Fig. 3: A microcircuit for directing sweeps.
From: Left–right-alternating theta sweeps in entorhinal–hippocampal maps of space

a, Identification of putative monosynaptic connections. Top, dots show locations of co-recorded internal direction cells (green), conjunctive grid cells (pink) and pure grid cells (blue) along each recording shank during an example session. Lines show detected connections between pairs of cells; the colour indicates the functional identity of the presynaptic cell. One example pair of connected cells (conjunctive grid to pure grid) is highlighted. Bottom, cross-correlogram between the firing rates of the two highlighted cells in the top panel. The blue rectangle indicates the time window (0.7–4.7 ms) used to detect putative monosynaptic connections. b, A cell pair consisting of an internal direction cell (ID, green) with a putative projection to a conjunctive grid cell (Conj., pink). Temporal cross-correlogram of firing rates (left), directional tuning (middle) and position tuning (right) (rate expressed by colour intensity, ranging from the 10th to the 99th percentile of each cell’s firing-rate map) for the putative pre- and postsynaptic cells. Note the similar preferred direction of the two cells (middle). c, Preferred directions for all pairs (dots) of putatively connected internal direction to conjunctive grid cells (144 pairs from 13 animals). d, Schematic of inferred connectivity between internal direction cells and conjunctive grid cells. Internal direction cells relay the direction signal to conjunctive grid cells with similar directional tuning. e, A cell pair consisting of a conjunctive grid cell (pink) projecting to a pure grid cell from the same grid module (blue). Panels are as in b. The firing fields of the pure grid cell are shifted along the preferred direction of the conjunctive cell. The direction and magnitude of this offset (green arrow) was found by cross-correlating the spatial rate maps of the cells. f, Preferred direction of presynaptic conjunctive grid cells and grid phase offset direction between pre- and postsynaptic cells for all pairs (dots) of putatively connected conjunctive grid to pure grid cells (85 pairs from 12 animals). g, Schematic of inferred connectivity between conjunctive grid cells and pure grid cells. Conjunctive grid cells project asymmetrically to pure grid cells, and excite pure grid cells with a grid phase offset that is aligned to the preferred direction of the conjunctive cell. In this connectivity scheme, activating a set of conjunctive cells with a particular preferred direction leads to a sweep-like trajectory in that direction in pure grid cells.