Published September 15, 2026 | Version 1.0.0

Study app, data, analysis code and materials for: What's the Difference? Voice Design and Confirmation Bias in AI-Assisted Self-Diagnosis: Anthropomorphic vs. Machine-Like Voice Output

  • 1. Technische Hochschule Ingolstadt

Description

Study application, pseudonymised data, analysis code and materials for a within-subjects online experiment (N = 81, 648 trials) on whether the voice design of an AI symptom checker moderates confirmation bias in lay self-diagnosis. Participants heard an AI recommendation that confirmed or refuted their initial diagnosis, spoken by an anthropomorphic or a machine-like voice, and then decided again. The direction of the recommendation affected decisions, and the voice did not moderate this effect.

Notes

If you use this software, data or materials, please cite them using these metadata.

Files

niroet/stud-voice-bias-initial_public_release.zip

Files (216.6 kB)

Name Size
md5:072cf241d41c95c20eb18080387cd716
216.6 kB Preview Download

Additional details