Programming generalization of social skills in adults with developmental disabilities: Effects on generalization and social validity

Although many studies have demonstrated the acquisition of social skills in persons with developmental disabilities in the training setting, in vivo generalization has been more difficult to obtain. In this study, squads comprising 3 to 4 adults with developmental disabilities from the same group home were randomly assigned to a ''game'' condition (n = 15) with three generalization strategies, a ''social life'' condition (n = 13), which added to the game condition four generalization strategies, or a ''no-training'' control group (n = 15). After training, the social life group was significantly superior to the no-training control group on all the generalization and social validity measures; the game condition was significantly better than the control group on one posttest generalization measure only. Furthermore, the social life condition was significantly superior to the game condition on percentage correct in the last game session and in a 3-month follow-up on the frequency of target skills displayed in the participants' natural settings. The findings indicate that meaningful generalization and social validity can be obtained when sufficient generalization strategies are explicitly programmed in the social skills game format.