Developmental trajectories for model-free and model-based reinforcement learning: computational and neural bases

Informations projet

DEVELOP‐LEARNING

N° de convention de subvention: 328822

Projet clôturé

Date de début 1 Janvier 2014

Date de fin 31 Decembre 2015

Financé au titre de

• Specific programme "People" implementing the Seventh Framework Programme of the European Community for research, technological development and demonstration activities (2007 to 2013)

Coût total

€ 231 283,20

Contribution de l’UE

€ 231 283,20

Coordonné par

UNIVERSITY COLLEGE LONDON

Objectif

Adolescence is defined as the period of life that starts with the biological changes of puberty and ends at the time at which the individual attains a stable, independent role in society. During this period, decisions are characterized by being impulsive and high-risk, and this can have serious consequences, for example accidents caused by dangerous driving and experimentation with alcohol and drugs. It is thus important to understand the neurocognitive processes that underlie decision-making in adolescence.
Decision-making depends on the interaction of several component processes, including the representation of value, response selection and learning. In reinforcement learning all these processes are integrated. Reinforcement learning deals with the ability of learning to improve one’s future choices in order to maximize the occurrence of pleasant events (rewards) and minimize the occurrence of unpleasant events (punishments).
Theoretical models for reinforcement learning postulate the existence of a dual controller for action selection: ‘Model-free’, which is computationally simple, more reflexive and inflexible and ‘Model-based’, which is more reflective and flexible, but computationally complex.
The development of reinforcement learning in human adolescence has been studied only recently. Studies to date have not attempted to disentangle model-free and model-based behaviour. The aims of the current project are: i. to map the development of reinforcement learning systems – via behavioural/computational analysis - and ii. to relate the maturation of model-based reinforcement learning to the functional maturation of specific neural circuits – via functional magnetic resonance experiment.
This study will provide further insights in adolescent reward-based decision-making and its neural bases, within the theoretical framework of reinforcement learning.

Champ scientifique (EuroSciVoc)

• sciences naturelles › informatique et science de l'information › intelligence artificielle › apprentissage automatique › apprentissage par renforcement

Programme(s)

• FP7-PEOPLE - Specific programme "People" implementing the Seventh Framework Programme of the European Community for research, technological development and demonstration activities (2007 to 2013)

Thème(s)

• FP7-PEOPLE-2012-IEF - Marie-Curie Action: "Intra-European fellowships for career development"

Appel à propositions

FP7-PEOPLE-2012-IEF
Voir d’autres projets de cet appel

Régime de financement

MC-IEF - Intra-European Fellowships (IEF)

Coordinateur

UNIVERSITY COLLEGE LONDON

• Contribution de l’UE: € 231 283,20
• Adresse: GOWER STREET ; WC1E 6BT LONDON ;
• Type d’activité: Higher or Secondary Education Establishments
• Coût total: Aucune donnée