Edward C. Tolman was the initiator of propositional behaviorism and a key figure for the introduction of cognitive variables in behavioral models.

Although the study of cognitive maps is Tolman’s best known contribution , the theory of this author is much broader and was a real turning point in scientific psychology.

Biography of Edward Tolman

Edward Chace Tolman was born in Newton, Massachusetts in 1886. Although his father wanted him to continue the family business, Tolman decided to study electrochemistry; however, after reading William James he discovered his vocation for philosophy and psychology, a discipline he would eventually pursue.

He graduated in Psychology and Philosophy at Harvard . Shortly afterwards he moved to Germany to continue his education on his way to his doctorate. There she studied with Kurt Koffka; through him she became familiar with Gestalt psychology, which analyzed perception by focusing on the global experience rather than on the individual elements.

Back at Harvard, Tolman researched the learning of nonsense syllables under Hugo Münsterberg, a pioneer of applied and organizational psychology. He obtained his PhD with a thesis on retroactive inhibition , a phenomenon consisting of the interference of new material in the recovery of previously learned memories.

After being expelled from Northwestern University, where he taught for three years, for publicly opposing American intervention in World War I, Tolman began teaching at the University of California at Berkeley. He spent the rest of his career there, from 1918 until his death in 1959.

Theoretical contributions to Psychology

Tolman was one of the first authors to study cognitive processes from the framework of behaviorism ; although he was based on behaviorist methodology, he wanted to demonstrate that animals could learn information about the world and use it in a flexible way, and not only automatic responses to certain environmental stimuli.

Tolman conceptualized cognitions and other mental contents (expectations, objectives…) as intervening variables that mediate between stimulus and response. The organism is not understood as passive, in the manner of classic behaviorism, but rather actively manages information.

This author was especially interested in the intentional aspect of behavior, that is, in goal-oriented behavior; that is why his proposals are categorized as “propositional behaviorism” .

The E-E and E-R learning models

In the mid-20th century there was a profound debate within the behaviouralist orientation about the nature of conditioning and the role of reinforcement. Thus, the Stimulus-Response (E-R) model, personified in authors such as Thorndike, Guthrie or Hull, and the Stimulus-Stimulus (E-E) paradigm, of which Tolman was the most important representative, were opposed.

According to the E-E model, learning is produced by the association between a conditioned stimulus and an unconditioned stimulus, which evokes the same conditioned response in the presence of the reinforcement; in contrast, from the E-R perspective, it was argued that learning consists of the association between a conditioned stimulus and a conditioned response .

Thus, Tolman and related authors considered that learning depends on the subject detecting the relationship between two stimuli, which will allow him to obtain a reward or avoid a punishment, as opposed to the representatives of the E-R model, who defined learning as the acquisition of a conditioned response to the appearance of a previously unconditioned stimulus.

The E-R paradigm proposed a mechanistic and passive vision of the behavior of living beings, while the E-E model affirmed that the role of the learner is active since it implies a component of voluntary cognitive processing, with a determined goal .

Experiments on latent learning

Hugh Blodgett had studied latent learning (which does not manifest itself as an immediately observable response) through experiments with rats and mazes. Tolman developed his famous proposal on cognitive maps and much of the rest of his work from this concept and Blodgett’s work.

In Tolman’s initial experiment three groups of rats were trained to walk through a maze . In the control group, the animals were given food (reinforcement) when they reached the end; in contrast, the rats in the first experimental group only got their reward from the seventh day of training, and those in the second experimental group from the third day.

Tolman found that the error rate of the control group rats dropped from day one, while those in the experimental groups dropped sharply from the introduction of the food. These results suggested that the rats learned the route in all cases, but only reached the end of the maze if they expected to get reinforcement.

Thus, this author theorized that the execution of a behaviour depends on the expectation of obtaining reinforcement or , but that nevertheless the learning of such behaviour can take place without the need for a reinforcement process.

The study of cognitive maps

Tolman proposed the concept of cognitive maps to explain the results of his and Blodgett’s experiments. According to this hypothesis, the rats constructed mental representations of the maze during the training sessions without the need of reinforcement, and therefore they knew how to reach the goal when it made sense.

The same would be true for people in everyday life : when we repeat a route frequently we learn the location of a large number of buildings and places; however, we will only address these if it is necessary to reach a certain goal.

To demonstrate the existence of the cognitive maps Tolman did another experiment similar to the previous one, but in which after the rats learned the route of the labyrinth it was filled with water. Despite this, the animals managed to get to the place where they knew they would find food.

This confirmed that rats did not learn to execute a chain of muscle movements , as advocated by the theorists of the E-R paradigm, but that cognitive variables, or at least non-observable ones, were necessary to explain the learning they had acquired, and the response used to achieve the objective could vary.