Table 2.

EEG biomarkers to predict treatment response.

Alpha asymmetry	Based on the approach-withdraw model (43), this measures relative alpha band activity between brain hemispheres (mainly in frontal regions; higher alpha may reflect lower brain activity). Alpha asymmetry has been proposed as a suitable prognostic biomarker related to anxious subtype and bipolar features (44).
Microstate abnormalities	Using polarity-insensitive k-mean clustering, we will segment resting-state high-density EEG data into microstates (45). The proportion, duration, occurrence, and transition of microstates will be studied as potential biomarkers of state and trait abnormalities and as predictors of treatment outcome.
Rostral anterior cingulate cortex theta activity	This is a robust marker that predicts greater improvement in selective serotonin reuptake inhibitor-induced depressive symptom (46).
Subgenual/prefrontal connectivity	Based on recent findings that suggest that changes in rTMS-induced within-network connectivity are a mediator of treatment response (47), eLORETA linear-lagged connectivity measures of theta (4-7.5 Hz) and alpha (8-13 Hz) frequency will be obtained between the following regions of interest: right and left DLPFC, dorsomedial prefrontal cortex, and subgenual cingulate cortex (as in Iseger et al, 2017).
Gamma-band power envelope connectivity	Orthogonalized power envelope correlation will measure EEG source connectivity (48). Large-scale connectivity patterns have been proposed as predictors of placebo/antidepressant outcomes.