UCL Logo

Organization of the entorhinal grid-cell network

May-Britt Moser

Kavli Institute for Systems Neuroscience, NTNU, Norway


Grid cells in the medial entorhinal cortex (MEC) exhibit a striking hexagonal spatial modulation when animals traverse open environments. The grid pattern has three main geometric properties which are thought to play an important role in their contribution to spatial processing. These are grid phase (spatial offset), grid spacing (spatial frequency), and grid orientation (the orientation of the grid axes). Within the MEC network, grid phase is represented randomly, with neighbouring cells showing any combination of spatial phases, whereas grid spacing is organized topographically along the dorsoventral axis of MEC, with the smallest grids expressed most dorsally and the largest most ventrally. The question of whether grid cells are organized according to orientation is not resolved, however. In the first part of my talk, I shall present data from grid-cell recordings at up to 7 ipsilateral locations simultaneously covering a substantial portion of the dorsal-to-intermediate MEC. These data show that grid cells express one orientation across widespread regions of MEC; however, there are systematic minor variations in the relative direction of the three grid axes across cell clusters. These variations give rise to elliptical, rather than a perfectly hexagonal, grid patterns in many cells. Preliminary observations suggest that grid cells with similar ellipse orientations cluster within tetrodes, suggesting that grid cells may be segregated into small functional networks according to orientation. In the second part of my talk, I shall discuss how grid cells interact with place cells in the hippocampus. The GABA-A agonist muscimol was infused locally in the hippocampus while place cells were recorded in CA1 and grid cells and head-direction cells were recorded in MEC in freely moving rats. Muscimol infusions eventually abolished the grid pattern in time-averaged rate maps for entorhinal neurons but temporal correlations between pairs of grid cells were maintained, suggesting that significant ensemble structure persists in the absence of hippocampal backprojections. The gradual deformation of grid fields after hippocampal inactivation implies, however, that the entorhinal-hippocampal network operates as a functional unit where the hippocampus may provide the entorhinal network with stored associations that anchor the grid to external landmarks. In the final part of the talk, I will ask to what extent this spatial map is present at the time when animals explore external environments for the first time during their life. I will show that a rudimentary map of space is already present when 2½-week old rat pups explore an open environment outside the nest for the first time. Head direction cells in the pre- and parasubiculum have adult-like properties from the beginning. Place and grid cells are also present but evolve more gradually. The presence of adult-like directional signals before independent movement raises the possibility that such signals are instrumental in setting up networks for place and grid representation.