THE PHENOMENON OF "ALGORITHM ADDICTION" ON SHORT VIDEO PLATFORMS: THE RELATIONSHIP BETWEEN STUDENTS’ USAGE TIME AND ALGORITHM RECOMMENDATIONS

The recommendation algorithms of short video platforms, by continuously pushing content that is "more likely to be viewed," may alter users' attention allocation and usage time structure, thus sparking the controversy surrounding "algorithm addiction." This paper takes university students as the research subject and proposes an operational empirical research framework focusing on the coupling relationship between usage time and algorithm recommendations. While preserving real-world usage scenarios, it records students' short video usage time, conversation frequency, and characteristics of recommended stream content, and performs correlation analysis with individual difference indicators to examine whether algorithm recommendations are significantly correlated with behaviors such as increased usage time, repeated visits, and deep immersion. Methodologically, the study emphasizes reproducibility and comparability: by using a unified recording template and a hierarchical indicator system, a structured explanation of the "algorithm addiction" phenomenon is formed, providing empirical evidence and quantifiable references for digital health education and platform governance.