Spatial navigation in rodents
In order to navigate around an environment, our brain needs to construct an internal representation of place. The aptly named place cells are active only when and animal is in a certain location within its environment. However, these are not the only cells involved in this internal map. Grid cells, whose activity patterns form hexagonal grids when viewed relative to the animals position in the environment are another key ccomponent that allow animals to learn their spatial location.
In spite of our knowledge of grid and place cells, together with a wide variety of other cells, such as speed and head direction cells, we do not yet now how information provided by these is integrated to give a robust location detector. For example, it was once thought that grid cells provide feed-forward input to the place cells, which would subsequently fire only when grid patterns overlapped. However, this does not account for the fact that place cells appear to develop earlier than grid cells in rodents.
In rodents, the grid cells are found in the entorhinal cortex, one of the main input areas to the hippocampus. Along the dorsoventral axis of this body, the spacing of grid cell firing patterns grows in a seemingly modular fashion, ranging from centimetres up to several metres (depending on the environment). Transgenic Alzheimer's mice show a prononced drop in this gradient of grid cell spacing, and this may account for their inability to solve maze tasks.
Our research considers two questions: firstly, how simple biophysical mechansims may give rise to this gradient of grid cell spacing and secondly, why these gradients disappear in the Alzheimer's mice. To answer these questions, we use a combination of phenomological neural field models, together with detailed network models of neurons found in the entorhinal cortex.