Theory of Mind in HRI

A key factor in human social interaction is our beliefs about others, a theory of mind. Our decisions to act are influenced by how we believe others will react. Whether we believe a message depends not only on its content but also on our model of the communicator. Giving its importance in human social interaction, modeling theory of mind can play a key role in enriching social simulations.

Typical approaches to modeling theory of mind in a computational framework have relied on first-order logic to represent beliefs and goals. However, such representations are often insensitive to the distinctions among conflicting goals that people must balance in a social interaction. For example, psychological research has identified a range of goals that motivate classroom bullies (e.g., peer approval, sadism, tangible rewards). All bullies share the same goals, but it is the relative priorities that they place on them that leads to the variations in their behavior. Resolving the ambiguity among equally possible, but unequally plausible or preferred, options requires a quantitative model of uncertainty and preference. Unfortunately, more quantitative frameworks, like decision theory and game theory, face their own difficulties in modeling human psychology. Game theoretic frameworks typically rely on concepts of equilibria that people rarely achieve in an unstructured social setting like a classroom. Decision theoretical frameworks typically rely on assumptions of rationality that people constantly violate.

We have developed a social simulation tool, PsychSim, that operationalizes existing psychological theories as boundedly rational computations to generate more plausibly human behavior. PsychSim allows a user to quickly construct a social scenario where a diverse set of entities, groups or individuals, interact and communicate. Each entity has its own preferences, relationships (e.g., friendship, hostility, authority) with other entities, private beliefs, and mental models about other entities. The simulation tool generates the behavior for these entities and provides explanations of the result in terms of each entity's preferences and beliefs. The richness of the entity models allows one to explore the potential consequences of minor variations on the scenario. A user can play different roles by specifying actions or messages for any entity to perform.

A central aspect of the PsychSim design is that agents have fully specified decision-theoretic models of others. Such quantitative recursive models give PsychSim a powerful mechanism to model a range of factors in a principled way. For instance, we exploit this recursive modeling to allow agents to form complex attributions about others, enrich the messages between agents to include the beliefs and preferences of other agents, model the impact such recursive models have on an agent's own behavior, model the influence observations of another's behavior have on the agent's model of that other, and enrich the explanations provided to the user. The decision-theoretic models in particular give our agents the ability to judge degree of credibility of messages in a subjective fashion that can consider the range of influences that sway such judgments in humans.