The competition of autistic stereotyped behavior with usual and specially assessed reinforcers.

This study was conducted to empirically assess a reinforcement theory of stereotyped behavior. Six students with autism were first presented with tasks, and no contingent reinforcers were provided for correct responding. Then, contingent reinforcers that were typically used with the students (usual reinforcers) were presented in a multiple baseline across subjects design. Three of the students evidenced decreases in stereotypy and increases in responding in the presence of usual reinforcers. The other three students required external suppression of stereotypy before increases in responding were shown. For these students, usual reinforcers and specially assessed reinforcers were then compared. The specially assessed reinforcers resulted in decreases in stereotypy and increases in responding and subjective measures of responsiveness. The results were discussed in terms of supporting a competing reinforcement hypothesis, such that powerful external reinforcers will successfully compete with and suppress reinforcers provided by stereotypy.