradict each other
(Goldman 2004). With respect to task variables, some
evidence suggests that comprehension and integration
can be facilitated by argument-centered writing tasks,
that is, by asking students to write argument essays
based on multiple documents (Perfetti et al. 1999;
Rouet 2006). However, research also indicates that
only students with relatively high prior knowledge or
the belief that knowledge is tentative may be able to
take advantage of instructions to construct arguments,
whereas students with low prior knowledge or the

belief that knowledge is certain may actually be more

hindered than helped by such task instructions (Braten
et al. 2011).
Although constructing meaning from multiple
information sources generally demands much of the
learner, research demonstrates that reading about
a controversial topic in multiple source documents
rather than a single source presenting the same content
may facilitate deep-level, integrated comprehension
(Britt and Aglinskas 2002). However, spontaneously,
many students at different educational levels have
been found to have great difficulty coping with this
complex activity and, therefore, need explicit instruction. Britt and Aglinskas (2002) addressed this instructional challenge and clearly demonstrated that students
can be successfully taught both integration and separation processes involved in learning from multiple documents. In their approach, a computer application
called the Sourcers Apprentice was used to present
separate documents about historical controversies, and
students learned to take note of both content and
source characteristics of each document. A series of
evaluation studies with North American high school
students demonstrated that those using the Sourcers
Apprentice improved their ability to integrate information across sources and paid more attention to the
sources themselves when compared with students in
control groups (Britt and Aglinskas 2002).
Despite the foundational scientific work referred to
above, much remains to be known about the construction of meaning from multiple information sources.
First, more online processing data are needed on how
individuals work with multiple documents to comprehend a particular issue, topic, or situation. Useful
approaches in this regard may be the collection and
analysis of concurrent verbal reports (i.e., thinkalouds), traces in the learning materials (e.g., software-logged activities), and eye movements during
task completion. Second, assessing the learning outcomes of working with multiple information sources
in reliable and valid ways remains a vital issue. Building
on preliminary work employing both questionnaires
and essay tasks (Braten et al. 2011), an important goal
is to define standard procedures for assessing individuals competence that can be adapted to different educational levels and subject areas. Finally, given that
only a small set of intervention studies have been carried out, more research on how individuals can be

1679

1680

Knowledge and Learning in Natural Language

effectively taught knowledge, beliefs, and strategies that

can promote their learning from multiple information
sources is greatly needed.

Cross-References
Beliefs about Learning
Learning Strategies for Digital Media
Learning with Multiple Goals and Representations
Literacy and Learning
Naturalistic Epistemology
Online Learning
Strategic Learning
Text Relevance

References
Braten, I., Gil, L., & Strms, H. I. (2011). The role of different task
instructions and reader characteristics when learning from multiple expository texts. In M. T. McCrudden, J. P. Magliano, &
G. Schraw (Eds.), Text relevance and learning from text. Greenwich: Information Age.
Britt, M. A., & Aglinskas, C. (2002). Improving students ability to
identify and use source information. Cognition and Instruction,
20, 485522.
Goldman, S. R. (2004). Cognitive aspects of constructing meaning
through and across multiple texts. In N. Shuart-Faris & D.
Bloome (Eds.), Uses of intertextuality in classroom and educational research (pp. 317351). Greenwich: Information Age.
Perfetti, C. A., Rouet, J. F., & Britt, M. A. (1999). Toward a theory of
documents representation. In H. Van Oostendorp & S. R.
Goldman (Eds.), The construction of mental representation during
reading (pp. 99122). Mahwah: Erlbaum.
Rouet, J. F. (2006). The skills of document use: From text comprehension
to web-based learning. Mahwah: Erlbaum.
Wineburg, S. (1991). Historical problem solving: A study of the
cognitive processes used in the evaluation of documentary and
pictorial evidence. Journal of Educational Psychology, 83, 7387.

Knowledge and Learning in

Natural Language
CHARLES YANG
Department of Linguistics & Computer Science,
University of Pennsylvania, Philadelphia, PA, USA

Definition
Knowledge and Learning in Natural Language (KLNL)
(Yang 2002) is a synthesis of computational learning

and theoretical linguistics to provide explanations for

child language development. It was a first effort to
integrate probabilistic learning mechanisms with the
theory of Universal Grammar, and has provided
a quantitative link between the statistical properties of
the input data to the learner and the developmental
patterns of the learners grammar.

Theoretical Background
The Principles and Parameter framework (Chomsky
1981) is a response to the challenges posed by language
learnability. By attributing the totality of language
variation to a finite set of parameters, the learners
hypothesis space may be effectively constrained to facilitate language acquisition. Even though the parameterbased approach is most closely identified with the
Chomskyan approach to language, most modern linguistic theories similarly admit only a finite range of
possible grammars; the learners task is to select those
used in his or her linguistic environment.
Thus, the acquisition problem becomes one of
parameter setting. The dominant approach follows
the conception of triggering (Chomsky 1981; Gibson
and Wexler 1994). The learner is identified with
a grammar (i.e., a string of parameter values) and
makes changes to that grammar (i.e., changing parameter values) as input data are processed. Aside from
learnability problems with this approach, the assumption of one grammar at a time fails to account for the
fact that childs language during acquisition generally
cannot be identified with a single adult-like grammar.
Moreover, child grammar typically develops gradually,
challenging the notion of triggering where the learner
makes categorical changes to his or her grammar
(Valian 1991).
KLNL makes a break with the tradition of triggering and much work in the tradition of Universal
Grammar. The learner is modeled as a population of
grammars, whose probabilistic distribution changes
in response to the input data in a Darwinian selectionist fashion. The learner nondeterministically
selects a string of parameter values with their associated probabilities, and rewards or punishes these
probabilities based on the success or failure of analyzing an input utterance (Bush and Mosteller 1951).
The rise of the target grammar is smooth, which provides an explanation for the gradualistic development
of child language.

Knowledge Change

Important Scientific Research and

Open Questions
Perhaps the most attractive feature of the KLNL learning model lies in its ability to connect the input data to
child grammar development. Under the competitionbased scheme, parameter values which are expressed
more frequently in the input data will be established
sooner than those expressed less frequently. This allows
the researcher to correlate the statistical properties of
the input data, which is available through collections of
child-directed speech, with the longitudinal development of child grammar. In addition, under the KLNL
model nontarget grammars (before their demise) can
be probabilistically accessed by the learner. This maintains the fruitful tradition in acquisition research that
the errors in child language may depart from the target
language but nevertheless is systematic and linguistically possible. Similar approaches have been applied to
the acquisition of morphology and phonology. Linguistic rules can be associated with probabilities
thereby capturing well-known developmental findings
such as frequency effects that previously had no place in
the traditional approach to child language under generative grammar.
An important direction for research is to further
clarify the formal properties of probabilistic learning
models similar to KLNL. Convergence proof has been
obtained, though the efficiency of learning depends on
the landscape of the parameter space and, given the
probabilistic nature of the learning model, the distribution of the input data. Parameters are designed to tie
together wide-ranging generalizations to linguistic
expressions that are readily available in the input.
Recent work exploring large and linguistically realistic
samples reveals that the space of parameters may be
sufficiently smooth to facilitate plausible language
acquisition. An open question here concerns the
scope and limit of parameters in linguistic description
and acquisition: every language has specific rules and
constructions that are results of history and cannot be
plausibly attributed to the innate endowment. Current
efforts are devoted to the development of an integrated
model that carries out probabilistic selection among
innate parametric choices while at the same time constructs language-particular rules and determines their
range of applications.
A final and broader consideration that emerges
from the KLNL approach lies in the connection

1681

between the language faculty and other cognitive systems (Chomsky 2005). The KLNL model opens up the
possibility that the mechanisms of language learning
are domain-general while still operating within
domain-specific constraints of language. On the one
hand, this suggests that individual variation in language acquisition, including certain clinical cases,
may not be the result of deficiencies in the affected
individuals linguistic ability, but in the ability to process and analyze the input data that may be reflected in
other cognitive and perceptual tasks. On the other
hand, the probabilistic learning mechanisms in parameter setting appear to be evolutionarily ancient (Bush
and Mosteller 1951), which raises interesting questions
for the evolution of language: specifically, how the ways
we learn might have shaped the organization of language in its attested form (e.g., parameters).

Cross-References
Formal Learning Theory
Infant Learning and Development
Language Acquisition and Development
Learnability
Learning and Instinct
Mathematical Models/Theories of Learning
Reinforcement Learning in Animals
Statistical Learning in Perception
Stochastic Models of Learning
Unlearning (The Nature of. . .)

References
Bush, R., & Mosteller, F. (1951). A mathematical model for simple
learning. Psychological Review, 68, 313323.
Chomsky, N. (1981). Lectures on government and binding.
Dordrectht: Foris.
Chomsky, N. (2005). Three factors in language design. Linguistic
Inquiry, 36(1), 122.
Gibson, E., & Wexler, K. (1994). Triggers. Linguistic Inquiry, 25,
355407.
Valian, V. (1991). Syntactic subjects in the early speech of American
and Italian children. Cognition, 40, 2182.
Yang, C. (2002). Knowledge and learning in natural language.
New York: Oxford University Press.

Knowledge Change
Cognitive Learning

1682

Knowledge Claim

Knowledge Claim
Belief Formation

Knowledge Compilation
Restructuring in Learning

Knowledge Creation Metaphor,

The
SAMI PAAVOLA
Institute of Behavioral Sciences, University of Helsinki,
Helsinki, Finland

Synonyms
Collaborative knowledge creation; The knowledge
creation approach; The knowledge creation metaphor
of learning

Definition
An overall term for such theories and views of learning
which emphasize learning and human cognition as
processes of developing and pursuing certain novelties
(artifacts, products, practices, concepts, activities, processes) collaboratively and with distributed means
where individuals initiative is embedded in fertile
social and institutional practices and processes. The
focus is on advancing knowledge, transforming social
practices, and developing expertise. It has been proposed that this view of learning is a third main metaphor of learning, which is becoming more and more
important in modern society (in contrast to the acquisition metaphor of learning and the participation metaphor of learning).

Theoretical Background
The knowledge creation metaphor is a claim that there
is an emerging trend of theories about human learning
and cognition which aim at understanding how people
organize their work and learning for developing and
creating things together. This metaphor is a sequel to

Anna Sfards (1988) famous distinction between acquisition and participation metaphors of learning. Schematically described, the acquisition metaphor focuses
on processes of adopting or constructing subjectmatter knowledge or conceptual knowledge within an
individuals mind, whereas theories representing the
participation metaphor focus on processes of socializing
in social communities and in social interaction and
practices. The acquisition metaphor emphasizes usually logically oriented epistemology, whereas the participation metaphor emphasizes such things as
communities, social identities, cultural mediation,
and the situatedness of human cognition. It can be
maintained that this distinction is a very apt characterization of basic theories of learning as such, but in
order to understand the emerging new phenomena
related to collaborative creativity and learning, a third
basic metaphor of learning should be defined where
change, an organized aim of developing something
new, and the role of mediating artifacts are emphasized.
According to Paavola et al. (2004); see also Hakkarainen
et al. (2004), influential representatives of the knowledge creation metaphor in learning sciences are the
knowledge building approach (Bereiter 2002), and
expansive learning as well as cultural-historical
activity theory (Engestrom and Sannino 2010), and,
within organizational sciences, Nonaka and Takeuchis
theory of knowledge-creating companies. The term
knowledge creation owes a lot to the theory of organizational knowledge creation (Nonaka and Takeuchi
1995), but nowadays it is used often in various fields of
research. Theories representing the knowledge creation
metaphor are quite different from each other. In spite
of clear and in some sense fundamental differences
among these theories, they have many features in common, that is, focuses that include (1) the pursuit of
newness; (2) processes of mediation to avoid Cartesian
dualisms (such as mind vs matter, or concepts vs material objects); (3) social processes, while also emphasizing the role of individual subjects in knowledge
creation; (4) going beyond propositional and conceptual knowledge as a sole locus of learning, while recognizing conceptualizations and conceptual artifacts as
central for knowledge creation; and (5) the interaction
around and through shared objects.
It has been maintained that various changes in
modern society form a basis for the knowledge creation metaphor of learning, such as: (1) the rapid

Knowledge Creation Metaphor, The

development of new technology, which has formed and

is all the time forming qualitatively new possibilities for
distributed interaction and collaboration; (2) the pressure to create and learn deliberately to create new
knowledge and transform existing practices in various
areas of life; and (3) the complexity of modern society,
which requires people to combine their expertise to
solve emerging and often unforeseen complex problems.

Important Scientific Research and

Open Questions
The concept of the knowledge creation metaphor
(KCM) of learning was first developed in relation to
computer-supported collaborative learning (CSCL)
in order to broaden or better understand the epistemological basis of technology-mediated learning and how
processes of innovative knowledge advancement and
discovery are taken into account (Paavola et al. 2002).
Partially because of this history, this concept has
been developed and used more in relation to CSCL
research and technology-mediated collaborative learning and inquiry learning. There are new methodological developments in how to research knowledge
creation processes and how to advance and support
collaborative knowledge creation with technology and
related knowledge practices. Quite often the KCM is
interpreted as a continuation or even a synonym for the
knowledge building approach because both of them
emphasize a new kind of approach needed for answering the challenges of modern knowledge work. The
original idea of the KCM has, however, been that it
includes different kinds of approaches where some of
them emphasize collaborative idea development (such
as knowledge building) and some collaborative practice
transformations (such as expansive learning). In the
CSCL context, it is maintained that the KCM provides
a basis for a trialogical approach to learning where an
emphasis on meaning making and dialogs is
complemented with a focus on jointly constructed
knowledge practices and knowledge artifacts.
The KCM has also been used in many contexts other
than CSCL, such as when reviewing various approaches
to workplace learning and organizational learning, or
doing research on teacher education and teacher communities, networked learning, and networked expertise, while defining a new digital epistemology for
schools and teacher education, or analyzing the nature
of open source communities. The KCM should not

only be seen in relation to new digital technology

because similar mediated processes which are organized for producing things together can be done without the use of any special technology. Very often the
KCM is, however, related to new technology that provides new means for collaboration, such as new collaborative learning environments, or Web 2.0 technology
in general, wikis, podcasts, and so on.
The KCM is proposed as a supplement or an alternative to theories belonging to the participation metaphor of learning with the idea that theories
representing the participation metaphor (or the acquisition metaphor) are not enough if the aim is to understand processes of collaborative knowledge creation. It
remains an open question as to what extent the participation metaphor differs from the knowledge creation
metaphor, or if the participatory approaches can be
developed so that they include knowledge creation
processes. Another basic criticism against the KCM is
that it aims at seeing commonalities between theories
which are epistemologically and methodologically very
different, and leads to eclecticism. According to this
view, better alternatives to the acquisition and the
participation approaches would be such metaphors as
expansion (Engestrom and Sannino 2010) or
knowledge building, but not both together. One
answer to these criticisms is that the KCM is not
meant to be used for overruling differences between
various approaches, but aims at pointing out some
important phenomena in areas related to collaborative
learning which can be used for enriching methodological and theoretical development and discourse in these
areas of research.

Cross-References
21st Century Skills
Collaborative Knowledge Building
Computer-Supported Collaborative Learning
Contradictions in Expansive Learning
Cultural-Historical Theory of Development
Learning Metaphors

References
Bereiter, C. (2002). Education and mind in the knowledge age.
Hillsdale: Erlbaum.
Engestrom, Y., & Sannino, A. (2010). Studies of expansive learning:
Foundations, findings and future challenges. Educational
Research Review, 5(1), 124.

1683

1684

Knowledge Embodiment

Hakkarainen, K., Palonen, T., Paavola, S., & Lehtinen, E. (2004).

Communities of networked expertise: Professional and educational
perspectives (Advances in learning and instruction series).
Amsterdam: Elsevier.
Nonaka, I., & Takeuchi, H. (1995). The knowledge-creating company:
How Japanese companies create the dynamics of innovation.
New York: Oxford University Press.
Paavola, S., Lipponen, L., & Hakkarainen, K. (2002). Epistemological
foundations for CSCL: A comparison of three models of innovative knowledge communities. In G. Stahl (Ed.), Computer
support for collaborative learning: Foundations for a CSCL community, proceedings of: CSCL 2002 (January 711, 2002, Boulder,
Colorado, USA). Hilldale: Lawrence Erlbaum.
Paavola, S., Lipponen, L., & Hakkarainen, K. (2004). Models
of innovative knowledge communities and three metaphors of
learning. Review of Educational Research, 74(4), 557576.
Sfard, A. (1998). On two metaphors for learning and the dangers of
choosing just one. Educational Researcher, 27(2), 413.

ideas. It may involve the generation of a groups idea

that is co-constituted by the interactions between members of a group discussing their own individual ideas.

Knowledge Integration
MICHAEL SCHNEIDER
Institute for Behavioral Sciences, ETH Zurich,
Zurich, Switzerland

Synonyms
Integrating knowledge; Merging knowledge structures

Definition

Knowledge Embodiment
Process of incorporating knowledge of a cultural process into ones physical bearing and actions, e.g., the
bodily understanding that people bring to religious
worship or a trade.

Knowledge Encoding
Routinization of Learning

Knowledge Gaps
Cognitive Dissonance in the Learning Processes

Knowledge Generation
Generative Learning

Knowledge Improvement
A process of learning in which learners are engaged in
constructively critiquing and improving each others

Knowledge integration refers to the process of merging

two or more originally unrelated knowledge structures
into a single structure. In the most general sense, it can
encompass the complexities of how two digital databases can be merged together or how two companies
can effectively combine the knowledge of their workers.
In the learning sciences, however, the term usually
refers to knowledge integration within persons memory. Learners pick up pieces of knowledge (e.g., experiences, observations, ideas, hypotheses, explanations)
in many different situations, for example, everyday life
observations, conversations with friends, the Internet,
and school instruction. Novices in a domain often do
not see which of these newly acquired pieces of knowledge relate to each other and why they should be related
at all. Recognizing relations usually depends on relevant prior knowledge, for instance, knowing an underlying rule or a scientific concept. Because they lack this
abstract background knowledge, novices tend to focus
on the superficial differences between pieces of knowledge they acquired in seemingly unrelated situations
(cf. Chi et al. 1981). They, then, store these pieces
independent of each other in their long-term memory,
which results in fragmented knowledge. Knowledge
integration takes place whenever learners connect previously unrelated pieces of knowledge together in their
memory, often as a result of instruction.

Theoretical Background
Research on knowledge integration is closely related
to research on conceptual change. Both investigate

Knowledge Integration

how learners acquire complex knowledge structures

in conceptually rich domains (diSessa 2006). Some
conceptual change researchers argue that conceptual
knowledge mostly takes the form of subjective naive
theories which are rather coherent and integrated. This
is the knowledge as theory perspective, from which
the problem of knowledge integration is negligible.
Other researchers argue that conceptual knowledge is
frequently fragmented, particularly for learners with
little expertise in a domain. This is the knowledge as
elements perspective on conceptual change. From this
perspective knowledge integration is an important
learning mechanism which should be activated and
strengthened by instruction.
One of the most widely recognized knowledge as
elements theories has been formulated by Andrea
diSessa and colleagues (diSessa et al. 2004). According
to this view, students acquire knowledge in pieces.
These pieces of knowledge are abstractions of common
experiences, for example, everyday life observations,
which diSessa calls phenomenological primitives or
p-prims. P-prims may or may not be conscious. They
are initially not related to each other, because the
learner does not have the necessary knowledge to relate
them to each other. Over time, that is, with increasing
expertise and knowledge in a domain, learners integrate p-prims into more organized knowledge structures. DiSessa emphasizes the importance of what
he calls coordination classes for this process. Coordination classes are systematic collections of strategies
for reading related information out from the world.
Coordination classes eventually lead to a different view
of the world, a view in which previously unconnected
p-prims are seen as aspects of the same more general
concept.
A related theoretical approach (Schneider and Stern
2009) emphasizes the roles of working memory and
long-term memory in knowledge fragmentation and
integration. Learners must create a complex, wellstructured, and well-integrated network of knowledge
in long-term memory in order to acquire expertise in
a domain. Long-term memory has a virtually unlimited capacity. However, it does not allow for the active
transformation and integration of knowledge. This
integration is possible only in working memory,
which can hold only a few pieces (also called chunks)
of knowledge at a time. Thus, building up wellintegrated knowledge structures in long-term memory

requires many cycles of loading some pieces of knowledge into working memory, integrating them when
appropriate, and storing the results back in long-term
memory. This process is complicated by the fact that
novices in a domain often do not understand which
pieces of knowledge are plausible candidates for knowledge integration and should be loaded into working
memory simultaneously.
Finally, Baroody (2003) and other researchers on
mathematics learning recommend that in addition to
integrating pieces of conceptual knowledge, learners
should also integrate their concepts and their problem-solving procedures. Conceptual knowledge can
help with the construction of new procedures, the
modification of existing ones, the transfer of procedures from well-known to new problem types, adaptive choices between alternative procedures, and
monitoring of the execution of a procedure. Procedural knowledge, in turn, is necessary to quickly and
efficiently apply conceptual knowledge to the solution
of problems. However, this is possible only when
learners realize how concepts and procedures relate
to each other.

Important Scientific Research and

Open Questions
Research on knowledge integration is a young field
which has emerged over the last 20 years from studies
on science learning (diSessa 2006; Linn 2006) and
mathematics learning (Baroody 2003; Schneider and
Stern 2009). It still needs to be generalized to other
content domains.
Marcia C. Linn (2006) and her colleagues conducted over 40 empirical case studies on the way
knowledge integration occurs and how it can be facilitated by learning environments. They found that the
processes of knowledge fragmentation and integration
are common in science learning. Successful instructional interventions which foster knowledge integration usually include some or all of the following
four activities. First, successful instruction stimulates
students to elicit their current ideas. Students profit
most when they are prompted to consider ideas from
many different contexts, for example, school, home,
recreation, and museum. Eliciting ideas in group settings helps learners to also consider the ideas of their
peers and to integrate knowledge by reflecting on these
ideas. Second, adding new, normative ideas can have

1685

1686

Knowledge Management

a positive effect on students. Learners sometimes prefer

learning something new over reflecting on their (partially incorrect) prior knowledge. The acquisition of
new ideas can increase knowledge integration, when
these ideas stimulate the reconsideration of existing
views, for example, by demonstrating the connection
between two everyday life experiences or by illustrating
an abstract idea. Third, instructions which foster
knowledge integration should help students to develop
criteria for the evaluation of their own ideas. Interventions helping students to acquire this metacognitive
competence will encourage students to judge ideas
based on their validity and plausibility. This is an
important competence, because in everyday life
learners encounter a mix of valid ideas and bogus
stories, for example, on the Internet. Looking for systematic connections between claims, empirical evidence, sources of information, persons perspectives
and intents can help to sort out valid ideas by seeing
them as parts of a bigger system of knowledge. Finally,
the integration of knowledge can require sorting out
ideas that are incorrect and contradict other parts of
the learners knowledge base. This sorting process is
based on the abovementioned processes, because it
requires students to already have acquired some normatively correct ideas and criteria for distinguishing
between more and less adequate ideas. The process of
sorting out ideas leads to a knowledge base that is more
focused and thus more easily comprehended. In summation, empirical research shows that knowledge integration is a challenging and time-consuming process
which is facilitated by well-prepared learning environments. Ideally, these are adapted to the learners specific
prior knowledge, their everyday life, and the broader
sociocultural environment so that knowledge from all
of these sources is integrated.

Cross-References
Complex Declarative Learning
Conceptual Change
Constructivist Learning
Knowledge Acquisition: Constructing Meaning from
Multiple Information Sources
Knowledge Organization
Learning and Understanding
Model-Based Learning
Role of Prior Knowledge in Learning Processes
Science, Art and Language Experiences

References
Baroody, A. J. (2003). The development of adaptive expertise and
flexibility: The integration of conceptual and procedural knowledge. In A. J. Baroody & A. Dowker (Eds.), The development of
arithmetic concepts and skills: Constructing adaptive expertise
(pp. 133). Mahwah: Erlbaum.
Chi, M. T. H., Feltovich, P. J., & Glaser, R. (1981). Categorization and
representation of physics problems by experts and novices. Cognitive Science, 5(2), 121152.
diSessa, A. A. (2006). A history of conceptual change research. In
R. K. Sawyer (Ed.), The Cambridge handbook of the learning
sciences (pp. 265280). Cambridge: Cambridge University Press.
diSessa, A. A., Gillespie, N. M., & Esterly, J. B. (2004). Coherence
versus fragmentation in the development of the concept of force.
Cognitive Science, 28, 843900.
Linn, M. C. (2006). The knowledge integration perspective on learning and instruction. In R. K. Sawyer (Ed.), The Cambridge handbook of the learning sciences (pp. 243264). Cambridge:
Cambridge University Press.
Schneider, M., & Stern, E. (2009). The inverse relation of addition and
subtraction: A knowledge integration perspective. Mathematical
Thinking and Learning, 11(1), 92101.

Knowledge Management
Acquiring Organizational Learning Norms

Knowledge Maps
Concept Maps

Knowledge Organization
GABI REINMANN
Learning and Teaching with Media, Universitat der
Bundeswehr Munich, Germany

Synonyms
Information classification; Information organization;
Knowledge representation; Knowledge structuring

Definition
The term knowledge organization has at least two
meanings depending on the concept of knowledge.

Knowledge Organization

If you concentrate on personal knowledge to which only

the individual has direct access, knowledge organization indicates a mental and therefore internal process of
structuring or transforming (e.g., visualizing) knowledge. If you concentrate on public knowledge (synonym: information) which is materialized in documents
(text, image, audio, video) so that principally all persons can access it, knowledge organization refers to the
technical and external processes of describing and classifying information. A third definition takes a sociological perspective and indicates such social entities as
knowledge organizations whose work primarily relies
on knowledge (like universities); this latter meaning
will not be included in this entry.

Theoretical Background
The theoretical background as well as the history of the
term knowledge organization has two disciplinary
roots: (1) learning and cognitive psychology and
(2) information and computer sciences. The preferred
perspective depends on the underlying conception of
knowledge. Therefore, the term knowledge organization can be applied only when considering the concept
of knowledge. There are many understandings of
knowledge which differentiate its special characteristics
and forms. Among others, Jean Piaget (18971980) has
had great influence on the conception of knowledge.
Piaget represents a theory of genetics of cognitive
structures giving the individual a crucial and active role
in constructing knowledge. But the theory also considers the strong interdependence between individual
constructions and external stimuli of the external world
(and their experience). These two complementary processes which balance this interdependence are called
accommodation and assimilation.
Against this background knowledge is a very personal and mental phenomenon to which only the individual has direct access. Under this perspective you can
speak about personal knowledge (Seiler 2004). Personal
knowledge can be of different specification (especially
conceptual, visual, and enactive) and is therefore more
or less easy to articulate and to make explicit. Nevertheless, persons can communicate and collaborate so
that one must assume that there is not only implicit or
idiosyncratic but also shared knowledge mostly conventionalized through language and often materialized
in written, spoken, or visualized documents like text,
audio, image, or video. Documented knowledge like

this is of principally public access. Under this perspective you can speak of public knowledge or information
(Seiler 2004).
To distinguish between personal and public knowledge is the only one alternative of many to structure the
domain of knowledge. But exactly this possibility can
aid one to make the decision to choose either the psychological or technical discipline as the leading one for
research of knowledge organization. Learning and cognitive psychology provide theoretical and empirical
insights referring to personal knowledge organization
which is always an internal process. Computer and
information sciences in contrast deliver the scientific
basis to organize public knowledge or rather information which always results in an external representation.
Internal knowledge organization: The growth of personal knowledge is part of human development and
does not happen randomly but in some organized
manner. There are three representational constructs
describing this internal knowledge organization:
semantic networks, theories, and schemas (see Chi
and Ohlsson 2005).
Knowledge can be represented in semantic networks

which consist of concepts (called nodes) and relations (called links). The organization of knowledge
in semantic networks indicates that everything is
related to everything. The quality of this organization is determined by the number and character of
relations (e.g., hierarchical, temporal, or causal)
between concepts as well as by content similarity
of concepts. So cognitive psychologists assume that
knowledge in semantic networks is grouped by
domain.
Domain-specific knowledge can also be represented
as theories: Theories in this sense can be wellarticulated structures with core knowledge elements (big ideas) in the center and peripheral
concepts around them. This attribute (i.e., different
importance of elements) characterizes theories in
expert knowledge as well as those in novice knowledge. The latter, however, are rather intuitive theories lacking the depth of scientific theories of
experts.
The representation of knowledge in schemas results
in the assumption that humans construct patterns
of experience. A schema is a set of attributes (called
slots) which can take on different values referring to

1687

1688

Knowledge Organization

phenomena in the external world. Schemas are typically abstract and organize knowledge about specific stimulus domains. They are retrieved as units
and are used to organize learning, thinking, and
acting.
To a certain extent internal representations like
semantic networks, theories, and schemas are the
results of natural processes of personal knowledge
organization which do not have to be externally
instructed and controlled. Nevertheless, there are
some possibilities to foster and optimize the internal
knowledge organization through educational support.
Nearly all proposals which do this stem from the
research of (cognitive) learning strategies and are
based upon the principle of producing external representations. These external representations can be verbal, pictorial, or something in between, for example,
maps and other logical pictures.
You can foster internal knowledge organization by

articulating knowledge verbally or in written form.

To think out loud can help to structure knowledge
because transforming implicit knowledge in
conventionalized words fosters the construction or
retrieval of schemas. Furthermore, specific programs have been developed to support knowledge
generation and organization through writing and
text production, respectively.
A widely used strategy in organizing knowledge
recurs to visualization. Psychological research on
memory shows that there are spontaneous productions of internal imageries probably influencing the
internal knowledge organization. So you can use
external imagery strategies to foster these processes.
But there is greater evidence in the surplus of logical
forms of visualizations like diagrams and maps.
Because of the structural similarity to assumed
semantic networks (as internal knowledge representations) concept mapping is a widespread
strategy or cognitive tool to produce effective external representations.
Processes and methods of internal knowledge organization are often integrated in educational settings (as
learning strategies), but they are also part of so-called
personal knowledge management strategies through
which adults should optimize their personal growth as
well as work performance (Holsapple 2004).

External knowledge organization: The growth of

public knowledge, or rather information, is part of
the sociocultural development and needs sophisticated forms of organization because of the increasing
amount of scientific knowledge focused in this context.
There are traditional and technology-based forms for
external knowledge organization in order to gather,
describe, index, classify, store, and find documents
(Hjrland 2008).
Information can manually be indexed and classified

like in former libraries or archives. This procedure is

slow but brings the advantage that persons (librarians) work with meaning in practice. However, today
all information institutions are using computer systems for archiving, identifying, and retrieving information relevant to specific purposes.
So information also can automatically be structured
in very different ways using insights of linguistics,
logic, mathematics, and philosophy. Information
scientists know a lot of methods for organizing
information, for example, from the general to the
specific, facet-analytical, bibliometric, user-oriented, and so forth.
Like personal knowledge information can be visually represented using new information technologies: Visualization presupposes well-structured
information and greatly facilitates the retrieval of
information which most notably is an advantage
with huge amounts of information.
For a short time even technology-based information can also be socially organized, for example,
through social tagging or social bookmarking.
Instead of expert taxonomies leading the external
knowledge organization, so-called folksonomies
are structuring the public knowledge on the
Internet.
Processes and methods of external knowledge organization are not only used in libraries and archives,
but also in knowledge-based organizations engaging
in organizational knowledge management where
knowledge organization covers not all but an important part of management tasks (Holsapple 2004).

Important Scientific Research and

Open Questions
Knowledge organization is an interdisciplinary challenge combining several disciplines which do not use

Knowledge Representation

the same concepts of knowledge or information. Communication between these disciplines is rare and difficult because there is little grounding referring to the
underlying core concept. As a consequence, different
scientific camps which widely ignore each other are
developing although psychological and technical processes of knowledge organization should correspond in
many contexts (e.g., education and work place).
Beneath this practical argument there are strong theoretical reasons for collaboration between scientists of
different origin: (a) Internal knowledge organization
often needs external representations as well as technical
tools to stimulate and support them. Further research
has to analyze the interrelation between methods based
upon cognitive and learning psychology on the one
hand and computer and information science on the
other hand. (b) External knowledge organization in
contrast presupposes objectives and criteria which are
possible only on the basis of internal representations of
persons engaging in articulation, visualization, or other
forms of externalization. Perhaps the field of visualization is a seminal research object which connects different disciplines working on knowledge organization
(see Tergan and Keller 2005). Information and knowledge visualization share the common goal to organize
information and knowledge for better access, search,
and understanding and use comparable techniques and
methods. Joint research can be noticed with concept
mapping as a method of graphical representation fostering internal and external forms of knowledge organization for different purposes.

Cross-References
Knowledge Integration
Knowledge Representation
Learning Strategies
Organizational Change and Learning
Schema(s)

References
Chi, M. T. H., & Ohlsson, S. (2005). Complex declarative learning. In
K. J. Holyak & R. G. Morrison (Eds.), The Cambridge handbook of
thinking and reasoning (pp. 371400). Cambridge: Cambridge
University Press.
Hjrland, B. (2008). What is knowledge organization (KO)? Knowledge Organization, 35(2/3), 86101 (International Journal
devoted to Concept Theory, Classification, Indexing and Knowledge Representation).

1689

Holsapple, C. W. (Ed.). (2004). Handbook of knowledge management.

Berlin: Springer.
Seiler, T. B. (2004). The human foundation of knowledge management. In J. Gadner, R. Buber, & L. Richards (Eds.), Organising
knowledge. Methods and case studies (pp. 4359). Hampshire:
Palgrave, Macmillan.
Tergan, S.-O., & Keller, T. (Eds.). (2005). Knowledge and information
visualization. Searching for synergies. Berlin: Springer.

Knowledge Representation
PABLO PIRNAY-DUMMER, DIRK IFENTHALER,
NORBERT M. SEEL
Department of Education, University of Freiburg,
Freiburg, Germany

Synonyms
Internal representation

Definition
Knowledge representation is a key concept in cognitive science and psychology. To understand this
theoretical term one has to distinguish between
knowledge and its representation. Intelligent
behaviors of a system, natural or artificial, are usually
explained by referring to the systems knowledge.
In other words: The capability of performing intelligent
behavior is associated with the existence of applicable
knowledge. By relating intelligence and knowledge, the
systems behavior becomes more or less reconstructible
and predictable. The most discussed distinction is
between declarative (knowing that) and procedural
(knowing how) knowledge (see Anderson 1983).
Declarative knowledge is defined as factual
knowledge, whereas procedural knowledge is defined
as the knowledge of specific functions and procedures
to perform a complex process, task, or activity.
Modern cognitive science sees cognition and learning as a complex process with many facets, including
symbolic representations of objects and events which
are not immediately present but exist only in imagination. Accordingly, most cognitive scientists agree on the
basic assumption that cognition and learning take place
in the use of mental representations, in which individuals organize symbols of experience or thought in such
a way that they effect a systematic representation of this

Knowledge Representation

experience or thought as means of understanding it or

of explaining it to others (Seel 1991). This author
describes the function of knowledge representation by
distinguishing three zones: the object zone W as part of
the physical world, the knowledge zone K, and the zone
of knowledge representation R (Seel 1991, p. 17).
The interplay between these zones can be depicted
as in Fig. 1.
As shown in Fig. 1, there exist two classes of functions: (1) fin as the function for the internal representation of the objects of the world (internalization), and
(2) fout as the function for the external re-representation back to the physical world (externalization).
In cognitive science usually a distinction has been
made between representation schemas which refer to
declarative or procedural knowledge. Conceptual graphs
and semantic networks are the most popular schemas
for representing declarative knowledge, whereas procedural knowledge is often represented by means of production systems. Both kinds of representational schemas
are discussed in particular entries of this encyclopedia.
Furthermore, frames must be mentioned as a representation schema which can be seen as a compound of
semantic networks and productions.

Theoretical Background
In Piagets tradition of semiotic functions (see the entry
on Semiotics and Learning), learning and thinking
are seen as a process of using and operating with systems of signs (gestures, images, language, or symbols).
These systems enable people to visualize and express

r1

r2

Representation R

Representation
Internalization: fin
Re-Representation
Externalization: fout

1690

Knowledge K

k1 k
2

w1
World W

w2

Knowledge Representation. Fig. 1 The interplay

between knowledge, its representation, and the physical
world

their subjective experiences, ideas, thoughts, and feelings. Consequently, the idea of mental representations
advanced to one of the most significant concepts of
cognitive science.

Types of Knowledge Representation

All work on knowledge representation is based on the
hypothesis that people memorize knowledge by means
of specific propositions about the conceptual information that is inherent in the issue to be remembered.
Alternatively, there is the conception that information
also can be memorized by means of images which
correspond as analogues to the original perception.
This basic alteration corresponds with semiotics
where, however, a third form of knowledge representation can be found. Actually, semiologists differentiate
between signs depending on whether they are used as an
index (e.g., smoke as a sign of fire, red as a sign of
danger), as an icon or pictorial sign (e.g., the line drawing of a face, a figure, a vase), or as a symbol (e.g., letters,
numerals, musical notes, mathematical symbols). Correspondingly, cognitive psychologists often differentiate basically between enactive (i.e., activity-based),
iconic, and symbolic types of knowledge representation
(Bruner 1964).
In a process which corresponds to the biological
development of the central nervous system, people
first develop the functions for perception, motor skills,
and coordination. Bruner has thus referred to enactive
representation as a mode of representing past events
through appropriate motor response . . . Such segments
of our environment bicycle riding, tying knots,
aspects of driving get represented in our muscles, so
to speak (Bruner 1964, p. 2). This anchoring of
enactive representation in the neurobiologically oldest
functions of the central nervous system allows this
representation format to maintain its central significance as a sign function over the entire lifespan.
The idea of an iconic representation of knowledge
corresponds to the idea of encoding-specificity, which
posits that content from memory is usually retrieved in
the same format in which it was coded when learned.
Thus, many psychologists (e.g., Larkin and Simon 1987)
use the old adage a picture speaks 1,000 words to
underline the significance of iconic representation and
highlight the fact that it can be influenced by graphics
and images. On the other hand, there are also cognitive
scientists (e.g., Pylyshyn 1984) who take up arguments

Knowledge Representation

of the Wurzburg School (from the beginning of the

twentieth century) and reject the idea of representing
knowledge by means of visualization. In the literature,
this controversial argumentation became popular as
imagery debate which dominated the discussion
about knowledge representation in the 1980s. The
iconic representation of knowledge is discussed in particular entries on imagery.
Finally, knowledge representation occurs by means
of symbols. Psychologists often speak of language as
a special form of symbolic representation. Indeed,
human language is symbolic because it is impossible
to infer the signified (a building in which people live)
from the signifier (e.g., the sounds or letters H + O +
U + S + E). As one and the same signified can be
expressed by any number of different signifiers (e.g.,
maison, Haus, casa), this type of sign is referred
to as arbitrary. By using linguistic means of expression,
humans can visualize ideas, thoughts, and feelings and
describe things that they can only express to a limited
extent in actions or ideational images. It is a distinctive
characteristic of humans that they can deal with their
own experiences through language. Linguistic expressions can trigger ideas (even very vivid ones) which can
even reconstruct facts which cannot be experienced
directly, complex connections, and abstract thought
constructions (e.g., scientific theories). Accordingly,
concepts are probably the most useful form of symbolic
knowledge representation. Concepts are categorizations as forms of generalized abstractions and require
the application of cognitive operators for class formation and abstraction. Other symbols which are used for
knowledge representation are musical notes, digits, and
conventionalized signs. Based on concepts specific representation schemas have been developed for
representing declarative knowledge.

Representation Schemas for

Declarative Knowledge
The idea of declarative knowledge representation
grounds on the assumption that one can consider the
issue of representation largely independent on the
methods of applying knowledge. Declarative knowledge
is considered as a set of facts that can be represented as
data structure. Mylopoulos and Levesque (1984) provide a taxonomy of representation schemes which
ground on the assumption that the world can be considered as a collection of individuals among which

manifold relationships exist. The collection of all individuals and relationships at any one time in any world
constitutes a state, and there can be state transformations that cause the creation of individuals or that can
change the relationship between them. Depending on
whether the starting point for a representation scheme
is individuals and relationships, or assertions about
states, a distinction can be made between semantic
networks and logical representation schemas.
Good examples for logical representation schemas
are knowledge bases that are constructed on the basis
of first-order logic. Statements about the world domain
to be represented are translated into formulas which
permit conclusions. This form of a representation
schema clearly presupposes appropriate inference
rules which must also be available. Possibly more popular are semantic networks as form of declarative
knowledge representation. In its most basic form
a semantic network represents knowledge in terms of
a collection of objects (nodes) and binary associations
(directed labeled edges); the former standing for individuals (or concepts of some sort), and the latter standing for binary relations over these. According to this
view, a knowledge base (or structure) is a collection of
objects and relations defined over them, and modifications to the knowledge base occur through the insertion or deletion of objects and the manipulation of
relations (more detailed information can be found in
the entry on Semantic Networks).
Another form of declarative knowledge representation is the Conceptual Dependency Structure from
Schank (1975) that centers on conceptualizations
which attribute cases to actions. Finally, also frames
must be mentioned in the context of declarative knowledge representation although frames aim at a combination of declarative and procedural knowledge
representation.

Representation Schemas for

Procedural Knowledge
Depending on whether the starting point for a representation schema is state transformations the application
of logical representation schemas and semantic networks is limited and demands for another type of
representation. Clearly, the most popular procedural
representation schemas operate with productions and
production systems. They are described in detail in the
entry on Production Systems and Operator Schemas.

1691

1692

Knowledge Representation and Reasoning and Learning

Important Scientific Research and

Open Questions
Knowledge representation is not only a central problem
in cognitive science but also in Artificial Intelligence
due to the fact that intelligent systems need the
availability of expert knowledge along with associated
knowledge handling facilities. Since the emergence of
cognitive science countless studies have focused on the
use of representational schemas by intelligent systems,
natural or artificial. Actually, the problem of knowledge
representation is at the center of most publications in
the field of cognitive psychology (e.g., Markman 1998)
and Artificial Intelligence (e.g., Brachman and
Levesque 1985; Davis et al. 1993).
With regard to cognitive psychology and its
focus on human knowledge representation the most
important result of research consists in the observation that people are able to use different forms
of representation of memorized information. People
can either recall an appropriate form of representation from memory or transform memorized information in an appropriate form of representation in
dependence on situational demands. However,
because it is not possible to assess directly internal
representations of knowledge one of the most important issues of research on knowledge representation is
concerned with reliable and valid measurements of
declarative and procedural knowledge (see for an
overview Ifenthaler et al. 2010). Here, it is argued
that different types of knowledge require different
types of representations.

References
Anderson, J. R. (1983). The architecture of cognition. Cambridge, MA:
Harvard University Press.
Brachman, R. J., & Levesque, H. L. (Eds.). (1985). Readings in knowledge representation. Los Altos: Kaufmann.
Bruner, J. S. (1964). The course of cognitive growth. The American
Psychologist, 19, 116.
Davis, R., Shrobe, H., & Szolovits, P. (1993). What is a knowledge
representation? Artificial Intelligence Magazine, 14(1), 1733.
Ifenthaler, D., Pirnay-Dummer, P., & Seel, N. M. (Eds.). (2010).
Computer-based diagnostics and systematic analysis of knowledge.
New York: Springer.
Larkin, J. H., & Simon, H. A. (1987). Why a diagram is (sometimes)
worth ten thousand words. Cognitive Science, 11, 6599.
Markman, A. B. (1998). Knowledge representation. Mahwah: Erlbaum.
Minsky, M. (1975). A framework for representing knowledge. In
P. Winston (Ed.), The psychology of computer vision. New York:
McGraw-Hill.
Mylopoulos, J., & Levesque, H. J. (1984). An overview of knowledge
representation. In M. L. Brodie, J. Mylopoulos, & J. W. Schmidt
(Eds.), On conceptual modelling. Perspectives from artificial intelligence, databases, and programming languages (pp. 317).
New York: Springer.
Pylyshyn, Z. (1984). Computation and cognition: Toward a foundation
for cognitive science. Cambridge, MA: MIT Press.
Schank, R. (1975). Conceptual information processing. New York:
Elsevier.
Seel, N. M. (1991). Weltwissen und mentale Modelle. Gottingen:
Hogrefe [World knowledge and mental models].

Knowledge Representation and

Reasoning and Learning
Semantic Technologies and Learning

Cross-References
ACT (Adaptice Control of Thought)
Activityand
Taxonomy-Based
Knowledge
Representation
Conceptual Clustering
Imagery and Learning
Knowledge Organization
Mental Imagery
Mental Representation
Pictorial Representation and Learning
Production Systems and Operator Schemas in
Procedural Learning
Representation, Presentation and Conceptual
Schemas
Semantic Networks
Wurzburg School

Knowledge Restructuring
Conceptual Change

Knowledge Storage
Routinization of Learning

Knowledge Structure
Concept Similarity in Multidisciplinary Learning

Kohlberg, Lawrence (19271987)

Knowledge Structuring
Knowledge Organization

Knowledge Transfer
Process of using knowledge from a learning situation in
a situation outside of the initial learning, e.g., applying
school mathematics to shopping tasks.

Cross-References
Transfer of Learning

Knowledge-Based Learning
Analytic Learning

Koffka, Kurt (18871941)

NORBERT M. SEEL
Department of Education, University of Freiburg,
Freiburg, Germany

Theoretical Background
Koffka was one of the founders of Gestalt theory (along
with Wertheimer and Kohler). Like Wertheimer and
Kohler, he was not only interested in research on perception but also in many philosophical issues as well
as in development, learning and thinking. Thus, he
published not only an important book on child psychology in 1921, but also a complete and systematic
overview on the principles of Gestalt in 1935. In consequence, Koffka focused on developmental aspects of
learning in accordance with Gestalt principles.

Contribution(s) to the Field of

Learning
Koffka believed that most of the childs early learning is
sensorimotor learning which occurs after a consequence.
For instance, a child who touches a hot stove will learn
not to touch it again. Furthermore, Koffka believed that a
lot of learning also occurs by imitation, which he saw as
a natural occurrence. However, the highest type of learning is ideational learning, which makes use of language.
Koffka notes that an important time in childrens development is when they understand that objects have names.

Cross-References
History of the Sciences of Learning
Kohler, Wolfgang
Wertheimer, Max

References
Life Dates
Kurt Koffka was born in Berlin in 1886. In 19041905,
he studied at the University of Edinburgh in Scotland,
where he met British scholars, and studied in English.
He completed a Ph.D. at the University of Berlin in
1908 under the supervision of Carl Stumpf. Koffka
served as research assistant at the universities of
Wurzburg and Frankfurt before moving to the University of Giessen, where he served as professor until 1924.
When Koffka was in Frankfurt he began his collaboration with Max Wertheimer and Wolfgang Kohler in
developing the foundation of Gestalt psychology. In
1924, Koffka came to the Unites States in order to
serve as visiting professor at Cornell University and
the University of Wisconsin. Then he accepted a position at Smith College in Northampton, Massachusetts,
where he remained until his death in 1941.

1693

Koffka, K. (1924). The growth of the mind. London: Routledge.

Reprinted 1999.

Kohlberg, Lawrence
(19271987)
DAMIAN GRACE, MICHAEL JACKSON
Department of Government and International
Relations, The University of Sydney, Sydney, Australia

Life Dates
Lawrence Kohlberg (October 25, 1927January 19,
1987) was an American psychologist whose theory of
the development of moral reasoning made him one of
the most significant psychological researchers of the

1694

Kohlberg, Lawrence (19271987)

twentieth century. Kohlberg was born in New York

City, attended a private school, and worked for a time
in the merchant marine. In this capacity he served the
Zionist cause by smuggling Jewish refugees into Palestine. In 1948 he was admitted to The University of
Chicago, which awarded him a bachelors degree after
just 1 year. He enrolled in graduate studies at Chicago
and began his academic career at Yale University
(19561961). From 1962 to 1968, he was a professor
at Chicago and thereafter at Harvard. In 1971,
Kohlberg contracted a debilitating parasitic disease
while working in Belize. For the rest of his life he
suffered from its effects, including depression. In
1987, he committed suicide by drowning at Winthrop,
just outside Boston (Walsh 2000).

Contribution(s) to the Field of

Learning
Kohlberg began investigating the moral development of
children in his doctoral dissertation (1958), and continued this line of research for the next 30 years. His
many publications include Essays on Moral Development: Vol. 1. The Philosophy of Moral Development
(vol. 1, 1981; vol. 2, 1984) and Child Psychology and
Childhood Education: A Cognitive-Developmental View
(1987). A Web of Knowledge search in 2010 returned
65 publications in journals, including the American
Journal of Education, American Journal of Mental Deficiency, American Journal of Orthospyschiatry, American
Sociological Review, Child development, Human Development, and Zygon. A parallel search returned more
than 750 citations of these works. In short, his work
was widely disseminated and has been very influential.
A search of dissertation titles and abstracts identified
more than 900 theses that refer to Kohlberg in the title
or abstract.
Kohlbergs initial research was to interview 72 white
working and middle class Chicago boys aged 10, 13,
and 16, and present them with this moral dilemma.
A man named Heinz cannot afford to buy a uniquely
efficacious drug to save his dying wife. The chemist
who discovered the drug demands an extortionate
price for it. Heinz cannot even borrow so large a sum,
so he steals the drug. Did he do the right or wrong thing
in stealing from the chemist?
Based on his subjects responses to this and other
dilemmas, Kohlberg built a classification of stages of
moral development. The characteristics of these stages

are, first that each is qualitatively different from the

others. Secondly, that they are structured wholes,
that is, responses to one dilemma will be consistent
with responses to others, reflecting patterns of thought
peculiar to each stage. Thirdly, children progress
through the stages in an invariant sequence, neither
skipping stages nor regressing. Fourthly, the stages are
integrated hierarchically, that is, learning at earlier stages
is not lost as subjects move into the later ones. Finally,
the stages are cross-cultural universals, found in all
societies. Kohlberg means by this not that all cultures
have the same beliefs but that the development of moral
reasoning occurs in all cultures according to his stages.
Development might not progress through all the stages,
for stages 5 and 6 demonstrate quite a high degree of
detachment and moral self-direction. Indeed, by 1976,
Kohlberg had stopped using stage 6, first because he
doubted the efficacy of his dilemmas in distinguishing
between stages 5 and 6; and secondly because he
doubted that many subjects reasoned consistently at
Stage 6 (Crain 2005, pp. 158, 165; Cortese 1990 p. 21).
Kohlberg placed his six stages into three levels. Level
1 contains two stages of Pre-conventional Morality. At
this level, children do not yet appreciate the social
nature of morality. At Stage 1, children regard morality
as a set of rules laid down by authority. That is, morality
is an externally imposed set of rules and deviation from
them brings punishment. In the Heinz case, the rules
dictate that it is wrong to steal and children asked
about whether Heinz would be right to steal the drug
typically respond, Its against the law, and youll get
punished. At this stage, egocentrism prevails and other
perspectives are not part of the childs way of thinking
(Crain 2005, p. 154). At Stage 2, moral reciprocity
emerges. Right is what parties reciprocally agree to be
fair, and morality is to this extent relativized. Children
can appreciate that Heinz might have sufficient reason
to steal the drug, but they can also see the chemists side
of things. Self-interest has emerged as a criterion of
right, though tempered by a sense of the chemists
unfairness in demanding too high a price for his drug.
At Level 2, Conventional Morality, children are
entering adolescence and have a grasp of morality as
necessary for social cohesion. The centrality of rules
gives way to the consideration of character and a concern for the welfare of others. Stage 3 reasoning reflects
empathic relationships with family and friends and the
obligations these entail. Children at this stage will

Kohlberg, Lawrence (19271987)

consider motives, such as the chemists selfishness and

greed, and Heinzs feelings of love and concern in
stealing the drug. At Stage 4, the moral stage is broadened beyond family and friends to include society. Now
there is a recognition that if everyone acted as Heinz did,
society would fall apart. There is recognition that Heinz
has good motives, but that this does not override the
obligation to consider society as a whole and its reliance
on obedience to the law. Responses at this stage resemble
those at Stage 1, but children at Stage 4 can give reasons
for their views, such as harm to society. At Stage, 1
stealing is just wrong (Crain 2005, pp. 155157).
At Level 3, Post-conventional Morality, Stage 5,
a person moves beyond concern for social cohesion to
questions about the nature of society and the principles
underlying its conventions. What might a good society,
irrespective of its setting, look like? It turns out to look
something like John Rawlss just contractarian political
association enshrining basic rights and using democratic
procedures (Crain 2005, pp. 157158). Stage 5 respondents take a critical position on social conventions,
distinguishing law and morality in the Heinz case.
A good society would protect the chemists property
rights but be sensitive to Heinzs wifes right to live.
Heinz would be justified morally in saving his wife,
and legally a judge should take account of this and
punish him but lightly (Crain 2005, p. 157). Stage 6
reasoners would exhibit a Rawlsian appreciation of the
principled requirements for social agreement in
a democracy and especially the centrality of justice conceived as fairness. This last stage follows logically from
Stage 5, but as indicated above, it lacks the empirical
support of the earlier stages.
Table 1 indicates Kohlbergs view of moral growth,
but note that ages for each stage are correlations.
Kohlbergs stages of moral development made
a considerable impact. In addition to his own publications and their influence on other researchers, the
stages also entered into the discourse of teaching in
psychology, education, and applied ethics. Publications
referring to Kohlberg in 2010 examined reasoning
across a range of fields beyond education, including
business ethics, nursing ethics, and more.

Important Scientific Research and

Open Questions
Kohlberg recognized that his theory had practical
implications. If the attainment of the higher stages of

1695

Kohlberg, Lawrence (19271987). Table 1 Kohlbergs

stages of moral development
Level

Age

Stage

Description

Level 3: Postconventional

Maturity

Stage 6

Kantian
universality

Maturity

Stage 5

Social contract

Level 2:
Conventional

Level 1: Preconventional

Adulthood Stage 4

Obedience to
authority

Teens

Stage 3

Good boy/girl

512

Stage2

Instrumental

05 years

Stage 1

Punishment
avoidance

While most individuals develop over the years to sat 4 above, only
a few continue on the stages 5 and even fewer to 6.

moral development is promoted by appropriate social

interactions, how could such interactions be facilitated?
His response was the so-called Cluster Schools that
afforded pupils the opportunity to discuss moral questions with more mature thinkers and to reflect on their
experience. Kohlbergs involvement in the Cluster
School movement began in 1974, when he was involved
in planning the first, and he then applied his vision to
other schools and a womens prison. He envisaged
small, democratic communities even schools
within schools where students and teachers participated equally in establishing norms. This participation
offered developing moral reasoners, the opportunity to
act on real issues, rather than just theorize about hypothetical dilemmas. The Cluster Schools did not survive
Kohlberg, but they did inspire others to embark on
similar innovative programs (Walsh 2000).

Criticisms of the Theory

Kohlbergs research looked at reasoning rather than
behavior, and this presents a difficulty. Someone
could offer a convincing discussion of ethical conduct
while behaving unethically. Equally, an ethical actor
might not be able or willing to articulate the reasoning
that explains and justifies the conduct.
Probably the most famous critique of Kohlberg was
that of Carol Gilligan, his colleague at Harvard. Her
book, In a Different Voice (1982) had wide influence in
education, gender studies, philosophy, and political
science. Gilligans argument was that Kohlberg had
studied only boys and that aspects of the reasoning of

1696

Kohler, Wolfgang (18871967)

girls are likely to overlooked or diminished in a study of

boys. Gilligans interviews of women faced with decisions about abortion found a strong emphasis on relationships and care rather than more abstract principles
such as justice. The concept of justice underpinning
Kohlbergs highest level of moral development emphasized rules, principles, rights, duties, and the like. In
contrast, Gilligan found that women faced with very
important choices considered consequences for a wide
variety of others and sought to pacify, support, accommodate, and respond to the needs of others. She argued
that the considerations of these women were moral in
a deep and abiding sense that was more profound than
Kohlbergs stages 5 and 6, and represented a different,
equally important moral voice.

Cross-References
Moral Learning
Piagets Learning Theory
Piaget, Jean (18961980)

References
Cortese, A. (1990). Ethnic ethics: The restructuring of moral theory.
Albany: SUNY Press.
Crain, W. C. (2005). Theories of development (5th ed.). Upper Saddle
River: Pearson Prentice-Hall.
Gilligan, C. (1982). In a different voice. Cambridge: Harvard University Press.
Walsh, C. (2000). Reconstructing Larry: Assessing the legacy of
Lawrence Kohlberg. Ed. Magazine, (Harvard Graduate
School of Education). http://www.gse.harvard.edu/news/features/
larry10012000_page1.html. Accessed 24 Oct 2010.

Kohler, Wolfgang (18871967)

NORBERT M. SEEL
Department of Education, University of Freiburg,
Freiburg, Germany

Carl Stumpf and earned his Ph.D. from the University

of Berlin in psycho-acoustics in 1909. Then he moved
to Frankfurt am Main, where he collaborated with Max
Wertheimer and Kurt Koffka from 1910 to 1913, working on the foundations of Gestalt theory. From 1913 to
1920, Kohler served as Director of the Anthropoid
Institute of the Prussian Academy of Sciences in Tenerife (Canary Islands), where he conducted research
with animals on insightful learning. In 1921, he was
appointed to the most prestigious position in German
psychology, Director of the Psychological Institute at
the Friedrich-Wilhelm University of Berlin, which
became the center of Gestalt psychology for the next
14 years. Like many other cultural institutions in Germany, the school of Gestalt psychology in Berlin was
destroyed by the Nazis. Wertheimer, Koffka, Duncker,
and Lewin immigrated due to their Jewish faith to the
United States or the United Kingdom. Kohler protested
publicly against the Nazis and their race policies against
Jewish scientists in 1935 (Henle 1978), and as
a consequence of his opposition to the Nazi government he also immigrated to the United States, where he
was appointed to the faculties of Swarthmore and
Dartmouth Colleges. Kohler was a fascinating personality. Not only did he bravely oppose the Nazis, he was
also open hearted and, unlike other leading psychologists of the time, could accept scientific criticism of his
work and respected alternative ideas. Due to his reputation he was President of the American Psychological
Association from 1956 to 1959. In 1962, he was
awarded with the honorary citizenship of the Free
University of Berlin. Kohler died in 1967 in Enfield,
New Hampshire. Until today, he is best known for cofounding the school of Gestalt psychology and for his
experimental research with animals. Closely connected
with his name is the Wolfgang Kohler Primate
Research Centre, a cooperation between the Max
Planck Institute for Evolutionary Anthropology and
the Leipzig Zoo (Germany).

Theoretical Background
Life Data
Wolfgang Kohler was born in 1887 in Reval, Estonia,
and grew up in Wolfenbuttel, Germany. Kohler studied
philosophy, science, and psychology at the Universities
of Tubingen, Bonn, and Berlin. One of his teachers in
science was Max Planck. He studied psychology with

Kohler is one of the founders of Gestalt psychology,

which opposed Wundts elementism as well as behaviorism with its focus on mechanical stimulusresponse
behavior and incremental learning. However, unlike
Koffka and Wertheimer, Kohler concentrated on animal research, especially during the six-year period in
which he served as Director of the anthropoid research

Kohler, Wolfgang (18871967)

facility on the island of Tenerife. This research contributed to a radical revision of learning theory.

Insight Learning Versus Incremental

Learning
When Wertheimer, Koffka, and Kohler started
founding Gestalt theory, Thorndike was the most influential learning researcher and had the prevailing
view on animal learning. Based on his studies of monkeys, dogs, rats, and especially cats, Thorndike concluded that learning is a trial-and-error process which
is dependent on reinforcement. Kohler opposed
Thorndikes functionalism and theory of incremental
learning and attempted to show that animals can solve
problems through insight as a kind of cognitive trial
and error rather than through behavioral trial and
error. His first experiments with dogs and cats
supported his assumptions, and encouraged by the
results he conducted a series of experiments in which
chimpanzees were situated in experimental settings
that confronted the subjects with a problem. Kohler
used four chimpanzees in his experiments, Chica,
Grande, Konsul, and Sultan. In one experiment, Kohler
placed bananas outside Sultans cage and two bamboo
sticks inside his cage. Neither stick was long enough to
reach the bananas, so the only way to reach them was to
put the sticks together. Kohler demonstrated the solution to Sultan by putting his fingers into the end of one
of the sticks (Hothersall 1995). However, this did not
help Sultan solve the problem. After some contemplation, Sultan put the two sticks together and created
a stick long enough to reach the bananas outside his
cage (Fig. 1).
Another study involved bananas suspended from
a roof. The chimps first tried to knock them down
using a stick. Then, they learned to stack boxes on top
of one another to climb up to the bananas (Fig. 2).
Some of the other experiments have been preserved
on film. In a typical sequence, a chimp jumps fruitlessly
at bananas that have been hung out of reach. Usually,
after a period of unsuccessful jumping, the chimp
apparently becomes angry or frustrated, walks away in
seeming disgust, pauses, then looks at the food in what
might be a more reflective way, then at the toys in the
enclosure, then back at the food, and then at the toys
again. Finally the animal begins to use the toys to get at
the food (http://www.iwf.de/iwf/do/mkat/details.aspx?
Signatur=D+1148).

1697

These experiments with chimpanzees led Kohler to

the conclusion that they did not solve the problem by
means of behavioral trial and error but rather by sudden comprehension of the situation (insight) and
intelligent use of available tools. Insight follows from
the particular characteristics of objects under consideration and occurs without reinforcement.

Kohler, Wolfgang (18871967). Fig. 1 Sultan making

a double stick (From Kohler 1925)

Kohler, Wolfgang (18871967). Fig. 2 Grande stacks

boxes to reach the goal (From Kohler 1925)

1698

Kolbs Learning Styles

Kohler proposed the view that insight follows from

the characteristics of objects under consideration.
Although the subjects varied in their ability to solve
a given problem, they all weighed different hypotheses
to achieve a solution and then succeeded in solving it
through a sudden leap of imagination or insight into
the features of the situation. In accordance with these
observations, Kohler formulated the theory that learning can occur by insight (defined in terms of sudden
comprehension) as opposed to incremental learning
and gradual understanding. Insightful learning can
occur without any reinforcement, but it is not necessarily observable by others. Based on the investigation
on Tenerife, Kohler published The Mentality of Apes in
1917, demonstrating that Gestalt theory could be
applied to animal learning and problem solving.
Another important field of Kohlers research
focused on the transfer of learning governed by the
law of transposition.

Contribution(s) to the Field of

Learning

Research on Transposition

Gestalt Psychology of Learning

History of the Sciences of Learning
Microgenetic Research on Insightful Problem Solving
Wertheimer, Max

In his research on the nature of learning, Kohler also

used chickens as subjects. In one experiment, he placed
a white and a gray sheet of paper on the ground and
covered both with grain. If a chicken pecked at the grain
on the white sheet, it was shooed away; but if it pecked
at the grain on the gray sheet, it was allowed to eat.
After many trials, the chickens learned to peck at the
grain on only the gray sheet What did the animals
learn? Behaviorists would say that reinforcement
played an important role in strengthening the chickens
response to the stimuli. However, when Kohler
replaced the white paper with a sheet of black paper
the vast majority of the chickens approached the black
sheet. Reinforcement theorists would have predicted
that the chickens would continue to approach the
gray paper, but Kohlers explanation was that the
chickens had not learned a stimulusresponse association but rather a relationship based on insight. The
chickens continued to approach the sheet on which
they had previously been fed because it was the darker
of the two. According to Gestalt theory, an organism
learns principles or relationships, not specific responses
to specific situations. Once a principle has been learned,
the organism applies it to similar situations. This was
called transposition, Kohlers explanation of learning
transfer that contrasts sharply with Thorndikes identical-elements theory of transfer.

Kohlers research methods and both his theory of insight

and the law of transposition revolutionized the understanding of learning in general, and of animal learning in
particular. His work directed the focus of research on
learning and problem solving away from behavioral trial
and error and insignificant elements of behavior toward
holistic aspects of consciousness and learning by insight
and transposition. In the opinion of the author of this
entry, Kohler can be considered as one of the most
important psychologists of the twentieth century. However, although Kohler was active in research and writing
until his death in 1967, Gestalt psychology was soon
characterized as an antiquated school of psychology.
However, Kohlers seminal work with animals,
especially chimps, has been continued until today, as
seen by the Kohler Institute in Leipzig, Germany.

Cross-References

References
Boring, E. G. (1950). A history of experimental psychology (2nd ed.).
Englewood Cliffs: Prentice Hall.
Henle, M. (1978). One man against the Nazis Wolfgang Kohler. The
American Psychologist, 33, 939944.
Hergenhahn, B. R. (2009). An introduction to the history of psychology
(6th ed.). Belmont: Wadsworth.
Hothersall, D. (1995). History of psychology. New York: McGraw-Hill.
Kohler, W. (1917/1925). The mentality of apes. London: Routledge &
Kegan Paul.
Kohler, W. (1929/1970). Gestalt psychology: An introduction to new
concepts in modern psychology. New York: Liveright.

Kolbs Learning Styles

ALICE KOLB, DAVID A. KOLB
Organization Behavior Department, Weatherhead
School of Management, Case Western Reserve
University, Cleveland, OH, USA

Synonyms
Learning mode; Learning preference

Kolbs Learning Styles

Definition
Kolbs learning styles are defined by an individuals relative preference for the four modes of the learning cycle
described in experiential learning theory: Concrete
Experience, Reflective Observation, Abstract Conceptualization, and Active Experimentation. These learning
styles can be assessed by the Kolb Learning Style Inventory (2005). Nine distinct learning style patterns have
been observed: experiencing, diverging, reflecting,
assimilating, thinking, converging, acting, accommodating, and balancing.

Theoretical Background
Learning style describes the unique ways that individuals spiral through the learning cycle based on their
preference for the four different learning modes: CE,
RO, AC, & AE. Because of our genetic makeup, our
particular life experiences, and the demands of our
present environment, we develop a preferred way of
choosing among these four learning modes. We resolve
the conflict between being concrete or abstract
and between being active or reflective in patterned,
characteristic ways. ELT posits that learning is the
major determinant of human development and how
individuals learn shapes the course of their personal development. Previous research has shown that
learning styles are influenced by culture, personality
type, educational specialization, career choice, and current job role and tasks (Table 1).
ELT argues that learning style is not a psychological
trait but a dynamic state resulting from synergistic
transactions between the person and the environment.
This dynamic state arises from an individuals

preferential resolution of the dual dialectics of

experiencing/conceptualizing and acting/reflecting.
The stability and endurance of these states in individuals comes not solely from fixed genetic qualities or
characteristics of human beings: nor, for that matter,
does it come from the stable fixed demands of environmental circumstances. Rather, stable and enduring
patterns of human individuality arise from consistent
patterns of transaction between the individual and his
or her environment.. . . The way we process the possibilities of each new emerging event determines the
range of choices and decisions we see. The choices
and decisions we make to some extent determine the
events we live through, and these events influence our
future choices. Thus, people create themselves
through the choice of actual occasions they live
through (Kolb 1984, pp. 6364).
Much of the research on ELT has focused on the
concept of learning style using the Kolb Learning Style
Inventory (KLSI) to assess individual learning styles.
While individuals who took the KLSI show many different patterns of scores, nine consistent styles have
been identified based on individuals relative preferences for the four learning modes. Four of these style
types emphasize one of the four learning modes
Experiencing (CE), Reflecting (RO), Thinking (AC),
and Acting (AE). Four others represent style types
that emphasize two learning modes, one from the grasping dimension and the other from the transforming
dimension of the ELT model Diverging (CE & RO),
Assimilating (AC & RO), Converging (AC &AE),
and Accommodating (CE &AE). The final style type
balances all four modes of the learning cycle Balancing

Kolbs Learning Styles. Table 1 Relationship between specialized learning styles and five levels of behavior
Behavior level

Diverging

Assimilating

Converging

Accommodating

Personality types

Introverted feeling

Introverted intuition

Extraverted thinking

Extraverted sensation

Educational
specialization

Arts, English, History,

Psychology

Mathematics,
Physical Science

Engineering, Medicine

Education
Communication Nursing

Professional
career

Social service Arts

Sciences Research
Information

Engineering Medicine
Technology

Sales Social service

Education

Current job

Personal jobs

Information jobs

Technical jobs

Executive jobs

Adaptive
competencies

Valuing skills

Thinking skills

Decision skills

Action skills

1699

1700

Kolbs Learning Styles

(CE, RO, AC, &AE). The following summaries of these

nine basic learning styles are based on research and
clinical observation of these patterns of KLSI scores.
Learners with an Experiencing style emphasize feeling (CE) while balancing acting (AE) and reflecting
(RO). Their greatest strengths reside in their ability to
deeply involve themselves in concrete experiences while
be equally comfortable in the outer world of action and
the inner world of reflection. They are particularly
adept in relationships with people. They learn by
actively involving themselves in new and challenging
situations and by stepping back and reflecting on their
experiences from differing points of view. They love
hands-on activities, but also learn by carefully observing the world around them. In the formal learning
situations, working in groups, role-playing, brainstorming, or fieldwork may appeal to them. Because
they place the least emphasis on Abstract Conceptualization they sometimes are disorganized, lacking plans
and theories to guide them.
Learners with a Reflecting style emphasize reflection
(RO) while balancing feeling (CE) and thinking (AC).
The learning strengths of this style are the capacity
for deep reflection informed by the ability to be both
feeling oriented and conceptual. They learn by combining the abilities of creative idea generation and putting
ideas into concise, logical form. They feel equally
at home in reflection on experiencing and thinking. As
a result, they have a rich and intuitive understanding of
matters of importance to them. They enjoy exploring
why things are the way they are, but also thrive in
uncovering what makes the world turn. They thrive in
learning environment rich in discussions, interactions,
and through readings that provide them with a deeper
understanding of themselves and the world around
them. Because of their low emphasis on Active Experimentation they have trouble in putting plans into
action, spending much time buried in thought. Because
action is short-circuited in the learning cycle, their
thoughts are about their feelings rather than direct
actions. This imbalanced cycle lacks the rejuvenation
provided by testing ideas in action.
Learners with a Thinking style emphasize thinking
(AC) while balancing reflecting (RO) and acting (AE).
They are deep thinkers who are able to inductively
develop a particular concept or idea and deductively
evaluate its validity and practicality by testing them in
the real world. They can draw both on the rich inner

world of reflection and abstraction and outer world of

action. They thrive on creating conceptual models that
can be applied or generalized to other situations.
Because they place little emphasis on feeling in their
style, they value being logical and unemotional. They
may be uncomfortable with personal relationships and
prefer working alone. They learn best in a wellstructured learning environment in which they can
design or conduct scientific experiments, or manipulate data.
Learners with an Acting style emphasize acting (AE)
while balancing feeling (CE) and thinking (AC). They
combine the ability to find solutions to questions or
problems based on their technical analysis with attention to the needs of people and sources of information
in concrete situations. They are equally comfortable in
functioning in a practical world that can make use of
their feelings and actions, as well as in a subjective
world that requires their thinking abilities. As a result,
they excel in identifying and integrating task and peoples needs. Their low emphasis on reflection can sometimes be a problem when they become overcommitted
to their idea of how things should be done. In formal
learning situations, they learn best through real-life
projects, field trips, and hands-on experiments.
Learners with Diverging style learn primarily
through feeling (CE) and reflecting (RO). They are
best at viewing concrete situations and exploring
them from many different points of view. Their
approach to situation is observing rather than taking
action. People with this style enjoy situations that call
for generating a wide range of feelings and ideas, such
as brainstorming sessions. They are imaginative, sensitive to feelings, have broad cultural interests, and like to
gather information. In formal learning situations, they
like to receive personalized attention and feedback.
They prefer working in groups to gather information
and listening with an open mind.
Learners with an Assimilating style learn primarily
through thinking (AC) and reflecting (RO). They are
best at understanding a wide range of information and
putting it into concise, logical form. They are less
focused on people and more interested in abstract
ideas and concepts. Generally, they find it more important that a theory have elegance and logical soundness
than practical value. Because they place less emphasis
on feeling and acting in their style, they may prefer to
work alone. They do not make quick decisions but

Kolbs Learning Styles

1701

NW
Accommodating

N
experiencing

NE
Diverging

W
Acting

C
Balancing

E
Reflecting

SW
Converging

S
Thinking

SE
Assimilating

Reflecting

Experiencing

Acting

think things through. In formal learning situations,

they may prefer lectures, readings, exploring analytical
models, and having time to think things through.
Learners with a Converging style emphasize thinking (AC) and acting (AE) in learning situations. People with this style are best at finding practical uses for
ideas and theories. They like to solve problems and
make decisions based on finding logical solutions to
issues or problems. They prefer dealing with technical
tasks and problems rather than with social and interpersonal issues. Because they place less emphasis on
feeling and reflection in their learning style, they can
be uncomfortable in ambiguous situations and interpersonal issues. In formal learning situations, they
may prefer to experiment with ideas and engage in
simulations, laboratory assignments, and practical
applications.
Learners with an Accommodating style learn primarily through acting (AE) and feeling (CE). They
have the ability to learn from hands-on experience
and function well in ambiguous and uncertain situations. They enjoy achieving goals and involving themselves in new and challenging experiences. Their
tendency may be to act on intuitive gut feelings
rather than on logical analysis. In solving problems,
individuals with an Accommodating learning style rely
more heavily on people for information than on their
own technical analysis. Because they place less emphasis on reflection and thinking in their approach to
learning they can sometimes be disorganized and act
before thinking. In formal learning situations, people
with this learning style prefer to work with others to get
assignments done, to set goals, to do field work, and to
test out different approaches for completing a project.
Learners with a Balancing style balance the extremes
of the dialectics of action/reflection and concrete/
abstract by finding a middle ground between them.
Their central position on the four learning modes
allows them to see many different perspectives on issues
and bridge differences between people with different
styles. They are often creative, but also experience difficulty in making decisions. They are able to change
their learning style to meet the learning demands of the
task they face. In a team they often adapt to fill in the
missing style needed to get the task done. In formal
learning environments they can change their learning
style to meet the learning demands of the task they
face (Fig. 1).

Thinking

Kolbs Learning Styles. Fig. 1 The nine learning styles of

the Kolb learning style inventory

Important Scientific Research and

Open Questions
Since its first statement in 1971 there have been many
studies of Kolbs learning styles. Since 2000, ELT
research in these fields around the world has more
than doubled. The current experiential learning theory
bibliographies (Kolb and Kolb 2010a, b) include over
2,500 entries. Since there are many learning style differences among different academic specialties, it is not
surprising to see that learning style research is highly
interdisciplinary, addressing learning and educational
issues in many fields. The following is a brief overview
of research activity in the various fields:
Education. The education category includes the
largest number of ELT/LSI studies. The bulk of studies
in education are in higher education (excluding professional education in the specific fields identified
below). Many of the studies in higher education use
ELT and the LSI as a framework for educational innovation. These include research on the matching of
learning style with instructional method and teaching
style. A number of studies in the education category
have been done in other cultures UK, Canada, Australia, Finland, Israel, Thailand, China, Melanesia,
Spain, Malta, and American Indian.
Management. LSI research was first published in
management and there has continued to be substantial
interest in the topic in the management literature.
Studies can be roughly grouped into four categories:
management and organizational processes, innovation
in management education, theoretical contributions to
ELT including critique, and psychometric studies of the

1702

Kolbs Learning Styles

LSI. Cross-cultural ELT/LSI research has been done in

Poland, New Zealand, Australia, Canada, UK, and Singapore. Another group of studies has examined the
relationship between learning style and management
style, decision making, and problem solving. Other
work has measured work-related learning environments and investigated the effect of a match between
learning style and learning environment on job satisfaction and performance.
Computer and Information Science. The LSI has
been used widely in computer and information science
particularly to study online learning, end-user software
use, and end-user training. Other studies have examined the relationship between learning style and problem solving and decision making, online search
behavior, and performance in computer training.
Psychology. Studies in psychology have shown
a large increase over time, with 77% of the studies in
the recent period. Many of these recent studies were on
LSI psychometrics. The first version of the LSI was
released in 1976 and received wide support for its
strong face validity and independence of the two ELT
dimensions of the learning process (Marshall and
Meritt 1985; Katz 1986). Although early critic of the
instrument focused on the internal consistency of
scales and testretest reliability, a study by Ferrell
(1983) showed that the LSI version 1 was the most
psychometrically sound among four learning instruments of that time. In 1985, version 2 of the LSI was
released and improved the internal consistency of the
scales (Veres et al. 1987; Sims et al. 1986). Critics of this
version focused their attention on the testretest reliability of the instrument, but a study by Veres, Sims,
and Locklear (1991) showed that randomizing the
order of the LSI version 2 items results in dramatic
improvement of testretest reliability. This finding led
to experimental research and finally to the latest LSI
revision, LSI Version 3 (Kolb 1999a). The LSI version 3
has significantly improved psychometric properties,
especially testretest reliability (see Kolb 1999b).
Other research includes factor analytic studies of
the LSI, construct validation studies of ELT using the
LSI, and comparison of the LSI with other learning style
and cognitive style measures. Another line of work uses
ELT as a model for personal growth and development,
including examination of counselor/client learning
style match and its impact on counseling outcomes.

Medicine
The majority of studies in medicine focus on learning
style analysis in many medical education specialties
residency training, anesthesia education, family medicine,
surgical training, and continuing medical education.
Of significance here is the program of research by Baker
and associates (e.g., Baker et al. 1986, 1988). Also Curry
(1999) has done a number of studies comparing different measures of learning styles. Other research has
examined clinical supervision and patientphysician
relationships, learning style and student performance
on examinations, and the relationship between learning style and medical specialty career choice.

Nursing
ELT/LSI research has also increased dramatically with
81% of the nursing studies in the recent period. In
1990, Laschinger reviewed the experiential learning
research in nursing and concluded, Kolbs theory of
experiential learning has been tested extensively in the
nursing population. Researchers have investigated relationships between learning style and learning preferences, decision-making skills, educational preparation,
nursing roles, nursing specialty, factors influencing
career choices and diagnostic abilities. As would be
expected in a human service profession, nursing
learning environments have been found to have a predominantly concrete learning press, matching the
predominating concrete styles of nurses . . . Kolbs
cycle of learning which requires the use of a variety of
learning modalities appears to be a valid and useful
model for instructional design in nursing education
(p. 991).

Accounting
There has been considerable interest in ELT/LSI
research in accounting education, where there have
been two streams of research activity. One is the comparative assessment of learning style preferences of
accounting majors and practitioners, including
changes in learning style over the stages of career in
accounting and the changing learning style demands of
the accounting profession primarily due to the introduction of computers. Other research has been focused
on using ELT to design instruction in accounting and
studying relationships between learning style and performance in accounting courses.

Kolbs Learning Styles

In 1991, Stout and Ruble reviewed ELT/LSI research

in accounting education. While reviewing the literature
on predicting the learning styles of accounting students they found mixed results and concluded that
low predictive and classification validity for the LSI
was a result of weak psychometric qualities of the
original LSI and response set problems in the LSI
1985. They tentatively recommended the use of the
randomized version proposed by Veres et al. (1991).
They write, researchers who wish to use the LSI for
predictive and classification purposes should consider
using a scrambled version of the instrument, and note,
it is important to keep in mind that assessing the
validity of the underlying theoretical model (ELT) is
separate from assessing the validity of the measuring
instrument (LSI). Thus, for example, the theory may be
valid even though the instrument has psychometric
limitations. In such a case, sensitivity to differences
in learning styles in instructional design may be
warranted, even though assessment of an individuals
learning style is problematic (p. 50).

Law
We are now seeing the beginning of significant research
programs in legal education, for example, the program
developed by Reese (1998) using learning style interventions to improve student learning at the University
of Denver Law School.

Cross-References
Experiential Learning Cycle
Experiential Learning Spaces
Experiential Learning Spiral
Experiential Learning Theory
Learning Identity
Learning Style
Meta-cognitive Experiential Learning

References
Baker, J. D., Wallace, C. T., Cooke, J. E., Alpert, C. C., & Ackerly, J. A.
(1986). Success in residency as a function of learning style.
Anesthesiology, 65(3A), A472.

Baker, J. D., III, Cooke, J. E., Conroy, J. M., Bromley, H. R.,

Hollon, M. F., & Alpert, C. C. (1988). Beyond career choice: the
role of learning style analysis n residency training. Medical Education, 22, 527532.
Curry, L. (1999). Cognitive and learning styles in medical education.
Academic Medicine, 74, 409413.
Ferrel, B. G. (1983). A factor analytic comparison of four learning styles
instruments. Journal of Educational Psychology, 75(1), 3339.
Katz, N. (1986). Construct validity of Kolbs Learning style inventory,
using factor analysis and guttmans smallest space analysis. Perceptual and Motor Skills, 63, 13231326.
Kolb, D. A. (1984). Experiential learning: Experience as the source of
learning and development. Englewood Cliffs: Prentice-Hall.
Kolb, D. A. (1999a). Learning style inventory, version 3. Boston: TRG
Hay/McBer, Training Resources Group. trg_mcber@haygroup.
com.
Kolb, D. A. (1999b). Learning style inventory, version 3: Technical
specifications. Boston: TRG Hay/McBer, Training Resources
Group. trg_mcber@haygroup.com.
Kolb, A. Y., & Kolb, D. A. (2005). The Kolb learning style inventory
version 3.1: 2005 technical specifications. Boston: Hay
Transforming Learning. www.haygroup.com/tl.
Kolb, A. Y., & Kolb, D. A. (2010a). Experiential learning theory bibliography: 19712005. Cleveland: Experience Based Learning Systems. www.learningfromexperience.com.
Kolb, A. Y., & Kolb, D. A. (2010b). Experiential learning theory
bibliography: Recent research 20052010. Cleveland: Experience
Based Learning Systems. www.learningfromexperience.com.
Marshall, J. C., & Merritt, S. L. (1985). Reliability and construct
validity of alternate forms of the learning style inventory. Educational and Psychological Measurement, 45, 931937.
Marshall, J. C., & Merritt, S. L. (1986). Reliability and construct
validity of the learning style questionnaire. Educational and
Psychological Measurement, 46, 257262.
Reese, J. (1998). Enhancing law student performance: Learning style
interventions. Unpublished report. Saratoga Springs, NY:
National Center on Adult Learning. Empire State College.
Sims, R. R., Watson, P., & Buckner, K. E. (1986). The reliability and
classification stability of the Learning Style Inventory. Educational and Psychological Measurement, 46, 753760.
Veres, J. G., Sims, R. R., & Shake, L. G. (1987). The reliability and
classification stability of the learning style inventory in corporate
settings. Educational and Psychological Measurement, 47(4),
11271133.
Veres, J. G., Sims, R. R., & Locklear, T. S. (1991). Improving the
reliability of Kolbs revised LSI. Educational and Psychological
Measurement, 51, 143150.

1703