Skip to Main content Skip to Navigation
Theses

Modélisation de la propagation de l'information cérébrale par graphes causaux qualitatifs

Abstract : In the domain of neuroscience, activation studies use functional brain imaging techniques in order to provide measures of human brain activity during the performance of cognitive tasks. These studies highlight the fact that large-scale networks of cerebral areas are involved in the processing of specifics cognitive functions. On the one hand, the cerebral information processing complexity, and on the other, the incompleteness of the measurements, reduce the understanding of the links between the networked organisation and the realised cognitive function. These limitations arise the necessity of a formal modelling tool and computer simulation. Moreover, a clear understanding of the functional mechanisms can only be achieved through their explicit representation. From this viewpoint, we propose an original formalism based on causal graphs associated to a qualitative representation of the information. The causal graphs fulfil the need for a representation of the double organisation of the brain: structural et functional. We propose a hierarchical organisation of graphs, based on components. This organisation reduces the complexity and extends the reusability of the models. Moreover, we use a homogeneous formalism for the description of both structural and functional levels. To have an explicit representation of the cerebral information, we propose a qualitative abstraction of it, by a couple: (numeric interval, symbol). With this dual representation, we can deal with the magnitude and the category of the information, although the magnitude only can be observed with the different imaging tools. Based on this formalism, we propose the BIOCAEN simulator. At each sampling time, the simulation principle consists in the propagation of the influences through the causal graph. This mechanism allows to express the brain's automatic, stimulus-driven, processes, which are the source of the major part of the cerebral activity. In the field of the exploration of the human brain function, our contribution is to offer to the neuroscientists a formal tool for the expression of hypotheses on cerebral information propagation, at a high level of abstraction.. Furthermore, the BIOCAEN simulation of the models can give an validation of the expressed hypotheses by a confrontation of the simulation results and neuroimaging data.
Document type :
Theses
Complete list of metadata

Cited literature [83 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00010176
Contributor : Marc Lafon <>
Submitted on : Friday, September 16, 2005 - 6:17:18 PM
Last modification on : Friday, January 10, 2020 - 9:09:02 PM
Long-term archiving on: : Friday, September 14, 2012 - 2:25:16 PM

Identifiers

  • HAL Id : tel-00010176, version 1

Collections

Citation

Marc Lafon. Modélisation de la propagation de l'information cérébrale par graphes causaux qualitatifs. Modélisation et simulation. Université Paul Sabatier - Toulouse III, 2000. Français. ⟨tel-00010176⟩

Share

Metrics

Record views

330

Files downloads

829