Abstract

The stability of resting EEG alpha has been studied on a time scale of weeks or months, often to match treatment protocols in clinical trials. However, few studies have examined EEG test-retest reliability on a time scale of years. The present study characterized the long term temporal stability of EEG alpha over a considerably greater time scale. Resting EEG was recorded from 73 participants in a study of familial risk for depression during testing sessions separated by a span of ten years. Low- and high-frequency EEG alpha factors were unequivocally identified, separated and quantified using reference-free methods that combine current source density (CSD) with principal components analysis (PCA). Measures of overall and net (eyes closed-minus-open) posterior alpha were compared across testing sessions. Overall alpha was remarkably stable (Spearman-Brown [SB] = 5.867; Pearson’s r = 5.765, Spearman’s rho q = 5.810, all p < .001) and net alpha was also reproducible (SB = 5.766; r = 5.621, q = 5.624, all p < .001). Although hemispheric asymmetries (right-minus-left hemisphere) were significantly correlated, they did not show strong reliability for overall (SB = 5.618; r = 5.448, q = 5.427, all p < .001) or net posterior alpha (SB = 5.588, r = 5.417, q = 5.388, all p < .001). Results verify the reliability of overall posterior EEG alpha at rest across prolonged time intervals, and are indicative of an individual trait.

Desciptors: EEG alpha, test-retest reliability, current source density (CSD)

[Supported by grants from NIMH (MH36197, MH106905) and John Templeton Foundation (#54679)].