The central aim of the proposed research project is to systematically explore the empirical implications of a novel theoretical approach to the early representational preconditions and the functional structure of the mechanisms dedicated for understanding other minds. We aim to explore and shed new theoretical light on the basic cognitive and brain mechanisms of the human social mind. One of these mechanisms that has received much attention by earlier approaches to ‘theory-of-mind’ research concerns the ability to infer and represent the mental states of others. Standard theories and research in the last twenty five years have suggested that representing other’s beliefs is an effortful and late developing capacity (Wellman et al., 2001) whose main function is to explain others’ behavior. Here we advance and propose to explore a new theoretical perspective according to which the mechanisms of mental state monitoring and representation involve primarily automatic and effortless processes grounded in on-line cooperative social interactions. Our approach is part of an on-going paradigm change in the field of theory-of-mind research motivated by the recent evidence showing that infants in their second year understand mental states (Onishi & Baillargeon, 2005). Our own research has gone a significant step further by demonstrating that these mechanisms are present as early as 7 month of age, and by showing that both young infants and adults seem to automatically compute others’ beliefs even in situations where they are not required to do so (Kovács et al., 2010). The present project explores the functional sub-components and triggering conditions of young infants’ powerful belief computation abilities and to chart their developmental unfolding. Furthermore, we shall explore the implications of the new theoretical proposal that this dedicated system presupposes as its proper domain the on-going collaborative and communicative interactions.