Resting-State Long-Range Functional Connectivity Density Reveals Sensorimotor Rhythm-Based BCI Performance Variations

Abstract

Sensorimotor rhythm-based brain-computer interface (SMR-BCI) control performance varies across subject. Functional connectivity density mapping (FCDM) analysis, based on resting-state fMRI data, was used to investigate the variation problem in current study. High performance group showed stronger long-range FCD in bilateral middle occipital gyrus, whereas low performance group exhibited stronger long-range FCD in insula and putamen. In addition, ROI analysis showed long-range FCD of left insula and putamen is strongly correlated with SMR-BCI performance (r = -0.72). The findings of current study may help understanding the brain network mechanisms of SMR-BCI control, and it also provides us a biomarker for predicting subject's SMR-BCI control performance prior to training.