Optimal location for gesture decoding in the sensorimotor cortex and implications for brain-computer interface research

Implantable brain-computer interfaces (iBCIs) aim to restore communication in individuals with severe motor impairments. For good iBCI performance, it is important to target an optimal location. In this study, we used high-resolution 7-Tesla functional magnetic resonance imaging (fMRI) to map the spatial distribution of brain activity that can discriminate between a large number of hand gestures. Ten able-bodied participants performed 20 different unimanual hand gestures. Using support vector machines, we measured decodability across the cortex. The highest decoding performance was achieved in the hand region of the sensorimotor cortex. Moreover, we found that a subset of six well-distinguishable gestures could predict the optimal decoding location for the full set, suggesting that a carefully chosen subset can effectively guide pre-implantation mapping. Furthermore, while significant decoding was possible from sulcal as well as gyral regions of the precentral cortex, our analyses revealed that the sulcal area did not contribute unique information beyond that found in adjacent gyral regions. Similarly, decoding in the postcentral cortex was primarily driven by the gyrus. This indicates that surface recordings may suffice for iBCIs. Together, these findings offer practical guidance for future iBCI electrode placement, with the potential to improve communication and autonomy for individuals with severe motor impairments.