The Effects of Programmed Contingencies and Social Conditions on Response Stereotypy With Human Subjects

Abstract

Reinforcement-induced stereotypy and social conditions involved in rule-following were investigated in two experiments. In both experiments the first phase was designed to investigate reinforcement-induced stereotypy. Subjects were presented with a computer task in which reinforcement was earned for any response pattern which moved a lighted square from the top left to the bottom right of a 5 × 5 matrix. In the second phase, a rule was introduced requiring subjects to perform a different pattern each trial in order to earn reinforcement. In Experiment 1, the third phase was a return to Phase 1 contingencies; in Experiment 2, the third phase was an extinction phase. Social conditions were manipulated by having the experimenter remain in the room during the task (experimenter present), or by having the subject complete the task alone. Data were analyzed by examining the percentage of novel response patterns occurring in each phase of the task. Sixty-five percent of subjects in Experiment 1 and 80% in Experiment 2 developed stereotyped response patterns in Phase 1. However, stereotypy was found to be more complex than just the development of one predominant response pattern. Manipulating social conditions was found to affect responding, with the experimenter-present subjects showing higher degrees of response variability. In Experiment 2 extinction was found’to increase response variability, while social conditions acted to maintain higher rates of responding for the experimenter-present subjects.