Curating Quality? How Twitter’s Timeline Algorithm Treats Different Types of News

This article explores how Twitter’s algorithmic timeline influences exposure to different types of external media. We use an agent-based testing method compare chronological timelines and for a group Twitter agents that emulated real-world archetypal users. first find exposed links at roughly half the rate timelines. Despite reduced exposure, proportional makeup remained fairly stable in terms source categories (major news brands, local news, new media, etc.). Notably, however, slightly increased proportion “junk news” websites link exposures. While our descriptive evidence does not fully exonerate algorithm, it characterize algorithm as playing minor, supporting role shifting media end users, especially considering upstream factors create algorithm’s input—factors such human behavior, platform incentives, content moderation. conclude by contextualizing within complex system consisting many deserve future research attention.