Evaluating the Determinism of Brain Signals Using Recurrence Chaotic Features in Positive, Negative and Neutral Emotional States in the Sources Achieved From ICA Algorithm

This study investigates electroencephalogram (EEG) signals in positive, negative and neutral emotion states.
It is assumed that the brain draws on several independent sources in any activity that are observable by independent component algorithm (ICA). To overcome the problem of ill-posedness of extracted components from ICA algorithm, first these sources are sorted out by Shannon entropy and then based on these sources, the features of trapping time and determinism of Recurrence Quantification Analysis (RQA) are extracted as representative of determination.
The results show that the degree of determinism of sorted sources related by emotions is significantly different over time and in three positive, negative and neutral states. The degree of determinism increases in neutral, positive and negative emotional states respectively.