STUDIES OF DISCRIMINATION LEARNING IN MENTALLY RETARDED CHILDREN : Effects of reinforcement rates on the learning rate

Abstract

This experiment was designed to examine the effects of reinforcement rates on simple discrimination learning in normals and retardates of same M.A. < Experiment I > A series of pictures was projected on the wall. Each picture had two figures in which different numbers of black circles were pictured. The Ss were asked to choose a figure in each picture which they thought was correct and to push the button in front of the fig ure. Each right response was reinforced at 100%, 75% 50% and 25% rate suited to four experimental groups. The Ss were run until the criterion of eight successive correct responses. The Ss were allowed to make 60 trials at the maximum. After S met the criterion, he was not reinforced all responses. The extinction process was discontinued after a wrong response, and after 30 correct trials. The results were as follows: (1) In the both groups of retarded and normal Ss, there was the significant reinforcement effect on acquisition process. (2) The average number of responses needed to extinct the correct response increased with decreasing percentage of reinforcement. Retarded Ss required more responses to extinct the correct response than normals. <Experiment II> Exp. II consisted of two tasks. The materials of Task 1 was the same with that of Exp. I. The materials of Task 2 was 12 cards consisted of two figures which differed along two dimensions: shape (circle, triangle and square) and size (big and small). In Task 2 the relevant dimension was size. Only Ss who met the criterion in Task 1 wese assigned Task 2. Each right response on Task 2 was reinforced at 100%, 66% and 33% rate suited to three experimental groups. The criterion for acquisition in both tasks were six successive correct responses. Ss were allowed to make 30 trials at the maximum. The results were as follows: (1) Only normal children who learned relatively faster Task 1 were not influenced their acquisition by different percentages of reinforcement. (2) The result on extinction process was the same with that in Exp. I.