Post Watsonian Behaviorism Contrary to traditional historical accounts, American psychology did not become predominantly behaviorist as the immediate result of Watson's program. Behaviorism did begin to take hold in the 1930s, however, partly because of Watson's continued propagandizing but also because full translations of Pavlov's research became available for the first time.
Logical positivism, which allowed for theories to include abstract concepts but insisted that these concepts be tied to observable events, created a fertile climate for the evolution of behaviorism. Operationism, originating in physics in the late 1920s, also helped to create an environment conducive to objective, behaviorist thinking. Operational definitions define concepts in terms of a set of operations, under the control of the researcher, that are assumed to bring about the term in question (e.g., 24 hours without food brings on hunger). Confidence in the generality of some research outcome increases when various studies, each using a slightly different operational definition, nonetheless converge on the same outcome.
Neobehaviorists disagreed on a number of issues, but agreed that (a) continuity between species allowed for general rules of behavior to be derived from nonhuman species, (b) understanding behavior required a thorough knowledge of how the organism learns, and (c) research results should have practical applications.
Edward C. Tolman (1886-1959): A Purposive Behaviorism
Much of Tolman's research used maze learning, and he investigated both the general reliability of the maze as an apparatus and the manner in which rats learn mazes. According to Tolman, rats in a maze do not learn a series of S-R connections; rather, they learn an overall cognitive map of the maze. This spatial ability can be shown in latent learning studies, in which animals can be shown to be learning a maze even though the learning is not reflected in their performance until reinforcement is made available, and in place learning, in which animals learn to go to a location more quickly than they learn to make a specific response.
Tolman believed that all important behavior was goal-directed or purposive and that molar rather than molecular behavior should be the unit of study. He did not think that reinforcement was necessary for learning to occur. He developed the concept of the intervening variable, a hypothetical factor internal to the organism that intervenes between stimulus and response and is defined operationally. Many of the intervening variables in Tolman's system (e.g., expectancy) were cognitive.
Clark Hull (1884-1952): A Hypothetico-Deductive Behaviorism
Although he is known primarily for his theory of learning based on animal studies, Hull also studied the development of concept learning in humans, aptitude testing, and experimental hypnosis, producing a doctoral dissertation on the first and books on the latter two topics.
Modeled on Newtonian physics and consistent with the dictates of logical positivism, Hull's hypothetico-deductive system of behavior involved the development of a theory in which specific experiments were created to test hypotheses that were derived from highly formalized postulates. Research outcomes would then strengthen faith in the postulates or bring about their revision.
Hull's learning theory is a drive-reduction theory. Postulate 4 proposes that learning (i.e., an increase in habit strength) involves stimulusresponse contiguity accompanied by reinforcement. Reinforcers are stimuli that reduce drives. They can be primary or biologically based (e.g., food), or secondary (i.e., learned through association with primary reinforcers).
Hull used a large number of intervening variables. The most important one was reaction potential, SE R, the probability that a response will occur at a given time. It was said to be influenced by a number of factors, including drive (D) and habit strength (sHR), both of which Hull believed were necessary for behavior to occur.
F. Skinner (1904-1990): Radical Behaviorism
Skinner rejected the more formal theories of both Tolman and Hull and argued for a more inductive, descriptive behaviorism that simply looked for evidence of behavior that could be predicted and controlled. He is best known for developing the distinction between classical (Type S) and operant (Type R) conditioning and for investigating the latter. To do so, he created the Skinner box, an experimental chamber in which the rate of some response (e.g., bar pressing) is recorded continuously by a cumulative recorder. Operant conditioning occurs when behavior is shaped by its immediate consequences. If the consequences are positive, the behavior occurring in a specific environment is more likely to occur in that environment in the future; if they are negative, the behavior becomes less likely to occur.
He rejected the use of what he called explanatory fictions, hypothetical factors that appear to explain a phenomenon but actually do nothing more than relabel it. Hence, he was critical of nervous system explanations of behavior, and he never accepted the idea that explanations for behavior would be found by cognitive psychologists.
Skinner called for a technology of behavior to improve child rearing, education, and society as a whole through the use of behavioral techniques. In Walden 71.vo, he outlined how an entire community could function according to operant principles | 
