Electroencephalography profiles as a biomarker of wellbeing: A twin study

Abstract

Alterations to electroencephalography (EEG) power have been reported for psychiatric conditions such as depression and anxiety, but not for mental wellbeing in a healthy population. This study examined the resting EEG profiles associated with mental wellbeing, and how genetics and environment contribute to these associations using twin modelling. Mental wellbeing was assessed using the COMPAS-W Wellbeing Scale which measures both subjective and psychological wellbeing. In 422 healthy adult monozygotic and dizygotic twins aged 18-61 years, we examined the association between mental wellbeing and EEG power (alpha, beta, theta, delta) using linear mixed models. This was followed by univariate and multivariate twin modelling to assess the heritability of wellbeing and EEG power, and whether the association was driven by shared genetics or environment. A significant association between wellbeing and an interaction of alpha, beta, and delta (ABD) power was found (β = -0.33, p < 0.001) whereby a profile of high alpha and delta and low beta was associated with higher wellbeing, independent of depression and anxiety symptoms. This finding was supported by a five-fold cross-validation analysis. A significant genetic correlation (rG = -0.43) was found to account for 94% of the association between wellbeing and the EEG power interaction. Together, this study has identified a novel EEG profile with a common genetic component that may be a potential biomarker of mental wellbeing. Future studies need to clarify the causal direction of this association.

Keywords: Alpha power; EEG; Heritability; Mental health; Twins; Well-being.

Fig. 1.

Association between the alpha, beta, delta (ABD) EEG interaction effect and wellbeing for the averaged central region in the eyes closed condition. Top panel (A) shows the interaction effect for different power levels in each EEG frequency band. High and low values for beta and delta are based on ± 1 standard deviation from the mean. Bottom panel (B) shows the strength of this association at each electrode site, whereby red indicates a larger standardised effect size and blue indicates smaller effect size using absolute values.

Fig. 2.

Correlated factors model for wellbeing, the alpha, beta, delta (ABD) EEG interaction, and depression/anxiety symptoms. Single-headed arrows indicate the influence of additive genetic (A) or unique environmental (E) influence on each variable with standardised path coefficients. Double-headed arrows indicate genetic and environmental correlations between variables, with standard errors provided in parentheses. Solid lines indicate significant correlations, whereas dashed lines indicate non-significant correlations. For point estimates and corresponding confidence intervals, please refer to Table S9.