Optimizing Stimulus Frequency Ranges for Building a High-Rate High Frequency SSVEP-BCI.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Steady-state visual evoked potential (SSVEP) based brain-computer interfaces (BCIs) offer high communication speed and short calibration times.
• Current SSVEP systems often use low-to-medium frequency stimuli, potentially impacting user comfort.
• High-frequency stimuli can enhance visual comfort but often exhibit lower performance.

Purpose of the Study:

• To investigate the distinguishability of 16-class SSVEPs using different high-frequency ranges.
• To compare classification accuracy and information transfer rate (ITR) across these ranges.
• To develop and validate an optimized online high-frequency SSVEP-BCI system.

Main Methods:

• Explored SSVEP distinguishability using three frequency ranges: 31-34.75 Hz (0.25 Hz interval), 31-38.5 Hz (0.5 Hz interval), and 31-46 Hz (1 Hz interval).
• Compared classification accuracy and ITR for each frequency range.
• Developed an online 16-target high-frequency SSVEP-BCI system based on the optimized range.
• Validated the system with 21 healthy subjects.

Main Results:

• The narrowest frequency range (31-34.5 Hz) demonstrated the highest performance.
• The optimized online BCI system achieved an average ITR of 153.79 ±6.39 bits/min.
• This frequency range proved effective for encoding 16 distinct SSVEP targets.

Conclusions:

• The narrowest high-frequency range (31-34.5 Hz) is optimal for SSVEP-BCI design, balancing performance and comfort.
• The developed online SSVEP-BCI system demonstrates high efficiency and feasibility.
• Findings contribute to the advancement of more comfortable and effective SSVEP-based BCIs.