On the Assessment of Functional Connectivity in an Immersive Brain-Computer Interface During Motor Imagery

Area of Science:

• Neuroscience
• Human-Computer Interaction
• Robotics

Background:

• Brain-computer interfaces (BCIs) are advancing, with new trends focusing on integration with immersive virtual reality (VR) to enhance user realism.
• Combining BCIs with VR offers novel ways to control robotic systems and explore human perception within virtual environments.

Purpose of the Study:

• To propose and evaluate an experimental BCI system that uses motor imagery (MI) to control an immersive telepresence robot from a first-person perspective (1PP).
• To analyze functional brain connectivity differences between 1PP and third-person perspective (3PP) control using graph theory metrics.

Main Methods:

• Electroencephalography (EEG) signals were recorded from two subjects performing MI to control a NAO humanoid robot in a telepresence system.
• Functional brain connectivity was analyzed using partial directed coherence (PDC) to construct binary directed networks.
• Graph theory properties (degree, betweenness centrality, efficiency) were applied to assess network organization.

Main Results:

• Preliminary assessment revealed greater efficiency in the alpha (α) brain rhythm at the prefrontal cortex in the 1PP condition compared to 3PP.
• A stronger influence of signals from the C3 EEG channel (primary motor cortex) to other regions was observed in the 1PP condition.
• Alpha (α) and beta (β) brain rhythms showed high indegree at the prefrontal cortex in 1PP, potentially linked to the sense of agency.

Conclusions:

• The integration of PDC and graph theory in an immersive BCI system for telepresence robot control provides insights into brain network organization.
• Results suggest that 1PP control in immersive BCI systems may enhance the sense of agency through specific brain network dynamics.