Answers are grounded in this paper and cite their sources.
Objective To test whether bipolar disorder (BD) is associated with predefined frontal/frontopolar low-frequency scalp electroencephalography (EEG) coherence differences relative to healthy controls (HC), depressive disorder (DD), and schizophrenia (SZ) in a covariate-controlled reanalysis of a public resting-state EEG feature dataset. Methods The covariate-complete sample included 472 participants (93 HC, 66 BD, 196 DD, and 117 SZ). Analysis of covariance (ANCOVA) models adjusted for age, sex, education, and intelligence quotient (IQ). Primary outcomes were predefined frontal/frontopolar delta, theta, and combined delta-theta coherence means. Electrode-pair, regional, all-frequency, distributional, global-reference, frontopolar, and age sensitivity analyses assessed robustness. Results BD had higher adjusted coherence than all comparison groups across the three primary measures (partial η 2 = .037–.040); Holm-adjusted BD–HC, BD–DD, and BD–SZ contrasts were significant for each measure. Features surviving false discovery rate correction were concentrated in the delta and theta bands. Findings were robust to distributional, outlier, and global-coherence analyses, but effects were strongest in frontopolar pairs and attenuated as anterior electrodes were excluded. Conclusions BD showed higher covariate-adjusted frontal/frontopolar low-frequency scalp coherence than HC within this dataset. Differences from DD and SZ were smaller and concentrated in frontopolar pairs susceptible to ocular artifacts. Because only scalp-level coherence features were available and participant-level artifact-rejection data were unavailable, the findings do not establish a diagnostic biomarker, diagnostic specificity, or source-level mechanism. Artifact-controlled replication is required.
No takes yet. Share an insight, caveat, or question.
Gyutae Jang (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: