Data mining task of identifying EEG reactions to emotionally significant videos

This empirical and statistical study is in line with the methodology of event-related potentials (ERPs) based on registration of bioelectrical brain responses to external stimulation or cognitive tasks (in the case of cognitive ERPs). The stimuli were videos with diverse emotionally rich content. Values of EEG periodogram under analysis were divided into three classes: «Eyes closed», «Relax» and «Emotions». The article shows that dynamic EEG range contain the information reflecting the effect of emotional impact the tested persons experience watching a number specially selected video clips. Data mining techniques allow locating such patterns in the dynamic EEG range that are much more frequent in case of visually emotional impact than the ones in case of emotionally neutral video clips. In order to determine which responses of test persons experience to videos with emotionally rich content it was enough to analyze the signal’s EEG with only one FP2 derivation. Apparently, additional analysis of signals’ EEG from other derivations will contribute to more rapid and accurate identification of emotional response...