Objectives: were to evaluate effects of a modified PECS protocol on requesting skills, social-communicative behaviors, and emerging speech.
Methods: Three boys, 9-11 years with severe autism and no functional speech participated. A multiple baseline design across participants evaluated treatment effects. The intervention was split into PECS and SGD phases, followed by a maintenance phase. Dependent measures were: (a) number of correct requests during a 20-trials session; (b) number of responses including eye contact, physical orientation towards communication partner, and displaying positive affect; (c) number of word vocalizations/approximations.
Intervention materials consisted of a traditional PECS book with PCS symbols for desired items and and two types of SGDs, the ProxTalker and the iPad with IPAAC app. The SGDs were taught at the stage of picture discrimination.
Reliability for dependent and independent variables was established by re-scoring 40% of sessions. Mean agreement was 99% for requesting, 90% for social-communicative behavior, and 92% for emerging speech. Treatment integrity was 98%.
Data were graphed using Excel and visually inspected for changes in level and trend. Effect sizes were estimated via Percentage of Non-overlapping Data (PND) and the Non-overlap of All Pairs index (NAP).
Results: Children 1 and 2 mastered all five phases of the modified protocol, whereas child 3 only achieved phase 3. Acquisition rates varied, with children 1 and 2 needing on average 6 sessions to mastery, and child 3 needing 7 sessions. The largest gain appeared for requesting: For child 1, requesting improved from a baseline mean of 0.25 to a maintenance mean of 17.75; for child 2 from 1.1 to 17.5; and for child 3 from 0.6 to 11.3. PND/NAP scores indicated “highly effective” for all. Varying results were obtained for social-communicative behavior: Child 1 improved from a baseline mean of .5 to a maintenance mean of 11.5; child 2 from 15.8 to 23.5; and child 3 from 7.3 to 13.6. PND/NAP scores were “fairly effective” to “highly effective” for children 1 and 2; but yielded “questionable effectiveness” for child 3. Results differed the most for emerging speech: only child 1 made noticeable gains. PND/NAP scores were “fairly effective” for child 1, and “ineffective” for the others.
Conclusions: Results suggest a smooth transition from the traditional PECS to an SGD is possible, if protocol is implemented with high fidelity. Children may not be able to complete all stages, and for child 3, this could be due to comorbid, severe intellectual impairment. Treatment effects are most noticeable for functional communication and social-communicative behaviors. A facilitative effect on speech development cannot necessarily be expected.