i recently read a few interesting papers from Jacob Bellmund, a german cognitive neuroscientist. He talks a lot about “Cognitive Spaces” and the navigation properties of grid and place cells in the Hippocampus and Entorhinal Cortex. I think a few papers could be very interessting to discuss because of the similaritis between the actual Numenta theory and his research.
For example here is an interessting review summary about navigating in cognitive spaces (like reference frames at numenta)
I’ve only read the first page so far, but I have two remarks. Maybe I’ll find corrections to this further down.
They talk about grid space in the hippocampus and the hippocampal-entorhinal region. Is there anywhere in other literature proof for grid cells in the neocortex?
In the example they show vehicules classified on a 2D graph showing weight vs engine power, and so this suggests a continuous data distribution. (So for instance a minicar, a fourseater, an SUV and a truck are represented on a continuous line). From Numenta’s research, I would have thought that each related unit of information is represented as a binary set. (So in my previous example, I would have sets (minicar, fourseater), (minicar, SUV), (minicar, truck), (fourseater, SUV), (fourseater, truck) and (SUV, truck). And for each of these sets I would imagine a specific grid cell module. Doesn’t that make more sense?
This grid-like signal is not unique to the entorhinal cortex, but can be measured during spatial navigation in prescribed parts of the medial frontal, medial parietal and lateral temporal cortices
In other words, the hub areas of each lobe, I see the hex-grid codes as being the lingua-franca between the very different functions of the various lobes.
