Pavlovian determiners of behavioral momentum

Abstract

Pigeons pecked keys on a six-component multiple schedule of food reinforcement arranged in successive pairs on three keys. On all three keys, the schedule in the first component was variable-interval 120 sec. On one key, the second component was variable-time 24 sec; on another, it was variable-time 120 sec; and on the third, extinction. Pairs of components were separated by timeout. In this arrangement, second-component stimuli were differentially correlated with food in their presence, and first-component stimuli were differentially correlated with second-component food in serial fashion. After baseline performance stabilized, resistance to change was assessed by prefeeding, by discontinuing food, and by presenting free food at random times throughout sessions. There was no consistent relation between baseline response rates in the first component and reinforcement rate in the second component. When assessed by prefeeding or extinction, resistance to change in the first component was positively related to reinforcement rate in the second component, demonstrating that serial stimulus-reinforcer contingencies affected resistance to change. © 1984 Psychonomic Society, Inc.