Cognitive Psychology
Edited by Christopher C. French and Andrew M. Colman
1995, London and New York: Longman. Pp. xiv + 106. ISBN 0-582-27810-4
Contents
Notes on editors and contributors
Series editor's preface
Introduction
1 Memory
Alan Baddeley MRC Applied Psychology Unit, Cambridge, England
2 Attention
Michael W. Eysenck Royal Holloway and Bedford New College, England
3 Psycholinguistics
Willem J. M. Levelt Max-Planck-Institut für Psycholinguistik, The Netherlands
4 Thinking and Reasoning
Jonathan St B. T. Evans University of the South West, England
5 Artificial Intelligence
Alan Garnham University of Sussex, England
Glossary
Index
Introduction
Christopher C. French and Andrew M. Colman
Many contemporary psychologists would describe
themselves as cognitive psychologists. This reflects not only the fact that
cognitive psychology is currently the dominant approach within academic
psychology, but also the increasing vagueness of the term cognitive.
After all, if the cognitive approach is seen to dominate, it is not surprising
that many different types of psychologists want to identify themselves with it.
However, it is probably fair to say that mainstream cognitive psychology is most
closely related to what used to be called experimental psychology. In
this volume, four of the five chapters deal with core components of mainstream
cognitive psychology -- memory, attention, psycholinguistics, thinking, and
reasoning -- largely from the perspective of human experimental psychology. The
final chapter discusses various cognitive topics from the particular perspective
of artificial intelligence which, along with experimental studies of cognition
in normal humans and the neuropsychological approach, constitutes one of the
main approaches within cognitive psychology.
The word cognition comes from the Latin cognoscere, which
means to apprehend. According to Chambers English Dictionary, to
apprehend means to be conscious of by the senses; to lay hold of by the
intellect; to catch the meaning of; to understand; to recognize; to consider.
These meanings are all covered in the following chapters. Increasingly, however,
it is recognized that cognitive psychologists must take into account the effects
of motivation, emotion, and a range of other psychological factors. So why not
just say that cognitive psychology is synonymous with psychology as a whole?
The main reason is that cognitive psychology is characterized by a
particular approach to its subject matter: the information-processing approach.
This approach views mental events in terms of information flow. It implies a
somewhat mechanistic view of the mind, which is seen as being enormously
flexible and adaptive but nevertheless as a rule-governed automaton.
Information-processing models of cognitive processes are often represented as
flowcharts (for example, Figure 1 of chapter 1). Flowcharts highlight the often
implicit assumption that mental operations occur in a clear sequence. Early
cognitive models tended to assume that each component (each of the boxes in the
flowcharts) had to complete its processing before passing on the results to the
next component (via the arrows). This serial processing assumption is
increasingly questioned and contrasted with parallel processing models
(for example, connectionist models, see chapter 5).
Early cognitive theories also tended to assume that stimuli impinge on a
passive organism and are processed in a fairly automatic manner. In fact it is
easy to demonstrate that the expectations of the organism influence what is
perceived and remembered. This is an example of what is known as a top-down
influence on processing (because it originates at a "higher" level
within the cognitive system), in contrast to the stimulus-driven or bottom-up
influences emphasized by early theories. Cognitive processes often involve
complex interplays of both top-down and bottom-up processes. At the heart of
cognitive psychology is the notion that cognition involves processes acting on
and transforming symbolic representations in ways that allow the organism to
model the external world internally. This internal model is based on the
interaction between stimulus information from the senses and what we already
believe about the way the world operates.
The cognitive approach may legitimately be applied to other traditional
subdivisions within psychology. With respect to developmental psychology, for
example, we can ask questions regarding how cognitive processes change, for
better or for worse, with ageing. We can also ask how cognitive representations
of the world develop during childhood and how this relates to children's
behaviour and abilities. Indeed, the approach of Piaget, one of the founders of
developmental psychology as we know it today, was very much in the spirit of
modern cognitive psychology. In social psychology, topics such as social
interaction, attitude formation, and belief systems, to name but a few, are
often approached from an information-processing perspective.
Recently, ideas from mainstream cognitive psychology have been applied
with considerable success in attempts to understand the traditional subject
matter of abnormal psychology, such as anxiety and depression. It has been
shown, for example, that anxiety and depression both influence cognitive
processing but in surprisingly different ways (Mathews & MacLeod, 1994). It
is appropriate at this point, however, to emphasize that many clinicians who
refer to themselves as cognitive therapists actually do very little that
corresponds in any way to the activities of mainstream experimental cognitive
psychologists. This is an example of the vagueness, referred to earlier, of the
term cognitive.
The relationship between cognitive psychology and neuropsychology is so
important that it merits special mention. In recent years, the relationship
between the two has led to the establishment of the field of cognitive
neuropsychology. Alongside human experimental psychology and artificial
intelligence, cognitive neuropsychology is one of the three main approaches to
cognition. Cognitive neuropsychology is concerned with the patterns of normal
and impaired functioning in brain-damaged patients. This can provide an
extremely useful way to test cognitive theories. One of the most fruitful
examples of this approach is in terms of cognitive models of reading and
writing. By studying the particular pattern of errors made by a brain-damaged
patient and contrasting it with the patterns of errors made by other patients,
the validity of different models of reading and writing can be tested. It is
assumed, for example, that different routes exist for translating written words
into speech (see chapter 3). Some brain-damaged patients seem able to read only
by translating each letter into its corresponding sound. These patients have
great difficulty reading orthographically irregular words such as yacht.
Others seem to have the ability to read only via whole word recognition, and
they have great difficulty reading simple non-words such as bink, because
they cannot translate letters into sounds. Such observations offer clear support
for the existence of more than one route from the printed to the spoken word
(Ellis, 1993).
It is important to be aware of the limitations of neuropsychological data
in providing explanations of normal cognition. Cognitive neuropsychologists
often assume that the cognitive system is basically modular, that is, that it
consists of numerous relatively independent cognitive processing subsystems.
Brain damage typically affects the functioning of some (but not all) of these
subsystems, and so it should ultimately be possible to use data gathered by
studying brain-damaged patients to specify the number and nature of cognitive
subsystems. In practice, brain damage is often fairly extensive, and this
complicates the enterprise. Little can be deduced if a patient shows general
impairments across a wide range of tasks. Furthermore, brain-damaged patients
may use different strategies from those used by other people to perform a task,
and this must also be taken into account. In spite of all these problems, such
data remain amongst the most useful in cognitive psychology.
In computer science, the term computation refers not just to
numerical calculation (addition, multiplication, and so on) but more generally
to symbol manipulation. The computer seems to offer a particularly appropriate
analogy for human cognition. Both computers and people take in information,
process it in a number of ways, and produce various types of output. It
therefore seems natural that psychologists should use computers to try to
simulate human cognition. Computer simulation is not synonymous with artificial
intelligence or AI (see Chapter 5). AI is concerned with making computers
perform tasks that we would normally think of as requiring intelligence, whereas
computer simulation has the rather more specific aim of making computers perform
such tasks in the same way that humans do. Naturally, this requires a
psychological theory as a starting point. If a theory of cognitive performance
is precise and explicit enough, it can be implemented on a computer. If the
computer performs in the same way as a human being (including making the same
errors), this provides good support for the model -- although it does not, of
course, prove that the model is correct.
As well as being part of psychology as a whole, cognitive psychology is
also part of the more general interdisciplinary subject of cognitive science.
The other major disciplines involved, in addition to neuropsychology and
computer science, are linguistics and philosophy. These other disciplines have
had a major influence on cognitive psychology and vice versa. Behaviourism, the
dominant approach within psychology before the rise of cognitive psychology, was
shown to be unable to explain the acquisition and use of language. The field of
linguistics was revolutionized by Noam Chomsky's ideas, and Chomsky was very
influential in attacking simplistic behaviourist explanations of language
behaviour. Philosophical issues such as the nature of mind and consciousness
have taxed great thinkers for centuries. Clearly, such issues cannot be avoided
in cognitive psychology (see chapter 5 for a discussion of some philosophical
issues raised by artificial intelligence).
The five chapters in this volume cover core topics of cognitive
psychology, and inevitably there is some overlap between their contents. In
chapter 1, which deals with memory, Alan Baddeley distinguishes between sensory
memory, working memory, and long-term memory, which have quite distinct
properties and functions, and he discusses at some length the real-world
implications of the major research findings on working memory and long-term
memory in particular. Among the issues he deals with are the unreliability of
eyewitness testimony, prospective memory (remembering to do things), and
autobiographical memory (remembering things that have happened in the past). He
concludes with some brief comments on the psychoanalytic (Freudian) theory of
forgetting as repression. For further discussion of memory see the suggestions
for further reading at the end of chapter 1 and also the books by Parkin (1993)
and Cohen, Kiss, and Le Voi (1993).
In chapter 2, Michael W. Eysenck provides a general overview of the
psychology of attention, including focused attention and visual attention
(attending to an auditory or visual message that is accompanied by distracting
signals). This chapter also deals with research into divided attention,
absent-mindedness, and vigilance (sustained attention during prolonged,
monotonous tasks, such as watching a radar screen for occasional unexpected
blips, or proofreading a book looking for typographical errors). Eysenck's
discussion of absent-mindedness, in particular, intersects with Baddeley's
remarks on the fallibility of memory in chapter 1. For more information on
attention, see Eysenck's suggestions for further reading.
Willem J. M. Levelt's survey of psycholinguistics in chapter 3 covers all
of the main areas of research into the psychology of language apart from
language acquisition. Chapter 3 deals with both speech production and the
understanding of speech, and it contains additional material on reading and sign
language. Some of these topics are also considered in chapter 5 in relation to
artificial intelligence. Levelt's suggestions for further reading will be
helpful to readers who wish to pursue psycholinguistics in greater depth.
In chapter 4 Jonathan St B. T. Evans discusses thinking and reasoning.
With the help of some fascinating logical puzzles that have been used in
experimental investigations, Evans summarizes what is known about problem
solving, reasoning, and subjective judgements of probability. For more
information on thinking and reasoning, in addition to the suggestions for
further reading at the end of chapter 4, see Kahney (1993), and Garnham and
Oakhill (1994). For highly readable accounts of the ways in which the
fallibility of human reasoning can have effects in everyday life, see Sutherland
(1992). Topics considered in chapters 1 to 4 of the current volume are also
dealt with by Smyth, Collins, Morris and Levy (1994).
Finally, in chapter 5, Alan Garnham discusses certain aspects of
thinking, reasoning, and problem solving, from the slightly different
perspective of artificial intelligence. Other topics considered from this
perspective include knowledge representation, vision, language, and learning.
The history of the discipline is also considered, as are applications and
philosophical issues. The suggestions for further reading at the end of chapter
5 contain further information on artificial intelligence. For more detail on
connectionism, see Bechtel and Abrahamsen (1991), and for a consideration of the
philosophical issues raised by artificial intelligence, see Copeland (1993).
REFERENCES
Bechtel, W., & Abrahamsen, A. (1991). Connectionism and the mind: An introduction to parallel processing in networks. Oxford: Blackwell.
Cohen, G., Kiss, G., & Le Voi, M. (1993). Memory: Current issues (2nd ed.). Milton Keynes: Open University Press.
Copeland, J. (1993). Artificial intelligence: A philosophical introduction. Oxford: Blackwell.
Ellis, A. W. (1993). Reading, writing and dyslexia: A cognitive analysis (2nd ed.). London: Erlbaum.
Garnham, A., & Oakhill, J. (1994). Thinking and reasoning. Oxford: Blackwell.
Kahney, H. (1993). Problem solving: Current issues (2nd ed.). Milton Keynes: Open University Press.
Mathews, A., & MacLeod, C. (1994). Cognitive approaches to emotion and emotional disorders. Annual Review of Psychology, 45, 25-50.
Parkin, A. (1993). Memory: Phenomena, experiment and theory. Oxford: Blackwell.
Smyth, M. M., Collins, A. F., Morris, P. E., & Levy, P. (1994). Cognition in action (2nd. ed.). Hove: Erlbaum.
Sutherland, S. (1992). Irrationality: The enemy within. London: Constable.