Real-Time Multi-Modal Estimation of Dynamically Evoked Emotions Using EEG, Heart Rate and Galvanic Skin Response

Emotion estimation systems based on brain and physiological signals such as electro encephalography (EEG), blood-volume pressure (BVP), and galvanic skin response (GSR) are gaining special attention in recent years due to the possibilities they offer. The field of human–robot interactions (HRIs) cou...

Full description

Saved in:
Bibliographic Details
Published in:International journal of neural systems 2020-04, Vol.30 (4), p.2050013
Main Authors: Val-Calvo, Mikel, Álvarez-Sánchez, José Ramón, Ferrández-Vicente, Jose Manuel, Díaz-Morcillo, Alejandro, Fernández-Jover, Eduardo
Format: Article
Language:English
Subjects:
Brain
Electroencephalography
Emotions
Experimentation
Feasibility studies
Galvanic skin response
Heart rate
Methodology
Physiology
Real time
Robots
Wearable technology
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Emotion estimation systems based on brain and physiological signals such as electro encephalography (EEG), blood-volume pressure (BVP), and galvanic skin response (GSR) are gaining special attention in recent years due to the possibilities they offer. The field of human–robot interactions (HRIs) could benefit from a broadened understanding of the brain and physiological emotion encoding, together with the use of lightweight software and cheap wearable devices, and thus improve the capabilities of robots to fully engage with the users emotional reactions. In this paper, a previously developed methodology for real-time emotion estimation aimed for its use in the field of HRI is tested under realistic circumstances using a self-generated database created using dynamically evoked emotions. Other state-of-the-art, real-time approaches address emotion estimation using constant stimuli to facilitate the analysis of the evoked responses, remaining far from real scenarios since emotions are dynamically evoked. The proposed approach studies the feasibility of the emotion estimation methodology previously developed, under an experimentation paradigm that imitates a more realistic scenario involving dynamically evoked emotions by using a dramatic film as the experimental paradigm. The emotion estimation methodology has proved to perform on real-time constraints while maintaining high accuracy on emotion estimation when using the self-produced dynamically evoked emotions multi-signal database.
ISSN:0129-0657
1793-6462
DOI:10.1142/S0129065720500136