Estimation of synaptic conductances and their variances from intracellular recordings of neocortical neurons in vivo.

Michael Rudolph (1, 2) , J. Guillaume Pelletier (3) , Denis Paré (3) , Alain Destexhe (1, 2)

1 UNIC - Unité de neurosciences intégratives et computationnelles
2 INAF - Institut de Neurobiologie Alfred Fessard
3 Center for Molecular & Behavioral Neuroscience

During intense network activity, neocortical neurons are in a "high-conductance" state. To estimate the respective contributions of excitatory and inhibitory conductances in generating such states, we combined computational models with intracellular recordings obtained in cat parietal cortex in vivo. Fitting a fluctuating-conductance model to the recordings revealed that inhibitory conductances are dominant (several times larger than excitation). Conductance variance (i.e., the "noise") was also larger for inhibition, indicating that inhibitory dynamics have a pronounced impact on membrane potential fluctuations. We conclude that the synaptic bombardment of neocortical neurons in vivo is not excitatory, but mostly determined by inhibitory conductances.

Mots clés: Cerebral cortex; Synaptic noise; Subthreshold activity; Stochastic systems

Michael Rudolph, J. Guillaume Pelletier, Denis Paré, Alain Destexhe. Estimation of synaptic conductances and their variances from intracellular recordings of neocortical neurons in vivo.. Neurocomputing, 2004, 58-60, pp.387. ⟨10.1016/j.neucom.2004.01.071⟩. ⟨hal-00018652⟩