In a field choked with seemingly impenetrable jargon, Philip N. Johnson-Laird has done the impossible: written a book about how the mind works that requires no advance knowledge of artificial intelligence, neurophysiology, or psychology. The mind, he says, depends on the brain in the same way as the execution of a program of symbolic instructions depends on a computer, and can thus be understood by anyone willing to start with basic principles of computation and follow his step-by-step explanations.

The author begins with a brief account of the history of psychology and the birth of cognitive science after World War II. He then describes clearly and simply the nature of symbols and the theory of computation, and follows with sections devoted to current computational models of how the mind carries out all its major tasks, including visual perception, learning, memory, the planning and control of actions, deductive and inductive reasoning, and the formation of new concepts and new ideas. Other sections discuss human communication, meaning, the progress that has been made in enabling computers to understand natural language, and finally the difficult problems of the conscious and unconscious mind, free will, needs and emotions, and self-awareness. In an envoi, the author responds to the critics of cognitive science and defends the computational view of the mind as an alternative to traditional dualism: cognitive science integrates mind and matter within the same explanatory framework.

This first single-authored introduction to cognitive science will command the attention of students of cognitive science at all levels including psychologists, linguists, computer scientists, philosophers, and neuroscientists--as well as all readers curious about recent knowledge on how the mind works.

Language and Perception lays foundations for a new branch of the psychological sciences—psycholexicology, the psychological study of the meaning of words. Although the basic argument is psychological, George Miller and Philip Johnson-Laird also draw on current work in artificial intelligence, linguistics, philosophy, and social anthropology. Their closely argued and lucid treatise will stimulate specialists in many fields to questions their assumptions and broaden their thinking about semantic problems.

Miller and Johnson-Laird explore an approach to word meaning that is procedural in orientation. The meaning of a word is construed as a set of mental procedures necessary to employ the word appropriately and respond sensibly to its use by others. Since the appropriate use of many words depends on a perceptual assessment of the situation to which the word applies, the authors begin by considering human perception in terms of the perceptual tests that it can apply to the environment.

As the argument advances, however, Miller and Johnson-Laird observe that the meaning of many words depends on functional as well as perceptual attributes and on the place that the word occupies within a system of conceptual relations between words. Ultimately, Miller and Johnson-Laird contend that perception and language are related only indirectly as alternative routes into a vastly complex conceptual world. Something of the shape of that world is inferred from the basic concepts that are important enough to be incorporated into the meanings of English words.

Mental Models offers nothing less than a unified theory of the major properties of mind: comprehension, inference, and consciousness. In spirited and graceful prose, Johnson-Laird argues that we apprehend the world by building inner mental replicas of the relations among objects and events that concern us. The mind is essentially a model-building device that can itself be modeled on a digital computer. This book provides both a blueprint for building such a model and numerous important illustrations of how to do it.

In several key areas of cognition, Johnson-Laird shows how an explanation based on mental modeling is clearly superior to previous theory. For example, he argues compellingly that deductive reasoning does not take place by tacitly applying the rules of logic, but by mentally manipulating models of the states of affairs from which inferences are drawn. Similarly, linguistic comprehension is best understood not as a matter of applying inference rules to propositions derived from sentences, but rather as the mind's effort to construct and update a model of the situation described by a text or a discourse. Most provocative, perhaps, is Johnson-Laird's theory of consciousness: the mind's necessarily incomplete model of itself allows only a partial control over the many unconscious and parallel processes of cognition.

This an extraordinarily rich book, providing a coherent account of much recent experimental work in cognitive psychology, along with lucid explanations of relevant theory in linguistics, computer science, and philosophy Not since Miller, Galanter, and Pribram's classic Plans and the Structure of Behavior has a book in cognitive science combined such sweep, style, and good sense. Like its distinguished predecessor, Mental Models may well serve to fix a point of view for a generation.