Cognitive Analysis of EEG Signals Induced by Visual Stimulation of Facial Emotion

This paper focuses on EEG signals induced by visual stimulation of facial emotion. We mainly research depressed and healthy subjects' cognition to different emotional visual stimulation on the dataset MODMA [1]. We also explore the problem of depression classification based on EEG signals. To solve these problems, we first preprocess EEG signals, and then extracted the LogFbank feature. We propose two methods of Channel Selection and Decision Fusion to analyze EEG signals. The channel selection algorithm considers the correlation between channels and selects the local optimal channel combination. The decision fusion strategy conducts fusion training on the model and improves the generalization of the model. Experimental results demonstrate that our proposed two methods are valid. We find that depressed patients are slower in emotional cognition than healthy controls, and EEG can be used to identify depression.