How are neuronal codes of the spatial environment generated at the
level of synapses, neurons, and neuronal circuits? Neurons in layer 2
of the medial entorhinal cortex (MEC2) fire spikes at regular spatial
intervals. The periodic hexagonal firing pattern of such grid cells
forms a neuronal map that may provide the brain with an accurate
metric of space. To understand how grid cell firing is generated at
the synaptic, cellular and network level, we have combined
computational modelling, in vivo and in vitro recordings. Whole-cell
patch-clamp recordings from MEC neurons in mice navigating in a
virtual-reality (VR) environment showed that firing of MEC2 neurons
was driven by sustained slow membrane potential depolarizations during
firing field crossings, which was best predicted by a continuous
attractor network model of grid cell firing. To understand how MEC2
neurons integrate their synaptic inputs to form the grid cell code, we
combined patterned two-photon glutamate uncaging in vitro with
simulations of grid cell firing. We find that the dendrites of MEC2
neurons are highly excitable, exhibiting dendritic spikes and
supralinear input-output curves. Modelling shows that these nonlinear
dynamics can sharpen the precision of the temporal code and enhance
the robustness of the rate code, thereby supporting a stable and
accurate representation of space under varying environmental
conditions.