@hlee leads a thorough discussion of the current literature on the functions of grid cells.
00:00 Introduction
00:08 Grid Cells Literature Review
06:01 Place Cells in the Hippocampus
13:51 Grid Cells in the Medial Entorhinal Cortex
19:50 Grid Cell Properties
41:32 How Grid Cells may give rise to Place Fields
01:06:20 Toroidal Topology of Population Activity in Grid Cells
Thank you Hojae Lee (@hlee) for this nice review and for sourcing this interesting discussion.
I think the paper Jeff (@jhawkins) mentioned around 1:04:48 could possibly be this one:
Will there be a follow-up on this video discussion? You mentioned you only covered 1/3 of what you prepared.
Around 1:00:30 in the video, Jeff explains how Numenta correctly predicted a process akin to grid cells in the neocortex. If the same mechanism for coding grid cells in the EC works in the lower levels of cortical columns too, does that mean that each cortical column has a number of grid cell-like modules, or at least intrinsically produces grid-like encoding for its own consumption? Or are there separate cortical areas with dedicated grid cell modules that feed grid encodings to nearby columns?
Hi @Falco ! Thanks so much for your kind words. 
Yes, I think that is the paper that Jeff mentioned as well. If you are interested, Numenta had recorded three videos discussing Hybrid Model of Grid Cells (OI/CAN): Part 1, Part 2, Part 3
Will there be a follow-up on this video discussion? You mentioned you only covered 1/3 of what you prepared.
Yes, we did a Part 2 the following week! I believe @brainwaves may know when it may be available. Also, I remember Jeff going over briefly the paper you linked in Part 2.
Around 1:00:30 in the video, Jeff explains how Numenta correctly predicted a process akin to grid cells in the neocortex. If the same mechanism for coding grid cells in the EC works in the lower levels of cortical columns too, does that mean that each cortical column has a number of grid cell-like modules, or at least intrinsically produces grid-like encoding for its own consumption? Or are there separate cortical areas with dedicated grid cell modules that feed grid encodings to nearby columns?
Currently, we think that it is the former. tbp.monty is built on the assumption that cortical columns learn models of objects in its own reference frame. Below is a snippet from Jeff’s paper, “A Framework for Intelligence and Cortical Function Based on Grid Cells in the Neocortex”.
Thanks @hlee
I’ll come back here and tag when the next part comes out.
