Behavioral self-control of stuttering using time-out from speaking.

Research has shown that stuttering may be attenuated by a variety of response-contingent consequences. To date, however, few attempts have been made to develop comprehensive clinical procedures based on the operant manipulation of stuttering. The present research examined the efficacy of self-initiated response-contingent time-out from speaking in two experiments involving a single subject. Multiple baseline designs were used in which the subject's speech was monitored in a variety of laboratory and community settings. In Experiment 1, the time-out procedure produced reliable reductions in disfluency across all observational settings. Experiment 2 combined a response-cost contingency with time-out in an attempt to increase the reliability with which the subject timed himself out, and to increase further the effectiveness of the procedure. Increases in reliability were produced, and the associated improvements in fluency were maintained at 6- and 12-months' follow-up.