Gamma entrainment modulates cognitive impaired subgroup of depression: Evidence from EEG microstate dynamics

Depression characterized by heterogeneous symptom profiles, has increasingly been recognized for its cognitive subtype. Gamma entrainment achieved through rhythmic sensory stimulation, has emerged as a promising method to restore neural synchrony. However, the effects and mechanisms of gamma entrainment to cognitive impairment in depression remain poorly understood.

We identified the cognitive impairment subgroup in depression by MATRICS Consensus Cognitive Battery. And then, we randomly classified these cognitively impaired patients into intervention and control group, gamma entrainment training was used. Microstates of neuronal oscillations were conducted through scalp electroencephalogram.

Microstate dynamics during eyes-closed conditions in the cognitive biotype of depression revealed distinct alterations across specific neural oscillatory parameters. Significant reductions were observed in Class-B global field power (GFP) and Class-D metrics, including GFP, duration, and coverage. Conversely, Class-C Occurrence frequency (Occ) exhibited increased activation in the cognitive biotype. Transition probabilities between Class B and D were also attenuated. After gamma entrainment， cognitive function improved in cognitive biotype of depression without affecting emotional symptoms. Moreover, significant temporal main effects emerged for Class-B microstate dynamics, including increase in global field power, duration, and coverage, alongside Class-D GFP.

Our findings delineate a distinct neurophysiological signature of the cognitive biotype of depression, marked by bidirectional dysregulation of large-scale neural synchrony. These results position gamma-driven neuromodulation as a potential therapeutic strategy for restoring network synchrony in cognitive-biased depression, while highlighting microstate metrics as sensitive biomarkers for biotype stratification. Future studies should validate these dynamics in longitudinal cohorts and assess their predictive value for treatment responsiveness.