A computational model for the formation of grid field based on path integration

Grid cells in the medial entorhinal cortex of rats have special firing characteristics. The cells fire repeatedly and regularly corresponding to a specific spatial location. This is a relatively small space, which is called grid cells' firing fields. Multiple firing fields overlap each other, which finally forms the grid nodes. When the rat reaches any grid nodes in the environment, there exists the corresponding grid cells which discharge maximally. Furthermore, an equilateral triangle is formed by connecting grid nodes and the triangles are covered with the entire environment. Since Hafting et al discovered grid cells in entorhinal cortex of the rat in 2005, researchers have paid more attention to the regular firing structure of grid cells. With the further studying on grid cells, it proved to be that entorhinal cortex plays an important role in path integration. Most importantly, grid cells act as the function of path integrator in the process of path integration. Based on what we have mentioned above, we put forward a computational model for the formation of grid field based on path integration. Our proposed model simulates grid cells' firing fields very well in that the simulation results of our proposed model are in agreement with grid cells' biological experimental electroencephalogram results.