Human-Aware Belief Revision: A Cognitively Inspired Framework for Explanation-Guided Revision of Human Models
Stylianos Loukas Vasileiou, William Yeoh
TL;DR
The paper argues that minimal changes are not sufficient to model real human belief revision. It proposes a cognitively inspired framework, the explanation-guided belief revision operator, which uses explanations to drive non-minimal revisions that preserve explanatory understanding. The operator is formalized via a correction kernel and a $\varphi$-preserving selection function, and its properties are axiomatized through postulates, with empirical validation from two human-subject studies showing a robust tendency toward non-minimal, explanation-driven revisions. These findings have practical implications for human-aware AI and model-reconciliation tasks by aligning AI reasoning with human cognitive patterns and explanations.
Abstract
Traditional belief revision frameworks often rely on the principle of minimalism, which advocates minimal changes to existing beliefs. However, research in human cognition suggests that people are inherently driven to seek explanations for inconsistencies, thereby striving for explanatory understanding rather than minimal changes when revising beliefs. Traditional frameworks often fail to account for these cognitive patterns, relying instead on formal principles that may not reflect actual human reasoning. To address this gap, we introduce Human-Aware Belief Revision, a cognitively-inspired framework for modeling human belief revision dynamics, where given a human model and an explanation for an explanandum, revises the model in a non-minimal way that aligns with human cognition. Finally, we conduct two human-subject studies to empirically evaluate our framework under real-world scenarios. Our findings support our hypotheses and provide insights into the strategies people employ when resolving inconsistencies, offering some guidance for developing more effective human-aware AI systems.