Using scalp electrical biosignals to control an object by concentration and relaxation tasks: design and evaluation

Area of Science:

• Neuroscience
• Human-Computer Interaction
• Biomedical Engineering

Background:

• Brain-computer interfaces (BCIs) offer novel interaction methods.
• Controlling video games with biosignals requires efficient signal processing and sensor arrays.
• Distinguishing between mental states like relaxation and concentration is key for BCI control.

Purpose of the Study:

• To compare two electroencephalographic (EEG) system designs for controlling a video game object.
• To evaluate the impact of electrode number and signal processing techniques on BCI performance.
• To investigate the role of muscular activity in differentiating mental states for BCI applications.

Main Methods:

• Comparison of a single-channel EEG system (alpha band power) with a multi-channel EEG system (16 electrodes) using machine learning.
• Evaluation of muscular activity using five facial and neck electrodes.
• Assessment of system performance through game control success rates and user subjective feedback.

Main Results:

• The multi-channel, machine learning-based system achieved 100% successful game control, compared to 70% with the single-channel system.
• Users reported a higher sense of control with the machine learning-based system.
• Analysis indicated that muscular activity, potentially captured by EEG electrodes, can also distinguish between relaxation and concentration states.

Conclusions:

• Multi-channel EEG combined with machine learning significantly enhances BCI performance for video game control.
• Muscular activity provides valuable data for differentiating mental states and may be inadvertently recorded by EEG sensors.
• Future BCI systems could benefit from integrating both brain and muscular activity for improved control and user experience.