When embodiment matters most: a confirmatory study on VR priming in motor imagery brain-computer interfaces training

Area of Science:

• Neuroscience
• Human-Computer Interaction
• Rehabilitation Technology

Background:

• Virtual Reality (VR) enhances Brain-Computer Interface (BCI) applications by improving the Sense of Embodiment (SoE) and Motor Imagery (MI).
• VR-based MI-BCIs show potential for motor rehabilitation, but neurofeedback quality is crucial.
• The impact of embodiment priming before VR-BCI interaction on MI-BCI outcomes remains under-explored.

Purpose of the Study:

• To investigate if prior embodiment induction influences subsequent MI-BCI performance.
• To analyze the effects of embodiment priming on event-related desynchronization (ERD) and BCI performance metrics.

Main Methods:

• A within-subject design combining pilot and extended experiments with 39 participants.
• Participants underwent an embodiment induction phase followed by EEG-recorded MI training.
• Analysis of ERD and lateralization indices to assess the impact of prior embodiment.

Main Results:

• Embodiment induction successfully increased SoE.
• No significant differences in ERD were observed between embodied and control conditions.
• Increased variability in lateralization indices in the embodied condition suggested individual differences in feedback integration.

Conclusions:

• Real-time VR-based feedback during training, not prior embodiment, significantly drives MI-BCI performance improvements.
• Real-time embodied feedback is the primary mechanism for performance gains in MI-BCI.
• Embodiment plays a complex role in VR-based MI-BCIs, with real-time feedback being the critical factor.