SCHEMAS IN THE COGNITIVE AND CLINICAL SCIENCES:

AN INTEGRATIVE CONSTRUCT

Dan J. Stein

Dept of Psychiatry,

College of Physicians and Surgeons and the

New York State Psychiatric Institute,

722 W 168 St,

NY, NY 10032

Correspondence concerning this article should be sent to Dan J.

Stein, M.B., Psychiatric Institute, 722 W 168 St, NY, NY,

10032.

Key Words: Schemas, Cognitive science, Psychotherapy

integration

Running Head: Schemas and Integration

Acknowledgements: Several anonymous reviewers deserve thanks

for their extremely helpful comments on earlier drafts of this

manuscript.

ABSTRACT

This paper is concerned in general with the

intersection of cognitive and clinical science and

in particular with schema theory. The use of

schema theory in the various subdisciplines of

cognitive science, as well as by cognitive-

behavioral clinicians and psychoanalytically

oriented clinicians is reviewed. It is argued

that schema theory, in both cognitive and clinical

sciences, allows a focus on mental structures,

their biological basis, their development and

change, and on the way in which they direct

psychological events. Schema theory not only

enables important advances in different clinical

schools, but it allows central clinical themes to

be tackled in convergent ways. It is concluded

that the schema construct allows integration

within cognitive science, within the clinic, and

between the two.

INTRODUCTION

Psychology has over the last few decades

witnessed a cognitive revolution, and the

multidisciplinary arena known as cognitive science

has increasingly come to the fore as a possible

unifying paradigm for the various psychological

sciences and schools. Workers within philosophy,

neuroscience, developmental psychology, cognitive

psychology, and social psychology have begun to

use similar constructs (Gardner, 1985).

Within the clinic, however, there remains

reluctance to accept a unified model of the mind.

Diverse theories and practices abound.

Nevertheless, some have argued that rigorous

research techniques and sophisticated theoretical

ideas have combined to foster a climate for a

unified, eclectic theory and practice of

psychotherapy (Beitman, Goldfried and Norcross,

1989).

There are some constructs, moreover, that are

being employed not only in the subdisciplines of

cognitive science but also in the clinic, and

indeed cognitive science appears to constitute a

paradigm to which diverse clinicians are attracted

(Bowers and Meichenbaum, 1984; Colby and Stoller,

1988; Ingram, 1986; Horowitz, 1988; Ruesch and

Bateson, 1968; Williams, Fraser, MacLeod, and

Matthews, 1988). Perhaps the most widely known of

the constructs used within both cognitive and

clinical science is that of schemas. It has been

argued that schemas are a heuristic and useful

concept in cognitive theory (Fiske and Linville,

1980). In this paper the suggestion is made that

schemas may be a heuristic and integrative notion

in clinical science. The argument procedes by

noting how cognitive scientists have used schema

theory to focus on the structures of the mind,

their biological basis, their development and

change, and the way in which they direct

psychological events. The paper then demonstrates

that schemas have been used by clinical scientists

to focus on analogous issues. It is concluded

that schemas constitute an important construct

that fosters integration within cognitive science,

within the clinic, and between the two.

COGNITIVE SCIENCE AND SCHEMA THEORY

The paper begins with a review of the use of

the schema concept in the various subdisciplines

of cognitive science. An historical perspective

is employed; the review begins with authors who

were important in laying the foundations for

cognitive science, and moves to a consideration of

contemporary workers.

a) Philosophy

Cognitive science may be viewed as an

enterprise that is concerned with the ways in

which the structures of the mind allow

representations of the world, and the ways in

which they process such representations. The

question of the mind's representation of the world

has long been posed by philosophy, and philosophy

may therefore be considered the founding

subdiscipline of cognitive science (Gardner,

1985).

One of the greatest philosophers, Kant,

employed the concept of the schema precisely in

order to discuss the possibility of knowledge.

Kant attempted to go beyond the impasse between

the empiricists, who argued that knowledge has its

origins in the external world, and the

rationalists, who argued that knowledge is a

product of the mind. He argued that schemas

interdigitate between properties of the mind (the

a priori categories) and raw sensory data (of a

posteriori experience). "This representation of a

universal procedure of the imagination in

providing an image for a concept, I entitle the

schema of the concept" (quoted in Gardner, 1985).

In more contemporary terms, mental schemas are

activated by the external world, and

simultaneously provide an interpretation of it.

b) Neuroscience

Cognitive science acknowledges that structures

of the mind have a biological basis, and

neuroscience is therefore an important

subdiscipline of cognitive science. Furthermore,

the neurologists Head and Holmes (1911) were among

the first to use the concept of the schema.

These workers were interested in the spatial

perceptions of patients of their bodies, and

referred to the basis of these as the postural

schema. The postural schema integrated sensations

which were triggered by postural change. In

lesions of the parietal lobe the schema may be

destroyed, with the possible outcome that patients

ignore part of their body, treating it as if it

were not their own. Conversely, an amputee may

have an intact brain schema, and therefore

experience movements in the missing phantom limb.

Today the more widely used term is the body schema

(Frederiks, 1969), and contemporary neuroscience

has advanced to the point where it can begin to

consider the biological underpinnings of more

complex schemas such as cognitive and affective

schemas (LeDoux, 1989).

c) Developmental Psychology

Although developmental psychology is not

usually considered one of the subdisciplines of

cognitive science, Piaget is one of the most

important figures in the prehistory of cognitive

science. The notion of the schema was central in

Piaget's work.

According to Piaget, the initial schemas of

the child comprise biologically based sensorimotor

reflexes which coordinate the child's interactions

with the environment. Gradually these biological

schemas allow adaptation to the environment by two

complementary processes. Via assimilation the

schema grasps some novel aspect of the

environnment, so modifying itself to cope with

the environment better. Via accomodation the

schema is differentiated and elaborated so as to

be consistent with the environment. With time,

the schemas are transformed to the point where the

organism reaches a new stage of development.

Piaget succeeds in providing a detailed

description of these transformations from the

point of sensorimotor schemas to the operations of

formal thought (Piaget, 1952).

d) Cognitive Psychology

The concept of the schema in contemporary

cognitive science is perhaps most directly

traceable to the work of the British cognitive

psychologist Bartlett (1932). Bartlett, a onetime

student of Head, was interested in memory, and in

particular in the notion that the context of an

experience had crucial effects on what was

retained and how well this was recalled.

Ebbinghaus had pioneered the experimental study of

memory using nonsense syllables, but this approach

did not seem adequate to Bartlett's concerns. A

conversation with Norbert Weiner, one of the

founding figures in cognitive science, gave

Bartlett an experimental methodology for

developing his ideas. Weiner's idea was to use

the Russian Scandal parlor game in which a story

is passed around the room, and then the original

and final versions compared. Bartlett found that

subjects showed consistent patterns of error in

the recall of narratives. Thus, for example, an

American Indian narrative would regularly be

revised by subjects until it came to resemble a

Western tale. Bartlett developed the construct of

the schema to explain this, describing a schema as

a component of memory which is formed from

encounters with the environment, and which

organizes information in specific ways. Such

schemas aid the recall of a typical (Western)

narrative, but systematically distort the recall

of an unusual (American Indian) narrative.

Bartlett (1932) wrote,

"Remembering is not the re-excitation of

innumerable fixed, lifeless, and

fragmentary traces. It is an

imaginative reconstruction, or

construction, built out of the relation

of our attitude towards a whole active

mass of past experience....It is thus

hardly ever really exact, even in the

most rudimentary cases of rote

recapitulation, and it is not at all

important that it should be so. The

attitude is literally an effect of the

organism's capacity to turn round up

upon its own "schemata" and is directly

a function of consciousness."

He defined a schema as

"an active organization of past

reactions or of past experiences which

must always be supposed to be operating

in any well-adapted organic repsonse.

Whenever there is any order or

regularity of behavior, a particular

response is possible only because it is

related to other similar responses which

have been serially organized, yet which

operate not singly as individual members

coming one after another, but as a

unitary mass."

A variety of definitions of schemas have been

offered subsequently. Thorndyke and Hayes-Roth

(1979) describe three universal assumptions made

by different authors: that a schema is an

organization of conceptually related elements

representing a prototypical abstraction of a

complex concept; that a schema gradually develops

from past experience; and that a schema guides the

organization of new information. A schema

comprises an architectural element (its structure)

and a propositional element (its content).

Cognitive schemas are involved in cognitive

operations (e.g. encoding, retrieval), in which

cognitive events (e.g. thoughts, images) are

produced and processed. Schemas are highly

interdependent and hierarchically organised, they

may involve verbal or nonverbal-elements, and they

may be more or less open to awareness (Craik and

Lockhart, 1972; Ingram and Kendall, 1986).

Schema theory has proven valuable in

accounting for a variety of psychological

phenomena. Schema theory has been used in

contemporary studies of memory (Schacter, 1989),

concept representation (Smith, 1989), problem

solving (VanLehn, 1989), movement (Jordan and

Rosenbaum, 1989) and language (Arbib et al, 1987).

Schemas have been found to facilitate recognition

and recall, to influence speed of information

processing and problem solving and allow for the

chunking of information into more meaningful units

to enable inference about missing data, and to

provide a basis for prediction and decision

making. It is thought that schemas engender bias

by relying on confirmatory evidence at the expense

of disconfirming evidence, but that this process

maintains schemas (Rumelhart, 1984; Meichenbaum

and Gilmore, 1984; Winfrey and Goldfried, 1986).

A variety of other constructs have been used

by cognitive scientists and their forerunners to

describe mental structures. Many bear a good deal

of family resemblance to the idea of schemas.

Notable examples include Abelson and Shank's

(1981) "scripts", Bandura's (1978) "self-systems",

Kelly's (1955) "personal constructs", Miller,

Galanter and Pribram's (1960) "plans", and

Minsky's (1975) "frames". The notion of the

connectionist network, currently extremely popular

in the field of artifical intelligence, has been

argued to represent a microlevel description of

the schema concept (Ben Zeev, 1988).

BETWEEN COGNITIVE AND CLINICAL SCIENCE

This review of the schema construct in

cognitive science demonstrates that schema theory

has been widely employed by cognitive scientists.

However, it may be objected that the very

diversity of the use of schema theory points to

the problematic nature of the schema concept.

Different cognitive scientists operationalize and

measure schemas in different ways. Similarly the

versatility of the concept may reflect only a lack

of theoretical rigor.

On the other hand the concept of schemas has

general heuristic value insofar as it allows

different cognitive scientists to theorize about

mental structures from the perspective of their

particular subdiscipline. The schema construct

allows different cognitive scientists to begin to

build an integrative framework that addresses such

questions as how the structures of the mind enable

representation, how they are based in biology, how

they develop and change, and how they account for

a variety of psychological phenomena.

Certainly there is room in schema research for

improvement in both empirical measurement and

theoretical rigor (Fiske and Linville, 1980;

Williams et al, 1988). Nevertheless, the

development of the schema concept as a broad

heuristic is important insofar as it represents a

move in cognitive science away from a molecular