The Analysis of Individuals Emotions Through Brain Signals Using Poincare Approach

Dynamic alterations of the brain are of high significance when it comes to analyze the human feelings. In this study, the hidden patterns corresponding for the emotional states have been investigated by adopting a certain Poincare’ map function inspired by the theory of chaos. The present study aimed to explore the significance relationship between the proposed methodology and the level of the emotional states. Methods and Materials: One of the well-known emotion dataset namely ‘DEAP’ was used to evaluate the performance of our methodology. DEAP is a multimodal dataset in which forty music videos lasting one minute were shown to 32 participants. Moreover, a set of 40 channels of biological signals including EEG and peripheral signals were recorded. Moreover, participants rated to each video in terms of ‘valence’, ‘arousal’, ‘dominance’, and ‘liking’. In this study we have investigated the dynamic variations of EEG signals in order to discriminate the levels of the emotional states. For describing the effect of the Poincare’ approach on the subjective ratings, we have adopted the correlation statistics by computing the Spearman correlated coefficients. The results indicate that the Poincare approach could properly discriminate the levels of valence (p<0.05), arousal (p<0.01), dominance (p<0.05), and liking (p<0.05). In other words, the statistical analysis showed the significant relationship between the levels of emotions and the Poincare’ method. Discussion and The result interpretations demonstrate and verify the success of the nonlinear and chaotic approaches in the analysis of human brain signals. Therefore, it could be a reliable methodology for studying the brain functions.