426 CHAPTER 13 directly against psychology—at least more or less, the tenets of logical positivism and all against the kind peddled by claiming scientific respectability. philosophers.… All sentences purporting Although there were major differences among to deal with psychical states are translatable the neobehaviorists, they all tended to believe the into sentences in the physical language. following: Two distinctly separate languages to de- ■ If theory is used, it must be used in ways de- scribe physics and psychology are therefore manded by logical positivism. not necessary.… It is the Logical Positivist’s way of saying that psychology ■ All theoretical terms must be operationally must be operational and behavioristic. defined. (Stevens, 1951, pp. 39–40) ■ Nonhuman animals should be used as research subjects for two reasons: (1) Relevant variables The “unity of science” movement and physi- are easier to control than they are for human calism went hand in hand: subjects. (2) Perceptual and learning processes How we get from Physicalism to the thesis occurring in nonhuman animals differ only in of the Unity of science is obvious indeed. If degree from those processes in humans; there- every sentence can be translated into the fore, the information gained from nonhuman physical language, then this language is an animals can be generalized to humans. all-inclusive language—a universal lan- ■ The learning process is of prime importance guage of science. And if the esoteric jar- because it is the primary mechanism by which gons of all the separate sciences can, upon organisms adjust to changing environments. demand, be reduced to a single coherent language, then all science possesses a fun- Not all psychologists followed the new ap- damental logical unity. (Stevens, 1951, proach. During the period from about 1930 to p. 40) about 1950, psychoanalysis (see Chapter 16) was becoming increasingly important in U.S. psychol- The science that was proposed as the model for ogy, as was Gestalt psychology (see Chapter 14), this “unified science” was physics. and psychologists embracing these viewpoints saw little need to follow the dictates of logical positiv- ism. With these exceptions and a few others, how- ever, neobehaviorism dominated the period. NEOBEH AVIORISM Edward Tolman was among the first to expand behaviorism by employing the tenets of logical pos- Neobehaviorism resulted when behaviorism was itivism, and it is to his version of neobehaviorism combined with logical positivism: “It is only a slight that we turn next. caricature to represent neobehaviorism as the prod- uct of the remarriage of psychology, in the guise of behaviorism, and philosophy, in the guise of logical positivism” (Toulmin and Leary, 1985, p. 603). EDWARD CHACE TOLMAN Logical positivism made many forms of behaviorism possible: “Objectivism in data collection was one Edward Chace Tolman (1886–1959) was born thing; agreement about specific modes of objectiv- on April 14 in West Newton, Massachusetts, the ism, and about the theoretical implications of ‘ob- son of a businessman who was a member of the first jective’ data, was something else” (Toulmin and graduating class of the Massachusetts Institute of Leary, 1985, p. 603). Thus, as we will see, a number Technology (MIT) and a member of its board of of versions of behaviorism emerged, all following, trustees. Tolman’s father, encouraged by his wife
NEOB EHA VIORIS M 427 Behavior: An Introduction to Comparative Psychology (1914) was used as the text: This worry about introspection is perhaps one reason why my introduction in Yerkes’ courses to Watson behaviorism came as a tremendous stimulus and relief. If objective measurement of behavior and Akron not introspection was the true method of of psychology I didn’t have to worry any University longer. (Tolman, 1952, p. 326) Archives–The work in philosophy and psychology at Harvard; In 1911 Tolman decided to pursue graduate Psychology once enrolled, his interest turned increasingly to psychology. After a year of study, Tolman decided © to improve his German by spending a summer in Germany. While in Germany, Tolman studied with Edward Chace Tolman the young Gestalt psychologist Kurt Koffka (whom we will meet in the next chapter). Although Gestalt psychology did not impress Tolman at the time, it who was raised in the Quaker religion, had a strong greatly influenced his later theorizing. Upon re- interest in social reform. Both sons, Edward and his turning to Harvard, Tolman studied the learning older brother Richard, earned their undergraduate of nonsense material under the supervision of degrees in experimental and theoretical chemistry at Hugo Münsterberg, and his doctoral dissertation MIT. Richard went on to become a prominent was on retroactive inhibition (Tolman, 1917). physicist after earning his doctorate at MIT. After attaining his doctorate from Harvard in Edward’s interests began to turn toward philosophy 1915, Tolman accepted an appointment at and psychology after taking summer school courses Northwestern University. Although he became a from Harvard philosopher Ralph Barton Perry compulsive researcher, he confessed to being (1876–1957) and Harvard psychologist Robert “self-conscious and inarticulate” as a teacher and Yerkes; most influential, however, was his reading frightened of his classes. Also, at about the time of James’s Principles. At this time, psychology was that the United States entered World War I, he dominated by Titchener and James, and psychology wrote an essay expressing his pacifism. In 1918 was still defined as the study of conscious experi- Tolman was dismissed for “lack of teaching suc- ence, a fact that bothered Tolman (1922): cess,” but more than likely his pacifism contributed The definition of psychology as the ex- to his dismissal. From Northwestern he went to the amination and analysis of private conscious University of California at Berkeley, where he re- contents has been something of a logical mained almost without interruption for the rest of sticker. For how can one build up a science his career. As we have seen, Tolman was raised in a upon elements which, by very definition, Quaker home, and pacifism was a constant theme are said to be private and noncommunic- throughout his life. He wrote a short book titled able? (p. 44) Drives Toward War (1942) to explain, from a psy- choanalytic viewpoint, the human motives respon- Tolman’s concern was put to rest in the course sible for warfare. In the preface of that book, he he took from Yerkes, in which J. B. Watson’s stated his reasons for writing it:
428 CHAPTER 13 As an American, a college professor, and especially psychology, are still immersed in one brought up in the pacifist tradition, I such tremendous realms of the uncertain am intensely biased against war. It is for me and the unknown, the best that any indi- stupid, interrupting, unnecessary, and un- vidual scientist, especially any psychologist, imaginably horrible. I write this essay can do seems to be to follow his own within that frame of reference. In short, I gleam and his own bent, however inade- am driven to discuss the psychology of war quate they may be. In fact, I suppose that and its possible abolition because I want actually this is what we all do. In the end, intensely to get rid of it. (p. xi) the only sure criterion is to have fun. And I have had fun. (p. 159) By the time the book came out, however, the United States was already involved in World War Tolman died in Berkeley, California, on II. The brutality of the war overcame even November 19, 1959. Tolman’s strong pacifism, and after receiving the approval of his brother Richard, he served for Purposive Behaviorism two years in the Office of Strategic Services (1944–1945). In the early 1920s, there were two dominant ex- After the war, Tolman’s social conscience was planations of learning: Watson’s explanation in tested once again. In the early 1950s, under the terms of the associative principles of contiguity influence of McCarthyism, the University of and frequency, and Thorndike’s, which emphasized California began to require its faculty members to the law of effect. Tolman (1952) explained why he sign a loyalty oath, and Tolman led a group of fac- could accept neither: ulty members who would rather resign than sign it. It was Watson’s denial of the law of effect They saw the requirement as an infringement of and his emphasis on frequency and recency their civil liberties and academic freedom. Tolman as the prime determiners of animal learning was suspended from his duties at California and which first attracted our attention. In this taught for a while at the University of Chicago we were on Watson’s side. But we got and Harvard University. Finally, the courts agreed ourselves—or at least I got myself—into a with Tolman, and he was reinstated at the sort of in-between position. On the one University of California. In 1959, upon his retire- hand I sided with Watson in not liking the ment and shortly before his death, the regents of law of effect. But, on the other hand, I also the university symbolically admitted that Tolman’s did not like Watson’s over-simplified no- position had been morally correct by awarding him tions of stimulus and response.… an honorary doctorate. According to Thorndike an animal Tolman was a kind, shy, honest person who learned, not because it achieved a wanted inspired affection and admiration from his students goal by a certain series of responses, but and colleagues. Although he was always willing to merely because a quite irrelevant “pleas- engage in intellectual dispute, he never took him- antness” or “unpleasantness” was, so to self or his work too seriously. In the final year of his speak, shot at it, as from a squirt gun, after life, Tolman (1959) reflected on his theoretical it had reached the given goal box. (p. 329) contributions: [My theory] may well not stand up to any Tolman (perhaps incorrectly) referred to final canons of scientific procedure. But I Watson’s psychology as “twitchism” because he do not much care. I have liked to think felt it concentrated on isolated responses to specific about psychology in ways that have proved stimuli. Watson contended that even the most congenial to me. Since all the sciences, and complex human behavior could be explained in
NEOB EHA VIORIS M 429 terms of S–R reflexes. Tolman referred to such re- child hiding from a stranger; a woman doing flexes as molecular behavior. Instead of taking as her washing or gossiping over the tele- his subject matter these “twitches,” Tolman de- phone; a pupil marking a mental-test sheet; cided to study purposive behavior. Although a psychologist reciting a list of nonsense Tolman’s approach differed from Watson’s in sev- syllables; my friend and I telling one another eral important ways, Tolman was still a behaviorist our thoughts and feelings—these are behaviors and was completely opposed to introspection and (Qua Molar). And it must be noted that in metaphysical explanations. In other words, Tolman mentioning no one of them have we re- agreed with Watson that behavior should be psy- ferred to, or, we blush to confess it, for the chology’s subject matter, but Tolman believed that most part even known, what were the exact Watson was focusing on the wrong type of behav- muscles and glands, sensory nerves, and ior. The question was how Tolman could employ a motor nerves involved. For these responses mentalistic term like purpose and still remain a somehow had other sufficiently identifying behaviorist. properties of their own. (p. 8) While at Harvard, Tolman learned from two of his professors, Edwin B. Holt and Ralph Barton Perry, that the purposive aspects of behavior could Tolman’s Use of Rats be studied without sacrificing scientific objectivity. Tolman did not engage in any animal research as a This was done by seeing purpose in the behavior graduate student at Harvard or as an instructor at itself and not inferring purpose from the behavior. Northwestern University. When he arrived at the Tolman accepted this contention and believed that University of California, he was asked to suggest a it pointed to a major distinction between his view new course to teach and, remembering fondly his of purpose and that of McDougall: “The funda- course with Yerkes, chose to teach comparative mental difference between [McDougall] and us psychology. It was teaching this course that stimu- arises in that he, being a ‘mentalist,’ merely infers lated Tolman’s interest in the rat as an experimental purpose from these aspects of behavior; whereas subject. He saw the use of rats as a way of guarding we, being behaviorists, identify purpose with such against even the possibility of indirect introspection aspects” (1925, p. 288). Tolman would later change that could occur if humans were used as experi- his position and use the terms purpose and cognition mental subjects. Tolman developed such a fondness more in accordance with the mentalistic tradition as for rats that he dedicated his Purposive Behavior to actual determinants of behavior. Tolman never be- the white rat, and in 1945 he said, lieved, however, that using concepts like purpose and cognition violated the tenets of behaviorism. Let it be noted that rats live in cages; they (For a discussion of Tolman’s use of mentalistic do not go on binges the night before one terms and how that use changed during his career, has planned an experiment; they do not see L. D. Smith, 1982.) kill each other off in wars; they do not Tolman called purposive behavior molar be- invent engines of destruction, and if they havior to contrast it with molecular behavior. did, they would not be so inept about Because Tolman chose to study molar behavior, controlling such engines; they do not go in his position is often referred to as purposive be- for either class conflicts or race conflicts; haviorism. In his major work, Purposive Behavior in they avoid politics, economics, and papers Animals and Men (1932), Tolman gave examples of on psychology. They are marvelous, pure, what he called purposive (molar) behavior: and delightful. (p. 166) A rat running a maze; a cat getting out of a About what could be learned by studying rats, puzzle box; a man driving home to dinner; a Tolman (1938) said,
430 CHAPTER 13 I believe that everything important in determining behavior (as McDougall believed). In psychology (except perhaps such matters as 1938 he decided how he would proceed: “I, in my the building up of a super-ego, that is, future work, intend to go ahead imagining how, if I everything save such matters as involve were a rat, I would behave” (p. 24). Clearly, Tolman society and words) can be investigated in was now embracing mentalism, and yet he still felt essence through the continued experi- strongly about remaining a behaviorist. For Tolman mental and theoretical analysis of the de- the solution to the dilemma was to treat cognitive terminers of rat behavior at a choice-point events as intervening variables—that is, variables in a maze. Herein I believe I agree with that intervene between environmental events and Professor Hull and also with Professor behavior. Following logical positivism, Tolman Thorndike. (p. 34) painstakingly tied all his intervening variables to ob- servable behavior. In other words, he operationally defined all his theoretical terms. Tolman’s final po- The Use of Intervening Variables sition was to regard purpose and cognition as theo- Tolman was not consistent in using mentalistic con- retical constructs that could be used to describe, cepts as only descriptions of behavior. By 1925 he predict, and explain behavior. was referring to purpose and cognition both as de- By introducing the use of intervening vari- scriptions and determinants of behavior. L. D. ables, Tolman brought abstract scientific theory Smith (1982) noted Tolman’s vacillation: into psychology. It was clear that environmental events influenced behavior; the problem was to Within a single paragraph of Purposive understand why they did. One could remain en- Behavior, Tolman described purposes and tirely descriptive and simply note what organisms cognitions on the one hand as “immanent” do in certain situations, but for Tolman this was in behavior, “in-lying,”“immediate,” and unsatisfactory. Here is a simplified diagram of “discovered” by observers, and on the Tolman’s approach: other hand as “determinants” and “causes” of behavior which are “invented” or “in- Independent Variables ferred” by observers. (p. 162) (Environmental Events) # In the following quotation, Tolman (1928) ap- Intervening Variables peared to believe that purposes were in the organ- (Theoretical Concepts) ism and were causally related to its behavior: # Our doctrine … is that behavior (except in Dependent Variables the case of the simplest reflexes) is not gov- (Behavior) erned by simple one to one stimulus- Thus, for Tolman, environmental experience response connections. It is governed by gives rise to internal, unobservable events, which, more or less complicated sets of patterns of in turn, cause behavior. To account fully for the adjustment which get set up within the or- behavior, one has to know both the environmental ganism. And in so far as these sets of adjust- events and the internal (or intervening) events that ments cause only those acts to persist and to they initiate. The most important intervening vari- get learned which end in getting the or- ables Tolman postulated are cognitive or mental in ganism to (or from) specific ends, these sets nature. Tolman, then, was a methodological rather or adjustments constitute purposes. (p. 526) than a radical behaviorist. What made Tolman a Increasingly, Tolman came to believe that cog- different type of mentalist was his insistence nitive processes really exist and are influential in that his intervening variables, even those that were
NEOB EHA VIORIS M 431 presumed to be mental, be operationally defined— has ever devised (for details, see Hergenhahn and that is, systematically tied to observable events. Olson, 2005). Hypotheses, Expectancies, Beliefs, and Cogni- Tolman’s Position on Reinforcement tive Maps. Although Tolman used several inter- Tolman rejected Watson’s and Thorndike’s expla- vening variables, we will discuss only those related nations of learning. In other words, he did not be- to the development of a cognitive map. Everyone lieve that learning is an automatic process based on knows that a rat learns to solve a maze; the question contiguity and frequency nor that it results from is, How does it do so? Tolman’s explanation was reinforcement (a pleasurable state of affairs). He be- mentalistic. As an example, when an animal is first lieved that learning occurs constantly, with or with- placed in the start box of a T-maze, the experience out reinforcement and with or without motivation. is entirely new, and therefore the animal can use no About as close as Tolman came to a concept of information from prior experience. As the animal reinforcement was confirmation. Through the runs the maze, it sometimes turns right at the confirmation of a hypothesis, expectancy, or be- choice point and sometimes left. Let us say that lief, a cognitive map develops or is maintained. The the experimenter has arranged the situation so that animal learns what leads to what in the environ- turning left is reinforced with food. At some point, ment—that if it does such and such, such and the animal formulates a weak hypothesis that turn- such will follow; or that if it sees one stimulus ing one way leads to food and turning another way (S 1 ), a second stimulus (S 2 ) will follow. Because does not. In the early stages of hypothesis forma- Tolman emphasized the learning of relationships tion, the animal may pause at the choice point as if among stimuli, his position is often called an S–S to “ponder” the alternatives. Tolman referred to theory rather than an S–R theory. this apparent pondering as vicarious trial and er- ror because, instead of behaving overtly in a trial- Learning versus Performance and-error fashion, the animal appears to be engaged in mental trial and error. If the early hypothesis “If I According to Tolman’s theory, an organism learns turn left, I will find food” is confirmed, the animal constantly as it observes its environment. But will develop the expectancy “When I turn left, I whether the organism uses what it learned—and if will find food.” If the expectancy is consistently so, how—is determined by the organism’s motiva- confirmed, the animal will develop the belief tional state. For example, a food-satiated rat might “Every time I turn left in this situation, I will find not leave the start box of a maze or might wander food.” Through this process, a cognitive map of casually through the maze even though it had pre- this situation develops—an awareness of all possibil- viously learned what had to be done to obtain food. ities in a situation—for example: If I leave the start Thus, for Tolman, motivation influences perfor- box, I will find the choice point; if I turn left at the mance but not learning. Tolman defined perfor- choice point, I will find food; if I turn right, I will mance as the translation of learning into behavior. not; and so on. The importance of motivation in Tolman’s theory For Tolman, hypotheses, expectations, beliefs, was due to the influence of Woodworth’s dynamic and finally a cognitive map intervene between ex- psychology. perience and behavior. Rather than just describing an organism’s behavior, these intervening variables Latent Learning. In one of his famous latent were thought to explain it. Tolman was careful, learning experiments, Tolman dramatically dem- however, to test his theoretical assumptions onstrated the distinction between learning and through experimentation. Tolman’s research pro- performance. Tolman and Honzik (1930) ran an ex- gram was one of the most creative any psychologist periment using three groups of rats as subjects.
432 CHAPTER 13 Subjects in group 1 were reinforced with food each Latent Extinction. Tolman explained both the time they correctly traversed a maze. Subjects in acquisition of a response tendency and its extinction group 2 wandered through the maze but were not in terms of changing expectations. In extinction, reinforced if they reached the goal box. Subjects in reinforcement no longer follows a goal response, group 3 were treated like subjects in group 2 until the and an animal’s expectation is modified accord- 11th day, when they began receiving reinforcement ingly. Tolman’s explanation of extinction has in the goal box. Subjects in all three groups were been supported by a number of latent extinction deprived of food before being placed in the maze. experiments (for example, Moltz, 1957; Seward Tolman’s hypothesis was that subjects in all groups and Levy, 1949). In the typical latent extinction were learning the maze as they wandered through it. experiment, one group of animals undergoes nor- If his hypothesis was correct, subjects in group 3 mal extinction, whereby a series of nonreinforced should perform as well as subjects in group 1 from responses gradually leads to extinction. A second the 12th day on. This was because, before the 11th group of animals is passively placed in the empty day, subjects in group 3 had already learned how to goal box a number of times before extinction trials arrive at the goal box, and finding food there on the begin. These experiments consistently found that 11th day had given them an incentive for acting on the second group of animals extinguished the be- this information. As Figure 13-1 shows, the experi- havior much more rapidly than the first. Tolman’s ment supported Tolman’s hypothesis. Learning ap- explanation was that animals in the second group peared to remain latent until the organism had a rea- “come to see” the absence of reinforcement, and son to use it. this influences both their expectations and their performance. 10 8 No Food Reward Average Errors 6 4 Regularly Rewarded 2 No Food Reward Until Day 11 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Days F I G U R E 13.1 The results of the Tolman and Honzik (1930) experiment on latent learning. (Used by permission.)
NEOB EHA VIORIS M 433 Tolman’s Influence interactions.… I would suggest that it was just this sort of difficulty that became L. D. Smith (1982) summarizes Tolman’s impor- tractable with the realization by psycholo- tance as follows: gists in the 1960s that computer programs In adopting and adapting the concepts of are highly suited for expressing complex purpose and cognition … Tolman helped interactions in models of cognitive pro- preserve and shape the tradition of cogni- cessing. If Tolman’s theoretical innovations tive psychology during a time when it was suffered from the limitations of the tech- nearly eclipsed by the ascendancy of clas- nology available in his time, they would sical behaviorism. He was able to do so by seem to suffer no longer. (p. 464) demonstrating that such concepts were compatible with a behaviorism of a more Clearly, Tolman viewed organisms as active sophisticated—clearly non- processors of information, and such a view is very Watsonian—variety. (p. 160) much in accordance with contemporary cognitive psychology. In Chapter 20, we will see much in With regard to Tolman’s use of mentalistic common between Tolman’s theory and both concepts, Innis (1999) says: information-processing psychology and Bandura’s social cognitive theory. Also (although space does Rather than get rid of them, he wanted to not permit discussion of it), Tolman was a pioneer give them objective, operational defini- in the currently popular field of behavior genetics tions. In place of the sterile mathematics (Innis, 1992). Tolman was the first to publish a and empty organisms of his competitors, study on selective breeding for maze-learning abil- Tolman proposed a rich theoretical struc- ity in rats (1924). And it was Tolman’s student, ture in which purpose and cognition Robert C. Tryon, whose name became most asso- played well-defined parts as potentially ciated with selective breeding because of his longi- measurable intervening variables. For him, tudinal study of maze-bright and maze-dull rats. actions were infused with meaning; be- In 1937 Tolman served as the 45th president of havior was goal-directed—that is, moti- the American Psychological Association (APA), and vated and purposive. However, adopting in 1957 he received the APA’s Distinguished this view did not mean that it was impos- Scientific Contribution Award with the following sible to develop mechanistic rules to ac- citation: count for the behavior observed. (p. 115) For the creative and sustained pursuit of a Once Tolman began postulating intervening theoretical integration of the multifaceted variables, his theory became extremely complex. data of psychology, not just its more cir- He postulated several independent variables and cumscribed and amenable aspects; for several intervening variables, and the possible inter- forcing theorizing out of the mechanical actions between the two types of variables were and peripheral into the center of psychol- enormous. Tolman expressed regret over this prac- ogy without the loss of objectivity and tical difficulty. L. D. Smith (1982) believes that discipline; for returning [the human being] Tolman’s theory was proposed before a technology to psychology by insisting upon molar was developed to evaluate it: behavior purposely organized as the unit of Tolman expressed despair over the im- analysis, most explicitly illustrated in his mense practical difficulty of determining purposive-cognitive theory of learning. intervening variables and their (American Psychologist, 1958, p. 155)
434 CHAPTER 13 C LARK L EONARD HU LL Clark Leonard Hull (1884–1952) was born on May 24 near Akron, New York, the son of an uneducated father and quiet mother who wed at the age of 15. It was Hull’s mother who taught Image not available due to copyright restrictions his father to read. Hull’s education in a rural one- room school was often interrupted by necessary chores on the family farm. After passing a teacher’s examination at the age of 17, Hull taught in a one- room school, but after a year of teaching returned to school, where he excelled in science and mathe- matics. While at school, Hull contracted typhoid fever from contaminated food. Although several of Hull’s fellow students died from the outbreak, Although Hull set a career in psychology as his he survived but, in Hull’s opinion, with his mem- goal, he was not financially able to pursue it. Instead, ory impaired. After his recuperation, he went to he became principal of the school he had attended as Alma College in Michigan to study mining engi- a child (which had expanded to two rooms). In his neering. Following his training, he obtained a job at spare time, he read James’s Principles to prepare a mining company in Minnesota, where his job was himself for his chosen profession. After two years, to evaluate the manganese content in iron ore. he had saved enough money to enter the University After only two months on the job, at the age of of Michigan as a junior. Among the courses that Hull 24, he contracted poliomyelitis, which left him par- took at Michigan was one in experimental psychol- tially paralyzed. At first he could walk only with ogy, which he loved, and one in logic, for which he crutches; for the rest of his life he used a cane. constructed a machine that could simulate syllogistic He needed to ponder a career that was less strenu- reasoning. After graduation from the University of ous than mining. Hull first considered becom- Michigan, Hull’s funds were again exhausted, and he ing a Unitarian minister. He was attracted to accepted a position in a school of education in Unitarianism because it was “a free, Godless reli- Kentucky. During this time, although not yet in gion,” but the idea of “attending an endless succes- graduate school, he began planning what would sion of ladies’ teas” caused him to abandon the idea. become his doctoral dissertation on concept forma- What he really wanted was to work in a field where tion. Hull applied for graduate study at Cornell and success could come relatively quickly and that Yale (where he ultimately would spend most of his would permit him to tinker with apparatus: professional career) and was rejected by both. He was, however, accepted at the University of [I wanted] an occupation in a field allied to Wisconsin. It took four years for Hull to complete philosophy in the sense of involving the- his dissertation on concept learning (1920). Although ory: one which was new enough to permit Hull believed that his research represented a break- rapid growth so that a young man would through in experimental psychology, it was essen- not need to wait for his predecessors to die tially ignored. Hilgard (1987) reminisced on Hull’s before his work could find recognition, experiences with his dissertation: and one which would provide an oppor- tunity to design and work with automatic Hull had struggled hard to complete his apparatus. Psychology seemed to satisfy dissertation, undergoing the trials of a baby this unique set of requirements. (Hull, daughter smearing the ink on charts he had 1952a, p. 145) so carefully laid out to dry, so that he had
NEOB EHA VIORIS M 435 to do them all over again. He felt proud of viewed the universe as a huge machine that could his dissertation because it moved experi- be described in precise mathematical terms. Hull mental psychology into the area of thought simply applied the Newtonian model to living or- processes by investigating the learning of ganisms. Another of Hull’s heroes was Pavlov. Hull concepts.… He told me how downcast he was deeply impressed by the English translation of had become when year after year no one Pavlov’s work that appeared in 1927. He began paid attention to it or cited it. He was fi- studying conditioned responses in humans while nally prepared to accept the fact that it had he was still at Wisconsin and continued his studies been “still-born” (his words). (p. 200) when he moved to Yale. At Yale, however, his experimental subjects were rats instead of humans. Hull received his doctorate from the University Hull’s many contributions were finally recog- of Wisconsin in 1918 and remained there as an nized when, in 1936, he served as 44th president of instructor until 1929. the APA. In his presidential address, he outlined his Perhaps disappointed over the reception of his goal of creating a theoretical psychology that would dissertation research on concept learning, Hull explain “purposive” behavior in terms of mechanis- moved into other research areas. For example, he tic, lawful principles. In creating his theoretical psy- accepted a research grant to study the influence of chology, Hull would employ the tenets of logical pipe smoking on mental and motor performance. positivism (and euclidean geometry) in that new Next, Hull was asked to teach a course in psycho- knowledge is deduced from what is already known. logical tests and measurements. He observed that In his autobiography, Hull said, “The study of ge- the existing bases for vocational guidance were ometry proved to be the most important event of not objective, and his efforts to improve the situa- my intellectual life; it opened to me an entirely new tion ultimately resulted in his book Aptitude Testing world—the fact that thought itself could generate (1928). As part of his work in this area, Hull in- and really prove new relationships from previously vented a machine that could automatically compute possessed elements” (1952a, p. 144). It is important intercorrelations among test scores. This machine, to note that neither Hull nor Tolman developed which was programmed by punching holes in a the theories they did because of logical positivism. tape, is now housed in the Smithsonian Institution Both reached their conclusions about theoretical in Washington, DC (Hilgard, 1987). In addition to psychology independently of logical positivism; his contributions to concept learning and aptitude when they discovered that philosophy of science testing, Hull also pursued his interests in suggestibil- in the 1930s, they simply assimilated its terminology ity and hypnosis while at the University of into their systems. In other words, Tolman and Wisconsin. Over about a 10-year period, Hull Hull used the language of logical positivism to ex- and his students published 32 papers on these topics. press their own ideas. They could do so because of This work culminated in Hull’s Hypnosis and the compatibility between the two. Suggestibility: An Experimental Approach (1933). Unlike Tolman, Hull found no need for men- In 1929 Hull accepted a professorship at Yale talistic concepts, whether they were considered University (one of the institutions that rejected his real entities or simply theoretical conveniences. graduate school application). At Yale, Hull pursued Like Watson, Hull believed that psychology’s pre- two interests: the creation of machines that could occupation with consciousness was derived from learn and think (like his correlation machine) and medieval metaphysics and theology. Although the study of the learning process. The two interests Hull’s interest in “psychic machines” was now sec- were entirely compatible because Hull viewed hu- ondary, he did demonstrate such a machine to his mans as machines that learn and think. Not surpris- APA audience, and he expressed the belief that if a ingly, one of Hull’s heroes was Newton, who machine could be built that performed adaptive
436 CHAPTER 13 behaviors, it would support his contention that the deductive theory, which he hoped would be self- adaptive behaviors of living organisms could be ex- correcting. Hull first reviewed the research that had plained in terms of mechanistic principles. been done on learning; then he summarized that Because of their willingness to speculate about research in the form of general statements, or pos- internal causes of behavior, both Hull and Tolman tulates. From these postulates, he inferred theorems were methodological behaviorists, and both eventu- that yielded testable propositions. Hull (1943) ex- ally employed logical positivism in their theorizing. plained why his system should be self-correcting: Philosophically, however, Hull was a mechanist and Empirical observation, supplemented by a materialist, and Tolman was, insofar as he believed shrewd conjecture, is the main source of mental events determined behavior, a dualist. the primary principles or postulates of a Supporters of Hull’s mechanistic behaviorism and science. Such formulations, when taken those of Tolman’s purposive behaviorism battled in various combinations together with with each other throughout the 1930s and 1940s. relevant antecedent conditions, yield in- This running debate resulted in one of the most pro- ferences or theorems, of which some ductive periods in psychology’s history. may agree with the empirical outcome of Between 1929 and 1950, Hull wrote 21 theo- the conditions in question, and some retical articles in the Psychological Review, and in may not. Primary propositions yielding 1940 he (with co-authors Hovland, Ross, Hall, logical deductions which consistently Perkins, and Fitch) published Mathematico-Deductive agree with the observed empirical out- Theory of Rote Learning. This book was an effort to come are retained, whereas those which show how rote learning could be explained in disagree are rejected or modified. As the terms of conditioning principles. In 1943 Hull pub- sifting of this trial-and-error process lished Principles of Behavior, one of the most influen- continues, there gradually emerges a tial books in psychology’s history; and A Behavior limited series of primary principles whose System (1952b) extended the ideas found in joint implications are progressively more Principles to more complex phenomena. In 1948, likely to agree with relevant observations. while preparing the manuscript for A Behavior Deductions made from these surviving System, Hull suffered a massive heart attack that ex- postulates, while never absolutely certain, acerbated his already frail physical condition. It took do at length become highly trustworthy. all the strength he could muster, but he finished the This is in fact the present status of the book four months before he died, on May 10, primary principles of the major physical 1952, of a second heart attack. Near his death, sciences. (p. 382) Hull expressed profound regret that a third book he had been planning would never be written. Whereas Watson believed that all behavior He believed that his third book would have been could be explained in terms of the associations be- his most important because it would have extended tween stimuli and responses, Hull concluded that a his system to human social behavior. number of intervening internal conditions had to be taken into consideration. Tolman had reached Hull’s Hypothetico-Deductive Theory the same conclusion. However, for Tolman, cogni- Hull borrowed the technique of using intervening tive events intervened between environmental ex- variables from Tolman, but he used them even perience and behavior; for Hull, the intervening more extensively than Tolman did. Hull was the events were primarily physiological. first (and last) psychologist to attempt to apply a In Hull’s final statement of his theory (1952b), he comprehensive, scientific theory to the study of listed 17 postulates and 133 theorems, but we review learning, creating a highly complex hypothetico- only a few of his more important concepts here.
NEOB EHA VIORIS M 437 Reinforcement definitions, Hull attempted to show how a number of internal events interact to cause overt behavior. Unlike Watson and Tolman, Hull was a reinforce- Hull’s theory is in the Darwinian tradition because it ment theorist. For Hull a biological need creates a associates reinforcement with those events that are drive in the organism, and the diminution of this conducivetoanorganism’s survival. His theory re- drive constitutes reinforcement. Thus, Hull had a flected the influence of Darwin, Woodworth, drive-reduction theory of reinforcement. For Watson, and logical positivism. Hull drive is one of the important events that inter- venes between a stimulus and a response. Hull’s Influence Habit Strength. If a response made in a certain Within 10 years of the publication of Principles of situation leads to drive reduction, habit strength Behavior (1943), 40% of all experimental studies in ( S H R ) is said to increase. Hull operationally defined the highly regarded Journal of Experimental Psychology habit strength, an intervening variable, as the num- and Journal of Comparative and Physiological Psychology ber of reinforced pairings between an environmen- referred to some aspect of Hull’s theory. The figure tal situation (S) and a response (R). For Hull an increases to 70% when only the fields of learning increase in habit strength constitutes learning. and motivation are considered (Spence, 1952). Hull’s influence went beyond these areas, however; Reaction Potential during the period between 1949 and 1952, there were 105 references to Hull’s Principles of Behavior Drive is not only a necessary condition for rein- in the Journal of Abnormal and Social Psychology, com- forcement but also an important energizer of be- pared to only 25 for the next most commonly cited havior. Hull called the probability of a learned work (Ruja, 1956). response reaction potential (SER), which is a In 1945 Hull was awarded the prestigious function of both the amount of drive (D) present Warren Medal by the Society of Experimental and the number of times the response had been Psychologists. It carried this inscription: previously reinforced in the situation. Hull ex- pressed this relationship as follows: To Clark L. Hull: For his careful devel- opment of a systematic theory of behavior. SE R ¼ SH R × D This theory has stimulated much research and it has been developed in a precise and If either S H R or D is zero, the probability of a quantitative form so as to permit predic- learned response being made is also zero. tions which can be tested empirically. The Hull postulated several other intervening vari- theory thus contains within itself the seeds ables, some of which contributed to S E R and some of its own ultimate verification and of its of which diminished it. The probability of a learned own possible final disproof. A truly unique response is the net effect of all these positive and achievement in the history of psychology negative influences, each intervening variable being to date. (Kendler, 1987, p. 305) carefully operationally defined. (For a more detailed account of Hull’s theory, see Bower and Hilgard, After Hull’s death in 1952, one of his former stu- 1981; Hergenhahn and Olson, 2005.) dents, Kenneth W. Spence (1907–1967), became the major spokesman for his theory (see Spence, 1956, 1960). The extensions and modifications Spence Hull’s Theory in General made in Hull’s theory were so substantial that the the- Hull’s theory can be seen as an elaboration of ory became known as the Hull–Spence theory. So Woodworth’sS–O–R concept. Using operational successful was Spence in perpetuating Hullian theory
438 CHAPTER 13 that a study showed that as late as the 1960s, Spence was the most cited psychologist in experimental psy- chology journals, with Hull himself in eighth place (Myers, 1970). Although Hull’s theory eventually won its bat- tle with Tolman’s and was extremely popular in the 1940s and 1950s—and under Spence’s influence, even into the 1960s—it is now generally thought of as having mainly historical value. Hull attempted Akron to create a general behavior theory that all social of sciences could use to explain human behavior, University and his program fit all the requirements of logical positivism (for example, all his theoretical concepts were operationally defined). However, although Archives–The Hull’s theory was scientifically respectable, it was relatively sterile. More and more, the testable de- Psychology ductions from his theory were criticized for being © of little value in explaining behavior beyond the laboratory. Psychologists began to feel hampered Edwin Ray Guthrie by the need to define their concepts operationally and to relate the outcomes of their experiments to a Guthrie graduated from the University of theory such as Hull’s. They realized that objective Nebraska in 1907 with a BA in mathematics and inquiry could take many forms and that the form a Phi Beta Kappa key. After graduation he taught suggested by logical positivism had led to a dead mathematics at a Lincoln high school while working end. In many ways, Hull’s approach was ultimately toward an MA in philosophy at the University of as unproductive as Titchener’s had been. Nebraska. He obtained his MA in 1910. That same year, Guthrie began work on his PhD at the University of Pennsylvania and, after obtaining it, EDWIN R AY GUTHRIE returned in 1912 to teaching high school mathemat- ics. In 1914 he accepted a position as instructor of philosophy at the University of Washington. In 1919 Edwin Ray Guthrie (1886–1959) was born on he became a member of the psychology department January 9 in Lincoln, Nebraska, the first of five at the University of Washington, where he remained children. His father owned a piano shop in until accepting the position of dean of the graduate Lincoln, where he also sold bicycles and furniture. school in 1943. In 1951 Guthrie attained emeritus His mother had been a school teacher before her status but continued to teach and involve himself in marriage. According to Guthrie’s son Peter, his fa- university affairs until his retirement in 1956. ther showed early academic promise: Guthrie’s basic work, The Psychology of Learning, He and a friend read Darwin’s Origin of was published in 1935 and revised in 1952. His writ- Species and The Expression of the Emotions in ing was nontechnical, humorous, and filled with nu- Man and Animals while they were in the merous homespun anecdotes. He believed strongly 8th grade. Edwin studied Greek and Latin that any scientific theory, including his own, should along with his other subjects and read be presented in such a way that it could be under- Xenophon in Greek. (Prenzel-Guthrie, stood by college undergraduates. He also placed 1996, p. 138) great emphasis on the practical application of his ideas. Although he had an experimental outlook
NEOB EHA VIORIS M 439 and orientation, he, along with George P. Horton, frequency of exposure was necessary. What made performed only one experiment related to his theory Guthrie’s theory of learning unique was his rejec- (discussed shortly). Guthrie was clearly a behaviorist, tion of the law of frequency, saying instead that “a but he argued with other behaviorists (such as stimulus pattern gains its full associative strength on Watson, Tolman, Hull, and Skinner), saying their the occasion of its first paring with a response” theories were unparsimonious and too subjective. (1942, p. 30). In other words, unlike any learning As we will see, Guthrie believed all learning phe- theorist before him, Guthrie postulated one-trial nomena could be explained by using only one learning. As Guthrie was aware, Aristotle had ob- of Aristotle’s laws of association—the law of served that learning can result from one experience. contiguity. Aristotle said, It is a fact that there are some movements, The One Law of Learning by a single experience of which persons take the impress of custom more deeply Guthrie’s one law of learning was the law of con- than they do by experiencing others many tiguity, which he stated as follows: “A combina- times; hence upon seeing some things but tion of stimuli which has accompanied a movement once we remember them better than will on its recurrence tend to be followed by that others which we may have seen fre- movement. Note that nothing is here said about quently. (Barnes, 1984, Vol. 1, p. 717) ‘confirmatory waves’ or reinforcement or pleasant effects” (1952, p. 23). In other words, according to However, Aristotle believed such learning to Guthrie, what you do last in a situation is what you be the exception and learning governed by the will tend to do if the situation recurs. Thus, Guthrie law of frequency to be the rule. accepted Watson’s recency principle. In his last publication before his death, Guthrie Why Practice Improves Performance (1959) revised his law of contiguity to read, “What is being noticed becomes a signal for what is being If learning occurs in one trial, why does practice done” (p. 186). This was Guthrie’s way of recog- appear to improve performance? To answer this nizing that an organism is confronted with so many question, Guthrie distinguished between acts and stimuli at any given time that it cannot possibly movements. A movement is a specific response form associations with all of them. Rather, the or- made to a specific configuration of stimuli. It is ganism responds selectively to only a small propor- this association that is learned at full strength after tion of the stimuli present, and it is that proportion one exposure. An act is a response made to varying that becomes associated with whatever response is stimulus configurations. For example, typing the made. letter “a” on a specific typewriter under specific stimulus conditions (such as under certain lighting and temperature conditions, and in a specific One-Trial Learning bodily position) is a movement. However, typing Learning theorists prior to Guthrie accepted “a” under varying conditions is an act. It is because Aristotle’s law of contiguity and his law of fre- learning an act involves learning a specific response quency. For example, Pavlov, Watson, Tolman, under varying conditions that practice improves Hull, and (as we will see later in this chapter) performance. Skinner theorized that associative strength increases Just as an act consists of many movements, a as a function of increased exposure to the learning skill consists of many acts. Thus, a skill such as typ- environment. Of course, they disagreed in their ex- ing, playing golf, or driving a car consists of many planation as to why an increase in associa- acts that, in turn, consist of thousands of move- tive strength took place, but they all agreed that ments. For example, the skill of playing golf consists
440 CHAPTER 13 of the acts of driving, putting, playing out of sand pre-escape conditions and the animal’s characteristic traps, and the like. Again, it is the fact that acts and response to those conditions. Guthrie’s claim that skills require the learning of so many S–R associa- reinforcement is merely a mechanical arrangement tions that their performance improves with practice. that prevents unlearning was confirmed. The Nature of Reinforcement Forgetting According to Thorndike, cats gradually became According to Guthrie, not only does learning occur more proficient at escaping from a puzzle box be- in one trial but so does forgetting. Forgetting occurs cause each time they did so they experienced a when an old S–R association is displaced by a new “satisfying state of affairs” (reinforcement). Guthrie one. Thus, for Guthrie, all forgetting involves new rejected this idea. Guthrie explained the effects of learning. Forgetting occurs only if an existing S–R “reinforcement” in terms of the recency principle. association is interfered with in some way. Guthrie He noted that when a cat in a puzzle box made a explained, response that allowed it to escape (moving a pole, for example), the entire stimulus configuration in The child who has left school at the end of the puzzle box changed. Thus we have one set of the seventh grade will recall many of the stimuli existing before the pole is moved and an- details of his last year for the rest of his life. other after it is moved. According to Guthrie, be- The child who has continued on in school cause moving the pole is the last thing the cat does has these associations of the schoolroom under the prereinforcement conditions, it is that and school life overlaid by others, and by response the cat will make when next placed in the time he is in college may be vary vague the puzzle box. For Guthrie “reinforcement” about the names and events of his seventh- changes the stimulating conditions thereby prevent- grade experience. ing unlearning. In other words, “reinforcement” When we are somehow protected preserves the association that preceded it. from established cues we are well aware The only systematic research ever performed that these may retain their connection with by Guthrie was done with Horton and was summa- a response indefinitely. A university faculty rized in a small book titled Cats in a Puzzle Box member’s wife recently visited Norway, (Guthrie and Horton, 1946). Guthrie and Horton the original home of her parents. She had observed approximately 800 escape responses by not spoken Norwegian since the death of cats in an apparatus similar to that used by her grandmother when she was five and Thorndike. Like Thorndike, Guthrie and Horton believed that she had forgotten the lan- observed that cats learned to move a pole to escape guage. But during her stay in Norway, she the apparatus. However, it was observed that each astonished herself by joining in the con- cat learned to move the pole in its own unique versation. The language and atmosphere of way. For example, one cat would hit the pole by her childhood revived words and phrases backing into it, another would push it with its head, she could not remember in her American and another would move it with its paws. This home. But her conversation caused much stereotyped behavior would be repeated by each cat amusement among her relatives because when it was replaced into the apparatus. This, of she was speaking with a facile Norwegian course, supported Guthrie’s claim that whatever “baby talk.” If her family in America had an animal does last in a situation will be repeated continued to use Norwegian, this “baby when the situation recurs (the recency principle). talk” would have been forgotten, its asso- Moving the pole changes the stimulating condi- ciation with the language destroyed by tions, thus preserving the association between the other phrases.
NEOB EHA VIORIS M 441 Forgetting is not a passive fading of the maintaining stimuli, that act becomes associated stimulus-response associations contingent with the maintaining stimuli. That is, because of upon the lapse of time, but requires active the recency principle, the last act performed in the unlearning, which consists in learning to presence of the maintaining stimuli will tend to be do something else under the circum- performed when those stimuli recur. Such acts stances. (1942, pp. 29–30) are referred to as intentions because they appear to have as their goal the removal of maintaining stimuli (drives). In fact, however, “intentional” be- Breaking Habits havior is explained by Guthrie as any other kind of behavior—that is, by the law of contiguity. A habit is an act that has become associated with a In 1945 Guthrie was elected president of the large number of stimuli. The more stimuli that elicit APA, and in the same year, his alma mater the the act, the stronger is the habit. Smoking, for ex- University of Nebraska awarded him an honorary ample, can be a strong habit because the act of doctorate. In 1958 the American Psychological smoking has become associated with so many sti- Foundation (APF) awarded Guthrie its gold medal muli. According to Guthrie, there is one general for distinguished contributions to the science of rule for breaking undesirable habits: Observe the psychology. Shortly thereafter, the University of stimuli that elicit the undesirable act and perform Washington named its new psychology building another act in the presence of those stimuli. Once Edwin Ray Guthrie Hall. Guthrie died of a heart this is done, the new, desirable act will be elicited attack in April 1959. by those stimuli instead of the old, undesirable act. The Formalization of Guthrie’s Theory Punishment Guthrie often presented his theory in terms too For Guthrie the effectiveness of punishment is de- general to be tested experimentally. An effort to termined not by the pain it causes but by what it make Guthrie’s theory more scientifically rigorous causes the organism to do it the presence of stimuli was made by Virginia W. Voeks (1921–1989), who that elicit undesirable behavior. If punishment eli- studied at the University of Washington when cits behavior incompatible with the undesirable be- Guthrie was influential there. After receiving her havior in the presence of these stimuli, it will be BA from the University of Washington in 1943, effective. If not, it will be ineffective. For example, Voeks went to Yale, where she was influenced by in attempting to discourage a dog from chasing cars, Hull. She obtained her PhD from Yale in 1947. In hitting it on the nose while it is chasing is likely to 1949 Voeks moved to San Diego State College, be effective. On the other hand, hitting it on its rear where she remained until her retirement in 1971. is likely to be ineffective, or perhaps even Voeks’s formalization of Guthrie’s theory strengthen the tendency to chase. In both cases, it (1950) consisted of four basic postulates, eight defi- can be assumed the amount of pain involved is the nitions, and eight theorems (testable deductions). same. Voeks tested a number of her deductions and found considerable support for Guthrie’s theory (see, for example, Voeks, 1954). Drives and Intentions Another attempt to formalize Guthrie’s theory For Guthrie, drives provide maintaining stimuli was made by William Kaye Estes (b. 1919). Early in that keep an organism active until a goal is reached. his career, Estes performed significant research on Maintaining stimuli can be internal (for example, the effects of punishment (1944). However, it is for hunger) or external (for example, a loud noise). his development of stimulus sampling theory (SST) When an organism performs an act that terminates that Estes is best known (1950, 1960, 1964). The
442 CHAPTER 13 cornerstone of SST was Guthrie’s law of contiguity with its assumption of one-trial learning. Estes’s SST showed that Guthrie’s theory, while appearing to be simple, was actually very complex. The model that Estes created (SST) effectively dealt with that complexity and launched a highly heuris- tic research program. Estes modified his theory, making it more compatible with cognitive psychol- ogy (see, for example, Estes, 1994). Even through its various revisions, however, Guthrie’s law of con- tiguity has remained at the core of Estes’s theoriz- ing. For an overview of Estes’s SST and its revisions through the years, see Hergenhahn and Olson, 2005. B. F. SKINNER © Bettmann/CORBIS As the complex theoretical systems of Tolman and Hull began to lose their popularity and Guthrie’s B. F. Skinner theory survived primarily through Estes’s relatively esoteric effort to create a mathematical model of water because I had used a bad word. My learning, another form of behaviorism was in its father never missed an opportunity, how- ascendancy. The version of behaviorism promoted ever, to inform me of the punishments by B. F. Skinner was contrary to logical positivism which were waiting if I turned out to have because it was antitheoretical, and yet it was in ac- a criminal mind. He once took me cordance with logical positivism because it insisted through the county jail, and on a summer that all its basic terms be operationally defined. As vacation I was taken to a lecture with we will see, Skinner’s version of behaviorism was colored slides describing life in Sing Sing. more in accordance with positivism than with logi- As a result I am afraid of the police and buy cal positivism. After World War II, Skinner’s ver- too many tickets to their annual dance. sion of behaviorism essentially displaced all other (Skinner, 1967, pp. 390–391) versions. In high school, Skinner did well in literature Burrhus Frederic Skinner (1904–1990) was born on March 20 in Susquehanna, Pennsylvania, but poorly in science, and he earned money by into a warm, stable, middle-class family. Skinner playing in a jazz band and with an orchestra. He went to Hamilton College, a small liberal arts had a younger brother who was a better athlete and more socially popular than he was but who school in Clinton, New York, where he majored died suddenly at the age of 16. Skinner was raised in English. Skinner did not fit well into college life, was terrible at sports, and felt “pushed around” by according to strict moral standards but was physi- cally punished only once: requirements such as daily chapel. By his senior year, Skinner viewed himself as “in open revolt” I was never physically punished by my fa- against the school. He, along with a friend, decided ther and only once by my mother. She to play a trick on their English composition profes- washed my mouth out with soap and sor, whom they disliked because he was “a great
NEOB EHA VIORIS M 443 name-dropper.” Skinner and his friend had posters humorous column of a local paper but wrote almost printed that read: “Charles Chaplin, the famous nothing else, and thought about seeing a psychia- cinema comedian, will deliver his lecture ‘Moving trist” (Skinner, 1967, p. 394). Next, Skinner tried Pictures as a Career’ in the Hamilton College writing in New York City’s Greenwich Village and chapel on Friday, October 9” (Skinner, 1967, p. then in Paris for a summer; these attempts also 393). The Chaplin visit was said to be under the failed. By this time, Skinner (1967) had developed auspices of the disliked English professor. The pos- a distaste for most literary pursuits: “I had failed as a ters were displayed all over town, and Skinner’s writer because I had had nothing important to say, friend called the newspaper in Utica with the but I could not accept that explanation. It was lit- news. By noon the prank was completely out of erature which must be at fault” (p. 395). hand. Police roadblocks were necessary to control Having failed to describe human behavior the crowds. The next day, the English professor to through literature, Skinner decided to describe it whom the hoax was directed wrote an editorial scientifically. While in Greenwich Village, Skinner lambasting the entire episode. Skinner said that it had read the works of Pavlov and Watson and was was the best thing the professor ever wrote. The greatly impressed. On his return from Europe in Chaplin prank was only the beginning of a mischie- 1928, he enrolled in the graduate program in psy- vous senior year for Skinner: chology at Harvard. Feeling that he at last found his niche, Skinner threw himself completely into his As a nihilistic gesture, the hoax was only studies: the beginning. Through the student pub- lications we began to attack the faculty and I would rise at six, study until breakfast, go various local sacred cows. I published a to classes, laboratories, and libraries with parody of the bumbling manner in which no more than fifteen minutes unscheduled the professor of public speaking would during the day, study until exactly nine review student performances at the end of o’clock at night and go to bed. I saw no the class. I wrote an editorial attacking Phi movies or plays, seldom went to concerts, Beta Kappa. At commencement … I had scarcely any dates and read nothing covered the walls with bitter caricatures of but psychology and physiology. (Skinner, the faculty … and we [Skinner and his 1967, p. 398) friends] made a shambles of the com- mencement ceremonies, and at intermis- This high degree of self-discipline typified sion the President warned us sternly that Skinner’s work habits throughout his long life. we would not get our degrees if we did Skinner earned his master’s degree in two years not settle down. (Skinner, 1967, p. 393) (1930) and his doctorate in three (1931) and then remained at Harvard for the next five years as a post- Skinner graduated from Hamilton College doctoral fellow. Skinner began his teaching career at with a bachelor’s degree in English literature and the University of Minnesota in 1937 and remained a Phi Beta Kappa key and without having had a there until 1945. While he was at Minnesota, course in psychology. He left college with a passion Skinner published The Behavior of Organisms (1938), to become a writer. This passion was encouraged in which established him as a nationally prominent ex- part by the fact that the famous poet Robert Frost perimental psychologist. In 1945 Skinner moved to favorably reviewed three of his short stories. Indiana University as chairman of the psychology Skinner’s first attempt at writing was in the attic department, where he remained until 1948 when of his parents’ home: “The results were disastrous. he returned to Harvard. He remained affiliated I frittered away my time. I read aimlessly … listened with Harvard until his death in 1990. In 1974 he to the newly invented radio, contributed to the became professor emeritus
444 CHAPTER 13 but continued for years to walk the two are related to each other. According to Mach, the miles between his home and his office in scientist determines how facts are related by doing a William James Hall to answer correspon- functional analysis. That is, by noting that if X oc- dence, to meet with scholars who paid him curs, Y also tends to occur. To ponder why such visits from around the world, and on oc- relationships exist is to enter the dangerous and un- casion to conduct research and supervise necessary realm of metaphysics. The job of science graduate students. (Fowler, 1990, p. 1203) is to describe empirical relationships, not explain them. Skinner followed Mach’s positivism explic- In addition to the short autobiography Skinner itly. By adopting Mach’s functional approach to wrote in 1967, he described the details of his life in science, Skinner (1931/1972) avoided the complex three more extensive volumes: Particulars of My Life problem of establishing causation in human (1976), The Shaping of a Behaviorist (1979), and A behavior: Matter of Consequences (1983). We may now take the more humble view of explanation and causation which seems Skinner’s Positivism to have been first suggested by Mach and is now a common characteristic of scientific In Chapter 4, we discussed the great Renaissance thought, wherein … the notion of func- thinker Francis Bacon. Bacon was intensely inter- tion [is] substituted for that of causation. ested in overcoming the mistakes of the past and (pp. 448–449) thus arriving at knowledge that was free of super- stition and prejudice. His solution to the problem As far as theory is concerned, Skinner was a was to stay very close to what was empirically ob- positivist, not a logical positivist. We examine servable and to avoid theorizing about it. Bacon Skinner’s positivism again when we review his atti- proposed that science be descriptive and inductive tude toward theory. rather than theoretical and deductive. Following Bacon’s suggestion, scientists would first gather em- Functional Analysis of Behavior pirical facts and then infer knowledge from those facts (instead of first developing abstract theories Like Watson, Skinner denied the existence of a from which facts are deduced). Bacon’s main point separate realm of conscious events. He believed was that in the formulation of theories, a scientist’s that what we call mental events are simply verbal biases, misconceptions, traditions, and beliefs (per- labels given to certain bodily processes: “[My] posi- haps false beliefs) could manifest themselves and tion can be stated as follows: What is felt or intro- that these very things inhibited a search for objec- spectively observed is not some nonphysical world tive knowledge. Skinner was deeply impressed by of consciousness, mind or mental life but the ob- Bacon and often referred to his influence on his life server’s own body” (Skinner, 1974, p. 17). But, said and work (L. D. Smith, 1992). Bacon can be seen as Skinner, even if there were mental events, nothing starting the positivistic tradition that was later fol- would be gained by studying them. He reasoned lowed by Comte and Mach. As he did with Bacon, that if environmental events give rise to conscious Skinner often acknowledged a debt to Mach (see, events, which, in turn, cause behavior, nothing is for example, Skinner, 1931/1972, 1979). For lost and a great deal is gained by simply doing a Mach, as we have previously noted, it was impor- functional analysis of the environmental and the tant that science rid itself of metaphysical concepts, behavioral events. Such an analysis avoids the many which, for him, were any concepts that refer to problems associated with the study of mental events that cannot be directly observed (causation events. These so-called mental events, said is such a concept). Mach and the other positi- Skinner, will someday be explained when we vists were interested only in facts and how facts learn which internal physiological events people
NEOB EHA VIORIS M 445 are responding to when they use such terms as called such behavior respondent behavior be- thinking, choosing, and willing to explain their own cause it was elicited by a known stimulus. Because behavior. Skinner, then, was a physical monist (ma- both Pavlov and Watson studied the relationship terialist) because he believed that consciousness as a between environmental stimuli (S) and responses nonphysical entity does not exist. Because we do (R), their endeavors represent S–R psychology. not at present know to which internal events peo- Thorndike, however, studied behavior that is con- ple are responding when they use mentalistic ter- trolled by its consequences. For example, behavior minology, we must be content simply to ignore that had been instrumental in allowing an animal to such terms. Skinner (1974) said, escape from a puzzle box tends to be repeated when the animal is next placed in the puzzle box. Using There is nothing in a science of behavior Thorndike’s experimental arrangement, a response or its philosophy which need alter feelings was instrumental in producing certain conse- or introspective observations. The bodily quences, and therefore the type of learning that states which are felt or observed are ac- he studied was called instrumental conditioning. knowledged, but there is an emphasis on Thorndike neither knew nor cared about the ori- the environmental conditions with which gins of the behavior that is controlled by its con- they are associated and an insistence that it sequences. What Thorndike called instrumental be- is the conditions rather than the feelings havior, Skinner called operant behavior because it which enable us to explain behavior. operates on the environment in such a way as to (p. 245) produce consequences. Unlike respondent behav- Skinner (1974) also said, “A completely inde- ior, which is elicited by known stimulation, operant pendent science of subjective experience would behavior is simply emitted by the organism. It is not have no more bearing on a science of behavior that operant behavior is not caused but that its than a science of what people feel about fire would causes are not known—nor is it important to have on the science of combustion” (pp. 220–221), know them. The most important aspect of operant and “There is no place in the scientific position for behavior is that it is controlled by its consequences a self as a true originator or initiator of action” (p. and not elicited by known stimulation. Skinner’s 225). Like Watson then, Skinner was a radical be- concentration on operant behavior is one major haviorist in that he refused to acknowledge any reason that his brand of behaviorism was much dif- causal role of mental events in human conduct. ferent from Watson’s. For Skinner, so-called mental events were nothing Although both Skinner and Thorndike studied but neurophysiological events to which we have behavior controlled by its consequences, how they assigned mentalistic labels. studied that behavior was different. Thorndike Skinner continued to attack cognitive psychol- measured how long it took an animal to make an ogy throughout his professional life, and toward the escape response as a function of successive, rein- end of his life he deeply regretted the increased forced trials. He found that as the number of rein- popularity of cognitive psychology. forced escapes increases, the time it takes for the animal to escape decreases. His dependent variable was the latency of the escape response. Skinner’s Operant Behavior procedure was to allow an animal to respond freely Whereas Watson modeled his psychology after the in an experimental chamber (called a Skinner box) Russian physiologists, Skinner modeled his after and to note the effect of reinforcement on response Thorndike. Watson and Pavlov attempted to cor- rate. For example, a lever-press response may occur relate behavior with environmental stimuli. That is, only 2 or 3 times a minute before it is reinforced they were interested in reflexive behavior. Skinner and 30 or 40 times a minute when it results in
446 CHAPTER 13 reinforcement. Rate of responding, then, was Thus, for Skinner, there is no talk of drive re- Skinner’s dependent variable. duction, satisfying states of affairs, or any other me- Despite the differences between them, how- chanisms of reinforcement. A reinforcer is anything ever, both Watson and Skinner exemplified radical that, when made contingent on a response, changes behaviorism because they believed that behavior the rate with which that response is made. For could be completely explained in terms of events Skinner nothing additional needs to be said. He external to the organism. For Watson, environ- accepted Thorndike’s law of effect but not the mental events elicit either learned or unlearned mentalism that the phrase “satisfying state of affairs” responses; for Skinner, the environment selects implies. behavior via reinforcement contingencies. For both, what goes on within the organism is rela- tively unimportant. As we have seen, the theories The Importance of the Environment of Tolman and Hull exemplified methodological Whereas the environment was important for behaviorism because they postulated a wealth of Watson and the Russian physiologists because it events that were supposed to intervene between elicited behavior, it was important for Skinner be- experience and behavior. cause it selected behavior. The reinforcement con- tingencies the environment provides determine The Nature of Reinforcement which behaviors are strengthened and which are not. Change reinforcement contingencies, and If an operant response leads to reinforcement, the you change behavior: rate of that response increases. Thus, those re- sponses an organism makes that result in reinforce- The environment is obviously important, ment are most likely to occur when the organism is but its role has remained obscure. It does next in that situation. This is what is meant by the not push or pull, it selects, and this function statement that operant behavior is controlled by its is difficult to discover and analyze. The consequences. According to Skinner, reinforcement role of natural selection in evolution was can be identified only through its effects on behav- formulated only a little more than a hun- ior. Just because something acts as a reinforcer for dred years ago, and the selective role of the one organism under one set of circumstances does environment in shaping and maintaining not mean that it will be a reinforcer for another the behavior of the individual is only be- organism or for the same organism under different ginning to be recognized and studied. As circumstances: the interaction between organism and en- vironment has come to be understood, In dealing with our fellow men in every- however, effects once assigned to states of day life and in the clinic and laboratory, we mind, feeling, and traits are beginning to may need to know just how reinforcing a be traced to accessible conditions, and a specific event is. We often begin by noting technology of behavior may therefore be- the extent to which our own behavior is come available. It will not solve our pro- reinforced by the same event. This practice blems, however, until it replaces traditional frequently miscarries; yet it is still com- prescientific views, and these views are monly believed that reinforcers can be strongly entrenched. (Skinner, 1971, p. 25) identified apart from their effects upon a particular organism. As the term is used Thus, Skinner applied Darwinian notions to his here, however, the only defining charac- analysis of behavior. In any given situation, an or- teristic of a reinforcing stimulus is that it ganism initially makes a wide variety of responses. reinforces. (Skinner, 1953, p. 71) Of those responses, only a few will be functional
NEOB EHA VIORIS M 447 (reinforcing). These effective responses survive and act in a given way. This result is no doubt become part of the organism’s response repertoire responsible for its widespread use. We “in- to be used when that situation next occurs. stinctively” attack anyone whose behavior According to Skinner, the fact that behavior is displeases us—perhaps not in physical as- governed by reinforcement contingencies provides sault, but with criticism, disapproval, blame, hope for the solution of a number of societal pro- or ridicule. Whether or not there is an in- blems. If it was the “mind” or the “self” that herited tendency to do this, the immediate needed to be understood instead of how the envi- effect of the practice is reinforcing enough ronment selects behavior, we would be in real to explain its currency. In the long run, trouble: however, punishment does not actually eliminate behavior from a repertoire, and its Fortunately, the point of attack is more temporary achievement is obtained at tre- readily accessible. It is the environment mendous cost in reducing the over-all effi- which must be changed. A way of life ciency and happiness of the group. (p. 190) which furthers the study of human be- havior in its relation to that environment The “tremendous cost” involved in the use of should be in the best possible position to punishment comes from the many negative by- solve its major problems. This is not jin- products associated with it, including the fact that goism, because the great problems are now it induces fear, it often elicits aggression, it justifies global. In the behavioristic view, man can inflicting pain on others, and it often replaces one now control his own destiny because he undesirable response with another, such as when a knows what must be done and how to do child spanked for a wrongdoing cries instead. it. (Skinner, 1974, p. 251) How then is undesirable behavior to be dealt with? Skinner (1953) said to ignore it: The most effective alternative process [to The Positive Control of Behavior punishment] is probably extinction. This Like Thorndike, Skinner (1971) found that the ef- takes time but is much more rapid than fects of reinforcement and punishment are not sym- allowing the response to be forgotten. The metrical; reinforcement strengthens behavior, but technique seems to be relatively free of punishment does not weaken behavior: objectional by-products. We recommend it, for example, when we suggest that a A child who has been severely punished parent “pay no attention” to objectionable for sex play is not necessarily less inclined behavior on the part of his child. If the to continue; and a man who has been child’s behavior is strong only because it imprisoned for violent assault is not nec- has been reinforced by “getting a rise out essarily less inclined toward violence. of” the parent, it will disappear when this Punished behavior is likely to reappear af- consequence is no longer forthcoming. ter the punitive contingencies are with- (p. 192) drawn. (p. 62) Because of the relative ineffectiveness of pun- Why, if punishment is ineffective as a modifier ishment and the many negative by-products associ- of behavior, is it so widely used? Because, said ated with its use, Skinner consistently urged that Skinner (1953), it is reinforcing to the punisher: behavior be modified positively through reinforce- Severe punishment unquestionably has an ment contingencies, not negatively through immediate effect in reducing a tendency to punishment.
448 CHAPTER 13 Skinner’s Attitude toward Theory from the logicians rather than empirical science, and describe thinking as necessar- Because Skinner’s position was nontheoretical, it ily involving stages of hypothesis, deduc- contrasted with the behavioristic positions of tion, experimental test, and confirmation. Tolman and Hull and, to a lesser extent, of But this is not the way most scientists ac- Guthrie. Skinner accepted operationism but re- tually work. It is possible to design signif- jected the theoretical aspects of logical positivism. icant experiments for other reasons, and He was content to manipulate environmental the possibility to be examined is that such events (such as reinforcement contingencies) and research will lead more directly to the kind note the effects of these manipulations on behavior, of information which a science usually believing that this functional analysis is all that is accumulates. (pp. 194–195) necessary. For this reason, Skinner’s approach is sometimes referred to as a descriptive behavior- In describing his nontheoretical approach, ism. There is, Skinner felt, no reason for looking Skinner (1956) said that if he tried something and “under the skin” for explanations of relationships if it seemed to be leading to something useful, he between the environment and behavior. Looking persisted. If what he was doing seemed to be lead- for physiological explanations of behavior is a waste ing to a dead end, he abandoned it and tried some- of time because overt behavior occurs whether or thing else. not we know its neurophysiological underpinnings. Some believe that Skinner’s article “Are We have already reviewed Skinner’s attitude to- Theories of Learning Necessary?” (1950) marked ward mentalistic explanations of behavior. Because the end of what Koch (1959) called the “age of Skinner did not care what was going on “under the theory” in psychology. skin” either physiologically or mentally, his ap- proach is often referred to as the empty-organism approach. Skinner knew, of course, that the organ- Applications of Skinnerian Principles ism is not empty, but he thought that nothing is lost Like Watson, Skinner and his followers sought to by ignoring events that intervene between the en- apply their principles to the solution of practical pro- vironment and the behavior it selects. blems. In all applications of Skinnerian principles, the Besides opposing physiological and mentalistic general rule is always the same: Change reinforcement explanations of behavior, Skinner (1950) opposed contingencies, and you change behavior. This principle has abstract theorizing like that of Tolman and Hull: been used to teach pigeons to play games like table Research designed with respect to theory tennis and basketball, and many animals trained is also likely to be wasteful. That a theory through the use of Skinnerian principles have per- generates research does not prove its value formed at tourist attractions throughout the United unless the research is valuable. Much use- States. In a defense effort, pigeons were even less experimentation results from theories, trained to guide missiles as the missiles sped toward and much energy and skill are absorbed by enemy targets (Skinner, 1960). In 1948 Skinner them. Most theories are eventually over- wrote a utopian novel entitled Walden Two, in which thrown, and the greater part of the asso- he demonstrated how his principles could be used in ciated research is discarded. This could be designing a model society. In Beyond Freedom and justified if it were true that productive re- Dignity (1971), Skinner reviewed the reasons that search requires a theory—as is, of course, cultural engineering, although possible, has been often claimed. It is argued that research largely rejected. would be aimless and disorganized without In the realm of education, Skinner developed a a theory to guide it. The view is supported teaching technique called programmed learning by psychological texts which take their cue (1954, 1958). With programmed learning, material
NEOB EHA VIORIS M 449 is presented to students in small steps; students are Hollon, and Rimm (1987), it is institutions without then tested on the material, given immediate feed- token economies that are unnatural and relatively back on the accuracy of their answers, and allowed ineffective: to proceed through the material at their own pace. Token economies are not really unnatural. Skinner had criticized U.S. education ever since Indeed, any national economy with a cur- 1953, when he visited his daughter’s classroom rency system is in every sense a token and concluded that the teacher was violating every- economy: any currency consists by defini- thing that was known about learning. Skinner tion of token or symbolic “reinforcers” that (1984) maintained that many of the problems in may be exchanged for items that constitute a our educational system could be solved through more direct form of reinforcement. the use of operant principles. Skinner’s main criti- Whereas the individual in society works to cism of U.S. educational practices was that the earn tokens (money) with which he pur- threat of punishment is used to force students to chases his dwelling place, food, recreation, learn and to behave instead of the careful manipu- and so on, most institutions provide such lation of reinforcement contingencies. This aversive comforts noncontingently and hence cease control, Skinner said, creates a negative attitude to- to encourage many adaptive behaviors that ward education. are appropriate and effective in the natural In 1983 Skinner, along with Margaret environment. (p. 222) Vaughan, wrote Enjoy Old Age: Living Fully Your Later Years, in which they addressed such topics as In general, the use of Skinnerian principles in diet, retirement, exercise, forgetfulness, sensory de- treating behavior problems has been very effective ficiencies, and fear of death. Interestingly, although (for example, see Ayllon and Azrin, 1968; Skinner counseled the elderly to avoid fatigue, he Craighead, Kazdin, and Mahoney, 1976; Kazdin, and Vaughan wrote the book in three months. 1989; Kazdin and Wilson, 1978; Leitenberg, 1976; Skinner and his followers have applied behav- Masters et al., 1987; Rimm and Masters, 1974; ior modification principles to helping individuals Ulrich, Stachnik, and Mabry, 1966). For his role in with problems ranging from psychosis to smoking, developing behavior modification procedures alcoholism, drug addiction, mental retardation, ju- used to improve the quality of life of the mentally venile delinquency, speech disorders, shyness, pho- retarded, Skinner was presented a Kennedy bias, obesity, and sexual disorders. The Skinnerian International Award in 1971. In 1972 he was named version of behavior therapy assumes that people “Humanist of the Year” by the American Humanist learn abnormal behavior in the same way that they Association. On August 10, 1990, the APA pre- learn normal behavior. Therefore, “treatment” is a sented Skinner with an unprecedented Lifetime matter of removing the reinforcers that are main- Contribution to Psychology Award. Eight days taining the undesirable behavior and arranging the later. he died of leukemia at the age of 86. As a reinforcement contingencies so that they strengthen further tribute to Skinner, the entire November desirable behavior. 1992 issue of the American Psychologist was dedicated Skinnerian principles have also been used to to his ideas and their influence. create token economies in a number of institu- tions, such as psychiatric hospitals. When partici- pants in such economies behave in desirable ways, they are reinforced with tokens that can be ex- BEHAVIORISM TODAY changed for such items as candy, cigarettes, coffee, or the exclusive use of a radio or television set. The work of all the neobehaviorists covered in this Token economies have been criticized as contrived chapter remains influential in contemporary psy- or unnatural but, according to Masters, Burish, chology. Tolman’s brand of behaviorism, with its
450 CHAPTER 13 emphasis on purposive behavior and mental con- survey Skinner ranked first, Piaget second, and structs, can be viewed as a major reason for the Freud third (Dittman, 2002, p. 28). As far as recog- current popularity of cognitive psychology. nition by the general public is concerned, Skinner is Although Hull did much to promote an objective perhaps second only to Freud. For an interesting behavioristic approach, his current influence is due account of how the popular press reacted to mainly to some of the esoteric features of his the- Skinner’s ideas, see Rutherford (2000). ory. His goal of developing a comprehensive be- Despite the current manifestations of behavior- havior theory, however, has given way to the goal ism and neobehaviorism in contemporary psychol- of developing theories designed to explain specific ogy, the influence of both has diminished. The phenomena (see, for example, Amsel, 1992; overwhelming interest in cognitive psychology to- Rashotte and Amsel, 1999). Guthrie’s theory has day runs counter to all brands of behaviorism ex- survived mainly through Estes’s efforts to create cept Tolman’s (see Chapter 20). Contrary to what mathematical models of learning and memory. the behaviorists believed, evolutionary psycholo- Skinner’s influence remains strong. In 1974 gists, and others, are providing evidence that Skinner wrote About Behaviorism, which attempted much animal behavior, including human social be- to correct 20 misconceptions about behaviorism. In havior, is genetically influenced (see Chapter 19). this book, Skinner traced a number of these mis- Also, the neobehaviorist’s insistence that all theoret- conceptions to Watson’s early writings—for exam- ical terms be operationally defined became a prob- ple, Watson’s dependence on reflexive behavior lem. Even the logical positivists abandoned a strict and his denial of the importance of genetic endow- operationism because it was too restrictive; it ex- ment. Skinner’s position rectified both “mistakes.” cluded from science concepts that were too nebu- Skinner also pointed out that he did not deny so- lous to be defined operationally but were still useful called mental processes but believed that ultimately in suggesting new avenues of research and methods they will be explained as verbal labels that we attach of inquiry: to certain bodily processes. As evidence of the re- If one were to criticize behaviorism, it cent popularity of Skinnerian behaviorism, fol- would not be for what it tried to accom- lowers of Skinner have formed their own division plish, but rather for the things it found of the APA (Division 25, the division of the necessary to deny. Fundamentally, it denied Experimental Analysis of Behavior) and have two the need for free theorizing, because all of their own journals in which to publish their re- theory had to be limited to observable search, the Journal of Applied Behavior Analysis and stimuli and responses. It denied all of the the Journal for Experimental Analysis of Behavior. commonsense constructs without which Korn, Davis, and Davis (1991) provide further none of us can get along in the world: evidence for the popularity of Skinner in contem- Conscious experience, thinking, knowl- porary psychology. Historians of psychology edge, images, feelings, and so on. In fact, it and chairpersons of graduate programs in psychol- rejected commonsense knowledge by fiat, ogy were asked to rank the 10 most important psy- rather than testing it and transcending it, as chologists of all time and the 10 most important the other sciences had done. (Baars, 1986, contemporary psychologists. On the “all time” pp. 82–83) list, historians ranked Wundt first and Skinner eighth. Chairpersons ranked Skinner first and Even the suggestions that logical positivism Wundt sixth. On the “contemporary” list, both made concerning theory construction eventually historians and chairpersons ranked Skinner first. In fell into disrepute. Perhaps the most important rea- another survey 1,725 members of the American son that logical positivism ultimately failed was the Psychological Society were asked to rank the most discovery that it did not accurately describe how eminent psychologists of the 20th century. In this science was practiced even by its most effective
NEOB EHA VIORIS M 451 practitioners. Individuals such as Thomas Kuhn (see Psychologists generally agree now that the subject Chapter 1) have shown that the behavior of scien- matter of psychology is overt behavior. Today, cog- tists is determined as much by beliefs, biases, and nitive psychology is very popular, but even the psy- emotions as by axioms, postulates, theories, or chologists studying cognitive events use behavior to logic. index those events. In that sense, most experimental One major legacy of behaviorism and neobe- psychologists today are behaviorists. haviorism still characterizes psychology, however. SUMMARY The positivism of Bacon, Comte, and Mach insisted concerning what leads to what in an environment, that only that which is directly observable be the to an expectancy, and, finally, to a belief. A set of object of scientific investigation. For the positivists, beliefs constitutes a cognitive map, which was all speculation about abstract entities should be Tolman’s most important intervening variable. In actively avoided. Watson and the Russian physiolo- Tolman’s theory, confirmation replaced the notion gists were positivists. The logical positivists had a of reinforcement, and an important distinction was more liberal view of scientific activity. For them, the- made between learning and performance. Tolman’s orizing about unobservable entities was allowed, general influence on contemporary psychology can provided those entities were directly linked to ob- be seen in the widespread popularity of cognitive servable events via operational definitions. psychology. Contemporary information-processing Operational definitions define abstract concepts in approaches to psychology also have much in com- terms of the procedures used to measure those con- mon with Tolman’s theory. cepts. The belief that all scientific concepts be opera- Using intervening variables even more ex- tionally defined was called operationism. Physicalism tensively than did Tolman, Hull developed an was the belief that all sciences should share com- open-ended, self-correcting, hypothetico-deductive mon assumptions, principles, and methodologies theory of learning. If experimentation supports the and should model themselves after physics. deductions from this theory, the theory gains Neobehaviorism resulted when behaviorism, with strength; if not, the part of the theory on which the its insistence that the subject matter of psychology deductions were based is revised or rejected. be overt behavior, merged with logical positivism, Equating reinforcement with drive reduction, Hull with its acceptance of theory and its insistence on defined habit strength as the number of reinforced operational definitions. By following the tenets of pairings between a stimulus and a response. He saw logical positivism, many neobehaviorists believed reaction potential as a function of the amount of habit they could be theoretical and still remain objective. strength and drive present. Hull’s theory was ex- Independently of logical positivism but in ac- tremely influential in the 1940s and 1950s, and be- cordance with it, Tolman introduced intervening cause of the efforts of Hull’s disciples such as Kenneth variables into psychology. Instead of studying re- Spence, the influence of his theory extended well flexive, or molecular, behavior, Tolman studied into the 1960s. Some particular aspects of Hull’s the- purposive, or molar, behavior; thus, his version of ory are still found in contemporary psychology, but psychology was called purposive behaviorism. To not his comprehensive approach to theory building; avoid even the possibility of introspection in his psychologists now seek theories of more limited research, Tolman used only rats as his experimental domain. subjects. According to Tolman, the learning pro- Guthrie created an extremely parsimonious cess progresses from the formation of hypotheses theory of learning. All learning was explained by
452 CHAPTER 13 the law of contiguity, which stated that when a operationism. Skinner distinguished between re- pattern of stimuli and a response occur together spondent behavior, which a known stimulus elicits, they become associated. Furthermore, the associa- and operant behavior, which an organism tion between the two occurs at full strength after emits. Skinner was concerned almost exclusively just one exposure. By postulating one-trial learning, with operant behavior. For Skinner reinforcement Guthrie rejected the law of frequency. To explain is anything that changes the rate or probability of a why practice improves performance, Guthrie differ- response. Nothing more needs to be known about entiated among movements, acts, and skills. A reinforcement, nor is an understanding of physiol- movement is a specific response made to a specific ogy necessary for an understanding of behavior. pattern of stimuli. It is the association between a Under the influence of Mach’s positivistic philoso- movement and a pattern of stimuli that is learned phy of science, Skinner urged a study of the func- in one trial. An act is a movement that has become tional relationship between behavior and the envi- associated with a number of stimuli patterns. A skill, ronment. Because such an analysis is correlational, it in turn, consists of many acts. It is because acts are avoided the complexities of determining causa- made up of many movements and skills are made tion in human behavior and eliminated the need up of many acts that practice improves perfor- to postulate unobserved cognitive or physiological mance. According to Guthrie, “reinforcement” is determinants of behavior. Watson and Skinner a mechanical arrangement that prevents unlearning. were radical behaviorists because they stressed envi- Forgetting occurs when one S–R relationship is ronmental influences on behavior to the exclusion displaced by another. Like learning, forgetting oc- of so-called mental events and physiological states. curs in one trial. Bad habits can be broken by caus- Tolman, Hull, and Guthrie were methodological ing a response, other than the undesirable one, to behaviorists because they were willing to theorize be made in the presence of the stimuli that previ- about internal causes of behavior (such as cognitive ously elicited the undesirable response. Like the maps and physiological drives). Many contempo- other procedures used to break bad habits, punish- rary psychologists label themselves Skinnerians and ment, to be effective, must cause behavior incom- are active in both research and the applied aspects of patible with the undesirable behavior in the pres- psychology. According to Skinnerian psychology, ence of the stimuli that previously elicited the behavior that is reinforced is strengthened (more undesirable behavior. What others called drives, probable), but behavior that is punished is not Guthrie called maintaining stimuli. Maintaining sti- necessarily weakened. It is best then to arrange muli, either internal or external, keep an organism reinforcement contingencies so that desirable be- active until the maintaining stimuli are terminated. havior is reinforced and undesirable behavior is It is behavior associated with maintaining stimuli not. No matter what type of behavior is under that appears to be intentional. Attempts to formalize consideration, the rule is always the same: Change Guthrie’s theory, thereby making it more testable, reinforcement contingencies and you change were made by Virginia Voeks and William Kaye behavior. Estes. Although the influence of behaviorism and In his approach to psychology, Skinner ac- neobehaviorism has diminished in contemporary cepted positivism instead of logical positivism. He psychology, some of their basic tenets have been can still be classified as a neobehaviorist, however, incorporated into all current brands of experimental because although he avoided theory he did accept psychology.
NEOB EHA VIORIS M 453 DISCUSSION QUESTIONS 1. Compare positivism to logical positivism. Horton observe that confirmed their view of “reinforcement”? 2. What is an operational definition? Give an example. What is operationism? 18. Summarize Guthrie’s explanation of forgetting. 3. What is physicalism? 19. According to Guthrie, under what circum- stances is punishment effective? Ineffective? 4. What is neobehaviorism? 20. In Guthrie’s theory, what is the function of 5. What convinced Tolman that he could study maintaining stimuli? For example, how were purposive behavior and still be an objective these stimuli used to explain what other the- behaviorist? orists called drives and intentions? 6. Explain how Tolman used intervening vari- 21. Was Skinner’s proposed functional analysis of ables in a way that was consistent with logical the relationship between environmental and positivism. behavioral events more in accordance with 7. How, according to Tolman, do early hypoth- positivistic or with logical positivistic eses concerning what leads to what in a situa- philosophy? tion evolve into a cognitive map? 22. Summarize Skinner’s arguments against cogni- 8. What did Tolman mean by vicarious trial and tive psychology. error? 23. How did Skinner distinguish between respon- 9. In Tolman’s theory, was reinforcement neces- dent and operant behavior? sary for learning to occur? What term in 24. What is meant by the statement that operant Tolman’s theory had some similarity to what behavior is controlled by its consequences? others called reinforcement? 25. Distinguish between radical and methodologi- 10. What evidence did Tolman provide for his cal behaviorism. contention that reinforcement influences per- formance but not learning? Also, how did he 26. For Skinner, what constitutes a reinforcer? explain extinction? 27. How did Skinner apply Darwinian concepts to 11. What influence did Tolman’s theory have on his analysis of behavior? contemporary psychology? 28. Why did Skinner argue that behavior should be 12. Why was Hull’s theory called a hypothetico- controlled by reinforcement contingencies deductive theory? Why did Hull consider his rather than by punishment? theory to be self-correcting? 29. Summarize Skinner’s argument against the use 13. With reference to Hull’s theory, define the of theory in psychology. following terms: reinforcement, habit strength, and 30. State the general rule that Skinnerians follow in reaction potential. modifying behavior. Give an example of how 14. What was Guthrie’s one law of learning? this rule could be applied in treating a behavior disorder. 15. Did Guthrie accept or reject the law of fre- quency? Explain. 31. Explain why the influence of behaviorism and neobehaviorism has diminished in contempo- 16. If learning occurs at full strength in one trial, rary psychology. how did Guthrie explain improvement in performance as a function of practice? 32. In what ways do the tenets of behaviorism re- main influential in contemporary psychology? 17. According to Guthrie, what is the function of “reinforcement”? What did Guthrie and
454 CHAPTER 13 SUGGESTIONS FOR FURTHER READING Bjork, D. W. (1997). B. F. Skinner: A life. Washington, Skinner, B. F. (1971). Beyond freedom and dignity. New DC: American Psychological Association. York: Knopf. Hull, C. L. (1952a). Clark L. Hull. In E. G. Boring, H. S. Skinner, B. F. (1974). About behaviorism. New York: Langfeld, H. Werner, & R. M. Yerkes (Eds.), A Knopf. history of psychology in autobiography (Vol. 4, pp. 143– Skinner, B. F. (1990). Can psychology be a science of 162). Worcester, MA: Clark University Press. mind? American Psychologist, 45, 1206–1210. Nye, R. D. (1992). The legacy of B. F. Skinner: Concepts Tolman, E. C. (1952). Edward C. Tolman. In E. G. and perspectives, controversies and misunderstandings. Boring, H. S. Langfeld, H. Werner, & R. M. Pacific Grove, CA: Brooks/Cole. Yerkes (Eds.), A history of psychology in autobiography Prenzel-Guthrie, P. (1996). Edwin Ray Guthrie: Pioneer (Vol. 4, pp. 323–339). Worcester, MA: Clark learning theorist. In G. A. Kimble, C. A. Boneau, & University Press. M. Wertheimer (Eds.), Portraits of pioneers in psy- Wiener, D. N. (1996). B. F. Skinner: Benign anarchist. chology (Vol. 2, pp. 137–149). Washington, DC: Needham Heights, MA: Allyn & Bacon. American Psychological Association. Skinner, B. F. (1953). Science and human behavior. New York: Macmillan. GLOSSARY Behavior therapy The use of learning principles to Guthrie, Edwin Ray (1886–1959) Accepted the law treat emotional or behavioral disorders. of contiguity but not the law of frequency. For him, Belief According to Tolman, an expectation that ex- learning occurs at full strength after just one association perience has consistently confirmed. between a pattern of stimuli and a response. (See also Law of contiguity.) Cognitive map According to Tolman, the mental re- presentation of the environment. Habit strength ( S H R ) For Hull, the strength of an as- sociation between a stimulus and response. This strength Confirmation According to Tolman, the verification depends on the number of reinforced pairings between of a hypothesis, expectancy, or belief. the two. Descriptive behaviorism Behaviorism that is positiv- istic in that it describes relationships between environ- Hull, Clark Leonard (1884–1952) Formulated a mental events and behavior rather than attempting to complex hypothetico-deductive theory in an attempt to explain those relationships. Skinner’s approach to psy- explain all learning phenomena. chology exemplified descriptive behaviorism. Hypothesis According to Tolman, an expectancy that occurs during the early stages of learning. Drive reduction Hull’s proposed mechanism of rein- forcement. For Hull anything that reduces a drive is Hypothetico-deductive theory A set of postulates reinforcing. from which empirical relationships are deduced (pre- dicted). If the empirical relationships are as predicted, the Expectancy According to Tolman, a hypothesis that theory gains strength; if not, the theory loses strength and has been tentatively confirmed. must be revised or abandoned. Functional analysis Skinner’s approach to research Instrumental conditioning The type of conditioning that involves studying the systematic relationship be- studied by Thorndike, wherein an organism learns to tween behavioral and environmental events. Such study make a response that is instrumental in producing focuses on the relationship between reinforcement con- tingencies and response rate or response probability. reinforcement.
NEOB EHA VIORIS M 455 Intervening variables Events believed to occur be- Physicalism A belief growing out of logical positivism tween environmental and behavioral events. Although that all sciences should share common assumptions, intervening variables cannot be observed directly, they principles, and methodologies and should model them- are thought to be causally related to behavior. Hull’s selves after physics. habit strength and Tolman’s cognitive map are examples Positivism The belief that science should study only of intervening variables. those objects or events that can be experienced directly. Latent extinction The finding that animals who pas- That is, all speculation about abstract entities should be sively experience a goal box no longer containing rein- avoided. forcement extinguish a previously learned response to Purposive behavior Behavior that is directed toward that goal box significantly faster than animals without some goal and that terminates when the goal is attained. such experience. Purposive behaviorism The type of behaviorism Latent learning According to Tolman, learning that Tolman pursued, which emphasizes molar rather than has occurred but is not translated into behavior. molecular behavior. Law of contiguity Guthrie’s one law of learning, which Reaction potential ( S E R ) For Hull, the probability of states that when a pattern of stimuli is experienced along a learned response being elicited in a given situation. with a response, the two become associated. In 1959 This probability is a function of the amount of drive and Guthrie revised the law of contiguity to read, “What is habit strength present. being noticed becomes a signal for what is being done.” Reinforcement For Hull, drive reduction; for Skinner, Logical positivism The philosophy of science accord- anything that increases the rate or the probability of a ing to which theoretical concepts are admissible if they response; for Tolman, the confirmation of a hypothesis, are tied to the observable world through operational expectation, or belief; for Guthrie, a mechanical ar- definitions. rangement that prevents unlearning. Maintaining stimuli According to Guthrie, the inter- Respondent behavior Behavior that is elicited by a nal or external stimuli that keep an organism active until known stimulus. a goal is reached. Skinner, Burrhus Frederic (1904–1990) A behaviorist Molar behavior (See Purposive behavior.) who believed that psychology should study the functional Molecular behavior A small segment of behavior such relationship between environmental events, such as rein- as a reflex or a habit that is isolated for study. forcement contingencies, and behavior. Skinner’swork Neobehaviorism Agreed with older forms of behav- exemplified positivism. (See also Positivism.) iorism that overt behavior should be psychology’s subject S–R psychology The type of psychology insisting that matter but disagreed that theoretical speculation con- environmental stimuli elicit most, if not all, behavior. cerning abstract entities must be avoided. Such specula- The Russian physiologists and Watson were S–R tion was accepted provided that the theoretical terms psychologists. employed are operationally defined and lead to testable Theoretical terms According to logical positivism, predictions about overt behavior. those terms that are employed to explain empirical Observational terms According to logical positivism, observations. terms that refer to empirical events. Token economies An arrangement within institutions One-trial learning Guthrie’s contention that the asso- whereby desirable behavior is strengthened using valu- ciation between a pattern of stimuli and a response de- able tokens as reinforcers. velops at full strength after just one pairing of the two. Tolman, Edward Chace (1886–1959) Created a Operant behavior Behavior that is emitted by an or- brand of behaviorism that used mental constructs and ganism rather than elicited by a known stimulus. emphasized purposive behavior. Although Tolman em- Operational definition A definition that relates an ployed many intervening variables, his most important abstract concept to the procedures used to measure it. was the cognitive map. Operationism The belief that all abstract scientific Vicarious trial and error According to Tolman, the concepts should be operationally defined. apparent pondering of behavioral choices in a learning situation. Performance The translation of learning into behavior.
14 ✵ Gestalt Psychology bout the same time that the behaviorists were rebelling against structural- A ism and functionalism in the United States, a group of young German psy- chologists was rebelling against Wundt’s experimental program that featured a search for the elements of consciousness. Whereas the focus of the behaviorists’ attack was the study of consciousness and the associated method of introspection, the German protesters focused their attack on Wundt’s elementism. Consciousness, said the German rebels, could not be reduced to elements without distorting the true meaning of the conscious experience. For them, the investigation of conscious experience through the introspective method was an essential part of psychology, but the type of conscious experience Wundt and the U.S. structuralists investigated was artificial. These young psychologists believed that we do not experience things in isolated pieces but in meaningful, intact configurations. We do not see patches of green, blue, and red; we see people, cars, trees, and clouds. These meaningful, intact, conscious experiences are what the introspective method should concentrate on. Because the German word for “configuration,”“form,” or “whole” is Gestalt, this new type of psychology was called Gestalt psychology. The Gestaltists were opposed to any type of elementism in psychology, whether it be the type Wundt and the structuralists practiced or the type the behaviorists practiced in their search for S–R associations. The attempt to reduce either consciousness or behavior to the basic elements is called the molecular approach to psychology, and psychologists such as Wundt (as experimentalist), Titchener, Pavlov, and Watson used this type of approach. The Gestaltists argued that a molar approach should be taken. Taking the molar approach in studying consciousness would mean concentrating on phenomenological experience (mental experience as it occurred to the naive observer, without further analysis). The 456
GESTALT PSYCHOLOGY 457 term phenomenon means “that which appears” or to failure. For Kant and the Gestaltists, an important “that which is given,” and so phenomenology, difference exists between perception and sensation. This difference arises because our minds (Kant) or the technique used by the Gestaltists, is the study our brains (the Gestaltists) change sensory experi- of that which naturally appears in consciousness. ence, making it more structured and organized Taking the molar, or phenomenological, approach and thus more meaningful than it otherwise would while studying behavior means concentrating on be. Accordingly, the world we perceive is never the goal-directed (purposive) behavior. We saw in the same as the world we sense. Because this embellish- last chapter that, under the influence of Gestalt ment of sensory information results from the nature of the mind (Kant) or the brain (Gestaltists), it is psychology, Tolman chose to study this type of be- independent of experience. havior. As we will see, the Gestaltists attempted to show that in every aspect of psychology, it is more Ernst Mach beneficial to concentrate on wholes (Gestalten, plu- ral of Gestalt) than on parts (atoms, elements). Ernst Mach (1838–1916), a physicist, postulated Those taking a molar approach to the study of be- (1886/1914) two perceptions that appeared to be independent of the particular elements that com- havior or psychological phenomena are called hol- pose them: space form and time form. For example, ists, in contrast to the elementists or atomists, who one experiences the form of a circle whether the study complex phenomena by seeking simpler actual circle presented is large, small, red, blue, components that compose those phenomena. The bright, or dull. The experience of “circleness” is Gestaltists were clearly holists. therefore an example of space form. The same would be true of any geometric form. Similarly, a melody is recognizable as the same no matter what key or tempo it is played in. Thus, a melody is an ANTECEDENTS OF GESTALT example of time form. Mach was making the PSYCHOLOGY important point that a wide variety of sensory ele- ments can give rise to the same perception; there- fore, at least some perceptions are independent of any particular cluster of sensory elements. Immanuel Kant Immanuel Kant (1724–1804) believed that con- Christian von Ehrenfels scious experience is the result of the interaction be- tween sensory stimulation and the actions of the Christian von Ehrenfels (1859–1932) studied in faculties of the mind. In other words, the mind Vienna with Brentano and in 1890 wrote a paper adds something to our conscious experience that titled “Uber ‘Gestaltqualitäten’” (On Gestalt Qualities). sensory stimulation does not contain. If the phrase About this paper Barry Smith (1994) says, “Almost faculties of the mind is replaced by characteristics of the all of the theoretical and conceptual issues which brain, there is considerable agreement between Kant subsequently came to be associated with the and the Gestaltists. Both believed that conscious Gestalt idea are treated at some point … at least experience cannot be reduced to sensory stimula- in passing” (pp. 246–247). Max Wertheimer, the tion, and for both conscious experience is different founder of Gestalt psychology, took several courses from the elements that compose it. Therefore, from Ehrenfels between 1898 and 1901 and no looking for a one-to-one correspondence between doubt was influenced by him. Elaborating on sensory events and conscious experience is doomed Mach’s notions of space and time forms, Ehrenfels
458 CHAPTER 14 said that our perceptions contain Gestaltqualitäten Act Psychology (form qualities) that are not contained in isolated We saw in Chapter 9 that Franz Brentano and Carl sensations. No matter what pattern dots are Stumpf favored the type of introspection that fo- arranged in, one recognizes the pattern, not the cuses on the acts of perceiving, sensing, or problem individual dots. Similarly, one cannot experience a solving. They were against using introspection to melody by attending to individual notes; only when search for mental elements, and they directed their one experiences several notes together does one more liberal brand of introspection toward mental experience the melody. For both Mach and phenomena. Thus, both the act psychologists and Ehrenfels, form is something that emerges from the the Gestaltists were phenomenologists. It should elements of sensation. Their position was similar to come as no surprise that act psychology influ- the one John Stuart Mill had taken many years ear- enced Gestalt psychology because all three founders lier. With his idea of “mental chemistry,” Mill had of Gestalt psychology (Wertheimer, Koffka, and suggested that when sensations fuse, a new sensa- Köhler), at one time or another, studied under tion totally unlike those of which it was composed Carl Stumpf. In 1920 Köhler even dedicated one could emerge. of his books to Stumpf. Like Mill, Mach and Ehrenfels believed that elements of sensation often combine and give rise to the experience of form. However, for Mach, Developments in Physics Ehrenfels, and Mill, the elements are still necessary Because properties of magnetic fields were difficult in determining the perception of the whole or the to understand in terms of the mechanistic- form. As we will see, the Gestaltists turned this re- elementistic view of Galilean–Newtonian physics, lationship completely around by saying that the some physicists turned to a study of force fields, in whole dominates the parts, not the other way which all events are interrelated. (Anything that around. happened in a force field in some way influences everything else in the field.) Köhler was well versed William James in physics and had even studied for a while with Max Planck, the father of quantum mechanics. In Because of his distaste for elementism in psychol- fact, it is accurate to say that Gestalt psychology ogy, William James (1842–1910) can also be represented an effort to model psychology after viewed as a precursor to Gestalt psychology. He field theory instead of Newtonian physics. We said that Wundt’s search for the elements of con- will say more about this effort shortly. sciousness depended on an artificial and distorted view of mental life. Instead of viewing the mind as consisting of isolated mental elements, James pro- TH E F OUNDING O F GESTALT posed a stream of consciousness. He believed that this stream should be the object of psychological PSYCHOLOGY inquiry, and any attempt to break it up for more detailed analysis must be avoided. The Gestaltists In 1910 Max Wertheimer was on a train, on his agreed with James’s antielementistic stand but way from Vienna to a vacation on the Rhineland, thought that he had gone too far. The mind, they when he had an idea that was to launch Gestalt believed, could indeed be divided for study; it was psychology. The idea was that our perceptions are just that in choosing the mental element for their structured in ways that sensory stimulation is not. object of study, Wundt and the structuralists had That is, our perceptions are different from the sen- made a bad choice. For the Gestaltists, the correct sations that comprise them. To further explore this choice was the study of mental Gestalten. notion, Wertheimer got off the train at Frankfurt,
GESTALT PSYCHOLOGY 459 bought a toy stroboscope (a device that allows still jects in his perception experiments. So closely are pictures to be flashed in such a way that makes Koffka and Köhler linked with the development of them appear to move), and began to experiment Gestalt psychology that they, along with in a hotel room. Clearly, Wertheimer was perceiv- Wertheimer, are usually considered cofounders of ing motion where none actually existed. To exam- the school. ine this phenomenon in more detail, he went to the University of Frankfurt, where a tachistoscope was Max Wertheimer made available to him. (A tachistoscope can flash lights on and off for measured fractions of a Max Wertheimer (1880–1943) was born on April second.) Flashing two lights successively, 15 in Prague and attended a Gymnasium (roughly Wertheimer found that if the time between the equivalent to a high school) until he was 18, at flashes was long (200 milliseconds or longer), the which time he went to the University of Prague observer perceived two lights flashing on and off to study law. While Wertheimer was attending successively—which was, in fact, the case. If the the University of Prague, his interest shifted from interval between flashes was very short (30 millise- law to philosophy, and during this time he attended conds or less), both lights appeared to be on simul- lectures by Ehrenfels. After spending some time at taneously. But if the interval between the flashes the University of Berlin (1901–1903), where he was about 60 milliseconds, it appeared that one light attended Stumpf’s classes, Wertheimer moved to was moving from one position to the other. the University of Würzburg, where in 1904 he re- Wertheimer called this apparent movement the ceived his doctorate, summa cum laude, under phi phenomenon, and his 1912 article Külpe’s supervision. His dissertation was on lie de- “Experimental Studies of the Perception of tection. Being at Würzburg at the time when Külpe Movement” describing this phenomenon is usually and others were locked in debate with Wundt over taken as the formal beginning of the school of the existence of “imageless thought” and over what Gestalt psychology. introspection should focus on no doubt affected It should be noted that Wertheimer was not Wertheimer’s thinking. the first to observe apparent motion. As early as Between 1904 and 1910, Wertheimer held ac- 1824, Peter Roget presented a paper on the topic ademic positions at the Universities of Prague, to the Royal Society of London (Boorstein, 1991). Vienna, and Berlin. He was at the University of The Prague physiologist Sigmund Exner, with Frankfurt from 1910 to 1916, the University of whom Wertheimer did postdoctoral research, pub- Berlin from 1916 to 1929, and again at the lished a paper on the phenomenon in 1875. The University of Frankfurt from 1929 to 1933. American psychologist George Stratton’s article on Because of the chaos caused by the Nazi movement the topic in 1911 preceded Wertheimer’s landmark in Germany, Wertheimer, who was 53 years old at article of 1912. Finally, by the time Wertheimer the time, decided to pursue his career elsewhere. published his article, motion pictures were com- Positions were offered to him at Cambridge monplace. However, although he did not discover University, Oxford University, and the University apparent motion, “It was Wertheimer who saw the of Jerusalem; but in 1933 he accepted a position at deeper significance of the phenomenon, relating it the New School for Social Research, and he, his to a coherent system of explanatory principles that wife Anne, and their three children (Valentin, gave it a central place in psychology” (Boynton and Michael, and Lise) sailed for New York. Smith, 2006, p. 131). Wertheimer knew only German, and his first classes Wertheimer’s research assistants at the were taught in that language. After only five University of Frankfurt were two recent Berlin months, however, he began teaching and publish- doctoral graduates—Kurt Koffka and Wolfgang ing in English. His second language posed a prob- Köhler—both of whom acted as Wertheimer’s sub- lem for Wertheimer because it sometimes interfered
460 CHAPTER 14 Medicine of Library National Medicine the of of Library © Courtesy National Kurt Koffka the of © Courtesy Born on March 18 in Berlin, Kurt Koffka (1886– Kurt Koffka 1941) received his doctorate from the University of Max Wertheimer Berlin in 1908, under the supervision of Stumpf. Koffka served as an assistant at Würzburg and at with his desire to express himself precisely. Michael Frankfurt before accepting a position at the Wertheimer and King (1994) give an example: “He University of Giessen in central Germany, where … had some problems with mathematical terms; his he remained until 1924. During his stay at the students were occasionally baffled before they real- University of Frankfurt, Koffka began his long as- ized that his references to obtuse and acute ‘angels’ sociation with Wertheimer and Köhler. In 1924 he had nothing to do with heavenly beings but with came to the United States, and after holding visiting trigonometric angles” (pp. 5–6). professorships at Cornell and the University of Wertheimer had wide interests and, after arriv- Wisconsin, he accepted a position at Smith ing in the United States, wrote (in English) articles College in Northampton, Massachusetts, where he on truth (1934), ethics (1935), democracy (1937), remained until his death. and freedom (1940). Wertheimer intended to pub- In 1922 Koffka wrote an article, in English, on lish these articles as a collection, and Albert Einstein Gestalt psychology. Published in the Psychological wrote a forward. Although the collection was never Bulletin, the article was titled “Perception: An published in English, it was eventually published in Introduction to Gestalt-Theorie.” This article is be- German under the editorship of Hans-Jürgen lieved to have been responsible for most U.S. psy- Walter (1991). Wertheimer wrote only one book, chologists erroneously assuming that the Gestaltists Productive Thinking, but he died suddenly on were interested only in perception. The truth was October 12, 1943, of a coronary embolism at his that, besides perception, the Gestaltists were inter- home in New Rochelle, New York, before it was ested in many philosophical issues as well as in published. Productive Thinking was published post- learning and thinking. The reason for their early humously in 1945. In October 1988, the German concentration on perception was that Wundt had Society for Psychology bestowed upon Wertheimer been concentrating on perception, and he was the its highest honor, the Wilhelm Wundt Plaque. primary focus of their attack.
GESTALT PSYCHOLOGY 461 Psychologist Ronald Ley (1990) suggests that Köhler did more than observe chimpanzees on Tenerife. The Canary Islands are an unlikely place to establish an anthropoid research station because Medicine chimpanzees are not native to the region. The of German Cameroons (a German colony in Africa) Library or a large zoo in Germany would have been more National logical locations. Ley speculates that Köhler’s reason for being in such a remote place was to observe the British shipping activity for the German military. of With a carefully concealed radio, Köhler informed Courtesy German military officials whether or not British vessels were in the vicinity. If they were not, Wolfgang Köhler German ships could safely be refueled by nearby fuel ships. These activities were confirmed by In 1921 Koffka published an important book on Manuel, the 87-year-old keeper, handler, and child psychology, later translated into English as The trainer of Köhler’s animals, and by two of Growth of the Mind: An Introduction to Child Psychology Köhler’s children. Ley also provides documents (1924). In 1935, Koffka published Principles of Gestalt from both German and British naval archives that Psychology, which was intended to be a complete, confirm an active espionage organization in the systematic presentation of Gestalt theory. The latter Canary Islands during World War I. Furthermore, book was dedicated to Köhler and Wertheimer in the British documents indicate that Köhler was gratitude for their friendship and inspiration. strongly suspected of being part of that organiza- tion. Several times Köhler’s home was searched by Spanish authorities on the orders of the British gov- Wolfgang Köhler ernment. If these charges are true, it indicates that, Wolfgang Köhler (1887–1967) was born on at the time, Köhler was a loyal citizen of Germany. January 21 in Reval (now Tallinn), Estonia, and As we shall see, this loyalty was to change dramati- received his doctorate in 1909 from the cally when the Nazis came to power. University of Berlin. Like Koffka, Köhler worked Upon his return to Germany, Köhler accepted under the supervision of Stumpf. In 1909 Köhler a professorship at the University of Göttingen went to the University of Frankfurt, where a year (1921–1922), and in 1922 he succeeded Stumpf as later he would participate with Wertheimer and director of the Psychological Institute at the Koffka in the research that was to launch the University of Berlin. This was a prestigious Gestalt movement. Köhler’s collaboration with appointment, and it gave Gestalt psychology inter- Koffka and Wertheimer was temporarily inter- national recognition. Köhler’s directorship was in- rupted when, in 1913, the Prussian Academy of terrupted twice by trips to the United States: He Sciences invited him to go to its anthropoid station was a visiting professor at Clark University (1925– on Tenerife, one of the Canary Islands, to study 1926), a William James lecturer at Harvard (1934– chimpanzees. Shortly after his arrival, World War I 1935), and a visiting professor at the University of began, and his stay on Tenerife was prolonged for Chicago (1935). His Gestalt Psychology (1929/1970) seven years. While at the anthropoid station, Köhler was written in English and was especially intended concentrated his study on the nature of learning in for U.S. psychologists. chimpanzees. He summarized his observations in the Like James, Köhler was highly critical of Mentality of Apes (1917/1925). Fechner and offered psychophysics as an example
462 CHAPTER 14 of what could happen if measurement precedes an understanding of what is being measured: Text not available due to copyright restrictions Text not available due to copyright restrictions Back in Germany, the Nazis were harassing in- stitutions of higher learning and professors, and Köhler believed that U.S. psychologists were Köhler’s attitude toward the fatherland changed making a similar mistake in their widespread accep- dramatically. Köhler complained bitterly and, on tance of operationism (see Chapter 13). He gave as April 28, 1933, published the last article that pub- an example the operational definition of intelli- licly criticized the Nazis. In the following excerpt gence in terms of performance on intelligence tests. from that article, Köhler, who was not Jewish, Here, he said, the measurements are precise (as they commented on the Nazis’ wholesale dismissal of were in Fechner’s work), but it is not clear exactly Jews from universities and other positions: what is being measured. In the quotation that fol- During our conversation, one of my lows, note the similarity between Köhler’s friends reached for the Psalms and read: (1929/1970) criticisms of the use of IQ tests and “The Lord is my shepherd, I shall not those of Binet (see Chapter 10): want.” He read the 90th Psalm and said, “It is hard to think of a German who has been able to move human hearts more deeply and so to console those who suffer. And these words we have received from the Jews.” Another reminded me that never had a man struggled more nobly for a clarifi- cation of his vision of the world than the Jew Spinoza, whose wisdom Goethe ad- mired. My friend did not hesitate to show Text not available due to copyright restrictions respect, as Goethe did. Lessing, too, would not have written his Nathan the Wise unless human nobility existed among the Jews.… It seems that nobody can think of the great work of Heinrich Hertz without an almost affectionate admiration for him. And Hertz had Jewish blood. One of my friends told me: “The greatest German experimental physicist of the present time is Franck; many believe that he is the greatest experimental
GESTALT PSYCHOLOGY 463 physicist of our age. Franck is a Jew, an In 1925, soon after my first arrival in this unusually kind human being. Until a few country, I had a curious experience. When days ago, he was professor at Göttingen, an once talking with a graduate student of honor to Germany and the envy of the psychology who was, of course, a behav- international scientific community.” iorist, I remarked that McDougall’s psy- [Perhaps the dismissal of Franck] shows the chology of striving seemed to me to be deepest reason why all these people are not associated with certain philosophical theses joining [the Party]: they feel a moral im- which I found it hard to accept; but that he position. They believe that only the qual- might nevertheless be right in insisting ity of a human being should determine his that, as a matter of simple observation, worth, that intellectual achievement, people do this or that in order to reach character, and obvious contributions to certain goals. Did not the student himself German culture retain their significance sometimes go to a post office in order to whether a person is Jewish or not. (Henle, buy stamps? And did he not just now 1978, p. 940) prepare himself for certain examinations to be held next Thursday? The answer was Eventually, the Nazi menace became too un- prompt: “I never do such things,” said the bearable, and in 1935 Köhler immigrated to the student. There is nothing like a solid sci- United States. After lecturing at Harvard for a entific conviction. (Henle, 1986, p. 120) year, he accepted an appointment at Swarthmore College, in Pennsylvania, where he remained until Köhler’s many honors included membership in his retirement in 1958. While at Swarthmore, he the American Philosophical Society, National published his William James lectures as The Place Academy of Sciences, and the American Academy of Value in a World of Facts (1938) and Dynamics in of Arts and Sciences; numerous honorary de- Psychology (1940), in which he discussed the rela- grees; being declared an Ehrenbürger (honorary citi- tionship between field theory in physics and zen) of the University of Berlin (an honor Gestalt psychology. After retiring, Köhler moved previously given to only two Americans—John F. to New Hampshire, where he continued his writ- Kennedy and Paul Hindesmith); the American ing and research at Dartmouth College. He also Psychological Association’s Distinguished Scientific spent considerable time lecturing at European uni- Contributions Award (1956); and the presidency of versities. Köhler died in Enfield, New Hampshire, the American Psychological Association (APA) on June 11, 1967. His last book, The Task of Gestalt (1959). Psychology (1969), was published posthumously. Gestalt psychology became highly influential in the United States. When it is realized that Koffka ISOMORPHISM AND THE LAW was at Smith College (an undergraduate institu- tion), Köhler was at Swarthmore (an undergraduate OF PRÄGNANZ institution), and Wertheimer was affiliated with the New School for Social Research (which was not A basic question Wertheimer had to answer was yet granting advanced degrees), the success of how only two stimuli could cause the perception Gestalt psychology in the United States is especially of motion. As previously noted, Wertheimer did impressive. Also, behaviorism was the dominant not discover apparent motion; however, his explana- theme in U.S. psychology as the Gestaltists were tion of the phenomenon was unique. As we have attempting to make inroads. Köhler described an seen, Mach, Ehrenfels, and J. S. Mill all recognized experience he had shortly after arriving in the that the whole was sometimes different from the United States: sum of its parts, but they all assumed that somehow
464 CHAPTER 14 peared to fall to the left and right simultaneously, and because the eyes could not move in two direc- tions at the same time, an explanation based on sensations from the eye muscles was untenable. Application of Field Theory If the experience of psychological phenomena could not be explained by sensory processes, infer- ences, or fusions, how could it be explained? The Gestaltists’ answer was that the brain contains struc- the whole (Gestalt) emerged from the characteristics tured fields of electrochemical forces that exist prior of the parts. That is, after the parts (elements) are at- to sensory stimulation. Upon entering such a field, tended to, they somehow fuse and give rise to the sensory data both modify the structure of the field whole experience. For example, attending to the and are modified by it. What we experience con- primary colors causes the sensation of white to sciously results from the interaction of the sensory emerge, and attending to several musical notes data and the force fields in the brain. The situation causes the sensation of melody to emerge. This is similar to one in which metal particles are placed viewpoint still depends on a form of elementism into a magnetic field. The nature of the field will and its related assumption of association. For exam- have a strong influence on how the particles are ple, Wundt’s explanation of apparent movement distributed, but the characteristics of the particles was that the fixation of the eyes changed with will also influence the distribution. For example, each successive presentation of the visual stimulus, larger, more numerous particles will be distributed and this causes the muscles controlling the eyes to differently within the field than smaller, less numer- give off sensations identical to those given off when ous particles. In the case of cognitive experience, real movement is experienced. Thus, because of past the important point is that fields of brain activity experience with such sensations (association), one transform sensory data and give that data character- experiences what appears to be movement. istics it otherwise would not possess. According to Because with apparent movement the sensation of this analysis, the whole (electrochemical force fields movement is not contained in the sensations in the brain) exists prior to the parts (individual that cause it, Wundt believed that the experience sensations), and it is the whole that gives the parts exemplifies creative synthesis. Similarly, Helmholtz their identity or meaning. explained the phenomenon as an unconscious infer- ence. Both Wundt and Helmholtz emphasized the Psychophysical Isomorphism role of learning in experiences like the phi phenomenon. To describe more fully the relationship between the Through an ingenious demonstration, however, field activity of the brain and conscious experience, Wertheimer showed that explanations based on the Gestaltists introduced the notion of psycho- learning were not plausible. Again using a tachisto- physical isomorphism, which Köhler described scope, he showed that the phi phenomenon could as follows: “Experienced order in space is always occur in two directions at the same time. Three lights structurally identical with a functional order in the were arranged as shown in the diagram below: distribution of underlying brain processes” (1929/ The center light was flashed on, and shortly 1970, p. 61). Elsewhere, Köhler said, “Psychological thereafter the two other lights were flashed on, facts and the underlying events in the brain resem- both at the same time. Wertheimer repeated this se- ble each other in all their structural characteristics” quence several times. The center slit of light ap- (1969, p. 66).
GESTALT PSYCHOLOGY 465 The Gestalt notion of isomorphism stresses that The Gestaltists totally disagreed with the con- the force fields in the brain transform incoming sen- ception of brain functioning implied by the con- sory data and that it is the transformed data that we stancy hypothesis. By rejecting the constancy experience consciously. The word isomorphism hypothesis, the Gestaltists rejected the empirical comes from the Greek iso (“similar”) and morphic philosophy on which the schools of structuralism, (“shape”). The patterns of brain activity and the pat- functionalism, and behaviorism were based. Instead, terns of conscious experience are structurally equiv- as we have seen, the Gestaltists employed field the- alent. The Gestaltists did not say that patterns of ory in their analysis of brain functioning. In any electrochemical brain activity are the same as pat- physical system, energy is distributed in a lawful terns of perceptual activity. Rather, they said that way, and the brain is a physical system. Köhler perceptual fields are always caused by underlying said, “According to several physicists the distribu- patterns of brain activity. It was believed that, al- tion of materials and processes in physical systems though the patterns of perceptual and brain activity tends to become regular, simple, and often sym- might have some similarity, the two represent two metrical when the systems approach a state of equi- totally different domains and certainly cannot be librium or a steady state” (1969, pp. 64–65). identical. The relationship is like that between a Michael Wertheimer (1987) elaborates this point: map of the United States and the actual United The Gestaltists argue that physical forces, States; although the two are related in important when released, do not produce chaos, but ways, they are hardly identical. their own internally determined organiza- tion. The nervous system, similarly, is not Opposition to the Constancy characterized by machinelike connections of tubes, grooves, wires, or switchboards, Hypothesis but the brain too, like almost all other With their notion of isomorphism, the Gestaltists physical systems, exhibits the dynamic self- opposed the constancy hypothesis, according to distribution of physical forces. (p. 137) which there is a one-to-one correspondence be- tween certain environmental stimuli and certain Thus, instead of viewing the brain as a passive sensations. This one-to-one correspondence did receiver and recorder of sensory information, the not mean that sensations necessarily reflect accu- Gestaltists viewed the brain as a dynamic configura- rately what is present physically. The psychophysi- tion of forces that transforms sensory information. cists, Helmholtz, Wundt, and the structuralists all They believed that incoming sensory data interacts accepted the constancy hypothesis while recogniz- with force fields within the brain to cause fields of ing that large discrepancies could exist between mental activity; and like the underlying physical psychological experiences and the physical events fields in the brain, these mental fields are organized that cause them. Rather, the constancy hypothesis configurations. The nature of the mental configura- contended that individual physical events cause in- tions depends on the totality of the incoming stim- dividual sensations and that these sensations remain ulation and the nature of the force fields within the isolated unless acted on by one or more of the laws brain, and any configurations that occur in the fields of association or, in Wundt’s case, are intentionally of brain activity would be experienced as percep- rearranged. This hypothesis was accepted by most tions (psychophysical isomorphism). British and French empiricists and was the corner- stone of Titchener’s structuralism. The structuralists, Analysis: Top Down, Not Bottom Up following in the tradition of empiricism, viewed mental events as the passive reflections of specific According to the Gestaltists, organized brain activ- environmental events. ity dominates our perceptions, not the stimuli that
466 CHAPTER 14 enter into that activity. For this reason, the whole is psychophysical isomorphism, mental experiences, more important than the parts, thus reversing one too, must be simple and symmetrical. The of psychology’s oldest traditions. The Gestaltists said Gestaltists summarized this relationship between that their analysis proceeded from the top to the bottom force fields in the brain and cognitive experience instead of from the bottom to the top, as had been the tra- with their law of Prägnanz, which was central to dition. In other words, they proceeded from the Gestalt psychology. The German word Prägnanz wholes to the parts instead of from the parts to the has no exact English counterpart, but an approxi- wholes. As Michael Wertheimer (1987) explains, mation is “essence.” Prägnanz refers to the essence or ultimate meaning of an experience. Sensory in- This formulation involved a radical reori- formation may be fragmented and incomplete, but entation: the nature of the parts is deter- when that information interacts with the force mined by the whole rather than vice versa; fields in the brain, the resultant cognitive experi- therefore analysis should go “from above ence is complete and organized. The law of down” rather than “from below up.” One Prägnanz states that psychological organization should not begin with elements and try to will always be as good as conditions allow because synthesize the whole from them, but study fields of brain activity will always distribute them- the whole to see what its natural parts are. selves in the simplest way possible under the pre- The parts of a whole are not neutral and vailing conditions, just as other physical force fields inert, but structurally intimately related to do. The law of Prägnanz asserts that all cognitive one another. That parts of a whole are not experiences will tend to be as organized, symmet- indifferent to one another was illustrated, rical, simple, and regular as they can be, given the for example, by a soap bubble: change of pattern of brain activity at any given moment. This one part results in a dramatic change in the is what “as good as conditions allow” means. entire configuration. This approach was It is tempting to categorize Gestalt psychology applied to the understanding of a wide as nativistic, but the Gestaltists themselves disagreed variety of phenomena in thinking, learn- with that. Köhler said, “Such concepts as genes, ing, problem solving, perception, and inherited, and innate should never be mentioned philosophy, and the movement developed when we refer to the basic … dynamic … processes and spread rapidly, with violent criticisms in the nervous system” (1969, p. 89). According to against it from outside, as well as equally Köhler, what governs brain activity are not geneti- vehement attacks on the outsiders from cally controlled programs but the invariant dynamics inside. (p. 136) that govern all physical systems. According to Henle (1986), it is time for psy- chology to follow the lead of the Gestaltists and The Law of Prägnanz stop attempting to explain everything in terms of the nativism-empiricism dichotomy: The Gestaltists believed that the same forces that create configurations such as soap bubbles and I do not know why we find it so difficult magnetic fields also create configurations in the to break out of the nativism–empiricism brain. The configurations of energy occurring dichotomy. Are we unable to think in in all physical systems always result from the total terms of trichotomies? If we are, we will field of interacting forces, and these physical continue to misinterpret Gestalt psychol- forces always distribute themselves in the most sim- ogy and—more serious—our explanations ple, symmetrical way possible under the circum- will not do justice to our subject matter. stances. Therefore, according to the principle of (p. 123)
GESTALT PSYCHOLOGY 467 PER CEPTUAL CONSTANCIES meaning, of objects, of shapes or patterns. Such is the traditional associationist view of Perceptual constancy (not to be confused with the matter. (pp. 105–106) the constancy hypothesis) refers to the way we re- spond to objects as if they are the same, even The Gestaltists disagreed. Köhler, for example, though the actual stimulation our senses receive asserted that the constancies are a direct reflection of may vary greatly: ongoing brain activity and not a result of sensation plus learning. The reason we experience an object as the same under varied conditions is that the rela- tionship between that object and other objects re- mains the same. Because this relationship is the same, the field of brain activity is also the same, and therefore the mental experience (perception) is the same. The Gestaltists’ explanation, then, is simply an extension of the notion of psychophysical isomorphism. Using brightness constancy as an ex- ample, Bruno (1972) nicely summarizes this point: Text not available due to copyright restrictions [Köhler] said that brightness constancy is due to the existence of a real constancy that is an existing Gestalt in the environ- ment. This Gestalt is physical—really there as a pattern. It is the ratio of brightness of the figure to the brightness of the ground. This ratio remains constant for sunlight and shade. Let us say that a light meter gives a reading of 10 (arbitrary units) for a bikini in the sun. A reading from the grass in the sun is 5. The ratio of figure to ground is 10/5; or 2. Assume now that the girl in the bikini is in the shade, and the light meter gives a reading of 4 for the bikini. The The empiricists explained perceptual constan- grass in the shade gives a reading of 2. The cies as the result of learning. The sensations pro- ratio of figure to ground is 4/2; or 2—the vided by objects seen at different angles, positions, same ratio as before. The ratio is a con- and levels of illumination are different, but through stant. The human nervous system responds experience we learn to correct for these differences directly to this constant ratio. The constant and to respond to the objects as the same. ratio in the environment gives rise to a Woodworth (1931) described what our perceptions pattern of excitation in the nervous system. would be like, according to the empiricists, if the As long as the ratio does not change, the influence of learning could be removed: characteristics of the pattern of excitation If we could for a moment lay aside all that do not change. Thus Köhler explained we had learned and see the field of view brightness constancy as a directly perceived just as the eyes present it, we should see a Gestalt not derived from learning or the mere mosaic of variegated spots, free of association of sensations.
468 CHAPTER 14 Köhler explained other perceptual The Gestaltists made the figure–ground rela- constancies involving color, shape, and size tionship a major component of their theoretical in a similar manner. (p. 151) system. Gestalt Principles of Perceptual Organization PERCEPTUAL GESTALTEN In addition to describing figure–ground perception, Through the years, the Gestaltists have isolated over the Gestaltists described the principles by which the 100 configurations (Gestalten) into which visual in- elements of perception are organized into config- formation is arranged. We will sample only a few of urations. For example, stimuli that have continuity them here. with one another will be experienced as a percep- tual unit. To describe this principle, Wertheimer used the terms intrinsic togetherness, imminent necessity, The Figure–Ground Relationship and good continuation. Figure 14.2a provides an ex- According to Danish psychologist Edgar Rubin ample of this principle of continuity. Note that (1886–1951), the most basic type of perception is the pattern that emerges cannot be found in any the division of the perceptual field into two parts: particular dot (element). Rather, because some the figure, which is clear and unified and is the ob- dots seem to be tending in the same direction, ject of attention, and the ground, which is diffuse one responds to them as a configuration (Gestalt). and consists of everything that is not being attended Most people would describe this figure as consisting to. Such a division creates what is called a figure– of two curved lines. ground relationship. Thus, what is the figure and When stimuli are close together, they tend to what is the ground can be changed by shifting one’s be grouped together as a perceptual unit. This is attention. Figure 14.1 demonstrates this phenome- known as the principle of proximity. In Figure non. When one focuses attention on the two pro- 14.2b, the Xs tend to be seen in groups of two, files, one cannot see the vase, and vice versa. instead of as individual Xs. The same is true of the Similarly, when one focuses attention on the black lines. cross, one cannot see the white cross, and vice According to the principle of inclusiveness, versa. when there is more than one figure, we are most F I G U R E 14.1 In each illustration, which is the figure and which is the ground? SOURCE: Adapted from Rubin, 1915/1921.
GESTALT PSYCHOLOGY 469 F I G U R E 14.2 Examples of (a) principle of continuity, (b) principle of proximity, (c) principle of inclusive- ness (Köhler, 1969), (d) principle of similarity, and (e) principle of closure. SOURCE: Sartain et al., 1973; used by permission of Prentice-Hall, Inc. likely to see the figure that contains the greatest Köhler observed that if perception is deter- number of stimuli. If, for example, a small figure mined by past experience (learning), then most is embedded in a larger one, we are most likely to people would perceive the familiar word “men” see the larger figure and not the smaller. The use of along with its mirror image in the figure. Instead, camouflage is an application of this principle. For however, most people perceive a less familiar figure, example, ships painted the color of water and tanks which somewhat resembles a horizontal row of painted the color of the terrain in which they op- heart-shaped forms. erate blend into the background and are thus less Objects that are similar in some way tend to susceptible to detection. In Figure 14.2c, the sym- form perceptual units. This is known as the princi- bol √16 is difficult to see because so many of its ple of similarity. Twins, for example, stand out in components are part of a larger stimulus complex. a crowd, and teams wearing different uniforms Köhler believed that the principle of inclusiveness stand out as two groups on the field. In Figure provided evidence against the empiricalistic expla- 14.2d, the stimuli that have something in common nation of perception. He said most people would stand out as perceptual units. clearly have much more experience with the sym- bol √16 than with the figure shown in Figure 14.2c. Yet, the stronger tendency is to perceive the more inclusive figure. Köhler (1969) made the same point with the following figure:
470 CHAPTER 14 As we have seen, the Gestaltists believed in stranger with surprise and asked him psychophysical isomorphism, according to which whence he came. The man pointed in the our conscious experience is directly related to pat- direction straight away from the inn, terns of brain activity, and the brain activity orga- whereupon the landlord, in a tone of awe nizes itself into patterns according to the law of and wonder, said: “Do you know that you Prägnanz. Thus, it is quite likely that the patterns have ridden across the Lake of of brain activity are often better organized than the Constance?” at which the rider dropped stimuli that enter them. This is clearly demonstrated stone dead at his feet. in the principle of closure, according to which In what environment, then, did the incomplete figures in the physical world are per- behavior of the stranger take place? The ceived as complete ones. As Figure 14.2e shows, Lake of Constance? Certainly, because it is even if figures have gaps in them—and thus are a true proposition that he rode across it. not truly circles, triangles, or rectangles—they are And yet, this is not the whole truth, for the nonetheless experienced as circles, triangles, or rec- fact that there was a frozen lake and not tangles. This is because the brain transforms the sti- ordinary solid ground did not affect his muli into organized configurations that are then behavior in the slightest. It is interesting for experienced cognitively. For the same reason, in the geographer that this behavior took Figure 14.2e we see a person on horseback. place in this particular locality, but not for the psychologist as the student of behavior; because the behavior would have been just the same had the man ridden across a SUBJECTIVE AND OBJECTIVE barren plain. But the psychologist knows something more: since the man died from REALITY sheer fright after having learned what he had “really” done, the psychologist must Because the brain acts on sensory information and conclude that had the stranger known be- arranges it into configurations, what we are con- fore, his riding behavior would have been scious of, and therefore what we act in accordance very different from what it actually was. with at any given moment, is more a product of the Therefore the psychologist will have to brain than of the physical world. Koffka used this say: there is a second sense to the word fact to distinguish between the geographical and the environment according to which our behavioral environments. For him, the geograph- horseman did not ride across the lake at all, ical environment is the physical environment, but across an ordinary snow-swept plain. whereas the behavioral environment is our sub- His behavior was a riding-over-a-plain, jective interpretation of the geographical environ- but not a riding-over-a-lake. ment. Koffka (1935/1963) used an old German What is true of the man who rode legend to illustrate the important difference be- across the Lake of Constance is true of tween the two environments: every behavior. Does the rat run in the On a winter evening amidst a driving maze the experimenter has set up? According snowstorm a man on horseback arrived at to the meaning of the word “in,” yes and an inn, happy to have reached a shelter no. Let us therefore distinguish between a after hours of riding over the wind-swept geographical and a behavioral environment. plain on which the blanket of snow had Do we all live in the same town? Yes, covered all paths and landmarks. The when we mean the geographical, no, landlord who came to the door viewed the when we mean the behavioral. (pp. 27–28)
GESTALT PSYCHOLOGY 471 that keeps the organism active until it solves the problem. Typically, an organism solves its problems perceptually by scanning the environment and cog- nitively trying one possible solution and then an- other until it reaches a solution. Thus, the Gestaltists emphasized cognitive trial and error as op- posed to behavioral trial and error. They believed that organisms come to see solutions to problems. Insightful Learning Köhler did much of his work on learning between 1913 and 1917 when he was on the island of Tenerife during World War I. In a typical experi- ment, using apes as subjects, Köhler suspended a desired object—for example, a banana—in the air just out of the animal’s reach. Then he placed ob- jects such as boxes and sticks, which the animal could use to obtain the banana, in the animal’s en- vironment. By stacking one or more boxes under the banana or by using a stick, the animal could obtain the banana. In one case, the animal needed a. An ape named Chica using a pole to obtain food (p. 72a). to join two sticks together in order to reach a ba- SOURCE: All photos from The Mentality of Apes by W. Köhler (1917/1925), London: Routledge and Kegan Paul Ltd. Reproduced by permission of Taylor & nana. The photographs on the next page depict the Francis Book UK. problem-solving activities of some of Köhler’s apes. In studying learning, Köhler also employed so- In other words, our own subjective reality gov- called detour problems, problems in which the erns our actions more than the physical environ- animal could see its goal but could not reach it ment does. directly. To solve the problem, the animal had to TH E GESTALT EXPLANATION OF LEARNING Cognitive Trial and Error As we have seen, the Gestaltists believed that brain activity tends toward a balance, or equilibrium, in accordance with the law of Prägnanz. This ten- dency toward equilibrium continues naturally un- less it is somehow disrupted. According to the Gestaltists, the existence of a problem is one such disruptive influence. If a problem is confronted, a F I G U R E 14.3 state of disequilibrium exists until the problem is A typical detour problem that Köhler used to study the solved. Because a state of disequilibrium is unnatu- learning process. ral, it creates a tension with motivational properties SOURCE: Köhler (1917/1925).
472 CHAPTER 14 b. Sultan putting two sticks together to obtain food (p. 128a). d. Chica beating down her objective with a pole (p. 146a). learn to take an indirect route to the goal. Figure 14.3 shows a typical detour problem. Köhler found that although chickens had great difficulty with such problems, apes solved them with ease. Köhler noted that during a problem’s presolu- tion period, the animals appeared to weigh the sit- uation—that is, to test various hypotheses. (This is what we referred to earlier as cognitive, or vicari- ous, trial and error.) Then, at some point, the ani- mal achieved insight into the solution and behaved according to that insight. For the Gestaltists, a prob- lem can exist in only two stages: It is either un- solved or solved—there is no in-between. According to the Gestaltists, the reason that Thorndike and others had found what appeared to be incremental learning (learning that occurs gradually) was that all ingredients necessary for the attainment of insight had not been available to the animal. But if a problem is presented to an organism c. An ape named Grande using a stack of boxes to obtain along with those things necessary for the problem’s food as Sultan watches (p. 138a). solution, insightful learning typically occurs.
GESTALT PSYCHOLOGY 473 According to the Gestaltists, insightful learning is the experiment, Köhler had presented a sheet of much more desirable than learning achieved paper of a lighter gray than the one on which the through either rote memorization or behavioral chickens had been reinforced, the chickens would trial and error. Hergenhahn and Olson (2005) sum- have continued to approach the sheet on which marize the conclusions that the Gestaltists reached they had previously been fed because it would about insightful learning: have been the darker of the two. Thus, for the Gestaltist, an organism learns Insightful learning is usually regarded as principles or relationships, not specific responses to having four characteristics: (1) the transi- specific situations. Once it learns a principle, the tion from presolution to solution is sudden organism applies it to similar situations. This was and complete; (2) performance based on a called transposition, Gestalt psychology’s explana- solution gained by insight is usually tion of transfer of training. The notion of transpo- smooth and free of errors; (3) a solution to sition is contrary to Thorndike’s identical-elements a problem gained by insight is retained for theory of transfer, according to which the similarity a considerable length of time; (4) a prin- (common elements) between two situations deter- ciple gained by insight is easily applied to mines the amount of transfer between them. other problems. (p. 276) The Behaviorists’ Explanation of Transposition. Transposition The Gestalt theory explanation of transposition did not go unchallenged. Kenneth Spence, the To explore further the nature of learning, Köhler major spokesman for Hullian psychology, came used chickens as subjects. In one experiment, he up with an ingenious alternative explanation. placed a white sheet and a gray sheet of paper on Hergenhahn and Olson (2005) summarize Spence’s the ground and covered both with grain. If a explanation: chicken pecked at the grain on the white sheet, it was shooed away; but if it pecked at the grain on Suppose, said Spence, that an animal is the gray sheet, it was allowed to eat. After many reinforced for approaching a box whose lid trials, the chickens learned to peck at the grain on measures 160 sq. cm., and not reinforced only the gray sheet. The question is, What did the for approaching a box whose lid measures animals learn? Thorndike, Hull, and Skinner would 100 sq. cm. Soon the animal will learn to say that reinforcement strengthened the response of approach the larger box exclusively. In eating off the gray paper. To answer the question, phase two of this experiment, the animal Köhler proceeded with phase two of the experi- chooses between the 160 sq. cm. box and ment: He replaced the white paper with a sheet the box whose lid is 256 sq. cm. The an- of black paper. Now the choice was between a imal will usually choose the larger box gray sheet of paper, the one for which the chickens (256 sq. cm.) even though the animal had had received reinforcement, and a black sheet. been reinforced specifically for choosing Given this choice, most reinforcement theorists the other one (160 sq. cm.) during phase would have predicted that the chickens would con- one. This finding seems to support the tinue to approach the gray paper. The vast majority relational learning point of view. of the chickens, however, approached the black pa- Spence’s behavioristic explanation of per. Köhler’s explanation was that the chickens had transposition is based on generalization.… not learned a stimulus-response association or a Spence assumed that the tendency to ap- specific response but a relationship. In this case, the proach the positive stimulus (160 sq. cm.) animals had learned to approach the darker of the generalizes to other related stimuli. two sheets of paper. If, in the second phase of Second, he assumed that the tendency to
474 CHAPTER 14 approach the positive stimulus (and the PRODUCTIVE THINKING generalization of this tendency) is stronger than the tendency to avoid the negative Wertheimer was concerned with the application of stimulus (and the generalization of this Gestalt theory to education. As mentioned, his book tendency). What behavior occurs will be Productive Thinking was published posthumously in determined by the algebraic summation of 1945. Under the editorship of Wertheimer’s son the positive and negative tendencies. Michael, this book was later revised and expanded, Whenever there is a choice between and it was republished in 1959. The conclusions two stimuli, the one eliciting the greatest Wertheimer reached about productive thinking net approach tendency will be chosen. In were based on personal experience, experimenta- the first phase of Spence’s experiment, the tion, and interviews with individuals considered ex- animal chose the 160 sq. cm. box over the cellent problem solvers, such as Einstein. 100 sq. cm. box because the net positive Those were wonderful days, beginning in tendency was 51.7 for the former and 29.7 1916, when for hours and hours I was for the latter. In phase two, the 256 sq. cm. fortunate enough to sit with Einstein, box was chosen over the 160 sq. cm. box alone in his study, and hear from him the because the net positive tendency was 72.1 story of the dramatic developments which for the former and still 51.7 for the latter. culminated in the theory of relativity. (pp. 279–280; see Figure 14.4) (Max Wertheimer, 1945/1959, p. 213) Spence’s explanation had the advantage of pre- Wertheimer contrasted learning according to dicting the circumstances under which transposition Gestalt principles with rote memorization governed would not occur. As the matter stands today, nei- by external reinforcement and the laws of associa- ther the Gestalt nor the behaviorist explanations can tion. The former is based on an understanding of account for all transpositional phenomena; there- the nature of the problem. As we have seen, the fore, a comprehensive explanation is still being existence of a problem creates a cognitive disequi- sought. librium that lasts until the problem is solved. The solution restores a cognitive harmony, and this res- toration is all the reinforcement the learner needs. F I G U R E 14.4 According to Spence, the algebraic sum of the positive and negative influences deter- mines which of two stimuli in a discrimination problem will be approached. SOURCE: Spence (1942).
GESTALT PSYCHOLOGY 475 Because learning and problem solving are person- Wertheimer’s Productive Thinking is filled with ally satisfying, they are governed by intrinsic delightful examples of productive problem solving. (internal) reinforcement rather than extrinsic One involves a childhood experience of Carl (external) reinforcement. Wertheimer thought Friedrich Gauss, who went on to become a famous that we are motivated to learn and to solve pro- mathematician. Gauss’s teacher asked the class to blems because it is personally satisfying to do so, add the numbers from 1 through 10 and report not because someone or something else reinforces the sum as soon as it was attained. While the other us for doing so. Because learning governed by students were just beginning to solve the problem, Gestalt principles is based on an understanding of Gauss raised his hand and correctly reported the the structure of the problem, it is easily remem- sum as 55. When the teacher asked Gauss how he bered and generalized to other relevant situations. arrived at the answer so quickly, he said, Wertheimer believed that some learning did [H]ad I done it by adding 1 and 2, then 3 occur when mental associations, memorization, to the sum, then 4 to the new result, and drill, and external reinforcement are employed but so on, it would have taken very long; and, that such learning is usually trivial. He gave as ex- trying to do it quickly, I would very likely amples of such learning associating a friend’s name have made mistakes. But you see, 1 and 10 with his or her telephone number, learning to an- make eleven, 2 and 9 are again—must be ticipate correctly a list of nonsense syllables, and a —11! And so on! There are 5 such pairs; 5 dog learning to salivate to a certain sound. times 11 makes 55. (Wertheimer, Unfortunately, according to Wertheimer, this is 1945/1959, p. 109) the type of learning that most schools emphasize. In Wertheimer’s analysis, teaching that empha- Gauss’s solution was based on a flexible, crea- sizes logic does not fare much better than rote tive approach to the problem rather than on stan- memorization. Supposedly, logic guarantees that dard, mechanical rules. Similarly, Michael one will reach correct conclusions. Teaching based Wertheimer (1980) describes an experiment that on such a notion, said Wertheimer, assumes that Katona originally performed in 1940. Katona there is a correct way to think and that everyone showed subjects the following 15 digits and told should think that way. But like rote memorization, them to study them for 15 seconds: learning and applying the rules of logic stifle pro- ductive thinking because neither activity is based on 149162536496481 the realization that problem solving involves the With only these instructions, most people at- total person and is unique to that person: tempt to memorize as many digits as possible in the According to Wertheimer, reaching an allotted time. Indeed, Katona found that most sub- understanding involves many aspects of jects could reproduce only a few of the numbers learners, such as their emotions, attitudes, correctly; and when tested a week later, most sub- and perceptions, as well as their intellects. jects remembered none. In gaining insight into the solution to a Katona asked another group of subjects to look problem, a student need not—in fact, for a pattern or theme running through the num- should not—be logical. Rather, the bers. Some individuals in this group realized that student should cognitively arrange and re- the 15 digits represented the squares of the digits arrange the components of the problem from 1 to 9. These subjects saw a principle that until a solution based on understanding is they could apply to the problem and were able to reached. Exactly how this process is done reproduce all numbers correctly, not only during will vary from student to student. the experiment but also for weeks after. In fact, (Hergenhahn and Olson, 2005, p. 281) those individuals could no doubt reproduce the
Search
Read the Text Version
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
- 19
- 20
- 21
- 22
- 23
- 24
- 25
- 26
- 27
- 28
- 29
- 30
- 31
- 32
- 33
- 34
- 35
- 36
- 37
- 38
- 39
- 40
- 41
- 42
- 43
- 44
- 45
- 46
- 47
- 48
- 49
- 50
- 51
- 52
- 53
- 54
- 55
- 56
- 57
- 58
- 59
- 60
- 61
- 62
- 63
- 64
- 65
- 66
- 67
- 68
- 69
- 70
- 71
- 72
- 73
- 74
- 75
- 76
- 77
- 78
- 79
- 80
- 81
- 82
- 83
- 84
- 85
- 86
- 87
- 88
- 89
- 90
- 91
- 92
- 93
- 94
- 95
- 96
- 97
- 98
- 99
- 100
- 101
- 102
- 103
- 104
- 105
- 106
- 107
- 108
- 109
- 110
- 111
- 112
- 113
- 114
- 115
- 116
- 117
- 118
- 119
- 120
- 121
- 122
- 123
- 124
- 125
- 126
- 127
- 128
- 129
- 130
- 131
- 132
- 133
- 134
- 135
- 136
- 137
- 138
- 139
- 140
- 141
- 142
- 143
- 144
- 145
- 146
- 147
- 148
- 149
- 150
- 151
- 152
- 153
- 154
- 155
- 156
- 157
- 158
- 159
- 160
- 161
- 162
- 163
- 164
- 165
- 166
- 167
- 168
- 169
- 170
- 171
- 172
- 173
- 174
- 175
- 176
- 177
- 178
- 179
- 180
- 181
- 182
- 183
- 184
- 185
- 186
- 187
- 188
- 189
- 190
- 191
- 192
- 193
- 194
- 195
- 196
- 197
- 198
- 199
- 200
- 201
- 202
- 203
- 204
- 205
- 206
- 207
- 208
- 209
- 210
- 211
- 212
- 213
- 214
- 215
- 216
- 217
- 218
- 219
- 220
- 221
- 222
- 223
- 224
- 225
- 226
- 227
- 228
- 229
- 230
- 231
- 232
- 233
- 234
- 235
- 236
- 237
- 238
- 239
- 240
- 241
- 242
- 243
- 244
- 245
- 246
- 247
- 248
- 249
- 250
- 251
- 252
- 253
- 254
- 255
- 256
- 257
- 258
- 259
- 260
- 261
- 262
- 263
- 264
- 265
- 266
- 267
- 268
- 269
- 270
- 271
- 272
- 273
- 274
- 275
- 276
- 277
- 278
- 279
- 280
- 281
- 282
- 283
- 284
- 285
- 286
- 287
- 288
- 289
- 290
- 291
- 292
- 293
- 294
- 295
- 296
- 297
- 298
- 299
- 300
- 301
- 302
- 303
- 304
- 305
- 306
- 307
- 308
- 309
- 310
- 311
- 312
- 313
- 314
- 315
- 316
- 317
- 318
- 319
- 320
- 321
- 322
- 323
- 324
- 325
- 326
- 327
- 328
- 329
- 330
- 331
- 332
- 333
- 334
- 335
- 336
- 337
- 338
- 339
- 340
- 341
- 342
- 343
- 344
- 345
- 346
- 347
- 348
- 349
- 350
- 351
- 352
- 353
- 354
- 355
- 356
- 357
- 358
- 359
- 360
- 361
- 362
- 363
- 364
- 365
- 366
- 367
- 368
- 369
- 370
- 371
- 372
- 373
- 374
- 375
- 376
- 377
- 378
- 379
- 380
- 381
- 382
- 383
- 384
- 385
- 386
- 387
- 388
- 389
- 390
- 391
- 392
- 393
- 394
- 395
- 396
- 397
- 398
- 399
- 400
- 401
- 402
- 403
- 404
- 405
- 406
- 407
- 408
- 409
- 410
- 411
- 412
- 413
- 414
- 415
- 416
- 417
- 418
- 419
- 420
- 421
- 422
- 423
- 424
- 425
- 426
- 427
- 428
- 429
- 430
- 431
- 432
- 433
- 434
- 435
- 436
- 437
- 438
- 439
- 440
- 441
- 442
- 443
- 444
- 445
- 446
- 447
- 448
- 449
- 450
- 451
- 452
- 453
- 454
- 455
- 456
- 457
- 458
- 459
- 460
- 461
- 462
- 463
- 464
- 465
- 466
- 467
- 468
- 469
- 470
- 471
- 472
- 473
- 474
- 475
- 476
- 477
- 478
- 479
- 480
- 481
- 482
- 483
- 484
- 485
- 486
- 487
- 488
- 489
- 490
- 491
- 492
- 493
- 494
- 495
- 496
- 497
- 498
- 499
- 500
- 501
- 502
- 503
- 504
- 505
- 506
- 507
- 508
- 509
- 510
- 511
- 512
- 513
- 514
- 515
- 516
- 517
- 518
- 519
- 520
- 521
- 522
- 523
- 524
- 525
- 526
- 527
- 528
- 529
- 530
- 531
- 532
- 533
- 534
- 535
- 536
- 537
- 538
- 539
- 540
- 541
- 542
- 543
- 544
- 545
- 546
- 547
- 548
- 549
- 550
- 551
- 552
- 553
- 554
- 555
- 556
- 557
- 558
- 559
- 560
- 561
- 562
- 563
- 564
- 565
- 566
- 567
- 568
- 569
- 570
- 571
- 572
- 573
- 574
- 575
- 576
- 577
- 578
- 579
- 580
- 581
- 582
- 583
- 584
- 585
- 586
- 587
- 588
- 589
- 590
- 591
- 592
- 593
- 594
- 595
- 596
- 597
- 598
- 599
- 600
- 601
- 602
- 603
- 604
- 605
- 606
- 607
- 608
- 609
- 610
- 611
- 612
- 613
- 614
- 615
- 616
- 617
- 618
- 619
- 620
- 621
- 622
- 623
- 624
- 625
- 626
- 627
- 628
- 629
- 630
- 631
- 632
- 633
- 634
- 635
- 636
- 637
- 638
- 639
- 640
- 641
- 642
- 643
- 644
- 645
- 646
- 647
- 648
- 649
- 650
- 651
- 652
- 653
- 654
- 655
- 656
- 657
- 658
- 659
- 660
- 661
- 662
- 663
- 664
- 665
- 666
- 667
- 668
- 669
- 670
- 671
- 672
- 673
- 674
- 675
- 676
- 677
- 678
- 679
- 680
- 681
- 682
- 683
- 684
- 685
- 686
- 687
- 688
- 689
- 690
- 691
- 692
- 693
- 694
- 695
- 696
- 697
- 698
- 699
- 700
- 701
- 702
- 703
- 704
- 705
- 706
- 707
- 708
- 709
- 710
- 711
- 712
- 713
- 714
- 715
- 716
- 717
- 718
- 719
- 720
- 721
- 722
- 723
- 724
- 725
- 726
- 727
- 728
- 729
- 730
- 731
- 732
- 733
- 734
- 735
- 736
- 737
- 738
- 739
- 740
- 741
- 742
- 743
- 744
- 745
- 746
- 747
- 748
- 749
- 750
- 751
- 752
- 753
- 1 - 50
- 51 - 100
- 101 - 150
- 151 - 200
- 201 - 250
- 251 - 300
- 301 - 350
- 351 - 400
- 401 - 450
- 451 - 500
- 501 - 550
- 551 - 600
- 601 - 650
- 651 - 700
- 701 - 750
- 751 - 753
Pages:
