Eye-state-dependent reorganization of EEG microstates in a pediatric transdiagnostic cohort

Abstract Resting electroencephalography (EEG) is sensitive to eye condition, yet the reorganization of pediatric EEG microstates between eyes-closed (EC) and eyes-open (EO) rest remains incompletely characterized. Paired EC and EO recordings from 110 children in the Healthy Brain Network were analyzed using polarity-invariant two-level clustering, backfitting, and within-participant tests of microstate duration, coverage, occurrence, and transition measures. Eye opening shortened A and C segments, increased the occurrence of all four K = 4 classes, and redistributed coverage from C toward B. All eight significant K = 4 temporal effects retained their direction after participant-wise duration matching and template re-estimation, and broad state sensitivity was also observed at K = 6. Change-score and baseline-adjusted analyses yielded comparable EC-to-EO response profiles among the comparison group without the target diagnoses, ADHD-only group, and ADHD+Anxiety group. The association between posterior alpha reactivity and switching varied in magnitude and direction across segmentation specifications. A stable fourth K = 4 map showed little correspondence with adult microstate D, whereas a D-like map emerged at K = 6. Pediatric microstate measures reflect both physiological context and analytical definition; eye condition, model order, temporal smoothing, and analyzed duration should therefore be specified explicitly in developmental and clinical studies.