Abnormal EEG-based Functional Connectivity under a Face-Word Stroop Task in Depression.

Zhenghao Guo,Hailiang Long,Li Yao,Xia Wu,Hanshu Cai

DOI: https://doi.org/10.1109/bibm.2017.8217784

2017-01-01

Abstract:Identifying and evaluating functionally connected regions in the brain has become a challenging problem to solve in many studies of neurological and psychiatric disorders. In particular, functional connectivity of brain mechanisms underlying disturbed cognition in depression is still not well understood in current neuroscience research. Based on the Stroop paradigm, specifically, the face-word Stroop task, we aimed to analyze task-based electroencephalography (EEG) functional connectivity in subjects with depression and in healthy controls, using concepts from time series clustering. In this study, EEG signals of 10 healthy subjects and 10 depressive patients were collected. Then EEG signals were segmented into temporal window corresponding to the event-related potentials (ERPs). For each duration, hierarchical clustering (HC) along with optimizations for the dynamic time warping (DTW) were employed to identify meaningful functionally connected regions and examine changes in depression. It was demonstrated that our method had the potential to become a useful tool for clinical investigators to identify the underlying impairments of brain functional connections in the patients with depression. One of the most representative functional connections obtained in the present study indicated that during the N450 component, the left and right frontal brain parts may discriminate depressive patients from healthy controls. This finding was interpreted by valence-hypothesis, which can prove the validity of the theory of emotional conflict in major depression.