People with autism show superior performance to controls on the Embedded Figures Test (EFT). However, studies examining the relationship between autistic-like traits and EFT performance in neurotypical individuals have yielded inconsistent findings. To examine the inconsistency, a meta-analysis was conducted of studies that (a) compared high and low Autism-Spectrum Quotient (AQ) groups, and (b) treated AQ as a continuous variable. Outcomes are consistent with superior visual search forming part of the broader autism phenotype, but in existing literature, this is evident only when comparing extreme groups. Reanalysis of data from previous studies suggests findings are unlikely to be driven by a small number of high scorers. Monte Carlo simulations are used to illustrate the effect of methodological differences on results.