Correlation structure of grid cells is preserved during sleep

The network of grid cells in the medial entorhinal cortex forms a fixed reference frame for mapping physical space. The mechanistic origin of the grid representation is unknown, but continuous attractor network (CAN) models explain multiple fundamental features of grid-cell activity. An untested prediction of CAN grid models is that the grid-cell network should exhibit an activity correlation structure that transcends behavioural or brain states. By recording from MEC cell ensembles during navigation and sleep, we found that spatial phase offsets of grid cells predict arousal-state-independent spike rate correlations. Similarly, state-invariant correlations between conjunctive grid-head-direction and pure head-direction cells were predicted by their head-direction tuning offsets. Spike rates of grid cells were only weakly correlated across modules, and module scale relationships disintegrated during slow-save sleep, suggesting that modules function as independent attractor networks. Collectively, our observations suggest that network states in MEC are expressed universally across brain and behaviour states.