TikTok’s For You Page (FYP) serves as the primary content feed for users, presenting a personalized stream of videos. The platform’s algorithm distinguishes itself by relying on both implicit signals—like the duration users spend watching specific videos—and explicit signals such as likes and follows. This approach generally enables TikTok to accurately predict user preferences.
However, some users have raised concerns regarding the algorithm’s handling of negative feedback. Reports suggest that even when users indicate disinterest in certain videos, those videos continue to appear on their FYP. To investigate these claims, a team of computer scientists from Northeastern University conducted a study.
Methodology of the Research
The researchers aimed to analyze user agency within TikTok’s algorithm, particularly in response to negative feedback mechanisms. Instead of relying on user data, they created bot accounts on the TikTok app to simulate user interactions. This method involved using emulated devices to generate accounts and manipulate the algorithm through automated code.
Co-author Piotr Sapiezynski explained that their approach was necessary due to limitations in existing data access options, which do not allow for individual user analysis. They found that while engagement signals do impact content recommendations, the effects are often temporary.
Findings on User Feedback
The study focused on three content categories: cooking, fitness, and sports betting. Results indicated that the “not interested” button was the most effective tool for reducing unwanted content, achieving an approximately 84 percent reduction in irrelevant videos. In contrast, merely skipping videos resulted in only a 48 percent reduction.
Despite this, the researchers noted that the “not interested” option is not prominently displayed, which may hinder user engagement with this feature. Furthermore, the algorithm tends to revert to previously unwanted content if users do not consistently provide negative feedback.
Implications for User Engagement
According to the findings, while TikTok’s algorithm initially responds to negative feedback, it can quickly revert to showing unwanted content if users engage with it again. This suggests that users must be proactive and consistent in their feedback to effectively curate their FYP.
The researchers expressed a desire to validate these findings with actual user data in future studies. They emphasized that while educating users on effective platform use is beneficial, the fundamental design choices of TikTok will ultimately shape user interactions.
This article was produced by NeonPulse.today using human and AI-assisted editorial processes, based on publicly available information. Content may be edited for clarity and style.
