The Cognitive Approach - Chapter 1, Selected Sections
Ulric Neisser (1923- )
It has been said that beauty is in the eye of the beholder. As a hypothesis about localization of function, the statement is not quite right - the brain and not the eye is surely the most important organ involved. Nevertheless it points clearly toward the central problem of cognition. Whether beautiful or ugly or just conveniently at hand, the world of experience is produced by the man who experiences it....
Physically, this page is an array of small mounds of ink, lying in certain positions on the more highly reflective surface of the paper. It is this physical page which Koffka (1935) and others would have called the "distal stimulus," and from which the reader is hopefully acquiring some information. But the sensory input is not the page itself, it is a pattern of light rays, originating in the sun or in some artificial source, that are reflected from the page and happen to reach the eye. Suitably focused by the lens and other ocular apparatus, the rays fall on the sensitive retina, where they can initiate the neural processes that eventually lead to seeing and reading and remembering. These patterns of light at the retina are the so-called "proximal stimuli." They are not the least bit like eidola. One-sided in their perspective, shifting radically several times each second, unique and novel at every moment, the proximal stimuli bear little resemblance to either the real object that gave rise to them or to the object of experience that the perceiver will construct as a result.
Visual cognition, then, deals with the processes by which a perceived, remembered, and thought-about world is brought into being from as unpromising a beginning as the retinal patterns. Similarly, auditory cognition is concerned with transformation of the fluctuating pressure pattern at the ear into the sounds and the speech and music that we hear. The problem of understanding these transformations may usefully be compared to a very different question that arises in another psychological context. One of Freud's papers on human motivation is entitled "Instincts and their
Vicissitudes" (1915). The title reflects a basic axiom of psychoanalysis: that man's fundamental motives suffer an intricate series of transformations, reformulations, and changes before they appear in either consciousness or action. Borrowing Freud's phrase - without intending any commitment to his theory of motivation - a book like this one might be called "Stimulus Information and its Vicissitudes." As used here, the term "cognition" refers to all the processes by which the sensory input is transformed, reduced, elaborated, stored, recovered, and used. It is concerned with these processes even when they operate in the absence of relevant stimulation, as in images and hallucinations Such terms as sensation, perception, imagery, retention, recall, problem-solving, and thinking, among many others, refer to hypothetical stages or aspects of cognition.
Given such a sweeping definition, it is apparent that cognition is involved in everything a human being might possibly do; that every psychological phenomenon is a cognitive phenomenon. But although cognitive psychology is concerned with all human activity rather than some fraction of it, the concern is from a particular point of view. Other viewpoints are equally legitimate and necessary. Dynamic psychology, which begins with motives rather than with sensory input, is a case in point. Instead of asking how a man's actions and experiences result from what he saw, remembered, or believed, the dynamic psychologist asks how they follow from the subject's goals, needs, or instincts. Both questions can be asked about any activity, whether it be normal or abnormal, spontaneous or induced, overt or covert, waking or dreaming. Asked why I did a certain thing, I may answer in dynamic terms, "Because I wanted . or, from the cognitive point of view, "Because it seemed to me.. ."
Recognition of the difference between cognitive and dynamic theory does not mean that we can afford to ignore motivation in a book like this one. Many cognitive phenomena are incomprehensible unless one takes some account of what the subject is trying to do. However, his purposes are treated here primarily as independent variables: we will note that they can affect one or another cognitive mechanism without inquiring closely into their origin. This strategy will break down in the final chapter; remembering and thinking are too "inner-directed" to be treated in such a fashion. As a consequence, the last chapter has a different format, and even a different purpose, from the others.
The cognitive and the dynamic viewpoints are by no means the only possible approaches to psychology. Behaviorism, for example, represents a very different tradition, which is essentially incompatible with both. From Watson (1913) to Skinner (1963), radical behaviorists have maintained that man's actions should be explained only in terms of observable variables, without any inner vicissitudes at all. The appeal to hypothetical mechanisms is said to be speculative at best, and deceptive at worst. For them, it is legitimate to speak of stimuli, responses, reinforcements, and hours of deprivation, but not of categories or images or ideas. A generation ago, a book like this one would have needed at least a chapter of selfdefense against the behaviorist position. Today, happily, the climate of opinion has changed, and little or no defense is necessary. Indeed, stimulus-response theorists themselves are inventing hypothetical mechanisms with vigor and enthusiasm and only faint twinges of conscience. The basic reason for studying cognitive processes has become as clear as the reason for studying anything else: because they are there. Cognitive processes surely exist, so it can hardly be unscientific to study them.
Another approach to psychological questions, a world apart from behaviorism, is that of the physiologist. Cognition, like other psychological processes, can validly be studied in terms of the underlying neural events. For my part, I do not doubt that human behavior and consciousness depend entirely on the activity of the brain, in interaction with other physical systems. Most readers of this book will probably have the same prejudice. Nevertheless, there is very little of physiology or biochemistry in the chapters ahead. At a time when these fields are making impressive advances, such an omission may seem strange. An example may help to justify it. For this purpose, let us consider recent work on the physical basis of memory.
No one would dispute that human beings store a great deal of information about their past experiences, and it seems obvious that this information must be physically embodied somewhere in the brain. Recent discoveries in biochemistry have opened up a promising possibility. Some experimental findings have hinted that the complex molecules of DNA and RNA, known to be involved in the transmission of inherited traits, may be the substrate of memory as well. Although the supporting evidence so far is shaky, this hypothesis has already gained many adherents. But psychology is not just something "to do until the biochemist comes" (as I have recently heard psychiatry described); the truth or falsity of this new hypothesis is only marginally relevant to psychological questions. A pair of analogies will show why this is so.
First, let us consider the familiar parallel between man and computer. Although it is an inadequate analogy in many ways, it may suffice for this purpose. The task of a psychologist trying to understand human cognition is analogous to that of a man trying to discover how a computer has been programmed. In particular, if the program seems to store and reuse information, he would like to know by what "routines" or "procedures" this is done. Given this purpose, he will not care much whether his particular computer stores information in magnetic cores or in thin films; he wants to understand the program, not the "hardware." By the same token, it would not help the psychologist to know that memory is carried by RNA as opposed to some other medium. He wants to understand its utilization, not its incarnation.
The same point can be illustrated with quite a different analogy, that between psychology and economics. The economist wishes to understand, say, the flow of capital. The object of his study must have some tangible representation, in the form of checks, gold, paper money, and so on, but these objects are not what he really cares about. The physical properties of money, its location in banks, its movement in armored cars, are of little interest to him. To be sure, the remarkable permanence of gold has some economic importance. The flow of capital would be markedly different if every medium of exchange were subject to rapid corrosion. Nevertheless, such matters are not the main concern of the economist, and knowledge of them does not much simplify economic theory.
Psychology, like economics, is a science concerned with the interdependence among certain events rather than with their physical nature. Although there are many disciplines of this sort (classical genetics is another good example), the most prominent ones today are probably the so-called "information sciences," which include the mathematical theory of communication, computer programming, systems analysis, and related fields. It seems obvious that these must be relevant to cognitive psychology, which is itself much concerned with information. However, their importance for psychologists has often been misunderstood, and deserves careful consideration.
Although information measurement may be of little value to the cognitive psychologist, another branch of the information sciences, computer programming, has much more to offer. A program is not a device for measuring information, but a recipe for selecting, storing, recovering, combining, outputting, and generally manipulating it. As pointed out by Newell, Shaw, and Simon (1958), this means that programs have much in common with theories of cognition. Both are descriptions of the vicissitudes of input information.
We must be careful not to confuse the program with the computer that it controls. Any single general-purpose computer can be "loaded" with an essentially infinite number of different programs. On the other hand, most programs can be run, with minor modifications, on many physically different kinds of computers. A program is not a machine; it is a series of instructions for dealing with symbols: "If the input has certain characteristics ... then carry out certain procedures .. . otherwise other procedures ... combine their results in various ways ... store or retrieve various items ... depending on prior results ... use them in further specified ways ... etc." The cognitive psychologist would like to give a similar account of the way information is processed by men.
This way of defining the cognitive problem is not really a new one. We are still asking "how the mind works." However, the "program analogy" (which may be a better term than "computer analogy") has several advantages over earlier conceptions. Most important is the philosophical reassurance which it provides. Although a program is nothing but a flow of symbols, it has reality enough to control the operation of very tangible machinery that executes very physical operations. A man who seeks to discover the program of a computer is surely not doing anything self-contradictory!
There were cognitive theorists long before the advent of the computer. Bartlett, whose influence on my own thinking will become obvious in later chapters, is a case in point. But, in the eyes of many psychologists, a theory which dealt with cognitive transformations, memory schemata, and the like was not about anything. One could understand theories that dealt with overt movements, or with physiology; one could even understand (and deplore) theories which dealt with the content of consciousness; but what kind of a thing is a schema? If memory consists of transformations, what is transformed? So long as cognitive psychology literally did not know what it was talking about, there was always a danger that it was talking about nothing at all. This is no longer a serious risk. Information is what is transformed, and the structured pattern of its transformations is what we want to understand.
A second advantage of the "program analogy" is that, like other analogies, it is a fruitful source of hypotheses. A field which is directly concerned with information processing should be at least as rich in ideas for psychology as other fields of science have been before. Just as we have borrowed atomic units, energy distributions, hydraulic pressures, and mechanical linkages from physics and engineering, so may we choose to adopt certain concepts from programming today. This will be done rather freely in some of the following chapters. Such notions as "parallel processing," "feature extraction," "analysis-by-synthesis," and "executive routine" have been borrowed from programmers, in the hope that they will prove theoretically useful. The test of their value, of course, is strictly psychological. We will have to see how well they fit the data....