Brenda (Siyue) Qiu¹, Kristin M. Woodard¹, Bridget Leonard¹, Hannah M. Rea², Sara J. Webb^1,3, Scott O. Murray¹; ¹University of Washington, ²West Virginia University, ³Seattle Children’s Research Institute

Sensory hypersensitivity is frequently reported in Autism Spectrum Disorder (ASD), yet neural evidence in early visual cortex is mixed. Studies have reported increased (Brieber et al., 2010), typical (Dinstein et al., 2012), or reduced (Kolodny et al., 2020) V1 responses in autistic individuals. It has been proposed that reduced V1 activity may be attributed to enhanced suppressive feedback from downstream regions such as MT+ (Kolodny et al., 2020), suggesting that atypical responsivity in ASD may reflect differences in how higher-order areas modulate early signals rather than alterations in V1 encoding itself. To evaluate this, we used fMRI and a spatial attention paradigm to examine responses in V1 and higher-order, attention-modulated regions in 50 non-autistic and 24 autistic participants performing a motion discrimination task. Following an attention cue, participants covertly attended to one of two bilaterally presented drifting Gabor gratings at 0% (stimulus absent), 6%, or 50% contrast. Participants judged the motion direction of the attended grating. Accuracy did not differ between groups. Consistent with prior work, non-autistic participants showed an additive effect of attention in V1 and MT+ (Murray, 2008). Here, we found that autistic participants demonstrated the same pattern. Additive attentional modulation was also observed in IPS and FEF for both groups. However, group differences emerged at different levels of the visual hierarchy. In V1, non-autistic participants showed stronger attentional modulation than autistic participants, despite comparable overall response magnitude. In contrast, MT+, IPS, and FEF showed similar attentional modulation across groups, but autistic participants showed higher responses at 50% contrast, reflecting increased responsivity to strong sensory drive. The emergence of this effect in higher-order regions, but not in contrast-sensitive V1, supports the notion that hypersensitivity in autism may arise from later integrative processes where sensory input and top-down signals interact, rather than from altered early sensory encoding.

Acknowledgements: This work was supported by funding from the National Institutes of Health (R01MH131595 to S.J.W. and S.O.M.)