THE EFFECT OF PRETRAINING REINFORCEMENT SCHEDULES ON CHILDREN'S LEARNING

Abstract

In a recent study Stevenson and Pirojnikoff (3) found that the rate with which children learned to discriminate among three forms was significantly affected by 20 pretraining trials in which responses to each form received 0 per cent, 50 per cent, or ioo per cent reinforcement. The Ss who received ioo per cent reinforcement for the choice of each form during pretraining later learned to discriminate among the forms with rapidity; Ss who received 50 per cent random reinforcement or o per cent reinforcement during pretraining showed a much slower rate of learning. The results were interpreted as giving support to the assumption that response tendencies developed on the basis of the different pretraining reinforcement schedules operated during the learning trials in such a manner as to interfere with or to facilitate the development of the correct response. It was noted, however, that pretraining appeared to affect the general behavior of the children. The Ss in the ioo per cent group performed in a smooth and efficient fashion, while Ss in the other groups appeared to be perturbed and frustrated. This would seem to indicate that the differences in the rate of learning among the groups may be attributable not only to the development of different response tendencies to the discriminative stimuli but also to the development of more general behavioral differences. The question asked in the present experiment is whether Ss will show differences in rate of learning when pretraining is given with stimuli different from those used in the discrimination task. It would be predicted that if Ss develop general response tendencies during pretraining which carry over to the learning task, the rate of learning of Ss receiving high * Institute of Child Development and Welfare, University of Minnesota, Minneapolis 14.