Gianluca Marsicano^1,2 (), David Melcher^1,2; ¹Psychology Program, Division of Science, New York University Abu Dhabi, Abu Dhabi, United Arab Emirates, ²Center for Brain and Health, NYUAD Research Institute, New York University Abu Dhabi, Abu Dhabi, United Arab Emirates

Causality judgments, which involve perceiving one event as causing another, are fundamental to understanding our sensory environment. Here, we investigated how perceptual history and individual differences in autistic and schizotypal traits shape visual causality judgments in the neurotypical population. Participants (n=90) reported causality judgments (causal/non-causal) for varying collision lags. We also measured autistic and schizotypal traits via questionnaires, and a data-driven cluster analysis divided individuals into three groups: high autistic traits (ASD-like), high schizotypal traits (SSD-like), and low traits (LT). Causal response rates were modelled as a function of collision lags, and serial dependence analysis examined how prior trial judgments influenced current decisions. Overall, perceptual history significantly shaped causality judgments, with trials following causal responses more likely to be judged causal, and vice versa. Consistent with prior research suggesting that ASD and SSD can represent two poles of a predictive continuum, the SSD-like group exhibited a stronger tendency toward causality judgments and greater serial dependence, while the ASD-like group showed reduced serial dependence compared to the LT group. We characterized these individual differences in terms of encoding and decoding mechanisms using a hierarchical drift diffusion model. Results revealed that the ASD-like group required longer sensory encoding times and adopted a more cautious response strategy (greater decision thresholds) while displaying faster evidence accumulation (higher drift rate). In contrast, for the SSD-like group the encoding processes did not differ from LT, but they exhibited an initial bias toward causal responses and a lower drift rate, reflecting greater reliance on prior models instead of immediate sensory evidence. Together, these findings highlight distinct atypicalities for ASD- and SSD-like profiles in the encoding and decoding mechanisms underlying visual causality judgments, providing insights into how sensory encoding, evidence accumulation, and perceptual history interact to shape causality perception.

Acknowledgements: This work was supported by the NYUAD Center for Brain and Health, funded by Tamkeen under NYU Abu Dhabi Research Institute grant CG012.