A comparison of nomothetic and individualized alpha frequency approaches to measuring frontal EEG alpha asymmetry

Abstract

Frontal alpha asymmetry (FAA) is considered to be a reliable marker of affective processing and psychopathology. Traditionally, the magnitude of alpha is calculated by taking the average over a nomothetic fixed frequency window (e.g., 8 to 13 Hz). Alternatively, methods have been proposed to extract individualized alpha frequency (IAF) peaks and windows in hopes of improving the reliability and validity of signal detection. However, no study has compared the nomothetic to IAF approaches to examine the reliability and validity of resting FAA in a large well‐characterized data set. In this study, we assessed the psychometric performance of the standard fixed window approach, a PZ‐alpha based IAF approach and a global‐alpha based IAF windows detection approach on a previously collected EEG data set (8 recordings per subject collected on four occasions across two weeks). Our results revealed that resting FAA calculated with these three different methods are highly correlated at all frontal regions (mean r = .98). The stability across the 8 recordings over the two weeks also showed no substantial difference between approaches as indicated by intraclass correlations. Moreover, internal‐consistency reliability, validity with respect to measures of emotion and emotion‐related psychopathology and state–trait Structure equation model (SEM) fitting were evaluated and yielded no significant differences across methods. Our results supported the overall reliability and validity of two different IAF approaches to assessing resting FAA but fail to find any incremental advantage over nomothetic approaches to defining alpha bands. Guidelines for methods selection for future research are provided.

Resting-state frontal, frontlateral, and parietal alpha asymmetry:A pilot study examining relations with depressive disorder type and severity

“…However, comparing these subdivisions in homologous frontal locations does not seem to come with a greater sensitivity than that of the asymmetry calculated on the conventional alpha band (Segrave et al, 2011). In addition, the measurements by individualized alpha frequency peak windows appear to be closely correlated with those of fixed frequency band, without bringing any additional reliability or validity to the calculation of fAA (Zhang and Allen, 2023), these results methodologically support any of the two approaches to alpha frequency, in fAA measurement.…”