Closed-Loop Systems for Large Scale Spatiotemporal Imaging and Actuation of Neural Activity in Freely Behaving Animals

摘要 神经科学的一个主要挑战是揭示大脑中已定义的神经回路是如何编码、存储、修改和检索信息的。增加这一挑战的是，神经功能不是在孤立的情况下运行，而是在活的、行为正常的动物中运行。在过去的几十年里，巨大的技术进步使研究人员能够开始光学测量和调节神经活动，但当研究与内部神经回路动力学相关的空间和时间尺度上的神经功能时，这些方法往往局限于头部固定动物。虽然已经取得了大量的科学和技术进步，但关于复杂的神经功能，特别是在自然行为的背景下，仍有许多需要学习。这种知识差距，至少在一定程度上是由于缺乏可获得的工具，可以同时以单细胞精度调制和观察自由行为动物的大规模神经电路。该项目将通过开发开源、头戴式微型显微镜来填补这一空白，该显微镜具有时空照明能力，可以在自由行为的动物中进行图案化光刺激和改善神经成像。我们将开发一个模块化的控制和采集平台，用于神经和行为设备的本地集成，以促进神经行为实验。最后，该平台将由一个新颖的、自动化的、闭环处理框架驱动，用于实时检测、解码和操纵神经和行为动力学。该平台的目标是：1)显著扩展和改进当前自由行为的神经成像和调制技术；2)提供一个在行为背景下观察、控制和操纵神经动力学的集成框架。这种方法有可能同时对大脑的任何区域进行“读”和“写”，从而能够对神经活动和行为之间的因果关系进行细粒度的研究。我们的开发将在开发和优化过程的每个阶段都以并行台式和体内测试为指导。为了最大限度地发挥我们的努力的影响，为这项提议开发的所有工具和技术都将是开源的，并将通过在线资源和技术研讨会与科学界广泛共享。

ABSTRACT A major challenge in neuroscience is to uncover how defined neural circuits in the brain encode, store, modify, and retrieve information. Adding to this challenge is the fact that neural function does not operate in isolation but rather within living, behaving animals. Great technological advances over the past decades have allowed researchers to begin to optically measure and modulate neural activity but these approaches are often limited to head-fix animals when studying neural function at spatial and temporal scales relevant to internal neural circuit dynamics. While a great deal of scientific and technological progress has been made, there is still much to learn concerning complex neural function, especially within the context of natural behavior. This knowledge gap, at least in part, is due to a lack of accessible tools for simultaneously modulating and observing large-scale neural circuits with single-cell precision in freely behaving animals. This project will fill this gap by developing open-source, head-mounted miniature microscopes with spatiotemporal illumination capabilities for both patterned photo-stimulation and improved neural imaging in freely behaving animals. We will develop a modular control and acquisition platform for native integration of neural and behavioral equipment to facilitate neural-behavioral experiments. Finally, this platform will be driven by a novel, automated, closed-loop processing framework for detecting, decoding, and manipulating neural and behavioral dynamics in real-time. The goal of this platform is to 1) significantly extend and improve upon current freely behaving neural imaging and modulation techniques and 2) provide an integrated framework for observing, controlling, and manipulating neural dynamics within the context of behavior. This approach has the potential to simultaneously “read” from and “write” into, potentially, any area of the brain, enabling fine-grained investigation of the causal role between neural activity and behavior. Our development will be guided by concurrent benchtop and in vivo testing at every stage of the development and optimization process. To maximize the impact of our efforts, all tools and technologies developed for this proposal will be open-source and shared widely with the scientific community through online resources and technical workshops.