The TAM-WG Model: Place Code
Theoretical background:
Single-unit recordings from CA3 and CA1 pyramidal cells in the hippocampus show a high correlation between spike rate and the location of the animal in the environment, hence the cells are called place cells. These cells were found to fire at an elevated rate over a continuous, compact area, called the place field of the cell (O'Keefe & Dostrovsky, 1971). However, place cells show direction-dependent firing fields in some specific types of tasks, particurlarly those with restricted movement paths as in a maze (McNaughton et al., 1983; Markus et al., 1995).
For information on place cells, please refer to the paper cited above or to the papers listed in the Hippocampus and Navigation Group homepage.
The place code ...
Placed atop the feature detector layers (head direction code and walls code) is the Place Layer. This layer, like the feature detector layers, uses a normalized Hebbian learning algorithm to learn the combination of features given by the feature detector layers. In the figure shown below, the place layer is composed of 400 nodes which are organized in a two dimensional matrix. The different colors represent the amount of activity being received in each cell of the matrix. Each node in the feature detector layers connects to all nodes in the place layer.
University
of Southern California Brain Simulation Lab
All contents copyright (C) 1994-1997. University of Southern California Brain Simulation Lab All rights reserved.Author: Alex Guazzelli <aguazzel@rana.usc.edu>