Exploring Training Effect in 42 Human Subjects Using a Non-invasive Sensorimotor Rhythm Based Online BCI

Area of Science:

• Neuroscience
• Biomedical Engineering
• Rehabilitation Technology

Background:

• Electroencephalography (EEG)-based brain-computer interfaces (BCIs) offer a novel communication pathway.
• BCI control is recognized as a learnable skill that improves with practice and training.

Purpose of the Study:

• To investigate the impact of three training sessions on BCI behavioral performance.
• To examine changes in electrophysiological properties during BCI use across training sessions.
• To identify differences in training effects between high and low BCI performers.

Main Methods:

• Utilized EEG-based BCIs with 42 human subjects across three training sessions.
• Monitored BCI accuracy and information transfer rates.
• Analyzed electrophysiological properties, including event-related desynchronization (ERD) and synchronization (ERS) at channels C3 and C4.
• Assessed ERD lateralization between left- and right-hand motor imagery tasks.

Main Results:

• Significant improvements in group average BCI accuracy and information transfer rate were observed by the third session compared to the first.
• Low BCI performers (<70% accuracy) showed particularly significant improvements.
• Significant differences in ERD lateralization for BCI control emerged in the second and third sessions, but not the first.
• No significant changes in R² values or ERD/ERS at C3/C4 were detected across sessions.

Conclusions:

• BCI training can rapidly enhance behavioral performance and electrophysiological characteristics within a few hours and three sessions.
• Multiple training sessions appear especially advantageous for individuals with lower initial BCI proficiency.
• Training induces measurable changes in BCI control, including altered lateralization patterns, suggesting neuroplastic adaptations.