What Do We Know About ABI for Young Children with Autism?

We know that autism is the fastest growing developmental disability in the United States (Autism Society of America, 2003). We know that children with autism spectrum disorders are being identified at earlier ages and that consequently increasing numbers of young children with autism are receiving early intervention in integrated early childhood settings. We know that the ABI approach appears to be gaining social validity with early interventionists and empirical support from researchers (Bricker, 2000; Hemmeter, 2000). We know that specific aspects of naturalistic instruction compatible with ABI have been used successfully with children with autism for some time (see, for example, Koegel, Dyer, & Bell, 1987; McGee, Krantz, & McClannahan, 1985).

Now we need to know how children with autism spectrum disorders currently are faring in these ABI environments. Although the total number of studies describing outcomes for children with autism in ABI settings is small, early results appear promising. At least four of the studies in the Pretti-Frontczak et al. (2003) review of ABI research included preschoolers with autism in their reports of positive outcomes. Recently two studies from the University of Washington have added to the emerging body of research on the effectiveness of an ABI approach with young children with autism spectrum disorders. Conclusions drawn from each of these six studies are described below.

Dunst, Bruder, Trivette, Hamby, Raab, and McLean (2001)
Child functioning in natural environments was enhanced when researchers used an activity matrix to amplify child engagement and taught parents how to respond to their child’s initiations. In addition, child functioning was enhanced when frequent and varying opportunities for participation in activities were provided. Three of the 63 children participating in the study had a diagnosis of autism. This study is part of an early childhood research institute examining methods of increasing the participation of young children with disabilities in natural learning opportunities. Results suggest that learning opportunities that are related to the interests of the child, promote active engagement, allow children to express competence, and encourage the development of a sense of mastery are more likely to result in positive behavioral change for children.

Kohler, Anthony, Steighner, and Hoyson (1998)
Four children with autism enrolled in inclusive preschool settings demonstrated increased social interaction skills after their teachers were trained to use naturalistic teaching strategies consistent with an activity-based approach to intervention. Following the child’s lead and enticing his or her interests led to increased success in facilitating social interactions with adults and peers. In addition, teachers were trained to use seven specific instructional strategies consistent with a naturalistic approach to intervention. Teachers received individualized instruction and daily feedback regarding their use of the naturalistic teaching strategies with their target students. When technical assistance was withdrawn in the maintenance phase of the study, teachers continued to use the naturalistic teaching strategies and two of the three children who participated in the maintenance phase continued to exhibit high rates of social interaction.

Kohler, Strain, Hoyson, and Jamieson (1997)
This study examined the effects of combining naturalistic teaching strategies with peer-mediated interventions implemented in integrated preschool settings during daily activities and routines. Participants included 10 children with autism, 3 to 5 years of age, and 22 typically developing preschool peers. In the initial phase, six teachers were trained to use naturalistic teaching methods to address the goals and objectives of the children with autism. In the next phase, teachers incorporated peer modeling, cooperative play, and verbal interactions with peers into embedded instructional episodes. The involvement of peers in the combined-treatment phase resulted in a substantial increase in the duration and rate of skills addressed during teaching episodes. Teachers continued to use peer-mediated strategies with naturalistic instruction in the brief maintenance phase of the study.

Schwartz, Sandall, Garfinkle, and Bauer (1998)
This article presents case studies of three children with autism who made impressive developmental and academic gains in an integrated early childhood program affiliated with a public school system. The program serves children 3 to 6 years of age with a wide range of abilities; about 30% of the children in the program who qualify for special education services have an autism spectrum diagnosis. Each preschool classroom includes 15 children, 6 of whom are typically developing. Early childhood classrooms are play-based, with naturalistic instruction embedded in daily activities for the most part, and direct instruction strategies provided as needed. Teachers use an activity matrix for children to facilitate planning and data collection in ongoing activities. Preschool classes operate for 2 hours and 15 minutes per day, 5 days per week, for 180 days a year. Kindergarten classes for children in this program follow an academic focus with an extended free-choice period each day. Kindergarten students attend class 6 hours per day, 180 days per year.

All three subjects of these case studies attended the early childhood program from age 3 until completing kindergarten at age 6. At the time of publication, all three children were reported to be thriving in inclusive first- or second-grade classrooms and one of the children was no longer receiving special education services. Based on their findings, the authors suggest that children with autism can make substantial gains in inclusive early childhood settings with “systematic and explicit instruction embedded in all classroom activities.” (p. 134). Nevertheless, the authors caution readers that their data are retrospective and that their subjects were not randomly selected.

McBride and Schwartz (2003)
Three children and their teachers participated in this study designed to evaluate the effects of teaching early childhood professionals to incorporate discrete trial instruction (DTI) with ABI. Two of the three preschoolers were diagnosed with an autism spectrum disorders. Results suggest that teacher training was successful in increasing the rate of teacher instruction, child engagement was maintained or increased, and all three children showed increased rates of correct responding to target instructional objectives when DTI was embedded into naturalistic classroom activities. All three teachers reported that they would continue to use the DTI strategy as a component of ABI.

Venn, Wolery, Werts, Morris, DeCesare, & Cuffs (1993)
This study evaluated the effects of embedding a strategy involving progressive time delay in order to teach imitation skills to 3 preschoolers with disabilities enrolled in a mainstreamed program for young children with autism spectrum disorders. One of the children had autism, one was described as having autistic features, and the third had visual impairment and developmental delays. Requests (5 per day) to imitate peers were embedded into teacher instructions for each child during daily art activities. Findings from the multiple probe design study suggest the procedure was effective in producing increased levels of imitation of peers for all 3 participants. Imitation also generalized to other fine motor activities.