Augmentative and alternative communication in autism: A comparison of the Picture Exchange Communication System and speech-output technology

The purpose of this comparative efficacy study was to investigate the Picture Exchange Communication System (PECS) and a speech-generating device (SGD) in developing requesting skills, social-communicative behavior, and speech for three elementary-age children with severe autism and little to no functional speech. Requesting was selected as the primary measure while social-communicative behavior (eye contact, smiling, and physical proximity) and speech were selected as ancillary measures. A multiple baseline design (Baer, Wolf, & Risley, 1968) was combined with an alternating treatment design (Barlow & Hayes, 1979) to evaluate the treatment effects across participants and between modalities for each condition. A 3-stage preference assessment was conducted to systematically identify food reinforcers prior to treatment. Training conditions for both treatment modalities included the implementation of PECS phase 1 (how to communicate), phase 2 (distance and persistence), and phase 3 (picture discrimination). PECS followed the traditional protocol as outlined by Bondy and Frost (1994) and the ProxTalker, a device developed to closely follow the principles of PECS, used a modified PECS protocol. Results demonstrated increases in requesting behavior for all participants across conditions and treatment modalities, however, difficulties were evident with picture discrimination. Statistical analysis did not reveal significant differences between PECS and the SGD for any participant. Results for social-communicative behavior were mixed although an increasing trend was present in phase 2 for all participants and treatment modalities. However, no significant differences between modalities were found. Data for speech outcomes did not reveal any increases across participants and no statistical differences between treatment modalities were found. Findings suggest PECS and the ProxTalker device are equally appropriate for developing initial requesting skills. Results on social-communicative behavior are inconclusive however; data patterns suggest phase 2 is conducive to encouraging social-communicative behavior. The lack of speech emergence supports previous research. Based on the current findings, successful implementation of either modality is achievable when appropriate teaching strategies are used. Further research is needed to address difficulties in picture discrimination, mixed results in social-communicative behavior, and the non-emergence of speech. Non-significant differences between modalities suggest the behavioral principles inherit in the PECS protocol are responsible for the findings in this study; therefore, future research directions are provided