Models of firing neurons in the brain with dynamic coupling to produce complex network synchronisation
This project presents Matlab code to simulate the following cases:
- One firing neuron cell, using three different models.
- A network of x number of firing neurons, coupled together using a static coupling matrix.
- A network of x number of firing neurons, coupled together using a dynamic coupling function based on a model of the synapse between neuron cells.
In the single_neuron_models directory are individual programs that run to model behaviour of a single firing neuron Each program has some example runs listed at the top.
- OneNeuronTau.m : A simple model based on the Tau constant
- OneNeuronIzhInF.m : Izkevich's famous integrate-and-fire neuron model
- OneNeuronExpInF.m :A more complex exponential neuron model
In the neuron_network_models directory run the program 'Neuron Simulations' to bring up a GUI that allows configuration and display of the neuron network.
- NeuronSimulations.m : A GUI that brings together all the different models and behaviours
- NeuronNetworkTau.m : Models a network of 'Tau firing' neurons connected together with a coupling matrix. The state of each neuron through time is calculated using the forward Euler method.
- NeuronNetworkInF.m : Models a network of 'Integrate-and-fire firing' neurons
- NeuronNetworkExp.m : Models a network of 'Exponential model firing' neurons
There are also a number of programs that help construct the network models
- NetworkCoupling.m : Constructs a few different coupling matrices - straight line network / nearest neighbour network / random coupling
- Synchronisation.m : Calculates level of synchronisation between all pairs of neuron in network at every time point in simulation, and takes average
- *plot.m : helper method to plot figures in GUI
- *.fig : figures for displaying result of Matlab programs
The final file
- NeuronNetworkInFwithPlasticity.m : Models a network of integrate and fire neurons, however this extended model calculates each link between two neurons dynamically using a synaptic weights model. The more spikes fired by a neuron feeding into the synapse, the stronger the link made by the synapse becomes.
The Docs folder contains sample run configurations and results in a Demo.doc