Pleasantness Recognition Induced by Different Odor Concentrations Using Olfactory Electroencephalogram Signals.

Area of Science:

• Neuroscience
• Sensory Science
• Computational Neuroscience

Background:

• Olfactory-induced emotions significantly impact human behavior and cognition.
• Electroencephalogram (EEG) offers a non-invasive method to study brain activity related to sensory perception.

Purpose of the Study:

• To investigate the feasibility of recognizing odor-induced pleasantness using EEG.
• To identify optimal EEG frequency bands and odor concentrations for accurate emotion classification.
• To evaluate machine learning classifiers for olfactory emotion recognition.

Main Methods:

• Collected EEG data from 13 participants exposed to varying concentrations of rose and rotten odors.
• Classified odor-induced pleasantness using power spectral density features and Support Vector Machine (SVM).
• Analyzed classification accuracy across different odor concentrations and EEG frequency bands.

Main Results:

• Achieved high classification accuracies: 93.5% for pleasant odors and 92.2% for disgusting odors.
• The gamma frequency band demonstrated superior performance compared to other EEG rhythms.
• Higher odor concentrations, particularly the highest, resulted in better classification accuracy.

Conclusions:

• EEG technology is a viable tool for recognizing odor-induced pleasantness.
• Gamma band activity is crucial for distinguishing emotional responses to odors.
• Odor concentration is a key factor influencing the accuracy of olfactory emotion recognition via EEG.