Some factors controlling the interaction between response-dependent and response-independent schedules of reinforcement

Abstract

Response rate interactions were studied in four rats following a shift from: (a) a multiple schedule in which two stimuli, S1 and S2, were both correlated with response-dependent reinforcement (mult VI 1 min VI 1 min) to (b) a multiple schedule in which S2 was now correlated with response-independent reinforcement (mult VI 1 min VT 1 min). Concomitant with the decrease in the rate of responding in the VT component, there was also a decrease in rate of responding during the VI component, i.e., negative induction. The nature of the observed interaction suggested that responding appropriate to the VT component may have generalized to the VI component. To test this interpretation, in a second experiment an effort was made to accentuate the differences between the components of the multiple schedules and consequently to reduce the degree of interaction. For four rats, two had a different reinforcer correlated with each component and two had a different response lever correlated with each component. The results of the separate-reinforcer treatment were not consistently different from those results obtained in Experiment I. The separate-lever treatment, however, resulted in decreased negative induction interaction and lower rates of responding during the VT component.