This is the large scale model of turtle visual cortex (the "NGU model") described in:
Nenadic, Z., Ghosh, B.K. and Ulinski. P. (2003) Propagating Waves in Visual Cortex: A Large Scale Model of Turtle Visual Cortex", J. Computational Neuroscience 14:161-184.
and
Nenadic, Z., Ghosh, B.K. and Ulinski. P. (2002) Modeling and Estimation Problems in the Turtle Visual Cortex, IEEE Trans. Bio-Med. Eng., 49:753-762
It is also described in considerable detail in the file TurtleVisCortex-descrip.pdf, which is included in this archive. This README file tells how to run the simulation with GENESIS.
With the default settings in the main simulation script TurtleVisCortex.g, a simple graphical interface appears, after a few minutes of setup time for the network. It's use is optional, and the main script may be modified to perform the simulation with no graphics, as for doing long simulation runs in batch mode. In either case, output is generated to plain text files in subdirectories of TurtleVisCortex/output.
6. Click on RUN in the CONTROL PANEL
The default stimulus applied is the "diffuse light flash", simulated with a 0.15 sec pulse of 0.2 nA injection current to the soma of all 201 LGN neurons. The four graphs show the network response by plotting the soma membrane potential of the medial and lateral layer pyramidal cell neurons /network_lateral/cell321, /network_medial/cell279, /network_lateral/cell1, and /network_medial/cell1. The window under the control panel shows the locations of the lateral (represented by squares) and medial (represented by triangles) pyramidal cells. During the simulation, they change color to represent the soma membrane potenial of the cells cells. For the first 0.6 seconds of the simulation, you will see the spreading wave of activation, due to the flash.
The TurtleVisCortex.g script includes various GENESIS scripts in ../lib that define functions that are used in the main script for setting up the network. It defines several global variables that may be changed to alter the simulation behavior. Some of the more important ones are:
Simulation duration and integration step size in seconds:
float tmax = 1.50 // Time in seconds for the simulation to run float dt = 0.00005 // Internal time stepsize
The stimulation flags:
int DIFFUSE_LIGHT_FLASH = 1
int STATIONARY_LIGHT_SPOT = 0
int MOVING_LIGHT_SPOT = 0
int DIRECT_NEURON_STIM_REC = 0
int DIRECT_NEURON_STIM_CIR = 0
These are for
The functions can be used individually, or in combinations. In fact, you can stimulate all five ways in the same run if you want to. Details for changing the stimulus are provided in TurtleVisCortex.g and in TurtleVisCortex/lib/stimulus.g.
The default name of the data directory for the output response files is defined in TurtleVisCortex.g as:
str responseName = "TurtleVisCortex_diffuse_"
Change this name for each different run. If files in a directory with this name exist, they will be overwritten the next time the simulation is run.
../lib/utilities_control.g defines the functions that are used to run the simulation with or without graphics. It can be used as the starting point for more sophisticated graphical interfaces to the simulation.
../coords contains plain text data files with the coordinates of neurons in the seven layers of the model. The figure in coords/Distribution_of_Neurons_Cortex.pdf shows the locations of the 368 lateral neurons (blue), 311 medial neurons (yellow), 20 horizontal neurons (green), 45 stellate neurons (red), and 44 subpial neurons (magenta) in a simulated 1.6 x 1.6 mm piece of cortex. There are also 201 LGN input neurons, represented as a line of equally spaced cells.
Davis (Data Viewing System) is a general-purpose data viewer designed for the simultaneous display of a large number of dynamic data sets. Davis was inspired by the need to explore computational models of the cerebral cortex, and may be used to provide a variety of visualizations of the output from this model.
You may find more about Davis from http:/visual.cs.utsa.edu/davis/ and instructions for using webstart to run it directly from a browser in order to further analyze the output of this model.