Decoding the Neural Basis of Sensory Phenotypes in Autism

Area of Science:

• Neuroscience
• Autism Research
• Computational Psychiatry

Background:

• Sensory processing differences affect up to 87% of autistic individuals, impacting cognition and daily life.
• Five distinct sensory processing phenotypes have been identified in autism, but their neural underpinnings are not well understood.
• Understanding the neural basis of sensory phenotypes is crucial for addressing the heterogeneity in autism.

Purpose of the Study:

• To investigate the neural basis of five identified sensory processing phenotypes in autism.
• To determine if unique patterns of functional brain connectivity differentiate these sensory phenotypes.
• To ground autism sensory phenotypes in measurable neural network characteristics.

Main Methods:

• Analyzed functional connectivity data from 146 autistic participants.
• Classified participants into five sensory phenotypes using k-means clustering of Short Sensory Profile scores.
• Utilized graph theory and machine learning to analyze brain connectivity measures and classify phenotypes.

Main Results:

• Successfully clustered participants into five distinct sensory phenotypes.
• Machine learning models differentiated seven out of ten pairs of sensory phenotypes using graph-theoretic measures (p < 0.005).
• Predictive connectivity patterns involved the somatomotor network, orbitofrontal cortex, parietal cortex, prefrontal cortex, and subcortical regions.

Conclusions:

• Autism sensory phenotypes are associated with specific functional connectivity differences at cortical, subcortical, and network levels.
• These findings demonstrate that sensory processing variability in autism is reflected in distinct neural patterns.
• This research supports refining models of autism sensory processing to better capture individual heterogeneity.