Description du projet
Déterminer des réseaux d’apprentissage causal du cerveau
Comprendre la manière dont nous apprenons les relations de cause à effet est une question qui remonte à l’Antiquité. Malgré les progrès actuels de la technologie et des neurosciences, entre autres, la voie exacte qu’utilise le cerveau pour construire ces relations reste hors de portée. L’hypothèse de travail du projet CausalBrain est que plusieurs régions cérébrales y participent à travers des interactions dirigées pour former des réseaux afin d’aider le cerveau à apprendre. En utilisant des données magnétoencéphalographiques recueillies pendant une tâche d’apprentissage causal, le projet vise à clarifier davantage cette voie et à analyser des théories actuelles d’apprentissage causal face à des données cérébrales et comportementales.
Objectif
Humans have an extraordinary capacity to infer cause-effect relations and form beliefs about the causal effect of actions. This ability provides the basis for rational decision-making and allows people to engage in meaningful life and social interactions. In fact, alterations of cognitive processes involved in causal learning have been found in patients affected by psychiatric disorders such as obsessive-compulsive disorder, schizophrenia and addiction. The formation of causal beliefs relies on learning rules determined by the contingency between actions and outcomes. Although fronto-striatal areas are known to be involved in the cogntive architecture of causal beliefs, it is still unknown how these brain regions interact to learn causal structures. This project aims to unravel the link between functional brain networks and causal reasoning. We hypothesize that causal representation are implemented in a dynamic distributed network of directed functional interactions between brain regions and that this network is shaped by learning. We will characterize the modulations of brain circuits involved in learning phases as well as the brain networks responsible of internal representations of contingency values and associated uncertainty. We are going to pursue these two aims by analyzing magneto-encephalografic and intracranial electro-encephalographic data collected during a causal reasoning task. We will use state-of-the-art methods for dynamic directed connectivity estimation. In addition, we will develop machine learning pipelines to found those subnetworks that implement the cognitive architecture of causal learning. Overall, we will be able to understand whether causal learning and the psychological internal variables predicted by rational theories are reflected in dynamically changing directional influences in whole-brain circuits.
Champ scientifique
Programme(s)
Régime de financement
MSCA-IF - Marie Skłodowska-Curie Individual Fellowships (IF)Coordinateur
75794 Paris
France