Verbal learning and reinforcement: A reexamination of the Premack hypothesis

A report of a replication of a previous study (See McIntire, R. "Reinforcement and verbal learning: A test of the Premack hypothesis," in Psychological Reports, 1963, 12.) which purportedly tested and confirmed the applicability of the Premack principle (reinforcement as a relative relation between differentially probable responses, i.e., more probable responses reinforce the less probable) to verbal learning. Ninety male and female Ss learned paired associate (PA) lists (comprised of words of intermediate response probabilities) in which the response words were reinforced (i.e., immediately followed) by words of high, intermediate, or low probabilites. Word probabilities were obtained from two standard word count sources. As was anticipated, no differential effects of the probability of the "reinforcement" words were obtained, as measured by trials to criterion for learning the PA lists. The failure of this experiment to support this application of the Premack principle was attributed to inappropriate methodologies used in identifying word probabilities. An appropriate method for testing the Premack principle in a PA learning task is described. Modified HA