When presenting this work at conferences, people often point out the striking resemblance of left-right-alternating sweeps and directed echolocation clicks in bats (Yovel et al 2010)

Is this left-right-alternating sweep pattern useful for anything? We hypothesized that alternation might signify an efficient strategy for spatial coverage. In line with this, @rjgardner.bsky.social found that stable alternation emerged in an artificial sweep-generating agent moving forwards:

Left-right-alternating sweeps were extremely persistent, and were expressed continuously as rats ran in open arenas, on mazes, in completely novel envs and during REM sleep. This video shows decoding during a period of REM sleep (animal is completely immobile).

Based on patterns of connectivity between cell types (inferred from CCGs), we propose a circuit model of how internal-direction input might be integrated by grid cells to drive sweeps in an experience-independent manner:

Grid-cell sweeps were aligned to an "internal” direction signal expressed by cells in parasubiculum, one of the main cortical inputs to MEC. These parasubicular cells (often regarded as weakly tuned HD cells) expressed a remarkably coherent population signal that alternated in step with sweeps:

By recording from hundreds of MEC cells in freely moving rats, we found that grid cells encode a position signal that sweeps outwards from the animal once per theta cycle, in a stereotyped left-right-alternating pattern across successive theta cycles.

Excited to share that our paper is now published in Nature! In this study, we uncovered a grid-cell-based circuit in medial entorhinal cortex that probes the surrounding environment with theta-paced, sweeping spatial representations 🧵
www.nature.com/articles/s41...