Analysis Of Physiological Characteristics Of Real-scene Psychological Stress

Stress is a kind of psychological nervousness and physical tense caused by selfcognition and interpretation to specific stressful stimulus or situation.Due to the dual pressures to complete the academic studies and to pursue a work position after graduation,college students have become a stress susceptible population.The early research on college students' stress focused on the questionnaire census.In recent years,researchers began to analyze the physiological signals of college students in the real stressful environment,e.g.academic examinations,graduation,economy,work,and so on.On the basis of previous studies,this study focuses on the stress physiological responses of college students in the context of thesis defense and academic examinations.The video data of the graduate students' thesis defense and the ECG data of the students in the semester final examination were collected and analyzed.The research work is divided into the following two aspects:(1)The facial expression data of the graduate students in the real scene thesis defense were collected and analyzed.The experiment obtained the facial expression data of 18 postgraduates in their thesis defense for the Master's degree.Then,valid data were selected by using self-evaluation of the subject and the average evaluation of 5 audience.Next,the paper used Active Shape Model(ASM)algorithm for feature extraction of 68 feature points in 5 facial zones.Moreover,the paper applied support vector machine algorithm to recognize high stress state according to the displacements of the feature points,and backward selection(BS)algorithm was used to find out the most important features in recognition.The result illustrated the validity of the method and obtained a best correct rate of 99% in the recognition of high stress state.Meanwhile,the maximum displacement zone in social anxiety state is jaw,and followed by mouth,eyebrow,eyes and nose.Obvious individual differences were found in the formation of high stress facial expression.(2)ECG data in the real-scene examination and the lab-simulated examination were collected and analyzed.The experiment collected 27 undergraduate students' ECG data,and RR interval time series were obtained after the preprocessing of these ECG data.38 linear and nonlinear features were extracted from the RR interval time series each of which was cut into the same length.Backward selection algorithm was applied for feature selection,and the optimal feature subset was obtained by this means.The self-reported stress scores and RR interval time series were analyzed to reveal the stress states in different experiment stages.Through the analysis of the data in the real-scene and lab-simulated examination,we found that the lab-simulated examination did induce the stress of the subjects,but the real-scene examination had better stress elicitation effect.Through the analysis of ECG data right before the examination,we found that stress experience and physiological response were prior to the stress elicitation events.Through the analysis of ECG data right after the examination,we found that stress experience and physiological response did not stop at the end of the examination,but the stress physiological response had a decrease trend in 6-minute time after the examination.RR interval data before,during and after the real-scene examination could not be well distinguished from each other,showing that the stress physiological response in these three stages kept in indistinguishable levels.