The author reviews the history of the behavioral approach in treatment of autistic children and contrasts aspects of the disease or illness model with the learning or behavioral model. He states that the disease model failed in identifying the cause and in providing effective treatment for autism. (CL)

This article describes several features important in the success of using a behavioral model when treating individuals with autism and presents remaining issues to be addressed for improving treatment effectiveness. It urges future research to include strict empirical designs that allow for sound inferences regarding the parameters of treatment…

The possibility of differential sex proportions as a confounding factor in the Lovaas (1987) study is raised. It is argued that the chi-square analysis was inadequate and that the appropriate comparison should be made between the experimental and the control group using population data to estimate expected cell frequencies. (Author/CR)