Some limitations must be kept in mind when interpreting our findings. Several key brain regions implicated in MDD pathophysiology are subcortical, and since eLORETA can only reliably estimate activity in cortical regions, we could not examine connectivity in these regions. In addition, although we observed abnormal rsFC in the beta-band in MDD, neural networks likely involve coordinated communication across frequencies [66] and an obvious extension of our work is to examine measures of lagged cross-frequency coupling. Finally, although we used several pre-requisite parameters for conducting functional connectivity on EEG source estimates, such as using a high-density EEG montage and a realistic head model [67, 68], due to the limitations and inherently low spatial resolution of eLORETA, we cannot rule out that synchrony between ROIs may be related to activity from regions adjacent to the ROIs. As such, our findings await replication using methods that have superior spatial resolution.
