Covert Intention to Answer "Yes" or "No" Can Be Decoded from Single-Trial Electroencephalograms (EEGs)

Area of Science:

• Neuroscience
• Cognitive Science
• Biomedical Engineering

Background:

• Interpersonal communication relies heavily on binary question-answer exchanges.
• Decoding intentions from brain activity is a key challenge in neuroscience.
• Electroencephalography (EEG) offers a non-invasive method to study brain signals.

Purpose of the Study:

• To demonstrate the feasibility of decoding binary ('yes'/'no') intentions from single-trial EEG.
• To identify optimal temporal and spectral ranges within EEG for intention decoding.
• To explore the neural correlates associated with 'yes' and 'no' responses.

Main Methods:

• Multichannel single-trial EEG data were recorded during covert 'yes'/'no' self-referential question answering.
• An intention decoding algorithm combined Common Spatial Pattern (CSP) for feature extraction and Support Vector Machine (SVM) for classification.
• Analysis involved time-frequency subwindows (200ms x 2Hz) to pinpoint effective decoding ranges.

Main Results:

• Decoding accuracy was highest in specific time-frequency bands: 800-1200ms (alpha band) and 200-400ms (theta band).
• Combining features from multiple subwindows significantly improved accuracy to approximately 86%.
• Key discriminating features were localized to the right frontal region (theta band) and right centroparietal region (alpha band).

Conclusions:

• Decoding binary intentions ('yes'/'no') directly from brain activity is feasible.
• Specific EEG frequency bands and brain regions are crucial for intention decoding.
• This research paves the way for advanced brain-computer interfaces and understanding neural mechanisms of intention.