Olfactory Perception and Neural Rhythms: A Simulation-Based EEG Analysis Using Power Spectral Density...

Area of Science:

• Neuroscience
• Cognitive Science
• Signal Processing

Background:

• Olfactory perception is a key cognitive function, influencing emotions, memory, and decision-making.
• Studying olfactory brain responses using electroencephalography (EEG) is challenging due to low signal-to-noise ratios and individual variability.
• A lack of established olfactory EEG databases hinders research progress.

Purpose of the Study:

• To propose a simulation-based framework for analyzing olfactory EEG signals.
• To enable researchers to study brain responses to odors using power spectral density (PSD) analysis.
• To create a reproducible methodological platform for olfactory EEG research.

Main Methods:

• A simulated olfactory EEG dataset was generated for 50 virtual participants across six olfactory conditions (rose/rotten, low/medium/high concentrations).
• EEG data (45 channels, 256 Hz sampling rate) underwent band-pass filtering (0.5-70 Hz).
• Welch's method estimated PSD features for five canonical EEG bands (delta, theta, alpha, beta, gamma), followed by multiclass support vector machine (SVM) classification using Stratified 10-fold cross-validation.

Main Results:

• The simulation framework successfully generated olfactory EEG data.
• PSD-based features effectively distinguished between different olfactory conditions.
• The classification model achieved 99.67% accuracy and a 0.99 macro-averaged F1-score.

Conclusions:

• The proposed simulation-based framework provides a validated platform for reproducible olfactory EEG studies.
• This research lays the groundwork for future studies using actual human olfactory EEG data.
• The findings support the use of PSD analysis and machine learning for olfactory EEG research.