Document Type



Four experiments compared the effects of self-rules and rules, and varied and specific schedules of reinforcement. Participants were first exposed to either several schedules (varied groups) or to one schedule (specific groups) and either were asked to generate rules (self-rule groups), were provided rules (rule groups), or were not asked nor provided rules (control groups). When exposed to FI, sensitivity was greater for the varied than for the specific self-rules and rules groups, regardless of reinforcement rate. Control groups showed intermediate sensitivity levels. When nondifferentiated response rates were obtained, sensitivity for the varied groups was similar to that observed for the specific groups. These results suggest that varied rules promote greater sensitivity than do specific ones as long as variable behavior patterns are obtained.