Towards the development of physiological models for emotions evaluation.

Area of Science:

• Psychophysiology
• Affective Computing
• Biomedical Engineering

Background:

• Emotions are linked to physiological responses, but automatic recognition remains difficult.
• Existing methods for emotion detection face challenges in accuracy and real-time application.

Purpose of the Study:

• To develop and validate a novel approach for estimating emotional valence, arousal, and dominance using physiological signals.
• To identify key biological parameters that best predict these emotional dimensions.

Main Methods:

• Utilized multiple linear regression models trained on physiological data (electrodermal activity, heart rate variability, electroencephalography).
• Employed the International Affective Picture System (IAPS) database for model training and validation.
• Applied stepwise regression to select optimal input features, including mean RR, EEG spectral power (Alpha, Beta, Theta), and mean EDA.

Main Results:

• Identified mean RR, EEG spectral power (Alpha, Beta, Theta), and mean EDA as significant predictors of emotional states.
• Developed multiple linear regression models demonstrating good performance in predicting valence, arousal, and dominance.
• The selected models effectively captured the relationship between physiological signals and subjective emotional evaluations.

Conclusions:

• The proposed method offers a promising approach for objective and automated emotion recognition.
• Physiological parameters like heart rate variability, EEG, and electrodermal activity are valuable indicators of emotional states.
• This research contributes to advancing the field of affective computing and its applications.