The Promise and Limits of Conversational Agents for Misinformation Correction: A Systematic Review of Experimental Evidence
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Conversational agents have been proposed as scalable tools for correcting misinformation, yet empirical findings remain fragmented. This systematic review synthesizes evidence from 15 randomized experimental studies examining conversational-agent interventions across diverse contexts. The review maps key design features and evaluates their effects on agent perceptions and misinformation-related outcomes. Overall, conversational agents reliably outperform no-intervention controls but show mixed advantages over non-conversational corrections. Effects are context- and user-dependent, with moderation by demographics or dispositions. Findings highlight both the promise and boundaries of conversational agents for misinformation correction.
