Exploring EEG resting state differences in autism: sparse findings from a large cohort.
Adam J O Dede, Wenyi Xiao, Nemanja Vaci and 2 others
PMID 39994801WHAT IT FOUND
Analysis of resting-state EEG in 776 participants found little evidence for reliable differences between autistic and neurotypical groups.
While some effects appeared, most failed replication testing. This suggests autism diagnosis alone is too heterogeneous to identify clear EEG biomarkers.
Key findings
01Only 11 of 174 diagnosis-related EEG variables with moderate effect sizes replicated with sufficient consistency.
02Diagnosis predicted none of the EEG variables with an effect size greater than 0.035 when all data were analysed collectively without age grouping.
03Small sample sizes produced more high effect sizes for diagnosis but low replicability, whereas larger samples reduced diagnosis-related findings towards zero.
STILL TO COME
How it was doneWhat they found
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The study focused on univariate resting-state EEG variables; differences might exist in task-based or multivariate analyses. Combining data from multiple labs and datasets may have obscured subtle group differences, although age-related patterns were successfully replicated. The autism group was highly heterogeneous, potentially masking specific subgroups with distinct neural profiles. Sample sizes for some subgroups (e.g., ASD in certain datasets) were small, limiting power for specific interaction effects.
Declared interests
The authors declare no competing interests. Funding details are not explicitly stated in the provided text, though data was obtained from the NIMH data archive.
The easy way to misread this
Do not interpret the 11 replicable variables as evidence of a general EEG biomarker for autism. These findings were exploratory, not pre-registered, and the overall conclusion is that diagnosis alone does not predict EEG patterns reliably in large samples.