Temporal Processing in Autism Spectrum Disorder: A Systematic Review and Meta-Analysis

Background: Studies of autism spectrum disorders (ASD) have reported deficits in processing rapidly timed unisensory information, as well as integrating sensory information from different sensory modalities (Zhou et al., 2018). Failure to temporally integrate sensory cues can lead to deficits in higher-order communication and social skills (Stevenson et al., 2014). However, it is unclear whether atypical processing of unisensory and multisensory temporal information is a consistent feature in ASD.

Objectives: This study used quantitative meta-analyses to examine two standard tests of temporal processing, the temporal order judgment (TOJ) and simultaneity judgment (SJ) tasks in ASD. The aims were to 1) determine if temporal processing shows consistent deficits in ASD, 2) assess whether atypical temporal processing exists in ASD for both unisensory processing and multisensory processing, and 3) compare the TOJ and SJ tasks to evaluate whether these two types of temporal judgment show similar effects in ASD.

Methods: An exhaustive search for ASD studies using TOJ and SJ tasks from 1980 to 2018 was conducted in five online databases (Proquest, PsycInfo, Pubmed, Scopus, and Web of Science). Data from 18 different studies (Nasd=369, Ntd=408) were retrieved and coded. A synthesis and analysis of the set of ASD studies was performed following the PRISMA guidelines (Moher et al., 2009). Analyses were performed on study effects (Hedges’ g) using random-effects models, and heterogeneity of variance was measured using the I2 measure (Higgins et al., 2003). Visual inspection of funnel plots and a trim-and-fill procedure were used to assess the presence of publication bias (Duval & Tweedie, 2000).

Results: Individuals with ASD showed lower temporal acuity in simultaneity and order judgments compared to typically developing (TD) controls (n=18, g=0.444, SE=0.119, p<.001, CI= 0.210-0.677). Separate analyses of unisensory and multisensory studies provided stronger evidence for deficits in multisensory temporal processing (n=12, g=0.484, SE=0.121, p<.001, CI= 0.248-0.720), as the analysis of unisensory studies failed to reach significance (n=7, g=0.463, SE=0.252, p=.066, CI= -0.031-0.957). Lastly, temporal acuity in ASD appears to be more impacted in SJ tasks (n=8, g=0.585, SE=0.147, p<.001, CI= 0.296-0.873), versus TOJ (n=10, g=0.321, SE=0.168, p=.057, CI=-0.009-0.650). Heterogeneity of variance for all analyses fell between low and moderate (4%-40%), and no evidence of publication bias was found.

Conclusions: ASD shows lower temporal acuity than TD, as assessed by two relative timing tasks. Poorer multisensory processing was found, but unisensory processing did not differ between ASD and TD. Finally, overall temporal processing differences are larger when considering only the SJ task, which involves a judgement about simultaneity rather than order of events, a task considered to be easier than TOJ. Future work will address the association between multisensory temporal processing in ASD and symptom severity, as well as whether performance on different tasks can identify individual differences in temporal processing.