Display Settings:


Send to:

Choose Destination
See comment in PubMed Commons below
J Neurophysiol. 2004 Jun;91(6):2884-96.

A method to estimate synaptic conductances from membrane potential fluctuations.

Author information

  • 1Integrative and Computational Neuroscience Unit (UNIC), Centre National de la Recherche Scientifique, 1 Avenue de la Terrasse, 91198 Gif-sur-Yvette, France.

Erratum in

  • J Neurophysiol. 2004 Dec;92(6):3604.


In neocortical neurons, network activity can activate a large number of synaptic inputs, resulting in highly irregular subthreshold membrane potential (V(m)) fluctuations, commonly called "synaptic noise." This activity contains information about the underlying network dynamics, but it is not easy to extract network properties from such complex and irregular activity. Here, we propose a method to estimate properties of network activity from intracellular recordings and test this method using theoretical and experimental approaches. The method is based on the analytic expression of the subthreshold V(m) distribution at steady state in conductance-based models. Fitting this analytic expression to V(m) distributions obtained from intracellular recordings provides estimates of the mean and variance of excitatory and inhibitory conductances. We test the accuracy of these estimates against computational models of increasing complexity. We also test the method using dynamic-clamp recordings of neocortical neurons in vitro. By using an on-line analysis procedure, we show that the measured conductances from spontaneous network activity can be used to re-create artificial states equivalent to real network activity. This approach should be applicable to intracellular recordings during different network states in vivo, providing a characterization of the global properties of synaptic conductances and possible insight into the underlying network mechanisms.

Comment in

[PubMed - indexed for MEDLINE]
Free full text
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for HighWire
    Loading ...
    Write to the Help Desk