A. I. Virtanen Institute, University of Eastern Finland, Kuopio, Finland.
Picower Institute for Learning and Memory, Massachusetts Institute of Technology, Cambridge, Massachusetts.
Hippocampus. 2018 Feb;28(2):81-96. doi: 10.1002/hipo.22810. Epub 2017 Nov 8.
For the sake of rigorous control of task variables, hippocampal place cells have been usually studied in relatively simple environments. To approach the situation of real-life navigation in an urban-like environment, we recorded CA1 place cells while rats performance a memory task in a "Townmaze" with two start locations, three alternate paths in the maze midsection, followed by a two-way choice that determined the trial outcome, access to a goal compartment. Further, to test the ability of place cells to update their spatial representation upon local changes in the environment while maintaining the integrity of the overall spatial map to allow effective navigation, we occasionally introduced barriers in the maze mid-section to force the rat to select a nonpreferred route. The "Townmaze" revealed many new interesting features of CA1 neurons. First, we found neurons with 3-5 fields that appear to represent segments on a single common route through the maze. Second, we found neurons with 3-5 fields similarly aligned along the longitudinal or transverse maze axis. Responses to the barriers were assessed separately near and far from the barriers. Appearance of new fields in response to the barriers took place almost exclusively only locally near the barrier, whereas in-field firing rate changes occurred throughout the maze. Further, field location changes did not correlate with the task performance, whereas firing rate changes did. These findings suggest that in a complex environment with blocked distal views, CA1 neurons code for the environment as sequences of significant nodes but are also capable of extracting and associating common elements across these sequences.
为了严格控制任务变量,通常在相对简单的环境中研究海马体位置细胞。为了接近在城市环境中进行实际导航的情况,我们在一个有两个起始位置、迷宫中段有三条交替路径的“Townmaze”中记录 CA1 位置细胞,然后进行双向选择,以确定试验结果,进入目标室。此外,为了测试位置细胞在环境局部变化时更新其空间表示的能力,同时保持整体空间图的完整性以允许有效导航,我们偶尔会在迷宫中段引入障碍物,迫使老鼠选择非首选路线。“Townmaze”揭示了 CA1 神经元的许多新的有趣特征。首先,我们发现了 3-5 个场的神经元,它们似乎代表了通过迷宫的单个共同路径上的片段。其次,我们发现了 3-5 个场的神经元沿着纵向或横向迷宫轴同样排列。分别评估靠近和远离障碍物的障碍物响应。响应障碍物出现新字段几乎完全仅在障碍物附近局部发生,而场内的发射率变化发生在整个迷宫中。此外,字段位置变化与任务表现无关,而发射率变化则有关。这些发现表明,在具有阻塞的远距离视图的复杂环境中,CA1 神经元将环境编码为重要节点的序列,但也能够提取和关联这些序列中的共同元素。