Effects of a delay-reinforcement procedure on performance under IRT>t schedules

Water-deprived rats were studied under a compound schedule that prescribed that responses terminating interresponse times (IRTs) greater than a fixed value t(1) (IRT > t(1) component schedule) initiated a delay of reinforcement interval t(2), at the end of which water was presented if the subject did not respond (R > t(2) component schedule). If the subject responded before the t(2) interval elapsed, the IRT > t(1) component schedule was re-initiated and water was not presented. The IRT > t(1) and R > t(2) component schedules were not differentially correlated with distinctive stimuli. Rate of responding during the IRT > t(1) component decreased as a function of the value of t(2). The magnitude of the decreases in response rate appeared to be proportional to the subject's rate under the IRT > t schedule with no delay of reinforcement (t(2) = 0 sec). The effects were independent of the parameter value of the IRT > t(1) component schedule and of the rate of reinforcement. The results suggested that "efficiency" of performance under IRT > t schedules can be increased by appropriately arranging brief delays of reinforcement.