Algorithm Recommendation and the Shaping of Youth Values: Mechanisms, Risks, and Governance Pathways—An Empirical Study Based on a Survey of University Students

Abstract

In the context of rapid advancements in artificial intelligence, algorithm recommendation technology has become deeply embedded in the information acquisition and cognitive construction processes of young people, emerging as a significant technological force shaping their value perceptions, identifications, and choices. This study, grounded in Social Construction of Technology (SCOT) theory and Selective Exposure Theory, constructs an analytical framework of “technological logic—influence mechanisms—governance pathways,” and employs a questionnaire survey method to empirically examine the current status of algorithm recommendation’s influence on youth values among 230 university students across four grade levels. The findings reveal that: (1) algorithm recommendation creates a significant “filter bubble” effect, with students generally perceiving content homogenization and limited exposure to diverse viewpoints; (2) algorithmic influences demonstrate dual effects—promoting knowledge acquisition while reinforcing cognitive biases, with notable grade-level variations; (3) the level of algorithmic literacy remains generally moderate, with insufficient critical awareness; (4) students exhibit a significant “awareness-behavior gap” in algorithmic coping strategies. The study further identifies value orientation drift, echo chamber formation, critical thinking erosion, and superficial value internalization as primary risks. Accordingly, governance pathways are proposed encompassing three dimensions: technological optimization, institutional regulation, and literacy enhancement.