Max Planck Institute for Dynamics and Self-Organization -- Department for Nonlinear Dynamics and Network Dynamics Group
Personal tools
Log in

BCCN Seminar

Wednesday, 06.07.2005 09 s.t.

Local rules of synaptic modifications in Spike-timing-dependent plasticity

by Ausra Saudargiene
from Vytautas Magnus University, Kaunas, Litauen

Contact person: Tobias Frederic Niemann

Location

Seminarraum Haus 2, 4. Stock (Bunsenstr.)

Abstract

In spike-timing-dependent plasticity (STDP) synapses are potentiated or depressed depending on the temporal order and time difference between the pre- and postsynaptic spikes. We developed a biophysical model of STDP based on a differential Hebbian rule which correlates the NMDA receptor-mediated synaptic conductance with the filtered postsynaptic signal at the location of the synapse as the pre- and postsynaptic quantities. The model reproduces the generic STDP timing rule for the change of synaptic weights. The actual shape of the weight change curve is strongly influenced by the amplitude and time course of the postsynaptic membrane potential depolarization which induces learning. These waveforms vary along the dendritic tree as a result of changes in the shape of backpropagating action potentials with distance from the soma, or due to the emergence of local dendritic spikes, and thus our model predicts that learning rules are dependent on synaptic location in the dendritic tree. The model has been extended to account for more natural spike patterns by including short-term depression mechanisms. We also implement a biophysically-motivated local learning rule approach in artificial neural networks.

back to overview