Effective Explanations Support Planning Under Uncertainty
For researchers in human-robot interaction and cognitive science, this work provides a utility-guided framework for evaluating and generating explanations that support planning under uncertainty.
The paper proposes a computational model that converts explanations into action plans using a large language model and a planning agent, and shows that higher-scored explanations improve navigation efficiency and are judged more helpful in experiments with 1,200 explanations over 24 maps.
Explaining how to get from A to B can be challenging. It requires mentally simulating what the listener will do based on what they are told. To capture this process, we propose a computational model that converts utterances into action plans: a large language model translates an explanation into program-like guidance (a policy prior and value map), and a planning agent executes it under partial observability. We score explanations by the efficiency and reliability of the resulting paths, penalizing replanning. Across four preregistered experiments, we collect a corpus of 1,200 explanations over 24 maps, elicit helpfulness judgments, measure baseline navigation, and test behavior with explanations of differing quality. Higher-scored explanations are judged more helpful and improve navigation: participants with explanations outperform those without, and high-scoring explanations help more than low-scoring ones. Together, these results show procedural explanation as utility-guided communication shaped by how language can be grounded into action under uncertainty.