Dynamics and Inequalities in Digital Social Networks: A Computational and Sociological Review
Pengjia Cui
TL;DR
The paper addresses how micro-level user actions and platform algorithms fortify macro-level patterns in digital networks, including virality, polarization, and inequality. It adopts a multidisciplinary bibliometric and network-science approach, synthesizing sociology, network theory, and computational social science to map trends, core themes, and influential works from 1982 to 2024 using tools like VOSviewer, CiteSpace, and Gephi. Key contributions include characterizing the field's core-periphery structure, identifying foundational theories (e.g., homophily, diffusion) and the pivotal role of algorithmic mediation in shaping micro-macro dynamics, and proposing practical strategies for decentralized architectures, algorithmic fairness, and digital inclusion. The findings offer a framework to guide design, policy, and further research aimed at mitigating polarization and misinformation in digitally mediated societies.
Abstract
Digital networks have profoundly transformed the ways in which individuals interact, exchange information, and establish connections, leading to the emergence of phenomena such as virality, misinformation cascades, and online polarization. This review conducts a thorough examination of the micro-macro linkages within digital social networks, analyzing how individual actions like liking, sharing, and commenting coalesce into broader systemic patterns and how these interactions are influenced by algorithmic mediation. Utilizing a multidisciplinary literature base, this study explores the interaction among user behaviors, network structures, and platform algorithms that intensify biases, strengthen homophily, and foster echo chambers. We delve into crucial dynamics including the scalability's impact on weak tie propagation, the amplification effects on influencers, and the rise of digital inequalities, employing both theoretical and empirical approaches. By synthesizing insights from sociology, network theory, and computational social science, this paper underscores the necessity for novel frameworks that integrate algorithmic processes into established micro-macro models. The conclusion presents practical strategies aimed at promoting fairer digital networks through decentralized architectures, algorithmic fairness, and improved digital inclusion, tackling significant challenges such as polarization and misinformation within networked societies.