Answers are grounded in this paper and cite their sources.
Objective. Intracortical microstimulation (ICMS) can be used to deliver information to the cortex from a bidirectional brain-machine interface (BMI), thereby improving control of neuroprosthetic devices. Primary somatosensory cortex (S1) is a common choice for delivering information, but S1-ICMS can transsynaptically affect neurons in distant cortical regions such as primary motor cortex (M1), potentially disrupting BMI decoders trained on those neurons as inputs. Other nodes of the cortical reach-to-grasp network recently have been identified as effective territory for delivering information with ICMS. However, the extent of distant modulation produced by ICMS delivered in cortical regions other than S1 remains largely unknown. Approach. In each of two monkeys, we implanted several regions of the cortical reach-to-grasp network, including ventral premotor cortex, dorsal premotor cortex, M1, S1, the anterior intraparietal area, and dorsal posterior parietal cortex, and then delivered ICMS in each region (except M1) separately. We investigated whether the ICMS pulses in each region produced time-locked modulation of spike activity in neurons located in each of the other implanted regions. Results. ICMS pulses delivered in each region produced time-locked, short latency modulation of neurons recorded from most other implanted regions, including those not known to have direct anatomical interconnections with the stimulated region, with some effects having surprisingly high amplitudes and short latencies. Distant modulation effects were more common in frontal than in parietal regions, with neurons in M1 receiving particularly powerful modulation. Nevertheless, decoders trained on activity recorded from those M1 neurons during ICMS still could achieve strong performance. Significance. As the repertoire of cortical regions in which ICMS can be used to deliver information for bidirectional BMIs expands, knowledge about distant cortical modulation produced by ICMS will be useful to maintain robust decoding, as well as to better understand how information conveyed by ICMS is integrated into the cortical network.
No takes yet. Share an insight, caveat, or question.
Ruszala et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: