Delayed partial reinforcement: Length of delay interval and the relationship of independence

Abstract

In Experiment 1, groups received a 0-, 15-, 30-, or 60-sec delay of reinforcement on the rewarded trials of a partial reinforcement schedule (delayed PR). A 15-sec reinforcement delay did not reduce acquisition speeds compared with no delay. With delays of 30 and 60 sec, decremental effects of delayed PR were obtained for the first time, with a 60-sec delay of reinforcement producing much slower running than the 0- or 15-sec delays and the 30-sec delay group intermediate. In Experiment 2, two groups received a 60-sec delay of reinforcement, combined in one case with PR and in the other with consistent reinforcement (CR). The delayed PR group ran faster than the delayed CR group in acquisition in all alley sections. The results were discussed with reference to the role of inhibition in producing behavioral decrements under delayed reinforcement.