Humans are an intensely social species, and this fact is evident even at the earliest points in development. Infants are highly attentive to their social partners and cognitively engaged with them. In particular, research over the last decade has revealed that young infants view others' behavior not as purely physical movements but rather as actions structured by the person's goals and states of attention. This sensitivity to the goal structure of action is fundamental to human experience and critical for social and cognitive development. However, current findings do not tell us whether and how infants recruit this sensitivity to others' goals in order to predict their actions in real time. Anticipating others' goal-directed actions is essential for many forms of social coordination, including one-on-one social interactions, collaborative activities and communication. This project will address this gap in the literature by investigating infants' ability to visually predict a person's next action based on information about his or her prior goals and focus of attention. By using a remote eye-tracking system to record infants' gaze shifts as they view actions on a video monitor, the studies will investigate the range of situations in which infants of different ages can generate action predictions. In addition, the studies will explore two potential contributors to the development of action prediction -- infants' own experience producing well-structured actions and infants' sensitivity to linguistic information about a person's goals. In this way, the project will both document the emergence of action prediction during infancy and provide insights into the developmental processes that contribute to this ability in infants. This project will provide foundational insights into a critical and largely unstudied aspect of social competence in infants. More broadly, this work will inform our understanding of social, cognitive and linguistic development because it will shed light on the social-cognitive processes that undergird young children's social learning in these domains. The experiments will also develop new research methods for assessing social information processing in infants. Because the eye-tracking methods developed in these studies can also be adapted for use with broader populations, this research will lay the foundation for comparative work across ages, populations, and possibly species. These eye-tracking methods may be particularly useful in participants who have limited abilities to respond to verbal instructions, such as individuals with autism. By elucidating the early development of an important aspect of social cognition, this work may also yield important insights into understanding clinical disorders that involve deficits in social reasoning, social attention, and the on-line interpretation of others' actions (for example, autism and conduct disorder).