The distracted robot: what happens when artificial agents behave like us

In everyday life, we are frequently exposed to different smart technologies. From our smartphones to avatars in computer games, and soon perhaps humanoid robots, we are surrounded by artificial agents created to interact with us. Already during the design phase of an artificial agent, engineers often endow it with functions aimed to promote the interaction and engagement with it, ranging from its “communicative” abilities to the movements it produces. Still, whether an artificial agent that can behave like a human could boost the spontaneity and naturalness of interaction is still an open question. Even during the interaction with conspecifics, humans rely partially on motion cues when they need to infer the mental states underpinning behavior. Similar processes may be activated during the interaction with embodied artificial agents, such as humanoid robots. At the same time, a humanoid robot that can faithfully reproduce human-like behavior may undermine the interaction, causing a shift in attribution: from being endearing to being uncanny. Furthermore, it is still not clear whether individual biases and prior knowledge related to artificial agents can override perceptual evidence of human-like traits. A relatively new area of research emerged in the context of investigating individuals’ reactions towards robots, widely referred to as Human-Robot Interaction (HRI). HRI is a multidisciplinary community that comprises psychologists, neuroscientists, philosophers as well as roboticists, and engineers. However, HRI research has been often based on explicit measures (i.e. self-report questionnaires, a-posteriori interviews), while more implicit social cognitive processes that are elicited during the interaction with artificial agents took second place behind more qualitative and anecdotal results. The present work aims to demonstrate the usefulness of combining the systematic approach of cognitive neuroscience with HRI paradigms to further investigate social cognition processes evoked by artificial agents. Thus, this thesis aimed at exploring human sensitivity to anthropomorphic characteristics of a humanoid robot's (i.e. iCub robot) behavior, based on motion cues, under different conditions of prior knowledge. To meet this aim, we manipulated the human-likeness of the behaviors displayed by the robot and the explicitness of instructions provided to the participants, in both screen-based and real-time interaction scenarios. Furthermore, we explored some of the individual differences that affect general attitudes towards robots, and the attribution of human-likeness consequently.