ºìÐÓÊÓƵ

A study led by Dr Fangfang Hou from the International Business School Suzhou (IBSS), Xi¡¯an Jiaotong-ºìÐÓÊÓƵ University, has been published in the leading international journal Decision Support Systems. The study, titled ¡°Unraveling the Dark Side of Danmaku: The Effects of Aggressive Danmaku and Original Work¡¯s Attributes on Condensed Clips¡¯ Performance¡±, examines the darker side of interactive viewing culture, exploring how toxic danmaku content and intrinsic attributes of original audio-visual works jointly shape the performance of condensed video clips.

Condensed video clips have emerged as a dominant user-generated content format in the digital media landscape, enabling audiences to efficiently access highlights from films and television programmes. As a core interactive feature, danmaku ¡ª real-time, ?on-screen scrolling comments ¡ª facilitates synchronous (shared) viewing experiences, strengthens social connection among audiences, and enhances engagement. However, the anonymity of online interaction reduces social constraints, contributing to the widespread presence of ?aggressive danmaku, including ?insults, vulgar language, ?and hostile remarks. While existing literature has explored the positive social functions of danmaku , the negative effects of toxic danmaku content remains underexamined. Drawing on large-scale behavioural data, this study investigates how aggressive danmaku undermines the clip performance, alongside the ?moderating influence of ?original copyrighted works.

The findings provide strong evidence that aggressive danmaku has significant negative effects on key performance metrics, including view counts, reposting activity, and ?viewer retention. Two key dimensions of toxic comments are identified. First, a higher proportion of aggressive danmaku significantly reduces clip performance; as hostile remarks trigger negative viewer sentiment, discourage continued viewing, and limit voluntary sharing behaviour. Secondly, the temporal concentration of aggressive comments amplifies these effects. When such comments cluster within short time periods, they intensify negative emotional contagion, damage the viewing experience, and accelerate audience drop-off (attrition).

The research further highlights the moderating role of original content attributes. For clips adapted from high-reputation productions, both frequent and scattered aggressive danmaku lead to more pronounced negative effects to clip performance. Conversely, strong brand reputation can ?partially mitigate these effects when aggressive comments are highly concentrated. For nostalgic or classic content, the impact is more consistently negative. Audiences with strong emotional and cultural attachment to such works are more sensitive to toxic comments, with both volume and concentration of aggressive danmaku leading to sharp declines in clip performance.

Additional ?analysis identifies a threshold effect in the relationship between aggressive danmaku and clip performance. A low level ?of negative comments may be tolerated, but once a critical threshold is exceeded, performance declines sharply in a non-linear manner. The study also categorises diverse forms of aggressive danmaku, including personal attacks, sarcasm and crude humour, and shows these vary in their level of harmful impact.

From a practical perspective, the findings offer clear implications platforms, regulators and content creators. Targeted governance strategies ¡ª such as automated risk detection, real-time content moderation and customised community management guidelines ¡ª ?are essential to limit the spread of harmful content. Platforms are advised to adopt differentiated supervision mechanisms, with stricter controls for high-reputation classic content. Strengthening the danmaku ecosystem in this way can protect user experience, preserve the value of original content, and support the sustainable development of short-form video platforms.

Dr Fangfang Hou is an Assistant Professor in the Department of Intelligent Operations and Marketing, International Business School Suzhou, Xi¡¯an Jiaotong-ºìÐÓÊÓƵ University. Her research interests include social media, digital innovation, and FinTech. She has published articles in journals including Information Systems Journal (ISJ), Production Planning & Control (PPC), Internet Research (IntR), Information & Management (I&M), Information Technology & People (ITP), and Industrial Management & Data Systems (IMDS). She serves on the editorial boards of several journals, including for IMDS and IntR, and contributes actively to the Association for Information Systems (AIS), serving as a track chair, associate editor, session chair, and discussant.

Decision Support Systems is a leading peer-reviewed international journal in the fields of information management, business analytics and digital economy. Indexed in JCR Q1, the journal features rigorous double-blind peer review standards and high academic recognition worldwide. The journal publishes high-quality theoretical and empirical research covering business intelligence, social media analytics, human¨Ccomputer interaction, digital platform governance and data-driven decision-making. As a core academic platform for advancing interdisciplinary research bridging information systems and business management, Decision Support Systems maintains substantial influence across academia and industrial communities globally.