Memory of virtual environment and neural mechanism of route learning

为了研究认知地图的神经基础，即环境空间布局的表征，我们训练了两只日本猴子在虚拟建筑中执行导航任务，并记录了顶叶皮质内侧部分7M区及其附近的神经元活动。利用计算机图形学创建了一座两层的虚拟建筑，并通过两台液晶投影仪将其立体投影到100英寸的切向屏幕上。1.行为方面：训练猴子通过操作操纵杆从起点导航到目标房间，而不是直接看到的地方。作为记忆引导导航的提示，每次试验开始时都会呈现目标房间的图像。猴子们已经知道了通往5个球门房间的路线。在从第一个起点学习任务后，他们被训练从第二个起点执行导航任务，而第二个起点是他们在第一次训练期间已经熟悉的。第二次训练完成得比第一次快。此外，我们还发现，在第二次训练后，从一个新的起点开始进行一次找路测试所需的时间缩短了。2.神经元方面：我们在内侧顶叶皮质记录了364个神经元，并分析了神经元活动与几个任务事件的相关性。在记录的神经元中，150个(41%)对至少三个运动中的一个有反应，56个(15%)表现出运动选择性，根据运动的类型改变它们的活动。在运动选择神经元中，42个(75%)表现出位置选择性，根据虚拟环境中的位置改变它们对运动的反应。当猴子在特定位置做出特定动作时，这些神经元就会变得活跃，可能代表着沿着路线到达目的地的重要信息。

To investigate neural basis of the cognitive map, the representation of the spatial layout of environments, we trained two Japanese monkeys to perform navigation task in virtual building and recorded neuronal activity in the mesial part of the parietal cortex, area 7m and its vicinity. A virtual building with two floors was created by computer graphics and projected stereoscopically on the 100 inches tangential screen by two liquid crystal projectors. 1.Behavioral aspect : The monkey was trained to navigate from a starting point to a goal room out of its direct view by operating a joystick. As the cue of memory guided navigation, the image of the goal room was presented at the beginning of each trial. The monkeys had learned the routes to 5 goal rooms. After learning the task from the first starting point, they were trained to perform the navigation task from a second starting point which is already familiar to them during the first training session. The second training was completed faster than the first one. Furthermore, we found that the time required for a free-way-finding test from a novel starting point was shortened after the second training. 2.Neuronal aspect : We recorded 364 neurons in the mesial parietal cortex and analyzed correlation between the neuronal activity and several task events. Of the recorded neurons, 150 (41%) responded to at least one of three movements, and 56 (15%) showed movement selectivity, changing their activity depending on the type of the movement. Of the movement-selective neurons, 42 (75%) showed location selectivity, changing their responses to the movements depending on the location in the virtual environment. These neurons became active when the monkey made a specific movement at a specific location and may represent important information to follow a route to a destination.