The Impact of Augmentative and Alternative Communication on the Development of Functional Communication Skills in Individuals with Autism Spectrum Disorders: a Meta-Analysis of Intervention Research from 1976-2008

Background:

Approximately 14-25% of children diagnosed with ASD present with little or no functional speech and are candidates for augmentative and alternative communication (AAC). AAC replaces or supplements natural speech and handwriting through aided or unaided, alternative forms of communication such as, for example, manual signs and gestures, graphic symbols, and speech-generating devices (SGDs). AAC interventions aim to develop communication skills such as requesting, natural speech production, and social communicative behaviors.

Objectives: To evaluate the existing experimental research literature and to review the evidence for the effectiveness of AAC in ASD.

Methods: Meta-analytic procedures were applied to aggregate and evaluate intervention studies. Meta-analysis is a statistical reviewing technique that provides a quantitative summary of findings across an entire body of research. This meta-analysis involved the following elements: (a) an extensive literature search containing computerized database searches, hand-searches in professional journals, and ancestral searches; (b) stringent inclusion criteria to ensure only experimental studies of methodological rigor entered the meta-analysis; (c) a pilot-tested coding protocol plus reliability analysis on both study inclusion and coding; (d) aggregating intervention outcomes statistically by conducting separate analyses for group and single-subject design data and implementing adequate effect size metrics for each, i.e., Cohen's d and Hedges' g for group studies and Percentage of Non-overlapping Data, Improvement Rate Difference and Percentage of Pairwise Data-Overlap for single-subject studies.

Results: A total of 43 single subject and 2 group studies met the inclusion criteria. These studies involved a total of 118 participants, the majority of which was between six and ten years old, diagnosed with autism and no functional speech. Meta-analysis of pooled effect sizes yielded the following results.


Functional requesting skills: AAC interventions based on unaided approaches such as gestures and manuals signs showed equal effectiveness when compared to interventions based on graphic symbols. Both are significantly more effective than using SGDs.


Natural speech production: When increasing vocalizations was the target outcome, manual signs, the Picture Exchange Communication System (PECS), and SGDs, yielded mixed results of effectiveness regardless of intervention type. Increasing imitative speech yielded higher effect sizes compared to spontaneous speech, and aided language stimulation appeared to be superior to PECS training in targeting these outcomes.


Social-communicative behaviors: PECS was found to be “highly effective” for increasing cooperative play, joint attention, and eye contact.

Conclusions: The use of graphical symbols for the teaching of requesting offers the most solid empirical evidence. Research however, has not reached a level at which it could productively inform choices of one graphic symbol type over others. Manual signs and gestures might play a bigger role if they were part of a multimodal communication system while being mindful of their limitations. Research is lacking on effective strategies for teaching conditional use of manual signs. Speech production data seem independent of how well a child has acquired the introduced AAC modality. Research needs to investigate how acquiring different AAC modalities is related to effectiveness in speech production.