Abstract As digital platforms increasingly structure political communication and public opinion, understanding how Bigtech firms discipline user behavior and emotional alignment has become critically important. This study introduces the concept of platform disciplinary mechanisms to analyze how platforms construct behavioral norms, cognitive dependencies, and affective structures through interface design, algorithmic filtering, and feedback loops. Framed within a triadic platform-user-community framework, we argue that Bigtech not only shapes individual actions but also systematically engineers emotional convergence and intergroup polarization. Empirically, we combine topic modeling and sentiment analysis based on a large-scale dataset comprising GDELT global sentiment data and technical discourse from the Web of Science (WoS). We further apply a vector autoregression (VAR) model to identify 2020 as a structural breakpoint, after which emotional polarization became significantly more persistent. Our analysis shows that technological simplification and emotion-driven interaction mechanisms jointly reinforce echo chambers and intensify affective division. These findings reveal that platforms have evolved from neutral infrastructures into active architects of social order and emotional governance. We argue that regulatory frameworks must move beyond content moderation to address platform-induced behavioral normalization and emotional manipulation. This study contributes a multidimensional theoretical framework for understanding platform power and provides empirical evidence for rethinking accountability and oversight in the governance of the digital public sphere.
Changrong Lu (2025) studied this question.