The effect of patterning of reinforcement on resistance to extinction in a free response situation

Forty albino rats were trained to press a bar in a free response situation on four different patterns of reinforcement: random, double, quadruple and sextuple alternation. All patterns were run under 50% reinforcement. The Ss were run on acquisition for 1O days, 96 responses per day, with 48 of these responses being reinforced. The Ss were then run for three days on extinction with a fixed amount of time allowed for each day; day one of extinction was 15 minutes, day two was 10 minutes, and day three was five minutes. An analysts of variance for the rate of responding for each day of extinction produced non significant F ratios. This indicated that there was no difference between any of the four groups. The results were discussed in relation to the theory of cognitive dissonance which properly predicted the outcome of the study and the discrimination hypothesis which was unable to predict the outcome. The recent studies of N-length and of patterns of reinforcement were also analyzed in relation to this study. Inferential predictions were made concerning the aftereffects hypothesis and the frustration hypothesis as to what might have been expected of these positions. Another experiment was proposed to answer some of the questions raised by this study.