A gaze independent hybrid-BCI based on visual spatial attention

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Brain-computer interfaces (BCIs) translate brain activity into commands, bypassing muscle movement.
• Hybrid BCIs, using multiple brain signal measures, offer higher accuracy, particularly for attention-based systems using electroencephalography (EEG).
• Previous attention-based BCIs utilized steady-state visual evoked potentials (SSVEP) and alpha band modulations.

Purpose of the Study:

• To investigate if incorporating P3 event-related potentials enhances an existing attention-based BCI.
• To evaluate the performance improvement of a hybrid BCI design combining P3, SSVEP, and alpha band features.
• To assess the robustness of the enhanced BCI design against individual subject variability.

Main Methods:

• Participants viewed stimuli and were cued to covertly attend to one of two visual streams.
• EEG signals were recorded, measuring P3 components, SSVEPs, and alpha band power modulations.
• Machine learning classifiers were used to differentiate between left- and right-attended trials based on these neural signals.

Main Results:

• Including P3 features alongside SSVEP and alpha band features significantly increased BCI classification accuracy by 9%.
• The hybrid BCI design demonstrated improved robustness, accommodating individual differences in neural responses.
• This multi-modal approach effectively leverages different neurophysiological indices of covert attention.

Conclusions:

• Combining P3 responses with SSVEP and alpha band activity represents a significant advancement in attention-based BCI design.
• The enhanced BCI system offers improved accuracy and reliability, paving the way for more effective brain-computer interfaces.
• This study underscores the potential of integrating multiple neurophysiological measures for superior BCI performance.