The Relationship Between Joint Attention and Language in Autism Spectrum Disorder and Typical Development: A Systematic Review and Meta-Regression Analysis

Background:  Research has shown that joint attention plays a crucial role in the development of expressive and receptive language. However, joint attention is a complex construct, and the relative correlations of all different types on development have not been sorted out. In children with autism spectrum disorder (ASD), parsing the impacts of of joint attention is especially important, as joint attention delays negatively impact language development. This study used a structured literature search and meta-regression procedures to synthesize Pearson’s r effect sizes measuring correlations between joint attention and language in typically developing (TD) children and children with ASD.

Objectives:  To determine: 1) summary associations between joint attention and receptive and expressive language, for TD and ASD children, 2) whether diagnostic group moderates effect sizes, 3) whether joint attention type moderates effect sizes, and 4) which moderators maintain significance when significant moderators from simple meta-regressions were entered together. 

Methods:  Studies were located using database searches, hand searches, and electronic requests for data from approximately 200 experts in the field. The search resulted in 71 reports or datasets and 605 Pearson’s r effect sizes, representing 1,859 participants with ASD and 1,835 TD participants (see Table 1 for sample and study characteristics). Robust variance estimation was used to account for clustering of effect sizes within studies. Meta-regression was used to answer research questions regarding potential moderators of effect sizes. Joint attention types were categorized into initiating joint attention, responding to joint attention (RJA), coordinated attention, supported joint engagement, and coordinated joint engagement. 

Results:  Summary effect sizes collapsed across joint attention types varied by diagnostic group and language outcome (receptive language, r = .27 and r = 43 for TD children and children with ASD respectively; expressive language, r = .22 and r = .45 for TD children and children with ASD respectively). Meta-regression analyses indicated that effect sizes were significantly higher for the ASD group as compared to the TD group, and for RJA as compared to non-RJA joint attention types for both expressive and receptive language. Approximate mental age trended toward significance for expressive language, even after controlling for diagnosis and RJA (see Table 2 for full results of regressions). 

Conclusions:  Joint attention may be more tightly tied to language in children with ASD as compared to TD children because TD children exhibit joint attention at sufficient thresholds so that language development is a ‘sure thing’ and untethered to variations and joint attention. Conversely, children with ASD who exhibit deficits in joint attention develop language contingent upon their joint attention abilities.  Because RJA was more strongly related to language than other types of joint attention, future research should consider carefully the operationalization and measurement of joint attention constructs. The results of this meta-analysis support intervention approaches that facilitate joint attention as a means to improve language outcomes in children with ASD.