Open AccessANALYSIS OF EEG RESPONSE FOR AUDIO-VISUAL STIMULI IN FRONTAL ELECTRODES AT THETA FREQUENCY BAND USING THE TOPOLOGICAL FEATURES
Author(s) -
Himanshu Kumar,
Subha D. Puthankattil,
Ramakrishnan Swaminathan
Publication year - 2021
Publication title -
biomedical sciences instrumentation
Language(s) - English
Resource type - Journals
ISSN - 1938-1158
DOI - 10.34107/yhpn9422.04333
Subject(s) - electroencephalography , stimulus (psychology) , frontal lobe , arousal , speech recognition , psychology , pattern recognition (psychology) , artificial intelligence , computer science , neuroscience , cognitive psychology
Emotions are the fundamental intellectual capacity of humans characterized by perception, attention, and behavior. Emotions are characterized by psychophysiological expressions. Studies have been performed by analyzing Electroencephalogram (EEG) responses from various lobes of the brain under all frequency bands. In this work, the EEG response of the theta band in the frontal lobe is analyzed extracting topological features during audio-visual stimulation. This study is carried out using the EEG signals from the public domain database. In this method, the signals are projected in higher dimensional space to find out the geometrical properties. Features, namely the center of gravity and perimeter of the boundary space, are used to quantify the changes in the geometrical properties of the signal, and the features are subject to the Wilcoxon rank-sum test for statistical significance. Different electrodes in the frontal region under the same audio-visual stimulus showed similar variations in the geometry of the boundary in higher-dimensional space. Further, the electrodes, Fp1 and F3, showed a statistical significance of p < 0.05 in differentiating arousal states, and the Fp1 electrode showed a statistical significance in differentiating valence emotional state. Thus, the topological features extracted from the frontal electrodes in theta band could differentiate arousal and valence emotional states and be of significant clinical relevance.