Tolman, Edward C. (1886-1959)
TOLMAN, EDWARD C. (1886-1959)
The American psychologist Edward Chace Tolman was a forerunner of modern cognitive psychology; he showed that animals in learning mazes acquire organized spatial and temporal information about the maze and about the consequences of various alternative behaviors. In developing this approach, he was combating the dominant views of his time, which emphasized the acquisition of conditioned reflexes rather than knowledge about environmental events. Although several short biographies or reviews of Tolman's contributions are available (Crutchfield, 1961; Crutchfield et al., 1960; Hilgard, 1980; Innes, 1999, 2000; McFarland, 1993; Ritchie, 1964; Tolman, 1952), it is especially appropriate that one be included in an encyclopedia of learning and memory because workers in this field today are using ideas that were initiated and often developed by Tolman, although they do not necessarily recognize the source. Tolman's concepts and findings have helped to shape modern understanding of learning, memory, and cognition.
Tolman was born in Newton, Massachusetts, on April 14, 1886, into a prosperous family that valued hard work, high thinking, and social responsibility. After high school he attended the Massachusetts Institute of Technology, where his father served on the board of trustees. In his autobiography Tolman comments, "I went to MIT not because I wanted to be an engineer but because I had been good at mathematics and physics in high school and because of family pressure. After graduating from Technology (in electrochemistry), I became more certain of my own wants and transferred to Harvard for graduate work in philosophy and psychology" (1952, p. 323).
Among the experiences at Harvard that Tolman mentions as having influenced his later life were Ralph Barton Perry's course in ethics, which, he wrote, "laid the basis for my later interest in motivation and indeed gave me the main concepts (reinforced by a reading of McDougall's Social Psychology as part of the requirement of the course) which I have retained ever since; … Holt's seminar in epistemology in which I was introduced to and excited by the 'New Realism'; and Yerkes' course in comparative, using Watson's Behavior: An Introduction to Comparative Psychology, which was just out, as a text" (p. 325). Tolman also spent the summer of 1912 at the University of Giessen in Germany, where he studied with Kurt Koffka, one of the founders of Gestalt psychology.
In 1915 Tolman married Kathleen Drew and received his Ph.D. He then spent three years as an instructor at Northwestern University before accepting a position at the University of California at Berkeley in 1918. Except for brief periods, Tolman spent the rest of his life at Berkeley, where he had a distinguished scientific career and was an intellectual leader in the university community.
Early Experiments in Animal Learning
The line of research that occupied most of Tolman's life started when, on arriving in Berkeley, he found, he later wrote, that "it was up to me to suggest a new course. Remembering Yerkes' course and Watson's textbook, I proposed 'Comparative Psychology,' and it was this that finally launched me down the behaviorist slope" (1952, p. 329). This slope may have been behaviorist, but it was of a new and unusual kind that reflected Tolman's education at Harvard.
In his early experiments and papers, Tolman focused on the the rat's behavior in the maze to the exclusion of other types of apparatuses because it gave opportunities for observing the animal's solution to problems in space, in getting from here to there. He believed that when a rat ran from the start of a maze to the goal, its behavior reflected a purpose—getting to the goal in order to get something—and knowledge about the spatial layout. In referring to such knowledge, Tolman used terms such as sign-gestalt-expectation, which referred to his assumption that if, in the presence of a certain sign (that is, the events at the start box and on into the maze), the rat behaved in a particular way, it would achieve certain goals. The term gestalt referred to Tolman's assumption that the rat was acquiring a "cognitive map" that would allow it to use its organized information in getting to the goal.
In Tolman's early writings, including his major book, Purposive Behavior in Animals and Men (1932), he maintained the neorealist argument that knowledge and purpose could be directly observed in the behavior of the rat in the maze. But by 1932 he was also working with a different idea: that knowledge and purpose were inferences from behavior rather than characteristics of behavior. These inferences Tolman came to call "intervening variables" to convey the idea that knowledge and purpose intervene between the stimulus and behavior and guide the behavior (Tolman, 1938). In his autobiography Tolman (1952) takes the position that such intervening variables not only serve as summary statements that bring together data but also refer to real, presumably causal events.
Latent Learning Experiments
Tolman and his students conducted a vigorous, broad program of research on learning and problem solving in rats that served both to test his ideas and to change them in the light of new data. Two lines of research will be mentioned briefly here. The first, latent learning experiments, showed that rats learn about the layout of a complex maze even though, in the absence of reward, they show little or no evidence of such learning. When, after some trials, they are first rewarded in the goal box, they show almost error-free behavior on the next trial. These latent learning experiments demonstrated several points. First, learning is different from performance and is occurring even when there is no clear evidence for it. Current reviews show that research of this sort continues to grow and prove fruitful. Second, the latent learning experiments showed that rats gain organized knowledge of the maze that transcended the conceptual framework of stimulus-response. Third, animals learn about rewards. This conclusion was inconsistent with the dominant view of the era: that rewards determine which behaviors are learned. Tolman's conclusion is consistent with much later research in Pavlovian conditioning (Rescorla, 1978).
A second line of research, closely related to the first, directed a variety of cleverly constructed experiments to the problem of whether the animal could use its knowledge of the maze to make inferences about what to do in new situations. Tolman's team guided rats to the goal along a circuitous route for a number of trial, then deprived it of that route, and then exposed it to a variety of alternatives, one of which would lead more directly to the goal. The results showed that the animal was able to use its knowledge about the spatial arrangements in the room to make the appropriate inference and take the direct route. Other research by Tolman and his students aimed at control processes such as selective testing of alternative possible solutions ("hypotheses" and "vicarious trial and error").
At a time when learning theorists were still trying to establish the theory of learning, Tolman (1949) published an article entitled "There Is More than One Kind of Learning." In it he proposed that some of the basic disputes about learning might be resolved if investigators agreed that there are a number of kinds of learning: "The theory and laws appropriate to one kind may well be different to those appropriate to other kinds" (p. 144). Some of the types of learning that Tolman proposed are still under investigation.
Although Tolman, like his contemporaries, thought mostly in terms of the plasticity of behavior, he did not ignore genetic influences. In fact, in 1924 he was the first to apply the technique of selective breeding to the study of genetics of behavior, obtaining "maze-bright" and "maze-dull" strains of rats. His student Robert Tryon then carried out a successful program of selective breeding for maze ability over several generations. This was replicated in other laboratories and extended to other kinds of behavior. This clear evidence for the influence of genes on behaviors was important in holding a place for behavior genetics during the period when environmentalism was dominant (McClearn and Foch, 1988).
All of Tolman's research showed a remarkably coherent but nevertheless broad-ranging character. Although he dissented from the animal-learning orthodoxy from the 1930s through 1950s, Tolman's position had become a dominant one in animal learning by the 1980s and 1990s.
Tolman received many honors, including election to the Society of Experimental Psychologists, the National Academy of Sciences, the American Philosophical Society, and the American Academy of Arts and Sciences. He was an honorary fellow of the British Psychological Society and was awarded honorary degrees by a number of universities. Tolman was president of the American Psychological Association in 1937, president of the Society for the Psychological Study of Social Issues in 1940, and vice president of the American Association for the Advancement of Science in 1942. The Fourteenth International Congress of Psychology was scheduled to be held in the United States in 1954, and Tolman was to be its president. When it became apparent that the United States, because of its anticommunist policy, was likely to refuse admission to many participants from abroad, the venue was changed to Canada, and Tolman became copresident along with Canadian psychologist Edward A. Bott.
In 1949, Tolman took a leadership role in the Berkeley faculty's resistance to the imposition of a loyalty oath by the university. Prevented from teaching, he spent the academic year of 1949-1950 away from Berkeley. The nonsigners finally won their case in court in 1953, gaining recognition of tenure at the university, and Tolman's professorship was restored.
See also:LEARNING THEORY: A HISTORY
Crutchfield, R. S. (1961). Edward Chace Tolman. American Journal of Psychology 74, 135-141.
Crutchfield, R. S., Krech, D., and Tryon, R. C. (1960). Edward Chace Tolman: A life of scientific and social purpose. Science 131 714-716.
Hilgard, E. R. (1980). Edward Chace Tolman. Dictionary of American Biography, Supp. 6. New York: Scribners.
—— (2000). Edward Chace Tolman. In A. E. Kazdin, ed., Encyclopedia of psychology, Vol. 8. Washington, DC: American Psychological Association.
McClearn, G. E., and Foch, T. T. (1988). Behavioral genetics. In R. C. Atkinson, R. J. Herrnstein, G. Lindzey, and R. D. Luce, eds., Steven's handbook of experimental psychology, 2nd edition, Vol. 1. New York: Wiley.
McFarland, D. (1993). Animal behaviour: Psychobiology, ethology, and evolution. New York: Wiley.
Rescorla, R. A. (1978). Some implications of a cognitive perspective on Pavlovian conditioning. In S. H. Hulse, H. Fowler, and W. K. Honig, eds., Cognitive processes in animal behavior. Hillsdale, NJ: Erlbaum.
Tolman, E. C. (1920). Instinct and purpose. Psychological Review 27, 217-233.
—— (1924). The inheritance of maze-learning ability in rats. Journal of Comparative Psychology 4, 1-18.
—— (1932). Purposive behavior in animals and men. New York: Century.
—— (1938). The determiners of behavior at a choice point. Psychological Review 45, 1-41.
—— (1949). There is more than one kind of learning. Psychological Review 27, 217-233.
—— (1952). Autobiography. In E. G. Boring et al., eds., A history of psychology in autobiography, Vol. 4. Worcester, MA: Clark University Press.