by Morton Hunt
That was only one of many bits of rat behavior that Tolman concluded could be explained solely in terms of processes going on in the rat’s head. Years earlier, he and a colleague had built a simple maze with three routes to the goal box. The shortest was a straight path from the start box to the goal box; the second, a little longer, made a short loop to the left, then rejoined the straight path partway toward the goal box; and the third, the longest, made a long loop to the right, then rejoined the straight path close to the goal box. In a series of trials the rat, as behaviorist theory predicted, found its way to the food by all three routes but learned to take the shortest, since that was the most easily established habit.
Tolman then put a barrier across the straight path halfway to the goal so that the rat could reach it only by the longest route. According to theory, when the rat ran down the straight path and came up against the barrier, it should have turned back and tried the next most easily established habit—the medium-length route—but it immediately took the long route. To Tolman this suggested that the rat had built up a sort of mental map of the entire maze and “realized” that the barrier blocked all but the longest route.70
Tolman conducted many similar experiments, most of them far more complicated, and all of which supported his belief that “something like a field map of the environment gets established in the rat’s brain.” Standard behaviorist theory, he said, offered only a partial explanation of maze learning: “We agree… that the rat in running a maze is exposed to stimuli and is finally led as a result of these stimuli to the responses which actually occur. We feel, however, that the intervening brain processes are more complicated, more patterned, and often, pragmatically speaking, more autonomous than do the stimulus-response psychologists.”71
These studies led Tolman to propound a theory he called “purposive behaviorism.” Its essence was that rats act not as automata, developing habits solely according to the number and kind of stimuli they experience, but as if, in addition, they are influenced by their own expectations, their knowledge of what leads to what in a given situation, their goals, and other internal processes or states.72 As one orthodox behaviorist derisively said, Tolman’s rats were “buried in thought.”73
Tolman called these internal factors “intervening variables” (they intervened between stimulus and response) and insisted that they were compatible with behaviorism. “For the behaviorist,” he wrote, “‘mental processes’ are to be identified and defined in terms of the behaviors to which they lead. [They are] naught but inferred determinants of behavior… Behavior and these inferred determinants are both objective, defined types of entity.”74 It was a valiant effort to remain faithful, but Tolman had, willy-nilly, breached the dike of behaviorism and let in a trickle of mind. In time it would be a flood.
If reward and repetition only partly explain rat behavior, they give an even more limited account of the determinants and workings of human behavior. Consider memory, for example. Behaviorists portrayed it in purely mathematical terms: the more trials and reinforcements, the greater the rewards, the closer in time the S and the R, the more certain it is that the S will produce the R. If the stimulus is the question “What comes after five?,” the response is “six.” If the stimulus is the question “What is your phone number?,” the answer is a sequence of seven digits (ten if you include the area code). The first digit is the response to the question but is also the stimulus that produces the response of the second digit, and so on, in a chain of associative links.
But even at the height of the behaviorist era, psychologists knew that human memory was more complicated than that. For one thing, we “chunk” some information: we remember area codes, for instance, as units, not as a series of linked responses. For another thing, we have different kinds of memory: we learn some phone numbers only for a moment—we look them up, hold them in “short-term memory” until we dial, and then instantly forget them, but make others a part of our “long-term memory” (the vast stockpile of stored knowledge we draw on as needed). Some things require inordinate amounts of repetition and reward to become fixed in memory (many people can’t seem to remember their own Social Security number, though they’ve looked it up scores of times); other things (the exorbitant price we paid for dinner at a particular restaurant, our baby’s first words) remain indelibly fixed in memory after only a single experience. These and many other characteristics of human memory cannot be explained by the confined and rigid formulas of behaviorism.
Throughout the behaviorist era, some psychologists continued to explore, in broader and deeper terms, not only human memory but a number of the psychological phenomena that behaviorism had ignored, like perception, motivation, personality traits, reasoning, problem solving, creativity, child development, the interplay of hereditary tendencies and experience, and interpersonal relations. Gradually, the new data gathered about these subjects, and the questions those data raised that behaviorism could not answer, prepared the way for what Thomas Kuhn, in his famous analysis of scientific revolutions, called a “paradigm shift”—a relatively abrupt switch to a new theory encompassing and making sense of a large accumulation of data that could be accounted for only with difficulty, if at all, by the reigning theory.75
Meanwhile, research in a number of other fields was beginning to cast new light on the workings of the mind. From anthropology came studies of how preliterate peoples think; from psycholinguistics came accounts of how human beings acquire and use language; and from computer science came a wholly new way to conceive of thinking—as information processing, proceeding step by step like a computer program.
The convergence of all these forces achieved intellectual critical mass during the 1960s, resulting in a knowledge explosion and a new conception of psychology. The former cognitive specialties within psychology regained their status, and cognitive science became the hot new interdisciplinary specialty or, more accurately, aggregation of specialties. It was a mind-based science relying on experimental methods by means of which reasonable inferences could be made about mental processes. By the 1980s, cognitive science studies were going on in the psychology departments of nearly every American university, and a handful of universities had created semi-independent institutes of cognitive science. We’ll look at this more closely later in this history.
With the advent of the paradigm shift, behaviorism rapidly lost its commanding position in psychology and its claim to be a sufficient explanation of all behavior. Gregory Kimble of Duke University sums up the disenchantment of psychologists with behaviorism:
Although the classical theories were formulated and tested in terms of simple learning, behind the scenes there was always the presumption that these theories could be applied to all behavior… [and] that most of the basic laws of learning had already been discovered and all that remained was the minor problem of resolving the few systematic issues that separated the main theorists… [But] by the middle of the century it had become clear that the classic theories of learning were limited in scope and that the stature of our scientific knowledge was pre-Galilean rather than post-Newtonian, as Hull and others had thought.76
Curiously, it was only when behaviorism was in its decline that its off-shoot, behavior therapy, became a widely used and reasonably successful form of treatment for a limited range of psychological disorders.
What is true of behavior therapy—its usefulness but limited applicability—is similar to what has proved true of its parent, behaviorism. A number of its findings have been put to practical use, an example being taste-aversion learning: To inhibit coyotes from killing sheep, researchers put toxic lamb burgers, wrapped in sheep fur, on the perimeters of fenced areas of sheep ranches; coyotes that eat this bait get sick, vomit, and develop an instant aversion to lamb meat and to sheep.77 Other conditioning mechanisms have been used to counter the aversion cancer patients develop to food if mealtime comes just before painful chemotherapy (the simple answer: separate mealtime widely from treatment).78 Behavior modificatio
n methods have been used in programs with mentally retarded children and adults, psychiatric patients, and prisoners: They earn “tokens” for good behavior that can be traded for privileges. Secondary reinforcement methods have proved useful in the workplace in such applications as giving employees bonus vouchers for getting to work on time.79 Phobias of various kinds, including extreme fear of snakes, have been successfully treated by step-by-step conditioning of the phobic person to thoughts of, the sight of, and eventually the handling of, snakes.80
More generally, behaviorism yielded a legacy often taken for granted but considered essential in most areas of psychology: the need for rigorous experimentation and carefully defined variables. Behavior analysis continues to attract some psychologists as a field of research and application; in addition to the 4,500 members of the Association for Behavior Analysis, another 5,000 or more psychologists are members of local chapters of ABA. But their interest in it is apparently more as an adjunct to other areas of research than a primary identification, an indication being the membership of Division 25 of APA, which peaked at some 1,600 members in the early 1970s, then slid steeply downhill to a little over 600 in the last half dozen years, about 7 percent of APA membership.
In any case, the kinds of studies being performed within the field of behavior analysis these days seem strangulated to cognitivists. Here, for instance, are the titles of typical articles in the January 2006 issue of the Journal of the Experimental Analysis of Behavior:
“The influence of prior choices on current choice.”
“Resistance to extinction following variable-interval reinforcement: Reinforcer rate and amount.”
“Second-order schedules of token reinforcement with pigeons: Implications for unit price.”
And here is a brief excerpt exemplifying much of the work being done by contemporary behaviorists:
Rats obtained food-pellet reinforcers by nose poking a lighted key. Experiment 1 examined resistance to extinction following single-schedule training with different variable-interval schedules, ranging from a mean interval of 16 min to 0.25 min. That is, for each schedule, the rats received 24 consecutive daily baseline sessions and then a session of extinction (i.e., no reinforcers). Resistance to extinction (decline in response rate relative to baseline) was negatively related to the rate of reinforcers obtained during baseline, a relation analogous to the partial-reinforcement-extinction effect. A positive relation between these variables emerged, however, when the unit of extinction was taken as the mean interreinforcer interval that had been in effect during training (i.e., as an omitted reinforcer during extinction)…81
It is time for us to move on to something more easily recognizable as psychology.
* William of Ockham, a fourteenth-century Franciscan, is supposed to have said, “Entities are not to be multiplied without necessity,” although some sources claim that what he actually said was “It is vain to do with more what can be done with fewer.” In either case, the message is that the best explanation is the simplest one.
* Observers (of their own conscious processes).
TEN
The
Gestaltists
A Visual Illusion Gives Rise to a New Psychology
On a train speeding through central Germany late in the summer of 1910, a young psychologist named Max Wertheimer stared at the landscape, intrigued by an illusion millions have taken for granted but that he felt, at that moment, required an explanation. Distant telegraph poles, houses, and hilltops, though stationary, seemed to be speeding along with the train. Why?1
The puzzle led him to think about another illusory motion—that produced by the stroboscope, a toy employing the same principle as motion pictures, which were just becoming popular. In both cases, the rapid exposure to the eye of a series of photographs taken at split-second intervals, or drawings showing the smallest of changes, created the impression of continuous motion.
The phenomenon, known for decades, had never been satisfactorily explained. Thomas Edison and others who had developed motion pictures in the 1890s were content to achieve the effect without understanding what caused it. But on the train that day Wertheimer had a sudden intuition about the answer. He had taken his doctorate at Würzburg, where, in defiance of Wundtian principles, a handful of psychologists had been using introspection to explore conscious thinking. It now occurred to him that the illusion of motion might be due to something happening not in the retina, as many psychologists thought, but in the mind, where some higher-level mental process supplied transitions between the successive pictures, thereby creating the perception of movement. He promptly lost interest in the problem of the moving landscape and never returned to it.
At the time, Wertheimer, who had been doing research at the University of Vienna on the inability to read, was on his way to the Rhineland for a vacation. But his idea so excited him that he got off the train at Frankfurt to consult Professor Friedrich Schumann, an expert on perception with whom he had studied at the University of Berlin before going to Würzburg, and who had recently moved to the University of Frankfurt.
In town, Wertheimer bought a stroboscope at a toy store and spent the evening working with it in his hotel room. (A stroboscope is a scientific instrument for seeing moving parts, as in machinery, slowed down or stationary, but in the nineteenth century and early in the twentieth the term referred to a popular toy that created the impression of motion.) The stroboscope came with pictures of a horse and boy; at the right speed of operation the horse appeared to trot and the boy to walk. Wertheimer replaced the pictures with pieces of paper on which he alternately drew lines in two locations, parallel to each other. He found that at one speed of operation he saw first one line and then the other in their different places; at another, both lines side by side; and at yet another, a single line moving from one position to another. He had a historic experiment and a theory of psychology in the making.
The next day Wertheimer called on Schumann at the university, told him what he had observed and what he guessed was the explanation, and asked his opinion. Schumann could cast no light on the matter but offered Wertheimer the use of his laboratory and equipment, including a new tachistoscope of his own design. With it a researcher, by regulating the speed of rotation of a wheel with slits in it, could expose a visual stimulus to the viewer for brief durations, and by using wheels with differently located slits and a prism could present the viewer with alternating images. The tachistoscope did with precision and control what the stroboscope did crudely.
Because Wertheimer would need volunteers to serve as experimental subjects, Schumann introduced him to one of his two assistants, Wolfgang Köhler, who shortly brought in the other assistant, Kurt Koffka.2They were somewhat younger than Wertheimer (he was thirty, Köhler twenty-two, and Koffka twenty-four), but all three were interested in the higher-level mental phenomena that the New Psychology of the physiologists and the followers of Wundt ignored, and they hit it off at once. They were to be friends and co-workers for their entire lives.
Wertheimer, single and possessed of an independent income—his father had been director of a successful commercial school in Prague— could do as he pleased; what he pleased was to abandon his vacation plans and remain in Frankfurt for nearly half a year conducting a series of experiments, with Köhler, Koffka, and Koffka’s wife serving as his subjects.
In his basic experiment, adapted from the hotel room try-out, Wertheimer alternately projected a three-centimeter horizontal line and another one about two centimeters below it. At a low rate of exposure, his volunteers (who did not know until much later what he was doing) all saw first one line and then the other; at a high rate of exposure, both lines simultaneously; and at intermediate rates, a single line smoothly moving from the upper to the lower position and back again.3
In a variation, Wertheimer used a vertical line and a horizontal line. At the right speed his subjects saw one line rotating back and forth through 90 degrees. In another variation he used lights; these, at th
e critical speed, appeared to be a single light moving. In still others he used multiple lines, different colors, and different shapes, and in every case was able to produce the illusion of motion. Even after he told his three subjects what was happening, they could not make themselves not see the motion. Through still other variations, Wertheimer ruled out any possibility that the phenomenon was due to eye movements or retinal afterimages.
The illusion, he concluded, was a “psychic state of affairs,” which he called the φ phenomenon. The letter phi, he said, “designates something that exists outside the perceptions of a and b,” resulting from a “psychological short-circuit” in the brain.4 The φ phenomenon, he suggested, resulted from “a psychological short-circuit” in the brain between the two areas stimulated by the nerve impulses coming from the retinal areas stimulated by a and b.
This physiological hypothesis did not stand up in later research; what did was Wertheimer’s theory that the illusion of motion takes place not at the level of sensation, in the retina, but of perception, in the mind, where incoming discrete sensations are seen as an organized unity with a meaning of its own. Wertheimer called such an overall perception a Gestalt, a German word that means form, shape, or configuration but that he used to mean a set of sensations perceived as a meaningful whole.
Seemingly, he had spent months of work to explain a trivial illusion. In actuality, he and his co-workers had sown the seed of the Gestalt school of psychology, a movement that would enrich and broaden psychology both in Germany and the United States.*
