Nature Communications (Jun 2023)

Neural correlates of hierarchical predictive processes in autistic adults

  • Laurie-Anne Sapey-Triomphe,
  • Lauren Pattyn,
  • Veith Weilnhammer,
  • Philipp Sterzer,
  • Johan Wagemans

DOI
https://doi.org/10.1038/s41467-023-38580-9
Journal volume & issue
Vol. 14, no. 1
pp. 1 – 14

Abstract

Read online

Abstract Bayesian theories of autism spectrum disorders (ASD) suggest that atypical predictive mechanisms could underlie the autistic symptomatology, but little is known about their neural correlates. Twenty-six neurotypical (NT) and 26 autistic adults participated in an fMRI study where they performed an associative learning task in a volatile environment. By inverting a model of perceptual inference, we characterized the neural correlates of hierarchically structured predictions and prediction errors in ASD. Behaviorally, the predictive abilities of autistic adults were intact. Neurally, predictions were encoded hierarchically in both NT and ASD participants and biased their percepts. High-level predictions were following activity levels in a set of regions more closely in ASD than NT. Prediction errors yielded activation in shared regions in NT and ASD, but group differences were found in the anterior cingulate cortex and putamen. This study sheds light on the neural specificities of ASD that might underlie atypical predictive processing.