Effect of alpha range activity on SSVEP decoding in brain-computer interfaces.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Brain-computer interfaces (BCIs) offer direct neural control of external devices.
• Commercialization of BCIs necessitates high accuracy, efficiency, and user consistency.
• Inter-participant variability in BCI performance is a significant challenge, often due to physiological differences.

Purpose of the Study:

• To investigate the impact of steady-state visual evoked potential (SSVEP) flicker on endogenous alpha power.
• To analyze how SSVEP flicker affects classification accuracy in BCIs.
• To identify specific frequency ranges that may cause performance degradation.

Main Methods:

• Analysis of a publicly available SSVEP dataset.
• Examination of the correlation between alpha power and classification accuracy.
• Identification of error patterns in decoding algorithms based on predicted frequencies.

Main Results:

• Participants with classification accuracy below 95% exhibited elevated alpha power.
• The decoding algorithm showed maximum incorrect predictions when the target frequency was between 9-12 Hz.
• Alpha power interference may negatively impact BCI accuracy for certain users.

Conclusions:

• Frequencies between 9-12 Hz can lead to reduced performance in SSVEP BCIs using canonical correlation analysis.
• Alpha-band frequencies used for SSVEP stimulation can interfere with endogenous alpha power, affecting EEG-based BCI accuracy.
• Addressing alpha power interference is crucial for improving BCI reliability and user consistency.