Memory savings through unified pre- and postsynaptic STDP (Costa et al 2015)

 Download zip file 
Help downloading and running models
Accession:184487
Although it is well known that long-term synaptic plasticity can be expressed both pre- and postsynaptically, the functional consequences of this arrangement have remained elusive. We show that spike-timing-dependent plasticity with both pre- and postsynaptic expression develops receptive fields with reduced variability and improved discriminability compared to postsynaptic plasticity alone. These long-term modifications in receptive field statistics match recent sensory perception experiments. In these simulations we demonstrate that learning with this form of plasticity leaves a hidden postsynaptic memory trace that enables fast relearning of previously stored information, providing a cellular substrate for memory savings. Our results reveal essential roles for presynaptic plasticity that are missed when only postsynaptic expression of long-term plasticity is considered, and suggest an experience-dependent distribution of pre- and postsynaptic strength changes.
Reference:
1 . Costa RP, Froemke RC, Sjöström PJ, van Rossum MCW (2015) Unified pre- and postsynaptic long-term plasticity enables reliable and flexible learning eLife
Model Information (Click on a link to find other models with that property)
Model Type: Synapse;
Brain Region(s)/Organism: Neocortex;
Cell Type(s): Neocortex V1 pyramidal corticothalamic L6 cell;
Channel(s):
Gap Junctions:
Receptor(s):
Gene(s):
Transmitter(s): NO; Glutamate; Endocannabinoid;
Simulation Environment: MATLAB; Brian; Python;
Model Concept(s): STDP;
Implementer(s): Costa, Rui Ponte [ruipontecosta at gmail.com];
Search NeuronDB for information about:  Neocortex V1 pyramidal corticothalamic L6 cell; NO; Glutamate; Endocannabinoid;
########################################################
 >> Costa et al. 2015 [1] <<
 >> Code for running savings simulation (Figure 3) <<

 url: http://modeldb.yale.edu/184487
########################################################


0. To run the model you need Brian neural simulator installed [2] and for plotting the figure as in the paper Matlab [3]. There is some preliminary plotting available within the Python script.

1. Run file prepostSTDP_savings.py in Python, which will store the simulated data in the folder fromBrian/

2. For plotting run file X.m in Matlab. This will use the data generated by the Python script, which is saved in folder fromBrian/

Note: This code is not optimized for efficiency :)


References:
[1] - Costa RP, Froemke RC, Sjöström PJ and van Rossum MCW - Unified pre- and postsynaptic long-term plasticity enables reliable and flexible learning - eLife 2015; 10.7554/eLife.09457
[2] - http://briansimulator.org/  plus numpy and matplotlib
[3] - http://uk.mathworks.com/products/matlab (tested on version 2014a, the plotting needs to be adjusted for later versions)

Loading data, please wait...