How do people understand social interaction and adapt behavior to the requirements of each unique interaction? This project tests the idea that the processes that support learning (predictive cues and rewards) also govern social behavior. It is proposed that people use social cues to make predictions about others' behavior, and that some social cues are rewarding and therefore bias people's behavior. To test these ideas, a set of social cue stimuli will be developed and used in computerized learning games. In one game, participants use social cues to guess which of 2 pictures a computerized person likes, earning money for correct choices. If social cues are predictive, participants should be able to use them to improve performance. A second game tests whether social cues are rewarding. Participants choose one of 2 faces positioned side-by-side on a computer screen. After each choice, the chosen face "responds" with either a pleasurable or a polite smile. If participants enjoy pleasurable smiles, they should develop response strategies that increase the frequency of these social rewards. Study results will advance understanding about how humans navigate the social world and allow the extension of these ideas to real social environments and to people with social difficulties.

Data were collected from individual participants in both computerized tasks and questionnaires. There are 3 datasets. Dataset 1 was collected to validate a social cue stimulus set. Dataset 2 involved a study of how people choose cues to predict behavior. Dataset 3 was from a study in which we examined how the reinforcement value of a smile biases behavior.