National trends indicate an increase in the number of children identified as having autism spectrum disorders (Autism Society of America, 1999). Diagnostic criteria for Autistic Disorder include qualitative impairment in social interaction and communication, a failure to develop peer relationships, and use of nonfunctional rituals and routines (DSM-IV, 1994). The challenges of this growing population have promoted an increased focus by educators and families on issues related to effective instruction and appropriate social and behavioral supports for students with autism within school and community settings. Research-validated, effective educational programming has been forthcoming with a consensus for (a) early and intensive one-to-one language intervention (e.g., Lovaas, 1987); (b) the use of small groups, peers, and individualized instruction in functional and academic skills (e.g., Gessler Werts, Caldwell, & Wolery, 1996; Kamps et al., 1995, 1997); and (c) inclusive education with accommodations and supports for students progressing in the academic curriculum (e.g., Koegel, Harrower, & Koegel, 1999).
A key to accommodating students with autism in public school settings is the provision of social and behavioral programming to develop meaningful participation with nondisabled peers (e.g., Koegel, Koegel, & Dunlap, 1996). The present investigations are part of a series of studies by the authors designed to investigate the benefits of peer mediation and social skills interventions for students with autism and other developmental disabilities. Prior reports provided empirical evidence supporting (a) peer tutoring and cooperative groups (Kamps, Barbetta, Leonard, & Delquadri, 1994); (b) social skills groups (Kamps et al., 1992); and (c) peer networks with academic and social components (Kamps et al., 1997).
These studies employed strategies targeting primarily the behavior of children with autism participating with peers. Yet to be examined, however, is the role of peers when trained in explicit interaction strategies with children with autism during and after specific treatment, and generalization of skills for target and peer participants to nontraining settings and with novel students.
The current studies investigated the role of peer training embedded within interventions to maximize participation for students with autism and the social benefits for all participants. Study 1 was an initial investigation of peer training within the contexts of social skills and cooperative learning groups in one school setting with a small number of students with autism. Study 2 replicated and refined these procedures with 34 student with autism and their peer groups across multiple school districts and multiple school years. Maintenance and skill generalization across participants (Stokes & Osnes, 1988) in both studies were measured via videotaped social behavior probes (Haring, Breen, Pitts-Conway, Lee, & Gaylord-Ross, 1987).
This study examined the effects and generalization of three conditions: (a) social skills, (b) cooperative learning, and (c) control groups in which forms of peer training were embedded within the intervention. The overall design was a single subject reversal design (Barlow & Herson, 1984) including a no treatment baseline and social skills or cooperative learning groups intervention. Fall and spring social interaction behavior probes (videotaped sessions) were used to monitor maintenance and generalization effects to nontraining settings. The primary analysis of dependent variables (frequency, mean length, and duration of interactions) indicated that students changed with increased interaction levels during intervention conditions (see Dugan et al., 1995). In a secondary analysis, generalization effects were examined beyond the intervention settings. Three peer groups were available for this analysis of generalization: (a) those who participated in cooperative learning groups with students with autism, (b) those who participated in social skills groups with students with autism, and (c) a group of peers who were familiar with the students with autism but had not received training. …