Abstract
Creator platforms allocate attention through complex interactions among content quality, author reputation, and algorithmic visibility. Prior research examines these mechanisms in isolation, yielding inconsistent prescriptions and omitted variable bias. We develop an integrative framework decomposing attention into three distinct pathways and testing their interactions using 315,000 posts from 94,467 authors on Virgool.io (2017–2025). Employing negative binomial regression with author fixed effects, instrumental variables, and a natural experiment exploiting Iran’s daylight saving time abolition, we establish causal relationships. Results reveal that content quality strongly predicts views (β = 6.57, p < 0.001), social capital exhibits diminishing returns (β = 4.87, β2 = −0.51), and algorithmic exposure independently drives attention. Our key contribution identifies interaction effects: quality and exposure are complementary (β = 2.33), meaning algorithms amplify intrinsic merit. However, social capital and exposure show reinforcement (β = 0.23), indicating platforms amplify existing inequalities rather than leveling the playing field. These findings advance attention economy theory, update social capital theory for algorithmic environments, and inform platform governance regarding fairness of exposure allocation.
Recommended Citation
Safari, Ali and Kim, Dan J., "Decomposing Attention in Creator Platforms: The Relative Effects of Content Quality, Social Capital, and Algorithmic Exposure" (2025). BIGS 2025 Conference. 3.
https://aisel.aisnet.org/bigs2025/3