The effect of grouping on numerosity judgment varies with the autism-spectrum personality traits, consistent with changes in local-global biases.

Individuals with autism spectrum disorder or with non-clinical autistic tendencies are thought to have a perceptual style privileging local detail over global integration. Here we investigated whether the effect of grouping-induced response biases in numerosity judgments depends on autistic-like personality traits in typically developing adults. Perceived numerosity was measured by two alternative forced choice (2AFC), where participants reported which of two clouds of dots appeared more numerous, guessing when uncertain. In one presentation 40% of dots were grouped into “single items” (dumbbell-like shapes) by connecting pairs of neighboring dots with thin lines. Four different numerosities were tested (15, 25, 50 & 100) in separate sessions. As has been previously reported, at lower numerosities (≤50) all participants underestimated the numerosity of the stimuli with connected dots, by about 20%. The effect was much reduced (and insignificant) at N=100, where the dots were crowded. Importantly, the amount of bias at lower numerosities varied inversely with AQ, being greater for those with lower autistic traits (r = 0.72, p < 0.001, Bayes Factor = 126). As it is thought that connecting dots activates global mechanisms that cause dot-pairs to be processed as an integrated whole rather than as individual dots, the results suggest that these global grouping mechanisms may be weaker in individuals self-reporting high levels of autistic-like traits. The effects of grouping on extracting high-order information from the visual field may add an important piece to the puzzle of how visual processing differs in autism. We conclude that the preference for processing local over global elements in numerosity judgments varies along the autism continuum in the general population and might also be affected in individuals with ASD.