Enhancing training performance for brain-computer interface with object-directed 3D visual guidance.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Accurate classification of user intentions is crucial for motor imagery (MI)-based brain-computer interfaces (BCI).
• Effective user training protocols are key to achieving reliable decision-making in MI tasks.
• Investigating factors that enhance BCI performance is essential for practical applications.

Purpose of the Study:

• To evaluate the impact of visual guidance on the classification performance of MI-based BCI.
• To compare different visual guidance scenarios (non-object-directed, static-object-directed, dynamic object-directed) in a 3D virtual environment.
• To assess the effectiveness of single-subject versus multi-subject BCI paradigms.

Main Methods:

• Trained and classified MI tasks (left-hand/right-hand movement imagination) using single-subject and multi-subject BCI paradigms.
• Utilized three distinct visual guidance scenarios within a 3D virtual environment.
• Collected and analyzed classification accuracy and response time data across scenarios and paradigms.

Main Results:

• Classification accuracy for MI tasks varied across the three visual guidance scenarios.
• Static- and dynamic-object-directed scenarios yielded superior classification accuracy compared to the non-object-directed scenario.
• Object-directed scenarios reduced response times and were effective with limited training data. The multi-subject paradigm outperformed the single-subject paradigm.

Conclusions:

• Appropriate visual guidance and BCI paradigms can significantly enhance MI-based BCI classification performance.
• Findings suggest potential for improving the reliability and practical applicability of MI-BCI systems.
• Further research into optimized visual cues and multi-subject approaches is warranted.