Anmerkungen:
Diese Datenquelle enthält auch Bestandsnachweise, die nicht zu einem Volltext führen.
Beschreibung:
The objective of the present dissertation is to investigate a long-lasting question of whether there is psychophysiological response patterns specificity to basic emotions, with innovative data analysis method. For the purpose of eliciting the five basic emotions (amusement, sadness, anger, fear and disgust), 15 standardized film clips in German were carefully selected. Moreover, three psychophysiological signals, trapezius Electromyography (tEMG), Electrodermal activity (EDA) and Electrocardiogram (ECG) were recorded during the period of emotion elicitation. The Support Vector Machines (SVMs) were utilized as the classification algorithm and Sequential Forward Selection was used as the feature selection method. Two kinds of validation method, 10-fold cross-validation and leave-one-subject-out cross-validation were utilized for obtaining the subject-dependent and the subject-independent models, respectively. The subject-independent models perform better than the subject-dependent models on totally unseen participants, which indicates individual responses patterns. The averages of classification rates of both models are above the chance level, which indicates psychophysiological response patterns specificity to basic emotions. SVM is an efficient classifier for the psychophysiological-based film induced emotion classification. 16 features are essential to the present question and tEMG contributes a lot among them.