In social interactions, inferring the interaction partner's hidden mental state is crucial for predicting their actions and optimizing our responses. Effective models for this inference must account for how these mental states evolve due to the interaction history and environmental changes. For example, recognizing someone's emotional state can help forecast their behavior. Our study investigates how making these latent states visible influences decision-making in social interactions. Using the repeated trust game paradigm, we show how to use hidden Markov models (HMM) to formally represent latent state dependent strategies of the players. HMMs fitted to human dyadic play in the trust game are then used to specify adaptive AI agents that simulate changes in mental dispositions of human players, such as the level of trust in the opponent, during a repeated interaction. Making these artificial HMM based agents take the role of the investor and interact with real human trustees, we then explore how displaying “emotion” cues to the opponent's latent state affects people's actions. We find that the presence of cues was associated with more cooperative behavior from the human trustees, and that patterns of behavior that promote the maintenance of cooperation emerged in the presence of latent state cues and were transferred to settings where the cues were subsequently hidden.

Authors:

Ismail Guennouni: University College London; Maarten Speekenbrink: University College London; Kaitlyn Ng: University College London