Skip to main content
Peer-reviewed Authoritative — Peer-reviewed venues, standards bodies, regulators, official government publications

Authority Over Sympathy: How AI Emotional Tone and Thinking Dispositions Shape Hallucination Detection in Health Communication

Wizard-of-Oz experiment (N=476) testing how the emotional tone of AI-generated health responses that contain embedded hallucinations affects whether people detect the false content, framed by the Misinformation Recognition and Response Model. Authoritative and sympathetic tones were manipulated independently. An authoritative tone significantly reduced participants' hallucination-detection sensitivity; a sympathetic tone had no significant effect; thinking dispositions did not moderate the tone effect.

Read at doi.org (opens the original source in a new tab)

Publisher

Journal of Broadcasting & Electronic Media (Taylor & Francis for the Broadcast Education Association)

Published

21 Sept 2026

Added

today

Key Findings

  • Wizard-of-Oz design with N=476 participants reading AI-generated health responses containing embedded hallucinations
  • Authoritative tone significantly reduced hallucination-detection sensitivity
  • Sympathetic tone showed no significant effect on detection
  • Thinking dispositions did not significantly moderate the effect of tone on detection sensitivity

Methodology Notes

Wizard-of-Oz experiment with independently manipulated authoritative and sympathetic tones; the abstract does not report effect sizes, participant demographics or recruitment. Authors at South China University of Technology. Verification route: the Taylor & Francis page is walled; record confirmed via Crossref (published online 2026-09-21) and the abstract via OpenAlex, which appeared between 2026-09-22 and 2026-09-24.

Topics

cognitive epistemic safetyhigh stakes guidancehealth safetyModel behavior

Authors

Lu Liu, Bing Hu

Tags

wizard-of-oztonehallucinationhealth-communicationmrrmauthority

Cite This

APA

Lu Liu, Bing Hu. (2026). Authority Over Sympathy: How AI Emotional Tone and Thinking Dispositions Shape Hallucination Detection in Health Communication. Journal of Broadcasting & Electronic Media (Taylor & Francis for the Broadcast Education Association). https://doi.org/10.1080/08838151.2026.2735592

Related Insights

Peer-reviewed

Fallibility, persuadability, and correctability of large language models under sustained conversational misinformation pressure

Scientific Reports (Nature Portfolio); University of Arizona · 1 Sept 2026

Preprint

When Rubrics Fail: Hallucinations Reveal Blind Spots in Medical AI Evaluation

arXiv (University of Oxford) · 11 Sept 2026