Development Of Emotional State Assessment And Feedback Regulation System Based On Eeg Network

In recent years,with the continuous breakthrough of computer technology and cognitive neuroscience related technologies,Brain-Computer Interface(BCI)technology has developed rapidly.At present,researchers have gradually shifted their focus from the traditional brain-controlled BCI system to the new cognitive regulation BCI system.Existing studies consistently show that developing affective Brain-Computer Interface technology is an important way to achieve emotional intelligence.In essence,emotion is an important cognitive function of the brain,and the processing of emotion by the brain is a dynamic interaction process involving multiple brain regions.Therefore,how to realize emotion recognition by extracting the spatial topological features of EEG network under different emotional states is of great significance for the study of affective computing.In addition,how to use neurofeedback technology to achieve accurate assessment and feedback regulation of individual emotional states is also of great significance for the improvement of individual emotional regulation ability.In response to the above research problems,thesis develops a set of emotional state assessment and feedback regulation system based on EEG network.This system aims to evaluate individual emotional state in real time through BCI technology,and achieve feedback and regulation of individual emotional state through a specific way.In the system developed in thesis,relevant research contents mainly include:1.Real-time decoding of individual emotional states based on EEG network.The system firstly constructs a corresponding EEG network by collecting EEG generated by individuals viewing video stimuli.Then,machine learning method is used to mine the spatial topological features of EEG networks under different emotional states,and Linear Discriminant Analysis(LDA),which can effectively decode emotional states,is trained to construct an individual emotional decoding model.Further,based on the learned emotional decoding model,the system can decode individuals’ emotional states and display the decoding results on a Graphical User Interface(GUI)in real time.At the same time,the system also designed an independent network display interface to display the constructed emotional EEG network in real time,showing the information interaction mode between brain regions under different emotional states.In order to verify the emotion decoding performance of the system,a total of 21 subjects were publicly recruited for online emotion recognition experiment.The experimental results showed that the system could decode individual emotional states online with the decoding accuracy of 80.68%.2.Individual negative emotion regulation based on neurofeedback technology.In order to realize the feedback and regulation of individual emotional states,thesis further develops the emotional feedback regulation system on the basis of the development of individual emotional state decoding system,so as to form a closed-loop BCI system of emotion decoding and regulation.Specifically,the system first assesses individual emotional states,uses simple text descriptions with cartoon expressions that can clearly represent emotional states as feedback information,and feeds the results of the system evaluation to the subjects in real time.In the process of emotional feedback regulation,the system realized phased comprehensive assessment of individual emotional state based on the comprehensive evaluation of 30 s,and implemented the set feedback regulation strategy according to the results of comprehensive evaluation.When the system evaluates the individual’s emotional state to be negative,the system will impose specific emotional feedback regulation on the individual to achieve the adjustment of the individual’s negative emotions to positive emotions.In order to verify the effectiveness of the system’s emotional feedback regulation,a total of 10 subjects were publicly recruited for the emotional feedback regulation experiment,and the performance of the system’s emotional feedback regulation was evaluated according to the self-evaluation of the emotional states of the subjects before and after feedback regulation.The experimental results show that the emotional titer self-score of the subjects can be significantly improved after feedback regulation,which further proves the effectiveness of the system in the task of individual emotion regulation.