Exploring EEG resting state differences in autism: sparse findings from a large cohort.

Area of Science:

• Neuroscience
• Autism Spectrum Disorder Research
• Electroencephalography (EEG) Analysis

Background:

• Autism is a complex neurodevelopmental condition with elusive neurobiological underpinnings.
• Previous resting-state EEG (rsEEG) studies in autism show inconsistent and unreproducible findings.
• Small sample sizes and limited variable reporting characterize prior rsEEG research in autism.

Purpose of the Study:

• To investigate group differences in rsEEG between autistic and neurotypical individuals using a large, combined dataset.
• To assess the replicability and effect sizes of numerous EEG variables.
• To examine the impact of sample size on effect size and replicability in autism rsEEG research.

Main Methods:

• Combined five datasets for a total of 776 autistic and neurotypical participants.
• Extracted 726 rsEEG variables per participant, controlling for age, sex, and IQ.
• Calculated effect sizes and split-half replication rates, with bootstrap analysis for sample size effects.

Main Results:

• Few EEG measures showed significant group differences between autistic and neurotypical individuals.
• Larger effect sizes for EEG differences were often not replicable under split-half testing.
• Bootstrap analysis revealed that smaller sample sizes yielded larger effect sizes but lower replication rates.

Conclusions:

• The study provides limited evidence for a distinct neurobiological signature in autism detectable via rsEEG.
• Findings underscore the heterogeneity within autism spectrum disorder.
• Results caution against using autism diagnosis alone to categorize diverse neurobiological profiles.