Multiscale modeling from biophysical neural network models to cognitive and behavioral ones; computational neuroscience; behavioral studies with Humans.

The AMAC team is interested in the development of models of perceptual, cognitive and motor functions, as well as the synthesis of control architectures from an integrative perspective. It thus has a dual purpose. On the one hand, the understanding of the living through a mathematical and computer modelling approach of cognitive and motor functions, and the synthesis of robotic control architectures. On the other hand, to provide robots with cognitive and motor capacities integrating decision making and learning.

The fields of application are biology and engineering sciences.

The methodologies developed pursue a common goal: the understanding of the mechanisms underlying perception, action, decision and their organisation, with biology serving as inspiration for robotic developments, and the engineering sciences proposing their methods and tools for modelling, synthesis or validation of the hypotheses considered.

Address :
Institut des Systèmes Intelligents et de Robotique Sorbonne Université – Campus Pierre et Marie Curie Pyramide - 4, place Jussieu 75005 Paris – France

Team leader :
Stéphane Doncieux
Name of co-team leader :
Benoît Girard
Administrative Contact Name :


Website : Cliquez ici
Key words : Modèles computationnels ; prise de décision, apprentissage par renforcement, navigation, sommeil Computational models, decision-making, reinforcement learning, navigation, sleep