Figure 3: Coding of location by phase. (a) Actual trajectory run by the rat is shown in gray. If phase is reset at start location, the inverse transform of phase at any position yields a return vector with angle and distance leading back to start (shown with black dashed line). (b) Plot of membrane potential oscillation phase in a single dendrite of a simulated grid cell, showing linear change in phase with one dimension of location (dendrite receives input from head direction cell with angle preference zero). (c) Phase of another dendrite receiving input from head direction with angle preference 120. (d) Performing the inverse transform of oscillation phases at each point in time effectively reconstructs the full trajectory.