Audiovisual multisensory integration in individuals with autism spectrum disorder: A systematic review and meta-analysis

Abstract

An ever-growing literature has aimed to determine how individuals with autism spectrum disorder (ASD) differ from their typically developing (TD) peers on measures of multisensory integration (MSI) and to ascertain the degree to which differences in MSI are associated with the broad range of symptoms associated with ASD. Findings, however, have been highly variable across the studies carried out to date. The present work systematically reviews and quantitatively synthesizes the large literature on audiovisual MSI in individuals with ASD to evaluate the cumulative evidence for (a) group differences between individuals with ASD and TD peers, (b) correlations between MSI and autism symptoms in individuals with ASD and (c) study level factors that may moderate findings (i.e., explain differential effects) observed across studies. To identify eligible studies, a comprehensive search strategy was employed using the ProQuest search engine, PubMed database, forwards and backwards citation searches, direct author contact, and hand-searching of select conference proceedings. A significant between-group difference in MSI was evident in the literature, with individuals with ASD demonstrating worse audiovisual integration on average across studies compared to TD controls. This effect was moderated by mean participant age, such that between-group differences were more pronounced in younger samples. The mean correlation between MSI and autism and related symptomatology was also significant, indicating that increased audiovisual integration in individuals with ASD is associated with better language/communication abilities and/or reduced autism symptom severity in the extant literature. This effect was moderated by whether the stimuli were linguistic versus non-linguistic in nature, such that correlation magnitudes tended to be significantly greater when linguistic stimuli were utilized in the measure of MSI. Limitations and future directions for primary and meta-analytic research are discussed.

Keywords: Audiovisual; Autism; Meta-analysis; Multisensory integration; Sensory.

Figure 1.

Flow chart of search results. ASD = autism spectrum disorder.

Figure 2.

Percent of studies included in the qualitative synthesis reporting each variable coded to characterize study samples. Only mean chronological age of participants was available for every study.

Figure 3.

Forrest plot for meta-analysis of group difference effect sizes. *Reference not in the peer-reviewed literature (k = 11; 7 dissertations or theses, 3 posters presented at academic conferences, and 1 manuscript in preparation). Results suggest a small-moderate sized effect for between group differences in audiovisual multisensory integration on average across studies. Note that this effect varied according to mean chronological age of study participants.

Figure 4.

Forrest plot for meta-analysis of correlational effect sizes. *Reference not in the peer-reviewed literature (k = 3; 1 dissertation or thesis and 2 posters presented at an academic conference). Results suggest that audiovisual multisensory integration shows a small to moderate association with autism and related symptomatology on average across studies. Note that the magnitude of associations varied according to whether linguistic versus non-linguistic stimuli were used in tasks tapping multisensory integration.

Figure 5.

Funnel plot for meta-analysis of group difference effect sizes (top; Figure 5A) and correlational effect sizes (bottom; Figure 5B). Results indicate no evidence of publication bias.