Simulative belief logic

Abstract

The ability of ascribing beliefs to others is crucial for human beings to explain and understand each other. Belief ascription has been studied intensively in philosophy and cognitive science. In this paper, we propose a formal framework for belief ascription by simulation. An agent first acquires information about another agentʼs beliefs by communication. She then inputs the information into her own belief system to generate more beliefs, which she will ascribe to the other agent. In this way, the agent uses her own as a model of others. We present a modal belief logic, which contains private announcement operators for agentsʼ communication, and simulative belief operators for beliefs ascribed to others. We give a complete axiomatic system for the logic.