Rate maps for all sessions in which the cells shown in
Figure 7B were recorded, and other simultaneously recorded grid and boundary cells. Rate maps are grouped by day of recording as in
Figure 7B (day 10, 11, 13, 14). Rate maps from manipulated conditions are annotated with the grid rotation and x, y components of the phase shifts measured in cm as elsewhere (see
Figure 2). In ROT70 and ROT45 the firing patterns of several boundary cells tracked the same geometric boundary, whereas grids remained more strongly anchored to the room (see
Figure 3A,D and
Figure 3—figure supplements 1 and
2), generating the grid-boundary dissociations highlighted in
Figure 7B. However, a number of exceptions were observed (e.g., r377d10t10c1, r377d11t10c2, r377d14t10c1) in which the firing pattern of the boundary cell seems to have spread across two walls in ROT70, perhaps consistently with the orientation concurrently expressed by the grids in accordance with the Boundary Vector Cell model (
Hartley et al., 2000;
Barry et al., 2006).