Rapid Network Reorganization With Learning

描述（由申请人提供）：神经集成动力学-神经元组如何在真实的时间内在功能组之间转移以优化神经网络性能-代表了脑功能研究中的临床重要前沿。灵活的合奏结构是许多神经回路的固有特征，我们推测，这种动力学的畸变可能有助于重要的神经和精神疾病的病理。不幸的是，技术问题长期以来阻碍了对网络结构中实时灵活性的研究。在这里，我们将应用新开发的成像和数据处理工具来研究灵活的合奏动力学在实验上易于处理的模型系统-的Aaplasia运动网络。中心假设是，在刺激引起的爬行运动程序的过程中，失智症的运动网络显示灵活的整体结构，然后持续编码记忆。具体目的是1）定义在单个运动程序的时间过程期间发生的神经系综结构的快速重组，2）确定在目标1中识别的网络系综在多大程度上持续编码用于敏化的记忆，以及3）测试目标1和2中表征的快速系综重组由已知的，内源性多巴胺能调节神经元。该方法将使用具有快速电压敏感染料的大规模成像和三种数据分析技术的强大组合，包括用于识别原始数据中神经元的全自动尖峰分选方法，以及两种新一代无监督尖峰序列相关方法，以定义和跟踪正常网络功能和学习期间运行的功能神经元集合。该项目的短期目标是提供一个单神经元分辨率视图，以了解神经网络结构中可能对健康大脑功能至关重要的时刻变化。长期目标是利用这些信息为神经和精神疾病的新疗法提供信息。

DESCRIPTION (provided by applicant): Neural ensemble dynamics - how groups of neurons shift between functional groups in real time to optimize neural network performance - represents a clinically important frontier in the study of brain function. Flexible ensemble structure is an inherent feature of many neural circuits, and we speculate that aberrations of such dynamics may contribute importantly to pathology in neurological and psychiatric disease. Unfortunately, technical issues have long hindered investigations of real-time flexibility in network structure. Here we will apply newly developed imaging and data processing tools to study flexible ensemble dynamics in an experimentally tractable model system - the Aplysia locomotion network. The central hypothesis is that Aplysia's locomotion network displays flexible ensemble structures during the course of its stimulus-elicited crawling motor program, which then persist to encode memory. The specific aims are to 1) define the rapid the reorganization of neural ensemble structures that occur during the time course of a single motor program, 2) determine to what extent the network ensembles identified in Aim 1 persist to encode the memory for sensitization, and 3) test the hypothesis that the rapid ensemble reorganization to be characterized in Aims 1 and 2 is mediated by known, intrinsic serotonergic modulatory neurons. The approach will use a powerful combination of large-scale imaging with fast voltage sensitive dyes and three data analysis techniques, including a fully automated spike-sorting method for identifying the neurons in the raw data, and two new-generation unsupervised spike train correlation methods to define and track the functional neuronal ensembles operating during normal network function and learning. The short term goal of this project is to provide a single-neuron resolution view of the moment-to-moment alterations in neural network structures that may be essential to healthy brain function. The long-term goal is to use such information to inform novel treatments for neurological and psychiatric disease.