Objective <p>To investigate the neurophysiological features underlying depression-related cognitive dysfunction via dual-task event-related potential (ERP) and task-phase-dependent electroencephalography (EEG) functional connectivity analysis.</p> Methods <p>Thirty patients with depression (DP group) and 30 matched healthy controls (HC group) underwent EEG recordings across pre-task resting state, face perception task (indexed by N170), face mismatch conflict monitoring task (indexed by N270), and post-task resting state. Functional connectivity was quantified using imaginary coherence (iCoh), with between-group comparisons corrected by Benjamini–Hochberg false discovery rate (FDR). Correlation analyses between electrophysiological metrics and clinical variables were performed in the DP group.</p> Results <p>The DP group exhibited prolonged latency and decreased amplitude of N170, prolonged latency and increased amplitude of N270, as well as task-phase-dependent multi-frequency brain network imbalances. In the DP group, longer N170 and N270 latencies were associated with more severe depressive symptoms, whereas task-phase-dependent functional connectivity abnormalities showed no significant associations with clinical variables.</p> Conclusion <p>Depression was associated with electrophysiological abnormalities in emotional processing and conflict monitoring, as well as task-phase-dependent brain network imbalances, providing exploratory neurophysiological evidence related to depression-associated cognitive dysfunction.</p> Clinical trial number <p>Not applicable.</p>