ABSTRACT
In the present state of research on reinforcement in human behavior, it is hard to remember that only a couple of decades ago the major issue was the adequacy of the interpretations of reward and punishment as direct actions of after effects on response strengths or probabilities (Estes, 1959; Postman, 1962). Up to the early 1960s the dominant view of reinforcement (the term being used here descriptively to encompass operations of reward and punishment) comprised the stimulus-response theories of Hull (1943, 1952) and Skinner (1938) at the level of animal learning and neo-Thorndikean interpretations of the law of effect at the level of human learning, the latter being epitomized in a review of the literature by Postman (1962). Though at a theoretical level, interpretations of both animal and human behavior were extensions and elaborations of the law of effect, major differences appeared at the level of application. The reinforcement theories led directly to the vast technology of operant conditioning and behavior modification (Buchwald, 1976; Kanfer & Phillips, 1970; Verhave, 1966). In contrast, research on human learning conducted under the influence of the law of effect led to some gradual resolution of problems having to do with the role of awareness but contributed very little in the way of applications to problems of human behavior.
