| Emotion has been recognized as an important impact on human decisions and behavior for a long time. When our emotions seem to be ill-matched to a given situation, we frequently try to regulate our emotional responses. It is important to improve the capability of emotion regulation by intervention and training, since emotion dysregulation is a predisposing risk factor for many psychiatric disorders,such as depression, anxiety. However, less is known about the temporal dynamics and the physiological mechanisms underlying the regulation process. Current intervention mainly based on the cognitive behavioral therapy(CBT), which refers to behavior therapy, cognitive therapy, and therapy based upon a combination of basic behavioral and cognitive principles. Studying the time variation characteristics of emotion regulation using scalp electroencephalogram(EEG) with high temporal resolution and finally applying these physiological indexes to the intervention of emotional regulation have high academic value and potential practical significance.We adopted an emotion related electrophysiological experiment with young healthy people as participants and investigated the neural activity and the characteristics of brain network in the process of emotion regulation. According to these results, we constructed an EEG-based neuro-feedback system for emotion regulation. In this study, each subject participated in two cognitive tasks, watching and regulation. In the ‘watching’ task, participants were asked to just watch the stimulus attentively. In the ‘regulation’ task, the participants were asked to engage in cognitive reappraisal and decrease the negative interpretation of the scenes. In this paper, we explored the brain activity in different time range during these two tasks using event-related potentials, spectrum analysis,event-related spectral perturbation, phase synchronization and complex network analysis.The findings can be summarized as follows:(1) Cognitive reappraisal of unpleasant stimuli modulated the late positive potential(LPP) later at 1500 ms and enhanced LPP amplitudes in left fronto-central region. In 1500-2500 ms, the regulation of sadness increased the LPP in left and right frontal-central regions. In 2500-4000 ms, the regulation of sadness and fear both enhanced the LPP in left frontal-central. This finding pointed to an important role of the left fronto-central region in down-regulation of negative emotion. In 500-1000 ms following stimuli onset, we only found fearful stimuli evoked larger LPP than sad ones in right frontal-central and parietal regions, while the reappraisal of negative stimuli had no significant effect on LPP amplitudes.(2) Emotional regulation decreased the slow-wave activity and increased the fast-wave activity. Compared with watching, the reappraisal of negative emotion reduced the power and event-related spectral perturbation in theta band. This decrease was significant in early stage(before 1000 ms). At the same time, the power of fast wave(beta band and gamma band) was increased after regulation. Different from theta and gamma band, the effect of reappraisal on the oscillation of beta band began at later stages(after 1000 ms). These oscillation patterns showed the separate of attention and enhanced cognitive controls during emotion regulation.(3) Emotion regulation reduced the fronto-parietal connectivity in left hemisphere in gamma band and increased the small-world property of brain network in alpha2 band(10-13Hz). The brain network showed the classic ‘small-world’ organization in watching and regulation task. And the small-world property was modulated by emotion and condition in different stages of emotion regulation. In theta,alpha2 and gamma networks, the small-worldness was larger for fearful stimuli in 500-1000 ms time epoch. While in 1000-4000 ms time epoch, the emotion difference was changed in regulation task. At the same time, in alpha2 band networks, the small-worldness was larger and the path length was smaller during regulation task. These results showed that emotion regulation had an effect on topological organization of brain network and need larger local connectivity.(4) A set of EEG biofeedback system was developed, which has functions of EEG signal acquisition, threshold setting, feature extraction and the presentation of parameters, and so on. Emotion regulation task was incorporated into this feedback system and this system can display the result of emotion regulation by adding noise to emotional stimuli. The result of system testing proves that the system could satisfy the requirements of biofeedback training.In conclusion, by means of EEG analysis, we explored the temporal dynamics of emotion regulation and characteristic of brain network during this process and constructed an EEG-based neuro-feedback system. All the results show that emotion regulation is a dynamic process. And different brain areas are involved in the regulation,which have their own roles in different stages of emotion regulation. Furthermore, emotion regulation also has an effect on oscillatory patterns in frequency bands and the brain network in emotion regulation had higher small-world property and stronger local connection. Overall, this work enriches the current understanding of the temporal dynamics of emotion regulation, which also provides a theoretical basis for the development of EEG-based neuro-feedback system. |
