| Emotion is very important to human beings,and is a critical way to convey information and express oneself.Since the last century,the study of emotion had been a hot field.The mechanism of emotion generation is very complex,and it can be divided into spontaneous and induced types according to the source.The spontaneous type is derived from internal performance of the individuals,while the induced type is described as emotional expression or feedback caused by the external environment.According to the differences in properties of the external environment,the induced emotion can be divided to two categories: covered Induced Emotion(cIE)and uncovered Induced Emotion(ucIE).This paper focuses on the study of induced emotion,and the research contents include emotion recognition of cIE and face recognition of ucIE.Electroencephalogram(EEG)is used to record the signal reflecting the state of the brain,and then by analyzing the EEG signal to study the induced emotion.The main contents of the study are divided into the following two aspects:Firstly,in the study of cIE,the subjects were asked to watch the video to induce pleasure,relaxation,sadness and disgust.After pre-processing of the original signal,Welch and Yule-Walker spectral estimation methods were adopted to extract the average value of the alpha and beta bands as the features followed by the discriminant analysis(DA)method and support vector machine(SVM)classifier for emotion classification.Nine subjects participated in the experiment,and highest classification accuracy reached 85.92%,which provides a complete and feasible process for the use of EEG for emotion recognition.Secondly,in the study of ucIE,12 healthy subjects were selected to recognize happy,neutral and sad face,and it could be inferred that the rationality of the experimental design from the overall average Event Related Potentials(ERPs)diagram containing obvious early components related to face recognition.Next,the same experiment and data analysis method were performed on 12 age-matched patients with social anxiety disorder(SAD)before and after treatment.The statisticaltest method was used to compare the amplitude and latency of N170 under three conditions(Pre-patients group,Post-patients group and Healthy group).Analysis of N170 amplitude revealed that a larger N170 for Pre-patients group in response to inverted and upright stimuli as compared to HP group in the right hemisphere was found.The result of intragroup t-test showed that N170 was delayed for inverted relative to upright faces only in Post-patients and HP groups.Particularly,the results of three-factor ANOVAs manifested that emotional expression cannot modulate N170 for SAD patients.Besides,the N170-based asymmetry index(AI)was introduced to analyze the left-and-right hemisphere dominance of N170 for three groups,and it was found that the change in value ofbasedNAI170-under inverted or upright condition can be used as an evaluation index about disease condition. |