Seeing the Reasoning: How LLM Rationales Influence User Trust and Decision-Making in Factual Verification Tasks

2026-03-07

AI Summary

This study investigates how reasoning rationales generated by large language models influence user trust and decision-making in fact-checking tasks. Through an online experiment (N=68), the research systematically manipulates the correctness, certainty expressions, and presentation formats of model-generated rationales, employing a human-AI interaction design combined with psychological measurement to conduct causal analysis. The findings reveal, for the first time, that users primarily leverage rationales to audit model outputs and calibrate their trust. Correct reasoning and expressions of certainty significantly enhance user trust and decision confidence, whereas uncertainty cues diminish them. Presentation format shows no significant effect. These results underscore the critical role of rationale interface design in trust calibration and provide empirical support for explainable AI.

Technology Category

• Knowledge Representation and Reasoning: Reasoning with Beliefs
• Reasoning under Uncertainty: Causality
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Large Language Models (LLMs) increasingly show reasoning rationales alongside their answers, turning"reasoning"into a user-interface element. While step-by-step rationales are typically associated with model performance, how they influence users'trust and decision-making in factual verification tasks remains unclear. We ran an online study (N=68) manipulating three properties of LLM reasoning rationales: presentation format (instant vs. delayed vs. on-demand), correctness (correct vs. incorrect), and certainty framing (none vs. certain vs. uncertain). We found that correct rationales and certainty cues increased trust, decision confidence, and AI advice adoption, whereas uncertainty cues reduced them. Presentation format did not have a significant effect, suggesting users were less sensitive to how reasoning was revealed than to its reliability. Participants indicated they use rationales to primarily audit outputs and calibrate trust, where they expected rationales in stepwise, adaptive forms with certainty indicators. Our work shows that user-facing rationales, if poorly designed, can both support decision-making yet miscalibrate trust.

Problem

• LLM rationales
• user trust
• decision-making
• factual verification
• reasoning

Innovation

• LLM rationales
• user trust
• factual verification
• certainty framing
• human-AI interaction