Assessing Generalization of PECS

Greenberg, A., Tomaino, M., & Charlop, M. (2012). Assessing generalization of the Picture Exchange Communication System in children with autism. Journal of Developmental and Physical Disabilities, 24, 539-558.

Reviewed by: Melissa Taylor, BCaBA, ASAT Extern Karen Fried, PsyD, BCBA-D, ASAT Board Member

Why research this topic?

The Picture Exchange Communication System (PECS) is a popular approach to teaching children with autism to communicate by selecting pictures and handing them to a communication partner. For PECS to be an effective communication system, the learner must be able to use it in a variety of situations and with a variety of people. When this variety is achieved, PECS use is considered “generalized”. Adults with developmental disabilities have been shown to achieve generalized use of PECS as a communication system in different settings and with novel people. In contrast, research remains limited on PECS generalization in children with autism. The goal of this study was to contribute to our knowledge of PECS generalization in children with autism and to explore a “train and probe” technique of assessing generalization in this population.

What did the researchers do?

Four nonverbal boys with severe autism between the ages of four and eight years old participated in the study. The boys received individual training to use PECS. The design used for this study, a “multiple baseline design across participants,” is an experimental method for evaluating the effects of an intervention on each participant. The researchers provided PECS training in a small room at the children’s after-school behavioral treatment program one afternoon a week for two hours. The researchers also tested the children in specific environments before and during training to probe whether the children used their PECS to request items without direct training in that environment. The researchers also tested whether the children used their PECS across people, including their therapist, their parent, and an unfamiliar store clerk.

What did the researchers find?

After the initial training, three of the four children used PECS much more frequently in the four different types of situations that the researchers tested. One of the four participants required additional training to reach this outcome. At follow-ups conducted 1-18 months after the intervention ended, the children continued to use PECS in all locations. The findings indicate that when learning PECS in one setting with one instructor, children with autism can generalize PECS across settings and people. The findings also support the use of a “train and probe” technique to assess generalization after each phase of PECS training.

What are the strengths and limitations of the study?

At the start of this study, the researchers documented that all four boys showed signs of severe autism (as measured by a widely used rating scale), were nonverbal and lacked PECS skills. To ensure the data were valid, the researchers had a second observer take data on a segment of the sessions to compare coding of the children’s responses and to assess procedural accuracy. One limitation of this study was that generalization probes all occurred with the same people (therapist, parent, store clerk) and in the same environments (treatment center, home, local store). For a child to use PECS fully functionally, assessment of generalization should span a larger variety of people and locations. Also, as with most other studies that use a single-subject design, the small number of participants is a limitation; conclusions are accepted with caution until results are replicated with other participants.

What do the results mean?

This study replicated previous studies demonstrating that children with autism can learn to use PECS as a functional communication system. The researchers showed that PECS skills generalized across settings and people by the end of training. The study highlighted the importance of assessing generalization after each phase of PECS training rather than waiting until the end. When PECS training is limited to specific settings, it is even more important to test for PECS generalization so that booster trainings can be delivered as needed to individual children. Frost and Bondy, the creators of the PECS approach, recommend that adults create communication opportunities throughout a child’s day by restricting access to preferred items. The current study supported this recommendation by designing the free-play environment to include restricted access to highly preferred items and frequent temptations with these items. More research should be conducted looking at generalization across other environments and communication partners (at school, with peers). Future research might also compare the commenting skills of PECS users to users of other modes of communication.

Citation for this article:

Taylor, M., & Fried, K. (2016). Review of Assessing generalization of the Picture Exchange Communication System in children with autism. Science in Autism Treatment, 13(1), 24-25.