Search for dynamic neuronal assemblies underlying memory processes for order information

(1)对于大鼠，我们开发了行为任务，以查看呈现的音调刺激的持续时间或顺序的记忆，并训练大鼠执行这两项任务。老鼠在三周内完成了这两项任务。我们记录了多神经元的活动。从他们的海马结构在执行任务。(2)对于小鼠，我们开发了一种行为任务，以查看呈现光刺激的顺序的记忆，并训练小鼠执行该任务。有些小鼠在长时间的训练中不能完全获得任务，我们需要将任务的参数更改为更合适的参数。(3)我们开发了新技术，即，电极，头放大器，微驱动器和尖峰分选系统，以记录和分析行为动物的多神经元活动。(4)新开发的技术中最引人注目的是一种新方法，可以检测自由行为动物神经元之间精确的亚毫秒级相互作用。这是一种能够实时和自动分选紧密相邻的单神经元的细胞外活动以检测它们的亚毫秒相互作用的体内电生理技术。该技术使用独立成分分析（伊卡）和新开发的多电极，十二极的组合。(5)在大鼠海马CA_1、CA_3区和皮层体感区植入两个电极，同时记录到10个相互作用的单神经元。(6)该技术还可以检测动作电位的树突反向传播和通过间隙连接的电耦合细胞。这些结果表明，我们的方法可以用来研究非常精确的神经元相互作用，以及树突状细胞的功能，在自由行为的动物学习和运动控制的基础。

(1) For rats, we developed behavioral tasks to see memory for duration or order of presented tone stimuli and trained rats to perform both tasks. The rats acquired both tasks in three weeks. We recorded multi-neuronal activity. from their hippocampal formation during the performance of the tasks.(2) For mice, we developed a behavioral task to see memory for order of presented light stimuli and trained mice to perform the task. Some mice could not fully acquire the task in a long training period and we needed to change parameters of the task to more adequate ones.(3) We developed new techniques, i.e., electrode, head-amp, microdrive, and spike-sorting system, to record and analyze multi-neuronal activity in behaving animals.(4) The most remarkable one of the newly developed techniques is a new method to detect precise sub-millisecond interactions among neurons in freely behaving animals. That is an in vivo electrophysiological technique capable of real-time and automatic sorting of extracellular activity of closely neighboring single-neurons to detect their sub-millisecond interactions. The technique uses a combination of independent component analysis (ICA) and newly developed multi-electrodes, dozentrodes.(5) Up to 10 interacting single-neurons were simultaneously recorded with two dozentrodes implanted in the hippocampal CA1 and CA3 regions and the somatosensory cortex in.the rats.(6) The technique also could detect dendritic backpropagation of action potentials and electrically coupled cells via gap junctions. These results demonstrate that our method can be employed to investigate extremely precise neuronal interactions, as well as the functions of dendritic compartments that underlie learning and motor control in freely behaving animals.