Relative spike time coding and STDP-based orientation selectivity in V1 (Masquelier 2012)

 Download zip file 
Help downloading and running models
Accession:141062
Phenomenological spiking model of the cat early visual system. We show how natural vision can drive spike time correlations on sufficiently fast time scales to lead to the acquisition of orientation-selective V1 neurons through STDP. This is possible without reference times such as stimulus onsets, or saccade landing times. But even when such reference times are available, we demonstrate that the relative spike times encode the images more robustly than the absolute ones.
Reference:
1 . Masquelier T (2012) Relative spike time coding and STDP-based orientation selectivity in the early visual system in natural continuous and saccadic vision: a computational model. J Comput Neurosci 32:425-41 [PubMed]
Model Information (Click on a link to find other models with that property)
Model Type: Realistic Network; Synapse;
Brain Region(s)/Organism:
Cell Type(s): Thalamus geniculate nucleus/lateral principal GLU cell; Neocortex L2/3 pyramidal GLU cell; Retina ganglion GLU cell; Retina photoreceptor cone GLU cell; Retina bipolar GLU cell; Abstract integrate-and-fire leaky neuron;
Channel(s):
Gap Junctions:
Receptor(s): AMPA;
Gene(s):
Transmitter(s):
Simulation Environment: C or C++ program; MATLAB;
Model Concept(s): Pattern Recognition; Coincidence Detection; Temporal Pattern Generation; Synchronization; Spatio-temporal Activity Patterns; Synaptic Plasticity; Long-term Synaptic Plasticity; Action Potentials; Learning; Unsupervised Learning; Winner-take-all; STDP; Development; Information transfer; Orientation selectivity; Vision;
Implementer(s): Masquelier, Tim [timothee.masquelier at alum.mit.edu];
Search NeuronDB for information about:  Thalamus geniculate nucleus/lateral principal GLU cell; Neocortex L2/3 pyramidal GLU cell; Retina ganglion GLU cell; Retina photoreceptor cone GLU cell; Retina bipolar GLU cell; AMPA;
/
ModelDB.JCNS
mat
src
README.txt
                            
This readme and the code were contributed by Timothee Masquelier
timothee.masquelier@alum.mit.edu
Sept 2011

This code was used in:
Masquelier T (2012) Relative spike time coding and STDP-based orientation
selectivity in the early visual system in natural continuous and saccadic
vision: a computational model. J Comput Neurosc (in press).

Feel free to use/modify but please cite us if appropriate.

The code only contains the LGN and V1 stages.

The retina simulator, developed by Adrien Wohrer, can be downloaded here:
http://www-sop.inria.fr/neuromathcomp/public/software/virtualretina/
Note that it only compiles on Linux.
Then it should be launched using for eg:
Retina video/*.png -ret cat.retina.xml -r 10 -outD . -nodisp
(the cat.retina.xml file we provide contains the retinal parameters used in our baseline simulation)
This produces a text file spikes.spk that contains the RGC spikes, and that should be placed in ../mat/
This file is then read by my code.
Please read the Virtual Retina documentation for more information.

All my code is in the ./src/ directory
It's been tested on Linux and Windows (I guess it should also work on Mac OS).
It contains Matlab scripts and functions, and 2 mex files (in C), that should be compiled first (from within Matlab):
>> mex STDPContinuous.c
>> mex pspKernel.c

Then the main script can be launched:
>> main

The script reads the ./mat/spikes.spk file produced by Virtual Retina.
It then formats it (cropping), and then launches LGN, V1 and plots the final RFs (Fig 6 in the paper)
See comments at the beginning of each script.

20130515 T Masquelier supplied an update in response to a modeler's
request for help where a problem where STDPContinuous.c was not
compiling on a particular platform was fixed by renaming min and max
to local_min and local_max within that program.

Loading data, please wait...