Edwin Ray Guthrie

Introduction

Edwin Ray Guthrie (January 9, 1886 – April 23, 1959) was an American psychologist and one of the principal learning theorists of the behaviorist era. Trained as a mathematician and philosopher, he did not enter psychology until his late twenties, and he brought to it a taste for parsimony that became his signature. Guthrie's contiguity theory, often called one-trial learning, held that a single pairing of a stimulus pattern with a movement is sufficient to establish their connection, and that reinforcement plays no necessary part in learning at all. The position put him directly at odds with Edward Thorndike's law of effect and, later, with B. F. Skinner's operant conditioning, and it made him the most economical alternative available within behaviorism. He spent nearly his entire career at the University of Washington, served as president of the American Psychological Association in 1945, and left behind a theory that was admired for its clarity, attacked for its thinness, and quietly built into the mathematical learning theories that followed.

Early Life and Education

Guthrie was born in Lincoln, Nebraska, the eldest of five children. His father ran a store selling pianos and bicycles; his mother was a schoolteacher. He was a voracious reader, and he later traced the origins of his thinking to an eighth-grade reading program he undertook with a friend, in which they worked through Charles Darwin's On the Origin of Species and The Expression of the Emotions in Man and Animals. He graduated from high school at seventeen with a senior thesis arguing that both science and religion, being dependent on words whose meanings depend on the experience of their users, had no chance of expressing absolute truth.

He entered the University of Nebraska in 1903 and studied mathematics, philosophy and psychology, taking his bachelor's degree in mathematics in 1907 and a master's degree in philosophy in 1910. Guthrie credited the university's near-absence of required courses with letting him range as widely as he liked, through Latin and Greek and up to calculus. As a graduate student he was, by his own account, the only student in a seminar taught by Harry Kirke Wolfe, a student of Wilhelm Wundt, where the talk ran to the philosophy of science rather than to laboratory technique. His single course in experimental psychology, a winter spent measuring the two-point threshold with an aesthesiometer, he said served to quench his interest in psychophysics.

He then went to the University of Pennsylvania as a Harrison fellow and received a doctorate in philosophy in 1912. He taught high school mathematics for several years, in Lincoln and then in Philadelphia, before moving west in 1914.

Academic and Professional Career

Guthrie joined the University of Washington in 1914 as an instructor in the department of philosophy. The psychology department there had been founded in 1917 by Stevenson Smith, and Guthrie transferred to it within a few years. The collaboration with Smith shaped the rest of his career: together they wrote Chapters in General Psychology in 1921, and it was in the course of arguing out the problems of that book that Guthrie began to work seriously on how learning actually happens. He rose to full professor in 1928.

A trip to France with his wife, Helen MacDonald, whom he married in 1920, brought him into contact with Pierre Janet. Janet's work impressed him enough that the two Guthries translated his Principles of Psychotherapy into English, and Guthrie grafted an objective theory of learning onto Janet's descriptive psychology. During the Second World War Guthrie served as a consultant to the overseas branch of the War Department general staff and to the Office of War Information. In 1943 he became dean of the graduate school at Washington, and in 1945 he was elected president of the American Psychological Association; his presidential address, "Psychological Facts and Psychological Theory," was published in the Psychological Bulletin the following year. He retired in 1956, received the American Psychological Foundation's Gold Medal in 1958, and died of a heart attack in Seattle on April 23, 1959. The University of Washington named a building Guthrie Hall in his honor.

Major Contributions to Psychology

The principle of contiguity

Guthrie's entire system rested on a single sentence: a combination of stimuli that has accompanied a movement will, on its recurrence, tend to be followed by that movement. Learning is the formation of a connection between a stimulus pattern and a movement, and the connection is formed at full strength on the first occasion. Repetition does not strengthen it. Guthrie's contemporaries regarded this as an affront, because it removed reward, punishment, drive and practice from the explanation of acquisition and left only temporal pairing.

Two features distinguished his version from other associationisms. First, the unit of behavior is not the response but the movement: what is learned is a specific patterned action, not a general category such as "pressing the lever." Second, the stimulus is not a single event but a complex situation, an entire array of elements present at the moment of action; a behavior recurs when enough of that array is present again. From this it followed that the apparent gradualness of most learning curves is an artifact. Complex skills are mosaics of many separately acquired movements, plus the accumulation of cues; the curve measures the assembly, not the strength of any one connection.

One-trial learning and the puzzle box

Guthrie knew that his theory needed empirical support against reinforcement accounts, and with George P. Horton he built a glass-fronted puzzle box in which a cat's escape was triggered by contact with a central pole. Photographs showed something striking: each cat repeated the exact idiosyncratic movement that had immediately preceded its previous escape, rather than converging on an efficient general solution. One cat learned to brush the pole with its rear flank while turning toward the glass wall, and repeated that stereotyped turn on every trial. The results, published as Cats in a Puzzle Box in 1946, were offered as evidence for what Guthrie called the principle of postremity: when a stimulus is followed by several responses, it becomes associated with the last of them.

Punishment and the breaking of habits

Guthrie's account of punishment followed from the same logic. Punishment changes behavior only if it produces a movement incompatible with the one being punished, in the presence of the cues that originally evoked it, and it must be delivered while those cues are still present. Punishment that merely follows the act, or that occurs after the cues have gone, changes nothing and may even strengthen the unwanted behavior by giving the old cues new practice. This gave him a practical prescription for habit breaking: identify the cues that trigger the habit, then arrange for those cues to evoke a different, incompatible response. He described three ways of doing so, by keeping the stimulus below the threshold at which it evokes the habit, by exhausting the response through repeated elicitation until the organism can no longer perform it, and by substituting a competing response.

Later formalization

Guthrie wrote in plain, anecdotal prose and distrusted elaborate apparatus, which made him popular with students and suspect to systematizers. His ideas were given formal shape by others. Virginia Voeks restated them as a small set of explicit postulates, adding a principle of dynamic situations to the principles of association, postremity and response probability, and making the theory testable in a way it had not been before. William K. Estes took the probabilistic element and built from it his stimulus sampling theory, published in 1950, which is generally credited with launching mathematical learning theory.

Major Publications

Chapters in General Psychology (1921), with Stevenson Smith
The Psychology of Learning (1935; revised edition 1952)
The Psychology of Human Conflict (1938)
Cats in a Puzzle Box (1946), with George P. Horton
"Psychological Facts and Psychological Theory" (1946), Psychological Bulletin, 43, 1-20
"Association by Contiguity" (1959), in S. Koch (ed.), Psychology: A Study of a Science, Vol. 2

Impact and Legacy

Guthrie's standing in the history of learning theory rests less on the survival of his specific claims than on what they forced others to answer. By denying that reinforcement was necessary for learning, he made the role of reward an open empirical question rather than an axiom, and the argument with Thorndike and Skinner over that question structured research on conditioning for two decades. His insistence that behavior be described as movements rather than as molar responses anticipated later worries about what counts as a unit of analysis. Through Estes his associationism fed directly into mathematical psychology, and his prescriptions for breaking habits survive in applied behavior analysis and in the "incompatible response" logic of behavior therapy.

The criticisms were as durable as the influence, and Guthrie heard them from the beginning. The most common was that his theory was too simple, a charge his defenders found unfair and his detractors found unanswerable. The psychologist who reviewed him most sharply put the objection in a phrase that has followed Guthrie ever since: many reviews in the literature mistook incompleteness for simplicity. The complaint was not that his writing was unclear, but that a single principle was being asked to do the work of a dozen and that the gaps it left were not acknowledged as gaps. Where reinforcement theorists explained extinction, spontaneous recovery, resistance to extinction and the partial-reinforcement effect, Guthrie had to resort to ad hoc appeals to changed stimulus conditions and to interference from newly formed associations.

Reinforcement theorists pressed the empirical case hardest. Experiments repeatedly showed that the delivery, magnitude, delay and schedule of a reward change both the rate of acquisition and the persistence of responding, effects that a pure contiguity account either cannot predict or must reinterpret as changes in the stimulus situation. Guthrie's answer, that reinforcement matters only because it rearranges the stimulus field and protects the last response from being displaced, struck many as unfalsifiable. Others objected that his theory had no account of motivation, no real account of forgetting, and no room for the cognitive processes such as expectancy and memory that Edward Tolman and, later, the cognitive revolution insisted were central. There is also the charge that his deliberate vagueness was tactical: Guthrie was said to have preferred to leave his formulations loose so that they could not be pinned down and refuted, and much of his writing was indeed clarified only after his death, by students rather than by him. The fair summary is the one his own field arrived at. One-trial contiguity describes a narrow but real class of associations well; as a general theory of learning, it was overtaken by evidence it was never designed to accommodate.

Learn More

References

Guthrie, E. R. (1935). The Psychology of Learning. Harper.