Concurrent Reinforcement of Topographically Different Responses and the Matching Relation

Abstract

Three rats responded under concurrent variable-interval schedules of food reinforcement. One schedule component provided reinforcement for the manipulation of an omnidirectional rod; the other schedule component provided reinforcement for pressing a lever. The same force was required to execute each response. Reinforcement rates provided by the schedules were varied over experimental conditions. Allocation of both responses and time between schedules undermatched the ratios of reinforcements obtained from the schedules for each rat. All rats displayed bias in favor of the lever with both response and time allocation, but no consistent difference in the magnitude of bias with response-based versus time-based measures of choice was observed. Previous research on choice between topographically different responses revealed substantially greater bias in response-based measures of choice than in time-based measures. However, response topography has been confounded with response force in previous research. Findings of the present study, in which force requirements for the two responses were equal, indicate that differences in response topography do not necessarily produce differences between response-based and time-based measures of bias. Rather, these findings suggest that differences in measures of bias may result from different force requirements of the responses.