Skip to Main content Skip to Navigation
Theses

Classes de dynamiques neuronales et correlations structurées par l'experience dans le cortex visuel.

Abstract : Neuronal activity is often considered in cognitive neuroscience by the evoked response but most the energy used by the brain is devoted to the sustaining of ongoing dynamics in cortical networks. A combination of classification algorithms (K means, Hierarchical tree, SOM) is used on intracellular recordings of the primary visual cortex of the cat to define classes of neuronal dynamics and to compare it with the activity evoked by a visual stimulus. Those dynamics can be studied with simplified models (FitzHugh Nagumo, hybrid dynamical systems, Wilson Cowan) for which an analysis is presented. Finally, with simulations of networks composed of columns of spiking neurons, we study the ongoing dynamics in a model of the primary visual cortex and their effect on the response evoked by a stimulus. After a learning period during which visual stimuli are presented, waves of depolarization propagate through the network. The study of correlations in this network shows that the ongoing dynamics reflect the functional properties acquired during the learning period.
Complete list of metadatas

Cited literature [201 references]  Display  Hide  Download

https://pastel.archives-ouvertes.fr/pastel-00676104
Contributor : David Colliaux <>
Submitted on : Saturday, March 3, 2012 - 5:35:17 AM
Last modification on : Wednesday, January 23, 2019 - 10:29:08 AM
Long-term archiving on: : Wednesday, December 14, 2016 - 9:59:13 AM

File

Identifiers

  • HAL Id : pastel-00676104, version 1

Collections

Citation

David Colliaux. Classes de dynamiques neuronales et correlations structurées par l'experience dans le cortex visuel.. Sciences cognitives. Ecole Polytechnique X, 2011. Français. ⟨pastel-00676104⟩

Share

Metrics

Record views

625

Files downloads

663