Modeling Social Causality and Social Judgment in Multi-Agent Interactions

 

Modeling Social Causality and Social Judgment in Multi-Agent Interactions

Sage 4101

Intelligent agents are typically situated in a multi-agent environment and must reason about social cause and effect. Social causal reasoning is qualitatively different from physical causal reasoning that underlies most current intelligent systems. Besides physical cause, the assessment of social causality emphasizes epistemic variables such as intentions, foreknowledge and perceived coercion. Modeling the cognitive process and inferences of social causality can enrich the capabilities of intelligent agents, and advance our understanding of human social intelligence. In this talk, I will present a general computational model for social causality and social judgment in the context of multi-agent interaction. Based on attribution theory in social psychology, the model formalizes the commonsense reasoning about the beliefs of attribution variables from natural language communication and task execution. In addition, we have designed and conducted experiments to validate the model. The empirical results show the consistency between the model predictions and actual human performance for the individual variables, the overall judgments as well as the inferential mechanism.
Add to calendar
Share|