Taylor L. Li¹ (), Andrew S. Persichetti¹, Cassandra M. Levesque¹, W. Dale Stevens², Adrian W. Gimore¹, Alex Martin¹; ¹NIMH (National Institute of Mental Health), ²York University

A small, but intriguing, behavioral literature on individuals with autism spectrum disorder (ASD) has shown that they are impaired when navigating using an allocentric reference frame (i.e., map-based navigation), but are not impaired when navigating using an egocentric reference frame (i.e., visually guided navigation). Meanwhile, there is some neuroimaging evidence in typically developing individuals (TD) demonstrating that the retrosplenial complex (RSC) is part of a map-based navigation system, while the occipital place area (OPA) is part of a visually-guided navigation system. The key identifying feature of the RSC and OPA (putative place-processing cortical regions) is that they respond significantly more to pictures of places compared to pictures of objects or faces (i.e., place selectivity). Therefore, we predicted that in individuals with ASD, the RSC would show significantly weaker place selectivity when compared to a TD control group, while the OPA would not show such a difference between the groups. To test our prediction, we used functional MRI (fMRI) to scan groups of ASD and TD individuals that were matched on age, IQ, in-scanner head motion, and tSNR. During the scan, participants viewed pictures of unfamiliar places, faces, and objects while performing a one-back task. We operationalized place selectivity as the difference between the brain responses to places and the average responses to faces and objects in each region. As predicted, independent-samples t-tests showed that place selectivity was significantly lower in the RSC, but not the OPA, in the ASD group compared to the TD group. Crucially, we also found a significant ROI x Group interaction in a two-way ANOVA. These results suggest that impaired map-based navigation in individuals with ASD may be due to atypical functioning in the RSC.