Electroencephalography source-space functional connectivity reveals frequency-specific brain network dysfunctions in...

Obsessive-compulsive disorder (OCD) is characterized by disruptions in large-scale brain networks. However, the role of high-frequency neural synchrony in these abnormalities remains unclear. Elucidating frequency-specific alterations may offer critical insights into the neurophysiological mechanisms underlying brain network dysfunction in OCD.

We analyzed resting-state electroencephalography data from 88 OCD patients and 100 healthy controls (HCs), and constructed source-space functional connectivity networks using weighted phase lag index. Graph-theoretical metrics including global and local efficiency were computed to quantify network topology. Machine learning with support vector machines was applied to classify OCD patients and HCs using multi-frequency network features.

Compared to HCs, OCD patients exhibited increased beta-band connectivity between the default mode network (DMN) and frontoparietal network (FPN). Beta-band connectivity within the FPN was correlated with compulsive symptoms. Additionally, reduced global efficiency and local efficiency in the right visual cortex was observed in the beta band. Low-frequency (dela and theta) network topology metrics also correlated with anxiety severity. Classification analysis integrating multi-frequency features achieved an accuracy of 87.9 %.

The majority of patients in our sample were receiving pharmacological treatment, and the effects of medication could not be fully controlled.

These findings reveal frequency-specific disruptions in brain networks underlying OCD symptoms and demonstrate the potential of electrophysiological connectome biomarkers for diagnosis. These results support frequency-targeted neuromodulation strategies for personalized OCD treatment.