Spence, Kenneth (1907-1967)
SPENCE, KENNETH (1907-1967)
Kenneth W. Spence (1907-1967) played a major role in psychology from the early 1930s until his untimely death. His impact is illustrated by the fact that from 1962 to 1967, he was the most cited author in a survey of the fourteen most prestigious psychological journals (Myers, 1970). Spence's influence resulted from achievements as experimentalist, theorist, methodologist, and teacher. In all of these roles, he operated as a natural science psychologist, one who believed that the science of psychology can employ the same methods of empirical inquiry and theory construction as physics, chemistry, and biology. In essence, he was asserting that psychology, in principle, is capable of producing a body of reliable scientific knowledge. To achieve this goal he deemed it necessary to conceptualize psychology as the science of behavior, not of the mind. That is, the basic observations of the science of psychology are the behavior of organisms, not the direct examination of conscious experience. This methodological position, Behaviorism, was initially expressed, in a radical form, by John B. Watson (Kendler, 1987), but since has matured into a more sophisticated version known as neobehaviorism.
Spence was born May 6, 1907, in Chicago. When he was 4, Western Electric transferred his father, an electrical engineer, to Montreal. He majored in psychology at McGill University, receiving a B.A. in 1929 and an M.A. in 1930. His Ph.D. was granted in 1933 by Yale University, where he served as a research assistant to Robert M. Yerkes, under whose direction he completed a dissertation on the visual acuity of chimpanzees. The dominant intellectual influence during his Yale days evolved from the inspirational ideas of Clark L. Hull, who set as his goal the formulation of a theoretical interpretation of behavior that emulated the conceptual structure of Newtonian physics.
Hull's general approach was shaped by both Ivan Pavlov and Edward Thorndike. Pavlovian conditioning, for Hull, was the simplest form of learning, and hence principles of conditioning could provide the premises from which more complex forms of behavior could be deduced (explained). Thorndike's law of effect represented, for Hull, another fundamental principle of behavior; rewards, technically known as reinforcements, are necessary for the formation and strengthening of a connection between a situation and behavior, or what became to be known as a stimulus-response association (e.g., in Pavlovian conditioning the sound of a tone became connected with salivation because it was reinforced by food).
While a graduate student, Spence prepared a paper (Spence, 1932) that illustrated Hull's general strategy. Based on assumptions about the effect of delayed reinforcement in conditioning, Spence predicted that in a complex maze, one with several successive choice points, entrances into blind alleys would be eliminated in a backward order, that is, the blinds are more difficult the farther they are from the goal. After completing his graduate work at Yale, Spence accepted a National Research Council fellowship to the Yale Laboratories of Primate Biology at Orange Park, Florida, where he published his classical theory of animal discrimination learning (Spence, 1936), a conception that is still influential (Kendler, 1992). In 1937 he moved to the University of Virginia as an assistant professor of psychology. The following year he was appointed associate professor at the University of Iowa, where, in 1942, he became professor and head of the Department of Psychology, a position he occupied until 1964, when he moved to the University of Texas.
Kenneth Spence's efforts in theoretical psychology are characterized by a consistent direction and a close relationship between facts and theory. He had (a) an ability to express ideas in a lucid prose that resulted from clear thinking and hard work; (b) a flair for designing clever experimental tests of competing hypotheses; (c) a knack for analyzing data so that their theoretical implications would become fully apparent; (d) an ingenious talent for theoretically integrating a range of experimental results with the added, and unusual, feature of being able to offer possible alternative explanations, even some from competing theories; and (e) a special aptitude, in spite of limited mathematical training, to coordinate his theoretical interpretation with a quantitative model that structured an empirical problem so that further theoretical clarification and empirical development could occur.
One excellent example of Spence's style of theoretical and empirical structuring is exhibited in his influential analysis of discrimination learning of animals (Spence, 1936), which contained one of the first mathematical simulations (without the aid of a computer!) of a theory. The model, which seeks to identify the psychological processes involved in how animals learn to choose between two stimuli when one is followed by reward and the other is not, postulates two basic theoretical mechanisms that Hull had employed in interpreting conditioning phenomena: excitation and inhibition. When the animal is reinforced for selecting one stimulus, that habit (stimulus-response association) is gradually strengthened; when a response to the other cue is not reinforced, that habit is gradually weakened. When the difference in strengths between the two competing habits reaches a certain value, the subject consistently chooses the reinforced stimulus. Spence had no illusion that his formulation represented a complete interpretation of the discrimination learning process. He readily acknowledged that certain fundamental processes, such as complex perceptual mechanisms, were ignored but insisted, as a matter of strategy, that discrimination learning, as well as all psychological problems, must be broken down into bare essentials to be investigated fruitfully. Reducing discrimination learning to the analysis of competing habits in a simple experimental situation was a fruitful strategy for dealing with fundamental principles of behavior.
Spence's theory of discrimination learning had many ramifications, including the formulation of an ingenious stimulus-response explanation (Spence, 1937) of the transposition phenomenon, the tendency for nonverbal organisms to transfer a relational choice from one problem to a subsequent one (e.g., after learning to choose a medium gray in preference to a lighter one, a rat will probably select a darker gray rather than the previously rewarded medium gray). In addition, his theory initiated a crucial controversy as to whether discrimination learning was a continuous or noncontinuous process, occurring gradually or suddenly (Kendler, 1987).
Although Spence's discrimination learning theory was designed for nonverbal organisms, he sought to discover how the acquisition of symbolic skills would influence discrimination learning. He encouraged a doctoral student, Margaret Kuenne (1946), to investigate this problem; she found that in a transposition problem the behavior of inarticulate youngsters was similar to that of rats, in that simple stimulus-response associations were formed but the acquisition of symbolic skills introduced a more complex pattern of stimulation and associative connections that enhanced relational responding. This led to a two-stage theory of discrimination learning (Kendler and Kendler, 1962, 1975) in which the lower stage (single-unit), based upon Spence's continuity model, hypothesized that the responses of subhuman animals were directly linked (associated) to stimuli (e.g., black, white), whereas higher-level functioning (mediational model) suggested that incoming stimulation is transformed into some internal conceptual representation (e.g., brightness) that guides subsequent behavior. This model, which had its origins in Spence's discrimination learning theory, could account for developmental changes in the discrimination learning exhibited by humans.
From 1950 to the end of his life, classical (Pavlovian) eyeblink conditioning with human subjects occupied Spence's interest. His efforts revealed basic principles of habit formation and the interaction effects between habits and drives. With the collaboration of Janet A. Taylor, who later became his wife, the role of anxiety as a motivational mechanism was empirically and theoretically analyzed (e.g., Spence and Taylor, 1951; Spence, 1956). Spence (1966) was able, by clever experimental manipulations, to get human subjects to respond either in a simple associative manner analogous to subhuman behavior or in a cognitive (mediational) manner. This was another example of his effort, as well as that of several of his students, to gain an experimental grip on evolutionary processes in behavior theory.
In 1955, Spence was invited to deliver the Silliman Lectures at Yale, a prestigious series that until then had been given by distinguished physical and biological scientists such as Ernest Rutherford, Enrico Fermi, and Charles Sherrington. The lectures, published under the title Behavior Theory and Conditioning (1956), can be characterized as a realistic and reasonable theoretical interpretation of a wide range of experimental data. Unlike many of the theoretical messiahs who have dotted the psychological landscape with grandiose theories, Spence's formulation was never far removed from experimental evidence. In essence, Spence was an experimental psychologist's theorist because his hypotheses were clearly testable.
Spence's skills as an experimentalist and theorist were matched by his talents as a methodologist. With the collaboration of the philosopher of science Gustav Bergmann, Spence contributed to the clarification of the meaning of psychological concepts and the logic of psychological measurement (Bergmann and Spence, 1941, 1944). His greatest contribution as a methodologist was his clarification of issues in theoretical psychology, particularly in relation to the comparative analysis of competing learning theories. The competing theory to the Hull-Spence model that interested Spence the most was Edward Tolman's cognitive theory of animal learning. This formulation generated much confusion among both behaviorists and antibehaviorists; the former could not reconcile Tolman's hypothesized mentalistic processes with methodological behaviorism, and the latter denied that Tolman could be labeled a behaviorist if he employed phenomenological experience as a metaphor for his theoretical constructs. Spence clearly distinguished between the strategies employed for conceptualizing theoretical processes—mechanistic, phenomenological, mathematical, or some other—and the theory's deductive, empirical consequences. Unfortunately this distinction between a theorist's thinking style and the explicit theory with its deductive empirical implications was not fully appreciated at the time of the cognitive revolution, and as a result needless disputation and misunderstanding were encouraged. Many segments of the psychological community failed to appreciate the linkage between neobehaviorism in general, and Spence's theoretical efforts in particular, in attempting to understand the relationship between associative and cognitive processes (Kendler, 1984).
Spence's contributions to psychology cannot be limited to his own publications. His inspired teaching must also be considered. Many of his seventy-five doctoral students, too many to mention, made solid contributions to the science of psychology: "All of his doctoral students carry with them some of Spence's ideas and commitments and a desire to achieve a level of quality in their own work that would be acceptable to their Professor" (Kendler, 1967, p. 341).
In addition to being the first and only psychologist to give the Silliman Lectures, Spence received many other honors, including membership in the National Academy of Sciences, the Howard Crosby Warren Medal of the Society of Experimental Psychologists in 1953, and the American Psychological Association's Scientific Contribution Award in 1956, the first year that it was presented. But perhaps the greatest honor Spence aspired to was a place in the history of psychology. Although it is difficult to penetrate the haze of the future, especially in relation to psychology, a discipline that is conceptualized in many antithetical ways, it is likely that Spence will be remembered as a clear and sophisticated exponent of natural science psychology and as a theorist and empiricist who, while appreciating the immaturity of the science of psychology, was nevertheless able to advance it with fruitful conceptions of learning and motivation.
Bergmann, G., and Spence, K. W. (1941). Operationism and theory construction. Psychological Review 48, 1-14.
—— (1944). The logic of psychological measurement. Psychological Review 51, 1-24.
Kendler, H. H. (1967). Kenneth W. Spence: 1907-1967. Psychological Review 74, 335-341.
—— (1984). Evolutions or revolutions? In K. M. B. Lagerspetz and P. Niemi, eds., Psychology in the 1990's. Amsterdam: North Holland.
—— (1987). Historical foundations of modern psychology. Pacific Grove, CA: Brooks/Cole.
Kendler, H. H., and Kendler, T. S. (1962). Vertical and horizontal processes in problem solving. Psychological Review 69, 1-16.
—— (1975). From discrimination learning to cognitive development: A neobehavioristic odyssey. In W. K. Estes, ed., Handbook of learning and cognitive processes, Vol. 1. Hillsdale, NJ: Erlbaum.
Kendler, T. S. (1992). Levels of cognitive development. Hillsdale, NJ: Erlbaum.
Kuenne, M. R. (1946). Experimental investigation of the relation of language to transposition behavior in young children. Journal of Experimental Psychology 36, 471-490.
Myers, C. R. (1970). Journal citations and scientific eminence in contemporary psychology. American Psychologist 25, 1,041-1,048.
Spence, K. W. (1932). The order of eliminating blinds in maze learning by the rat. Journal of Comparative Psychology 14, 9-27.
—— (1936). The nature of discrimination learning in animals. Psychological Review 43, 427-429.
—— (1937). The differential response in animals to stimuli varying within a single dimension. Psychological Review 44, 430-444.
—— (1966). Cognitive and drive factors in the extinction of the conditioned eye blink in human subjects. Psychological Review 73, 445-451.
Spence, K. W., and Taylor, J. A. (1951). Anxiety and strength of the UCS as determiners of the amount of eyelid conditioning. Journal of Experimental Psychology 42, 183-186.