Improving the performance of brain-computer interface through meditation practicing.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Cognitive Science

Background:

• Brain-computer interfaces (BCIs) commonly utilize motor imagery for electroencephalography (EEG) control.
• Factors like concentration and task difficulty can affect EEG signal reliability in BCIs.
• Consistent control of EEG signals is crucial for effective BCI system development.

Purpose of the Study:

• To investigate if meditation practice can enhance mind controllability for mental tasks in a BCI system.
• To assess the impact of meditation on EEG signal generation and control during cognitive tasks.
• To improve classification accuracy of EEG patterns for BCI applications.

Main Methods:

• Experiments were conducted on two groups: a meditation group and a control group.
• Subjects performed mental tasks involving imaginative hand movement and an idle state.
• Time-frequency analysis of EEG signals was used to analyze brain activity patterns.

Main Results:

• Meditation practitioners exhibited event-related desynchronization (ERD) of beta rhythm before imagination.
• Strong event-related synchronization (ERS) of beta rhythm (around 25 Hz) was observed during hand motor imagery in the meditation group.
• The meditation group achieved an average classification accuracy of 88.73%, significantly higher than the control group's 70.28%.

Conclusions:

• Meditation practice enhances mind controllability, leading to improved EEG signal generation and control for BCI.
• Meditation practice can significantly improve the classification accuracy of EEG patterns in BCI systems.
• High classification accuracies, up to 98.0%, are achievable with meditation practice in BCI users.