An attractor network model of grid cells and theta-nested gamma oscillations (Pastoll et al 2013)

 Download zip file 
Help downloading and running models
A two population spiking continuous attractor model of grid cells. This model combines the attractor dynamics with theta-nested gamma oscillatory activity. It reproduces the behavioural response of grid cells (grid fields) in medial entorhinal cortex, while at the same time allowing for nested gamma oscillations of post-synaptic currents.
1 . Pastoll H, Solanka L, van Rossum MC, Nolan MF (2013) Feedback inhibition enables theta-nested gamma oscillations and grid firing fields. Neuron 77:141-54 [PubMed]
Model Information (Click on a link to find other models with that property)
Model Type: Realistic Network;
Brain Region(s)/Organism: Entorhinal cortex;
Cell Type(s): Entorhinal cortex stellate cell;
Gap Junctions:
Receptor(s): GabaA; NMDA; Glutamate;
Transmitter(s): Gaba; Glutamate;
Simulation Environment: Brian; Python;
Model Concept(s): Oscillations; Attractor Neural Network; Grid cell;
Implementer(s): Solanka, Lukas [l.solanka at];
Search NeuronDB for information about:  GabaA; NMDA; Glutamate; Gaba; Glutamate;
Installation instructions for the grid cell model

In fact the model does not need any installation, it is recommended to work
directly from the source directory. However there are a number of
Prerequisites_ that the user must install in order to start simulation.


- Python          2.6/2.7
- numpy           >= 1.6.2
- scipy           >= 0.11.0
- matplotlib      >= 1.1.1
- brian simulator >= 1.4.0

All of these should be possible to install with your favorite installer (pip,
easy_install, ...)

Running a simulation

All the simulation scripts are in the grid_cell_model/ directory. It is
advisable to inspect the simulation/submitting files before running.

The main files are described below:

    A script that submits a particular simulation run(s). It can run a single
    simulation or one can define a parameter sweep over a given range of

    A script that performs the actual simulation run. It sets up the model and
    the parameters, runs the simulation and extracts all the necessary output
    data, and usually saves the data into a user-specified file format
    (currently a MAT file) or plots them and saves the figures.

    Contains default parameters for this particular set of simulations. Each
    specific setup within the simulation directory can then override these
    default parameters.

The submitting scripts currently support submitting to a workstation or to Sun
Grid Engine cluster (using qsub). This can be selected inside the submit_*.py
script. Also, the workstation runs can be submitted in a blocking mode (each
successive simulation waits until the previous one has finished) or in a
non-blocking mode (all simulations are run at once concurrently). See the
blocking parameter in these scripts.

Loading data, please wait...