Authors:
G. Portnova
1
;
D. Stebakova
2
and
G. Ivanitsky
2
Affiliations:
1
Institute of Higher Nervous Activity and Neurophysiology of RAS, 5A Butlerova St., Moscow 117485, Russia, Pushkin State Russian Language Institute and Russia
;
2
Institute of Higher Nervous Activity and Neurophysiology of RAS, 5A Butlerova St., Moscow 117485 and Russia
Keyword(s):
EEG, Classification, Acoustic, Visual, Tactile and Olfactory Modalities, Pleasant and Unpleasant Stimuli.
Related
Ontology
Subjects/Areas/Topics:
Affective Computing
;
Biomedical Engineering
;
Biomedical Instruments and Devices
;
Brain-Computer Interfaces
;
Devices
;
Health Information Systems
;
Human-Computer Interaction
;
Physiological Computing Systems
Abstract:
We perceive pleasant and unpleasant stimuli using different modality systems, such as visual and acoustic tactile and olfactory modalities. In our study we investigated the specificity of emotional perception in four modalities using EEG. 20 healthy participants were instructed to assess the stimuli using emotional scales. We used power spectrum density, alpha-peak frequency, wavelet analysis and method of "emotional spaces" for EEG data and DNN classifier for modality specific and non-specific classification of pleasant and unpleasant stimuli. We found, that difference of EEG power spectrum density and alpha-peak frequency between states of pleasant and unpleasant stimulation varied from one modality to another. Meanwhile, the above-stated differences were more similar between tactile and olfactory modalities and acoustic and visual modalities. the method of "emotional spaces" and DNN classification showed general, modality nonspecific features of pleasantness evaluation.