What a Multimodal Information Density Shapes Audience Engagement: Evidence from Sports Event Mascot Short-Form Videos
Grounded in cognitive load theory, this study analyzes 173 sports mascot short videos from Bilibili to reveal that visual, textual, and audio information densities exhibit inverted U-shaped relationships with audience liking, where textual density is most influential overall but its impact is moderated by video source type, with official content favoring text and user-generated content favoring audio and visual elements.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In the crowded landscape of modern digital life, short videos have become the primary way people consume stories, news, and entertainment. These brief clips, often lasting only seconds or a few minutes, pack a surprising amount of sensory input into a tiny window of time. They rely on a mix of moving images, written words, and sound to capture attention. However, human brains have a limited capacity for processing information at any given moment. If a video presents too much at once, the viewer can become overwhelmed, a state researchers call cognitive overload, where the mind struggles to make sense of the flood of data. This creates a delicate balancing act for content creators: they must provide enough detail to be interesting and clear, but not so much that the audience tunes out. Understanding where that line lies is crucial for anyone trying to communicate effectively in a fast-paced digital world.
A team of researchers from Chengdu University of Technology set out to find the sweet spot for this balance, specifically looking at how sports event mascots are presented in short videos. Mascots, once static symbols found on posters and merchandise, have evolved into dynamic characters that star in their own digital content. The researchers wanted to know how the sheer volume of visual, textual, and audio information in these videos affects whether people click the "like" button. They focused on 173 videos featuring official sports mascots, collected from Bilibili, a popular video platform in China known for its young, active user base. The study examined three types of information density: how many visual changes happen per second, how much text appears in titles, subtitles, and scrolling comments, and how intense the audio is in terms of rhythm and volume changes.
The researchers discovered that all three types of information play a role in making a video popular, but they do not work in a simple, straight line. Initially, adding more visual details, more text, or more dynamic sound makes a video more engaging. Viewers appreciate the richness and the extra context these elements provide. However, this positive effect has a limit. The study found that once the information density crosses a certain threshold, the relationship flips. Adding even more information after this point actually causes the number of likes to drop. This creates an inverted U-shaped curve, suggesting that there is an optimal amount of information for each type of media. Too little leaves the viewer bored or confused, but too much causes mental fatigue, making the video less enjoyable.
Interestingly, the type of information matters more than the others. Textual information, which includes titles, descriptions, and the scrolling comments known as bullet comments, had the strongest influence on whether a video was liked. This suggests that for mascot content, explaining the story or cultural meaning behind the character is the most powerful tool for engagement. Visual information, while important, showed the sharpest decline in popularity when it became too dense. This indicates that while people enjoy watching the mascot move and change, their brains can only process so many visual shifts before it becomes distracting. Audio density followed a similar pattern, where a moderate amount of rhythmic sound and volume variation helped, but excessive noise or complexity hurt the viewer's experience.
The study also revealed that the source of the video changes how these rules apply. When a video is created by an official organization, such as a sports committee, viewers respond best when there is a high density of text. In these official contexts, the audience is looking for accurate information and cultural context, so detailed written explanations are highly valued. Conversely, when a video is created by a regular user, the audience responds more strongly to visual and audio cues. User-generated content is often watched for entertainment and emotional connection, so the energy of the music and the movement of the image matter more than the written facts. This distinction means that a one-size-fits-all approach to creating mascot videos does not work; the strategy must change depending on who is making the video and what the audience expects from them.
Ultimately, the research provides a clear guide for content creators. To maximize engagement, creators should aim for a moderate level of information density, avoiding the temptation to pack every second with maximum intensity. For official accounts, the priority should be on clear, informative text that helps viewers understand the mascot's significance. For independent creators, the focus should shift to the sensory experience, using dynamic visuals and rhythmic audio to create an emotional connection. By respecting the limits of human attention and tailoring the content to the source, creators can ensure their mascot videos remain engaging rather than overwhelming. The findings confirm that in the world of short-form video, quality and timing matter far more than simply piling on more content.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.