My opinion about why grid cell mechanism are formed in the brain

Before reading, it is just my opinion and I haven’t tried to find a paper about this particular topic. I would be glad to read those paper or to be criticized.

Grid cells are awesome. They have some advantage for identifying current location and so on. but, I can hardly find about ‘why’ a hexagonal grid is formed in the brain rather than other location encoding mechanism Ex) rectangular coordinate

I think, the hexagonal grid is formed because the relative distance between two dots are important and is often used in the brain.
By moving one point to another point in particular distance(for human, a good example is a one step, length of a thumb…etc) the potential location pool forms a circle. when they move again, the intersection between the two circle will start to form a hexagonal grid.
Also, to explain why hexagonal grid doesn’t regenerate every step and can form a regular representation while moving continuously and in the dark, we can use the idea about motor signal. the information about currently moving direction and speed will do.

thanks for reading my vague idea and i hope i can read some research papers related to this topic

1 Like

I think, the hexagonal grid is formed because the relative distance between two dots are important and is often used in the brain.

I think the answer is very simple. It’s how equally sized bubbles, cells, particles and spheres of any kind pack together when attracted to each other on a 2D surface.

A square or rectangular spacing has wasted space between rows, which gets filled by the rows shifting one radius. This causes the rows to mesh together to become close packed, a stable geometry. Each unit then contacts 6 neighbors.

Although cortical columns can contain many cells it’s still an interconnected system that must stay together as one unit. After units close pack together there are again 6 neighbors.

This is one of my experiments:

1 Like

If my understanding is correct, the triangular lattice pattern of neuron activations in the grid cell modules exist because of local inhibition effects. The set of winning/activated cells suppress the firing of their neighbors. The overlapping spheres of inhibition are what give rise to the lattice pattern. As the sensor is shifted (translated/rotated), there is a subtle change in the proximal inputs that result in a shift of the firing pattern that is proportional to the sensor shift. These same inhibitory effects then generate a similar (but shifted) lattice pattern. If after some number of these shifts, a sufficient number of the original nodes become active again, then the original lattice pattern will reestablish itself. This periodic behavior (amongst individual neurons) is perhaps what is being observed in the experiments with mice moving about their environment. (I go into a little more detail on these ideas here.)

2 Likes

I guess you misunderstood my point. I’m talking about grid cells, not grid of cells. I totally agree with you with how cortical columns are formed in the brain and why is it forms a hexagonal grid. But what I’m talking here is why the grid cell encodes location in hexagonal grid(I think the triangular grid is more suitable tho) this grid is what im talking here.

Also, I love your work. thanks for replying
:smile: Oscillatory “Thousand Brains” Mind’s Eye For HTM?

So the grid cell firing pattern is derived by the triangular lattice-like inhibition mechanism in the internal brain? Interesting. I’ll look foreword to it

1 Like

I guess you misunderstood my point. I’m talking about grid cells, not grid of cells.

I should have gone one step further and included this part of the video CollinsEM presented in another topic. In this case the equally sized units are the “doughnut of inhibitory connectivity” around the neurons.

I’m trying to find a good example to explain this kind of grid formation, but it’s not an easy one to find an example for. Best I can think of are equally powerful radio stations operating at the same frequency (by inhibition) not being allowed to interfere with the signal from neighboring stations. The “visualization of this inhibition generated pattern” that Eric Collins is working on would be useful right now!

1 Like

i hope i can read some research papers related to this topic

We might have to start with this paper from October 2017 that helps show how little is known about grid field formation:

1 Like