Resting-state EEG power differences in autism spectrum disorder: a systematic review and meta-analysis.

Researchers analyzed brain wave patterns in autistic people using EEG technology. They found that autistic individuals had different patterns in two specific brain wave types: reduced alpha waves (associated with relaxed awareness) and increased gamma waves (linked to information processing). These differences were consistent across 41 studies involving nearly 3,000 participants. The findings suggest these brain wave patterns might help identify autism, though more research is needed.

This systematic review and meta-analysis examined resting-state EEG power differences in autism spectrum disorder across 41 studies involving 1,246 autistic and 1,455 neurotypical individuals. The analysis revealed that autistic individuals showed reduced relative alpha power (g = -0.35) and increased gamma power (absolute: g = 0.37, relative: g = 1.06) compared to neurotypical controls. No significant differences were found in delta, theta, absolute alpha, or beta power. Substantial heterogeneity was observed across studies, with resting-state paradigm type (eyes-closed vs eyes-open) and recording duration identified as moderating factors for some frequency bands.

The findings suggest alpha and gamma power may serve as potential biomarkers for autism.

Resting-state EEG alpha and gamma power show promise as potential autism biomarkers. However, substantial between-study heterogeneity and limited moderating factors suggest clinical translation requires standardized protocols and further validation studies before implementation in diagnostic practice.

Substantial heterogeneity between studies (36-84%) limits interpretation. Moderator analyses found limited factors explaining variability. Study quality and methodological differences may impact results. Clinical utility as biomarkers requires further validation in diagnostic settings.

Narrative reviews have described various resting-state EEG power differences in autism across all five canonical frequency bands, with increased power for low and high frequencies and reduced power for middle frequencies. However, these differences have yet to be quantified using effect sizes and probed robustly for consistency, which are critical next steps for clinical translation. Following PRISMA guidelines, we conducted a systematic review of published and gray literature on resting-state EEG power in autism. We performed 10 meta-analyses to synthesize and quantify differences in absolute and relative resting-state delta, theta, alpha, beta, and gamma EEG power in autism.

We also conducted moderator analyses to determine whether demographic characteristics, methodological details, and risk-of-bias indicators might account for heterogeneous study effect sizes. Our literature search and study selection processes yielded 41 studies involving 1,246 autistic and 1,455 neurotypical individuals. Meta-analytic models of 135 effect sizes demonstrated that autistic individuals exhibited reduced relative alpha (g = -0.35) and increased gamma (absolute: g = 0.37, relative: g = 1.06) power, but similar delta (absolute: g = 0.06, relative: g = 0.10), theta (absolute: g = -0.03, relative: g = -0.15), absolute alpha (g = -0.17), and beta (absolute: g = 0.01, relative: g = 0.08) power. Substantial heterogeneity in effect sizes was observed across all absolute (I: 36.1-81.9%) and relative (I: 64.6-84.4%) frequency bands.

Moderator analyses revealed that age, biological sex, IQ, referencing scheme, epoch duration, and use of gold-standard autism diagnostic instruments did not moderate study effect sizes. In contrast, resting-state paradigm type (eyes-closed versus eyes-open) moderated absolute beta, relative delta, and relative alpha power effect sizes, and resting-state recording duration moderated relative alpha power effect sizes. These findings support further investigation of resting-state alpha and gamma power as potential biomarkers for autism.