Large-scale brain network alterations in young individuals with comorbid social anxiety and depression: Evidence from resting-state EEG spectral and microstate analyses

Social anxiety and depression have each been linked to alterations in large-scale brain network connectivity, yet the neural mechanisms underlying their comorbidity remain poorly understood. To address this gap, we analyzed resting-state EEG data from 420 young adults classified as healthy controls (HC), individuals with social anxiety without depression (SA-noDp), and those with comorbid social anxiety and depression (SA-Dp). Social anxiety and depressive symptoms were assessed using the Liebowitz Social Anxiety Scale (LSAS) and the Patient Health Questionnaire-9 (PHQ-9), respectively. We applied both spectral and microstate analyses to characterize large-scale brain network dynamics during eyes-open resting-state. Spectral analysis revealed reduced delta amplitude in the SA-Dp group relative to HC (p < .05), suggesting that decreased delta may be a neural marker of SA-Dp and diminished approach motivation. Microstate analysis showed reduced microstate F in SA-noDp relative to HC, while SA-Dp exhibited higher occurrence than SA-noDp despite lower delta activity, indicating a shift from anxiety-related vigilance to depression-related maladaptive self-focus linked to DMN activity. Additionally, SA-noDp displayed reduced transitions between microstates C to F or G relative to HC (all ps < 0.05). These findings reveal differential disruptions in resting-state brain dynamics in social anxiety with and without comorbid depression, highlighting diminished approach and elevated avoidance motivation in comorbid individuals versus predominantly elevated avoidance in non-comorbid social anxiety. Together, spectral and microstate measures may serve as neurobiological markers for comorbid social anxiety and depression, supporting early identification and tailored interventions, and emphasizing the importance of approach-avoidance motivational balance in these populations.