Kim Robin, Liu Yuxuan, Zhang Jiaao, Xie Chong, Luan Lan
Department of Electrical and Computer Engineering, Rice University, Houston, TX 77005, USA.
Rice Neuroengineering Initiative, Rice University, Houston, TX 77005, USA.
Npj Flex Electron. 2025;9(1). doi: 10.1038/s41528-025-00447-y. Epub 2025 Jul 14.
Neural representations arise from the spatiotemporally structured activity of neuron populations, inherently residing in high-dimensional spaces. Writing specific information into the central nervous system requires precisely manipulating neural states within this framework. However, current neuromodulation methods lack the precision to fully address this complexity, presenting a significant challenge for advancing effective bidirectional interfaces. In this perspective, we advocate for high-dimensional stimulation as a systematic approach capable of approximating the high dimensionality of natural neural code for brain-machine interface applications. We outline key technological requirements on resolution, coverage, and safety, review recent advances in critical application areas, and highlight the promise of flexible electrode technology in enabling a transformative leap towards precise synthetic neural codes.
神经表征源自神经元群体的时空结构化活动,其本身存在于高维空间中。将特定信息写入中枢神经系统需要在此框架内精确操纵神经状态。然而,当前的神经调节方法缺乏充分应对这种复杂性的精度,这对推进有效的双向接口构成了重大挑战。从这个角度来看,我们主张将高维刺激作为一种系统方法,能够在脑机接口应用中逼近自然神经编码的高维性。我们概述了在分辨率、覆盖范围和安全性方面的关键技术要求,回顾了关键应用领域的最新进展,并强调了柔性电极技术在实现向精确合成神经编码的变革性飞跃方面的前景。