A Meta-Analysis of Gaze Differences to Social and Nonsocial Information Between Individuals With and Without Autism.
Level 4 - case-series / case-control
Systematic review and meta-analysis of observational case-control eye-tracking studies
PubMed 28647006 · doi:10.1016/j.jaac.2017.05.005
What was done
Authors conducted a systematic search of PubMed and 1,132 publications to identify eye-tracking studies comparing gaze-fixation behavior between individuals with autism and controls using social and nonsocial stimuli. Sample features, eye-tracking methods, stimulus characteristics, and regions of interest (ROIs) were coded. Multivariate mixed-effects meta-regression models analyzed effect sizes across 1,155 comparisons from 122 independent studies.
What was found
Overall effect sizes were small to medium and varied by stimulus and ROI. Nonsocial ROIs yielded larger effect sizes than social ROIs overall; however, the largest effect sizes were for eye regions (Hedges g = 0.47) and whole-face regions (Hedges g = 0.50) in stimuli depicting human interaction. Studies with weaker designs or reporting yielded larger effects, but key effects remained statistically significant and medium in size in high-rigor designs.
Why it matters
This meta-analysis demonstrates that gaze abnormalities in autism are robust across ages and most pronounced when viewing interactive human scenes, providing benchmark effect sizes for social attention deficits.
Limits
All included studies were observational case-control designs. Effect sizes were inflated in studies with weaker designs and incomplete reporting. Total participant counts, specific age ranges, and precision metrics (such as confidence intervals) were not reported in the abstract.