Towards an Understanding of the Impact of Uncertainty and Emotional Content on Users' Decision-making About Misinformation

Search for this publication on Google Scholar
Karduni, A. (2020). Towards an Understanding of the Impact of Uncertainty and Emotional Content on Users' Decision-making About Misinformation. Unc Charlotte Electronic Theses And Dissertations.
Download PDF

Analytics

30 views ◎
7 downloads ⇓


Abstract

Misinformation on social media is a phenomenon with considerable impacts on our societies. To effectively mitigate its effects, we need to employ a cross-disciplinary effort to holistically address the critical elements involved in the process, including the source, the content, and the consumers of misinformation. However, the vast majority of computational approaches to addressing this problem focus on detecting and flagging misinformation content. Other important aspects, such as behaviors and intents of sources and the consumers’ decision-making processes, are often neglected. In this thesis, we address this gap through a series of studies around how users make decisions about content and sources of misinformation facilitated by Visual Analytics. First, We introduce a Visual Analytic system, Verifi, that combines temporal, language, and network analysis features and enables users to assess the veracity of multiple news sources. Using Verifi, we conducted a controlled experiment that highlighted how uncertainty and conflicting cues in information impacts users’ perceptions of source credibility. Next, we extended Verifi to a more comprehensive multi-modal system enabling users to study sources through social network analysis, text, and images. Through a qualitative domain expert study conducted on Verifi, we learned valuable lessons about the importance of users’ trust in sources and how emotional content in images might impact users’ judgments and perceptions. Inspired by the lessons learned from our work on Verifi, we present two studies on the effects of emotions in images on users’ perception of content bias and source credibility. In the first study, we investigated the impact of happy and angry portrayals in facial imagery on users’ decisions. In the second study, we explore the interaction between users’ prior attitude towards multiple personalities and how those personalities are portrayed visually on their decisions. Our results show that the systematic usage of angry facial emotions in images increases users’ perceived content bias and decreases the perceived source credibility. These results highlight how implicit visual propositions by news sources impact our judgements and pave the way for visual analytic systems that are sensitive to users' individual and group interactions with such visual information.

Details
Author

Karduni, Alireza

Title

Towards an Understanding of the Impact of Uncertainty and Emotional Content on Users' Decision-making About Misinformation

Physical Description

1 online resource (176 pages) : PDF

Date

2020

Degree Granting Institution

University of North Carolina at Charlotte

Abstract

Misinformation on social media is a phenomenon with considerable impacts on our societies. To effectively mitigate its effects, we need to employ a cross-disciplinary effort to holistically address the critical elements involved in the process, including the source, the content, and the consumers of misinformation. However, the vast majority of computational approaches to addressing this problem focus on detecting and flagging misinformation content. Other important aspects, such as behaviors and intents of sources and the consumers’ decision-making processes, are often neglected. In this thesis, we address this gap through a series of studies around how users make decisions about content and sources of misinformation facilitated by Visual Analytics. First, We introduce a Visual Analytic system, Verifi, that combines temporal, language, and network analysis features and enables users to assess the veracity of multiple news sources. Using Verifi, we conducted a controlled experiment that highlighted how uncertainty and conflicting cues in information impacts users’ perceptions of source credibility. Next, we extended Verifi to a more comprehensive multi-modal system enabling users to study sources through social network analysis, text, and images. Through a qualitative domain expert study conducted on Verifi, we learned valuable lessons about the importance of users’ trust in sources and how emotional content in images might impact users’ judgments and perceptions. Inspired by the lessons learned from our work on Verifi, we present two studies on the effects of emotions in images on users’ perception of content bias and source credibility. In the first study, we investigated the impact of happy and angry portrayals in facial imagery on users’ decisions. In the second study, we explore the interaction between users’ prior attitude towards multiple personalities and how those personalities are portrayed visually on their decisions. Our results show that the systematic usage of angry facial emotions in images increases users’ perceived content bias and decreases the perceived source credibility. These results highlight how implicit visual propositions by news sources impact our judgements and pave the way for visual analytic systems that are sensitive to users' individual and group interactions with such visual information.

Genre

doctoral dissertations

Subjects--Topics

Computer science
Communication
Cognitive psychology

Degree

Ph.D.

Keywords

Emotion
Misinformation
Social Media
Uncertainty
Visualization

Subject Area

Computer Science

Advisor(s)

Dou, Wenwen

Committee Members

Markant, Dougla
Levens, Sara
Shaikh, Samira
Cho, Isaac
Sauda, Eric

Degree Note

Thesis (Ph.D.)--University of North Carolina at Charlotte, 2020.

Rights Statement

This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s). For additional information, see http://rightsstatements.org/page/InC/1.0/.

Rights Holder Information

Copyright is held by the author unless otherwise indicated.

Identifier

Karduni_uncc_0694D_12680

Permalink

http://hdl.handle.net/20.500.13093/etd:2295