How different immersive environments affect intracortical brain computer interfaces

Ariana F Tortolani ¹, Nicolas G Kunigk ^2,3, Anton R Sobinov ⁴

⁰Committee on Computational Neuroscience, University of Chicago, Chicago, IL, United States of America.

Journal of Neural Engineering +

|

January 30, 2025

View abstract on PubMed

Summary

Brain-computer interface (BCI) decoder performance depends more on user experience with a virtual environment than its immersiveness. Decoders generalized well between virtual reality and monitor environments, with session order impacting results.

Area Of Science

Neuroscience
Human-Computer Interaction
Rehabilitation Engineering

Background

Advancements in brain-computer interface (BCI) technology enable diverse applications.
Seamless environment switching is crucial for the clinical utility of BCI devices.
Investigating virtual environment factors is key to optimizing BCI performance.

Purpose Of The Study

To assess the impact of virtual environment immersiveness on BCI decoder training and generalizability.
To determine if virtual reality (VR) or monitor-based environments yield better BCI performance.
To understand how BCI decoder performance transfers between different virtual environments.

Main Methods

Two participants with intracortical electrodes in the precentral gyrus controlled a virtual arm using a BCI.
BCI decoders were trained and tested in both immersive VR and non-immersive monitor environments.
Neural tuning and decoder performance were analyzed across different environmental conditions.

Main Results

Participant performance was superior in the environment with greater prior user experience.
Virtual environment immersiveness had minimal influence on neural tuning for movement.
BCI decoders demonstrated good generalizability between environments, though session order affected outcomes.

Conclusions

Prior user experience with a virtual environment is more critical than its immersiveness for BCI performance.
BCI performance generalizes effectively across different virtual environments when accounting for user experience.
Findings inform the design of more adaptable and effective BCI systems for clinical use.

Keywords:

intracortical BCI motor cortex motor learning virtual environment