In recent years, traditional methods such as power spectrum and amplitude analysis have been used to research the emotional electroencephalogram (EEG). The brain network method is also used in emotional EEG research, which can better reflect the activity of brains. A minimum spanning tree (MST) represents the key information flow in the weighted brain network, and it provides a sensitive method to capture subtle information in network organization while effectively avoiding the shortcomings of traditional brain networks. The DEAP dataset provides electroencephalogram (EEG) data for four categories of emotions: high arousal and high valence (HAHV), high arousal and low valence (HALV), low arousal and high valence (LAHV), and low arousal and low valence (LALV). Phase lag index (PLI) weighted matrices were calculated in five frequency bands. On this basis, the minimum spanning trees were constructed. At the same valence level in the gamma (γ) band, HAHV and HALV showed significant higher mean PLI (MPLI), maximum degree (Degree_max) and leaf fraction and significant lower diameter and eccentricity than LAHV and LALV. At the same arousal level in the γ band, HALV showed significant higher MPLI, Degree_max and leaf fraction and significant lower diameter and eccentricity than HAHV. These results indicate that the low-arousal showed more line-shaped configurations than the high-arousal. Additionally, in the high-arousal condition, a shift toward more star-shaped trees from high-valence to low-valence supports the trend toward randomness of the brain network with negative emotions and that the brain is more activated when faced with negative emotions. From a brain network perspective, this phenomenon provides a theoretical basis for negative bias.

Despite many studies reporting hemispheric asymmetry in the representation and processing of emotions, the essence of the asymmetry remains controversial. Brain network analysis based on electroencephalography (EEG) is a useful biological method to study brain function. Here, EEG data were recorded while participants watched different emotional videos. According to the videos’ emotional categories, the data were divided into four categories: high arousal high valence (HAHV), low arousal high valence (LAHV), low arousal low valence (LALV) and high arousal low valence (HALV). The phase lag index as a connectivity index was calculated in theta (4–7 Hz), alpha (8–13 Hz), beta (14–30 Hz) and gamma (31–45 Hz) bands. Hemispheric networks were constructed for each trial, and graph theory was applied to quantify the hemispheric networks’ topological properties. Statistical analyses showed significant topological differences in the gamma band. The left hemispheric network showed significantly higher clustering coefficient (C_p), global efficiency (E_g) and local efficiency (E_loc) and lower characteristic path length (L_p) under HAHV emotion. The right hemispheric network showed significantly higher C_p and E_loc and lower L_p under HALV emotion. The results showed that the left hemisphere was dominant for HAHV emotion, while the right hemisphere was dominant for HALV emotion. The research revealed the relationship between emotion and hemispheric asymmetry from the perspective of brain networks.