Different spatial scales in mapping from grid cells to place cells: A neural network model

Medial entorhinal cortex (MEC), which is known to be the hub of the brain network for navigation and spatial representation, is commonly perceived to be the major input and output structure of hippocampal formation. Grid cells, the principal cells of MEC, show multiple firing fields arranged in a triangular grid, tessellating the environment. Place cells of hippocampus have a single localized pattern of activity. The spatial scale in both MEC and hippocampus increase systematically along dorsoventral axis, which is seemingly due to a systematic variation in the gain of a movement-speed signal, generated outside the hippocampus. In this article, an artificial neural network model has been proposed, allowing for the single confined place fields of hippocampal pyramidal cells to be emerged from the activities of grid cells. The important point is that this model considers a movement-speed signal which determines the activation of a portion of grid cells with specific spatial scales. This might establish the scale of space representation in place cells. Place fields in this model could be formed considering a modest number of grid cells (for example, 60) with diverse spatial phase and spacing which is consistent with physiological experiments.