AND

We studied behavioral functions associated with stereotypical responses for students with autism. In Study 1, analogue functional analyses (attention, demand, no-attention, and recreation conditions) were conducted for 5 students. Results suggested that stereotypy was multiply determined or occurred across all assessment conditions. For 2 students, stereotypy was associated with positive and negative reinforcement and the absence of environmental stimulation. For 2 other students, stereotypy occurred at high levels across all experimental conditions. For the 5th student, stereotypy was associated with negative reinforcement and the absence of environmental stimulation. In Study 2, the stereotypy of 1 student was further analyzed on a function-by-function basis. Within a concurrent-schedules procedure, alternative responses were taught to the student using functional communication training. The results of Study 2 showed that similar topographies of stereotypy, based on qualitatively different reinforcers, were reduced only when differential reinforcement contingencies for alternative forms of communication were implemented for specific response–reinforcer relations. Our results suggest that the causes of stereotypy for students with autism are complex and that the presumed association between respons