The self-management program described here was used to teach individuals with autism (chronologically 9-14 years old but functioning in the 2- to 5-year old range according to standardized test scores) to self-manage their stereotypic behaviors (R.L. Koegel & Koegel, 1990). Some of the individuals' stereotypic behaviors were so disruptive and conspicuous that the children were being considered for more restrictive and nonintegrated placements (e.g., one boy's mother was considering a residential placement for her son and another boy's private school teacher was recommending his dismissal). The individuals in this study wore a wristwatch alarm to signal self-recording opportunities. After they were taught to self-manage their behavior in a controlled environment, dramatic decreases in stereotypic behavior (typically to 0%) occurred. The amount of time between opportunities to self-record was gradually lengthened until each of the participants was able to independently self-manage his or her behavior for several consecutive 15- to 20-minute time periods before exchanging their check cards for a reinforcer.

After the children learned to independently utilize the self-management skills they were taught, self-management was programmed to occur in their natural environments where alternate and more restrictive environments were being considered. To do this, the provider merely needed to enter the second environment for a few minutes to tell the children to begin self-management procedures and then gradually fade out of the environment by leaving for increasingly longer periods of time. The children then continued to self-manage their behaviors in the absence of the provider and without the need for constant vigilance on the part of others, such as parents or a teacher. Most important, appropriate behavior was maintained in future environments simply by allowing the self-management materials to remain with the children in these new environments. For example, after being taught to successfully self-manage stereotypic behavior in his classroom, one student was told by his classroom teacher (a person not associated with intervention) to take the self-management materials with him to the park where his class ate lunch. Prior to this, stereotypic behavior was occurring at a very high level (typically 100%) in the park. However, after being instructed to take his self-management materials with him, this student's stereotypic behavior decreased to 0%-20% in the park without a provider ever having to enter that environment. In other words, the child was essentially conducting his own intervention, indicating that a "teach the individual" model of generalization had evolved. By the conclusion of the study, all of the children had decreased their levels of stereotypic behavior in their natural environments to such negligible levels that plans for more restrictive placements were terminated.