Understanding the relationship between rosemary odor and mental workload through deep learning

Area of Science:

• Neuroscience
• Cognitive Science
• Artificial Intelligence

Background:

• Mental workload assessment is crucial for optimizing performance and preventing errors.
• Electroencephalogram (EEG) signals offer a direct measure of brain activity.
• Traditional EEG analysis often requires manual feature extraction, which can be complex and time-consuming.

Purpose of the Study:

• To investigate the effect of rosemary aroma on mental workload.
• To develop and validate deep learning models for classifying mental workload from raw EEG signals.
• To assess the efficacy of olfactory stimulation in reducing mental workload.

Main Methods:

• Thirty volunteers performed neuropsychological tests under rosemary aroma exposure and a control condition.
• Electroencephalogram (EEG) data were recorded during task performance.
• Deep learning models, including Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNN), were applied to raw EEG signals for mental workload classification without feature extraction.

Main Results:

• Volunteers exposed to rosemary odor exhibited reduced error rates and improved test success and learning scores.
• Deep learning models achieved high accuracy (97.11%) in classifying mental workload under rosemary odor.
• The study demonstrated successful mental workload classification directly from raw EEG signals using deep learning.

Conclusions:

• Rosemary odor shows potential in reducing mental workload.
• Deep learning models can effectively analyze raw EEG signals for mental workload classification, bypassing the need for manual feature engineering.
• Combining olfactory stimulation with advanced AI techniques offers a novel approach to understanding and managing cognitive states.