Effect of Transparency and Trust on Acceptance of Automatic Online Comment Moderation Systems

Brunk, Jens; Mattern, Jana; Riehle, Dennis M.

doi:10.1109/cbi.2019.00056

Cited by 21 publications

(8 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, this supplementary information may implicitly increase the perception of transparency in the model and positively affect user confidence. This interpretation aligns with (Brunk et al, 2019), who showed that the interpretability of black-box algorithms is perceived to be more transparent and trustworthy if additional information is present. However, the adjacency comparisons (Figure 5) indicate no significant difference between levels of MC when explanation data are presented adjacently.…”

Section: The Conditional Effect Of Morphological Clarity On User Conf...supporting

confidence: 88%

Designing for Confidence: The Impact of Visualizing Artificial Intelligence Decisions

et al. 2022

View full text Add to dashboard Cite

Explainable artificial intelligence aims to bring transparency to artificial intelligence (AI) systems by translating, simplifying, and visualizing its decisions. While society remains skeptical about AI systems, studies show that transparent and explainable AI systems can help improve the Human-AI trust relationship. This manuscript presents two studies that assess three AI decision visualization attribution models that manipulate morphological clarity (MC) and two information presentation-order methods to determine each visualization’s impact on the Human-AI trust relationship through increased confidence and cognitive fit (CF). The first study, N = 206 (Avg. age = 37.87 ± 10.51, Male = 123), utilized information presentation methods and visualizations delivered through an online experiment to explore trust in AI by asking participants to complete a visual decision-making task. The second study, N = 19 (24.9 ± 8.3 years old, Male = 10), utilized eye-tracking technology and the same stimuli presentation methods to investigate if cognitive load, inferred through pupillometry measures, mediated the confidence-trust relationship. The results indicate that low MC positively impacts Human-AI trust and that the presentation order of information within an interface in terms of adjacency further influences user trust in AI. We conclude that while adjacency and MC significantly affect cognitive load, cognitive load alone does not mediate the confidence-trust relationship. Our findings interpreted through a combination of CF, situation awareness, and ecological interface design have implications for the design of future AI systems, which may facilitate better collaboration between humans and AI-based decision agents.

show abstract

Section: The Conditional Effect Of Morphological Clarity On User Conf...supporting

confidence: 88%

Designing for Confidence: The Impact of Visualizing Artificial Intelligence Decisions

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Additionally, trust moderated the relationship between FAT and user satisfaction in their study. Other research found that transparency of an automatic moderation system is a prerequisite for trust in that system (Brunk et al, 2019). Yet, an important question is whether trust and perceived fairness of the moderation decision vary by type of moderation source (AI vs. user-based) and its visibility.…”

Section: Sources Of Moderationmentioning

confidence: 99%

Shall AI moderators be made visible? Perception of accountability and trust in moderation systems on social media platforms

et al. 2022

View full text Add to dashboard Cite

This study examines how visibility of a content moderator and ambiguity of moderated content influence perception of the moderation system in a social media environment. In the course of a two-day pre-registered experiment conducted in a realistic social media simulation, participants encountered moderated comments that were either unequivocally harsh or ambiguously worded, and the source of moderation was either unidentified, or attributed to other users or an automated system (AI). The results show that when comments were moderated by an AI versus other users, users perceived less accountability in the moderation system and had less trust in the moderation decision, especially for ambiguously worded harassments, as opposed to clear harassment cases. However, no differences emerged in the perceived moderation fairness, objectivity, and participants confidence in their understanding of the moderation process. Overall, our study demonstrates that users tend to question the moderation decision and system more when an AI moderator is visible, which highlights the complexity of effectively managing the visibility of automatic content moderation in the social media environment.

show abstract

“…Commenters openly rebuked TikTok for sharing educational materials addressing fake news while failing to curb the spread of misinformation, which resulted in a diminished level of trust. Providing explicit reasoning around content moderation practices promotes a higher level of user trust (Brunk, et al, 2019), which might help with the effectiveness of such educational initiatives. Platforms should aim for increased transparency in communicating how decisions are made regarding user reports of misinformation or content violations.…”

Section: Implications and Recommendationsmentioning

confidence: 99%

Likes, Sarcasm and Politics: Youth Responses to a Platform-Initiated Media Literacy Campaign on Social Media

Literat

Abdelbagi

Law

et al. 2021

SPIR

View full text Add to dashboard Cite

When it comes to youth and media literacy, the focus of both research and practice has been on the integration of media literacy instruction into young people’s various educational experiences, rather than studying, implementing and evaluating such initiatives in situ, on social media—which is known to be youths’ central news source. Furthermore, existing studies focus on older platforms like Facebook or Twitter, and we lack a nuanced understanding of these dynamics on the platforms most popular with youth today. The present study aims to fill this gap, by analyzing how users responded to a media literacy education campaign on the popular youth-oriented social media app TikTok, with a focus on the lessons that we might draw in terms of targeted media literacy education initiatives for youth. Our large-scale qualitative content analysis—which included 11,449 public comments posted on 5 educational TikTok videos about media literacy—facilitated a naturalistic study of young people’s attitudes towards such initiatives. We found that reactions to the campaign were mixed, and highly political in nature. While many users appreciated the media literacy campaign and TikTok’s role in implementing it, there was also resistance to the campaign, due to factors related to video content, dissemination approach, and the very identity of TikTok as a platform. Across the board, comments were highly political, illustrating the politicized nature of media literacy education today. These reactions serve as rich feedback that can usefully inform future media literacy campaigns on social media, and especially those targeting youth audiences.

show abstract

Effect of Transparency and Trust on Acceptance of Automatic Online Comment Moderation Systems

Cited by 21 publications

References 29 publications

Designing for Confidence: The Impact of Visualizing Artificial Intelligence Decisions

Designing for Confidence: The Impact of Visualizing Artificial Intelligence Decisions

Shall AI moderators be made visible? Perception of accountability and trust in moderation systems on social media platforms

Likes, Sarcasm and Politics: Youth Responses to a Platform-Initiated Media Literacy Campaign on Social Media

Contact Info

Product

Resources

About