Spatial Encoding Differences Between Horizontal and Vertical Planes

Post by Deborah Joye

What's the science?

The brain helps us navigate the world by creating a spatial map of our surroundings, combining sensory information about how our body is moving with inputs from our external environment. The brain has specialized cells, called hippocampal place cells, that track where we are in the environment by only firing when we are in the cell’s ‘preferred’ location, or ‘firing field’. Place cells work together with grid cells of the entorhinal cortex, which track distance travelled by creating a grid-like array of firing fields across the entire environment. When we move, place and grid cells use external landmarks to estimate location, and self-motion cues, like where the body is in relation to the surface and movement speed, to constantly update those estimations. However, the function of these cells has mostly been studied as animals move horizontally. Do these cells integrate sensory information the same way when we move vertically? This week in PNAS, Casali and colleagues record grid cells and place cells in freely behaving rats to demonstrate that the brain differently encodes movement in the vertical and horizontal planes.

How did they do it?

The authors recorded electrical activity from 148 grid cells in the medial entorhinal cortex of 11 rats and 72 place cells from the hippocampus of 3 rats as they explored an open field arena (horizontal surface only) or over a floor with an adjoining climbing wall (horizontal and vertical surfaces). To ensure that recorded brain activity was not due to the novelty of a new climbing environment, the authors recorded from rats that had extensive climbing experience. Since self-motion cues, such as running speed, and local field potentials are also important for grid cell and place cell function, the authors also recorded from separate cells that encode speed and measured local field potentials. If the grid plane is defined by gravity, then walking on a vertical surface instead of a horizontal plane should produce “stripes” (firing fields would be aligned in vertical stripes) in the recording of grid cell activity. Alternatively, if the grid plane is defined by the body plane, then firing fields should be grid-like (circular and evenly spaced) on the wall, just as they are on the floor. This is because even though the rat is moving vertically, the body is still parallel to the movement surface, as it would be if the rat was moving horizontally across the floor.

What did they find?

The authors found that grid cell firing patterns were different when rats climbed on the wall versus walking across the floor. During climbing, grid cells showed an overall reduction in firing activity with fewer, larger firing fields than those seen during horizontal movement. Grid cells also produced discrete firing fields during climbing, rather than the vertical “stripes,” that might be expected if grid cell firing while climbing on a vertical wall were due to gravity, suggesting that grid cell firing is adjusted by considering the rat’s body plane in relation to their movement surface. In contrast to grid cells, fewer hippocampal place cells were active during climbing but firing characteristics were otherwise not different between the horizontal and vertical planes. Lastly, recordings of firing rates and local field potentials from speed cells revealed that the brain’s encoding of movement speed was consistently underestimated during vertical climbing, which may contribute to the enlarged grid cell firing fields observed when rats were climbing.


What's the impact?

This study is the first to demonstrate that spatial representation in the brain is determined by an interaction between the body-plane alignment and the gravity axis; grid cells track distance differently when movement is over a vertical surface rather than a horizontal one. The speed-coding analysis suggests that this difference may result from underestimation of movement speed on the wall - the grid cells behave as though the animal is moving more slowly than it really is, thus producing larger, more widely spaced firing fields. Overall, this study suggests that the neural encoding of space is can distinguish horizontal from vertical movement, which may have adaptive consequences for animals that move over surface terrain.


Casali et al., Altered neural odometry in the vertical dimension, PNAS (2019). Access the original scientific publication here.