Effects of a Group Contingency on the Frequency of Social Interactions among Autistic and Nonhandicapped Preschool Children: Making LRE Efficacious

Abstract

This study was designed to evaluate the use of a group-oriented reinforcement contingency to (a) increase the frequency of verbal interaction between severely withdrawn, autistic preschool children and nonhandicapped peers, and (b) reduce the levels of teacher prompts needed to maintain interaction. After baseline, both handicapped and nonhandicapped children were taught initiation and response strategies and the frequency and types of teacher prompts were restricted. Initially, access to the group reinforcer depended upon all three play triads achieving an established criterion for appropriate interactions. The teacher awarded a token for each appropriate interaction, allowing children to monitor their progress. Marginal increases in appropriate interactions occurred under this condition. In a subsequent interdependent contingency system, differentially reinforcing each triad that met the established criterion produced reliable increases in appropriate social initiations and responses for both peers and handicapped children. Rates of teacher prompts remained low. Withdrawal of the group-oriented reinforcement contingency substantially reduced levels of appropriate initiations and responses by both peer confederates and handicapped children. Reinstatement of the group contingency resulted in immediate and significant increases in appropriate initiations and responses by both peers and handicapped children. The results suggest that (a) peers can be effectively taught strategies to increase the appropriate social behaviors of their withdrawn classmates, and (b) a group-oriented reinforcement contingency is an effective complement to teacher-prompted interventions and may be an effective means of systematically fading teacher prompts.