University of Illinois at Urbana-Champaign
Using chatbots to correct COVID-19 misinformation: a controlled experiment
Abstract
dc:descriptionConversational agents, otherwise known as chatbots, have been increasingly used by organizations worldwide to provide health-related information and correct misinformation in the COVID-19 pandemic. However, previous studies have neither empirically investigated the effectiveness of correcting misinformation using chatbots, nor have they provided theory-based guidelines for the designing of such chatbots. To address these concerns, we developed our misinformation-correcting chatbots and conducted two experiments. In the pre-test experiment (n = 42), we examined whether corrective information provided through chatbot interactions can correct COVID-related misperception and behavioral intention compared to not receiving such information. In the official study (n = 791), we investigate whether and how chatbots with interactive conversational cues and/or empathetic conversational cues can outperform the basic chatbot in correcting users’ COVID-19-vaccination misperception and enhancing users’ vaccination intention. Our results indicate that chatbots can be effective in fighting against misinformation and interactive and empathetic conversational cues do affect users’ psychological perception towards the chatbot and in turn leads to better interaction outcomes. We also identified different mechanisms of people’s misperception change and behavioral intention change. Theoretical and practical implications are discussed.
Degree
thesis:*- Name thesis:degree_name
- M.S.
- Level thesis:degree_level
- Thesis
- Discipline thesis:degree_discipline
- Advertising
- Grantor
- University of Illinois at Urbana-Champaign
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Gong, Ziyang
- Contributors dc:contributor
-
- Su, Leona Yi-Fan
- Wise, Kevin
- Sar, Sela
Subjects
dc:subject × 7Rights
dc:rights- Statement dc:rights
-
- Copyright 2022 Ziyang Gong
- Language dc:language
- en, eng
Identifiers
dc:identifier.*- Handle dc:identifier
- https://hdl.handle.net/2142/115956