文献类型：期刊文献

英文题名：The roles of internal dynamics and proprioceptive feedback in motor cortex during movement execution

作者：Jiang, Hongru[1];Bu, Xiangdong[1];Zheng, Zhiyan[1];Tang, Huajin[2];Pan, Xiaochuan[3];Chen, Yao[1]

机构：[1]Shanghai Jiao Tong Univ, Sch Biomed Engn, Shanghai 200240, Peoples R China;[2]Zhejiang Univ, Coll Comp Sci & Technol, State Key Lab Brain Machine Intelligence, Hangzhou 310027, Peoples R China;[3]East China Univ Sci & Technol, Sch Math, Shanghai 200237, Peoples R China

年份：2025

卷号：629

外文期刊名：NEUROCOMPUTING

收录：;EI(收录号：20250817908742);WOS:【SCI-EXPANDED(收录号:WOS:001430048300001)】；

基金：This work was funded by the STI 2030-Major Projects (2022ZD0208604), and the National Natural Science Foundation of China (62176151, 61773259 and 32441113). We are grateful to Guillaume Hennequin and Ta-Chu Kao for modeling instructions; to Juan Gallego and Matthew Perich for sharing the monkey data, to Xinyu Chai, Liming Li, and Xiaohong Sui for guidance, and to Jieji Ren and Di Zhu for discussions.

语种：英文

外文关键词：Motor control; Motor cortex; Proprioceptive feedback; Inhibitory stabilized network; Neural population dynamics

摘要：The motor cortex controls arm movements by sending commands to lower motor centers. The synaptic connections within the motor cortex (internal dynamics) are vital to generate motor commands during movement execution. However, recent studies suggest that proprioception also contribute to motor cortex processing. To investigate the contributions of internal dynamics and proprioceptive feedback to movement execution, we built a recurrent neural network model of the motor cortex; the model receives proprioceptive feedback from a virtual arm performing a delayed-reach task. We found that both internal dynamics and proprioceptive feedback contribute to the resemblance to the real motor cortex data. We then dissected their contributions by disrupting them separately. Internal dynamics dominate the generation of neural population activity, while proprioceptive feedback modulates neural responses and controls movement deceleration. Additionally, proprioceptive feedback improves the network's robustness in noisy initial conditions. We further investigated the relative importance of the components in proprioceptive feedback and found that hand velocity is most important. Our findings could inform the development of neural prosthetics that can replicate the sensorimotor control of the biological system.