Algorithmic Curation Shapes Cultural Heritage Visibility on Digital Platforms
This study analyzes over 107,000 heritage videos across TikTok, YouTube, and Xiaohongshu to demonstrate that algorithmic curation on digital platforms favors emotionally engaging and visually spectacular content while systematically disadvantaging indigenous knowledge, minority traditions, and non-visual intangible practices, resulting in highly concentrated cultural visibility.
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 digital age, the way we remember our past has shifted dramatically. For generations, museums, libraries, and schools acted as the gatekeepers of cultural history, deciding which stories were important enough to tell and which artifacts deserved a place in the public eye. Today, that role has largely been taken over by the invisible engines of social media platforms. These are the computer programs that decide what you see when you scroll through your feed. They do not choose based on historical importance or scholarly accuracy; instead, they prioritize content that grabs your attention quickly, keeps you watching, and encourages you to click, like, or share. This creates a new kind of cultural landscape where the past is not just preserved, but filtered through the lens of what is most entertaining in the moment. The question facing researchers is whether this shift helps us connect with our heritage or if it quietly erases the parts of our history that are harder to package as a quick spectacle.
A researcher set out to investigate exactly how these digital algorithms shape our view of cultural heritage. They focused on three major platforms: TikTok, known for its short, fast-paced videos; YouTube, which hosts a mix of long and short content; and Xiaohongshu, a popular Chinese platform that blends lifestyle images with video. The researcher gathered a massive collection of 107,950 public videos related to cultural heritage, covering everything from ancient monuments and traditional festivals to indigenous knowledge and minority traditions. They analyzed these videos over a two-year period, from January 2022 to December 2023, looking for patterns in what gets seen and what gets ignored. Their goal was to understand if the videos that perform best on these platforms are the ones that are most culturally significant, or if they are simply the ones that fit the specific rules of the algorithm.
The study found a clear and consistent pattern across all three platforms. Videos that were emotionally charged, visually striking, and created by popular accounts were far more likely to be seen by a large audience. Specifically, content that triggered strong feelings, such as awe or amusement, and videos that were bright, colorful, and full of movement, received significantly more views and engagement in their first few days. This early burst of attention acts as a signal to the platform's system, telling it to show the video to even more people. In contrast, heritage content that was subtle, relied on complex explanations, or focused on non-visual traditions like oral storytelling or specific community rituals struggled to gain traction. Even when a video was about a highly respected site, such as a UNESCO World Heritage location, it did not automatically receive more views if it lacked these high-energy features. In fact, the data showed that once the researcher accounted for the creator's popularity and the video's visual style, the official status of the heritage site did not guarantee visibility.
The researcher also discovered that this inequality is not just a static snapshot but a growing problem. They observed that the gap between the most-viewed videos and the least-viewed ones widened over the two years of their study. This phenomenon, often called the Matthew Effect, means that content which gets a little bit of early attention tends to accumulate more and more views over time, while content that starts with low visibility gets even less exposure as time goes on. This creates a cycle where the most spectacular and emotionally stimulating heritage stories dominate the digital landscape, while the quieter, more complex, and culturally specific traditions of minority groups and indigenous communities are pushed further to the margins. The study suggests that this is not because these stories are less valuable, but because the digital systems that distribute them are designed to reward speed and spectacle over depth and context.
This research highlights a fundamental tension in how we preserve and share our history today. While digital platforms have made it easier for anyone to share their culture, the rules that govern what gets seen are favoring a specific type of storytelling. The findings indicate that the digital version of our collective memory is becoming increasingly skewed toward what is entertaining and visually loud. The study does not claim that the algorithms are intentionally trying to erase certain cultures, but rather that the way these systems are built naturally filters out anything that does not fit their engagement models. As a result, the heritage that survives and thrives in the digital world may look very different from the heritage that scholars and communities have worked to preserve. The author suggests that understanding these patterns is the first step toward ensuring that the digital future of our past remains diverse and inclusive, rather than becoming a narrow reflection of what is most exciting to watch for a few seconds.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.