"This project aims to advance the science and technology of meta-cognition in social robotics. During social interactions, humans use powerful cognitive mechanisms to represent and interpret what others know or intend. Doing so helps us respond appropriately in social interactions: we use a combination of sensory perception together with internal models of what others are (and will be) doing and thinking. We believe these capabilities to be equally important for social robots, a type of robot which is intended to interact with people. Endowing robot with skills to “read the minds” of others is not only key to build next-generation assistive and industrial robots, but also presents a considerable research challenge. A challenge with tremendous benefits but few successes to date.Relying both on a novel formal approach based on modal logics, and on a strong experimental component in real-world conditions, this proposal suggests an ambitious interdisciplinary approach to better comprehend and formalize higher-order artificial cognition. The aim of which is to advance the design and practical implementation of socially interactive robots, in line with the Horizon 2020 Future and Emerging Technologies priority of moving "cognition beyond problem solving"."