Preventing users from going down rabbit holes of extreme video content: A study of the role played by different modes of autoplay

Cheng Chen, Jingshi Kang, Pejman Sajjadi, S. Shyam Sundar

Research output: Contribution to journalArticlepeer-review

Abstract

The autoplay feature of video platforms is often blamed for users going down rabbit holes of binge-watching extreme content. However, autoplay is not necessarily a passive experience, because users can toggle the feature off if they want. While the automation aspect is passive, the toggle option signals interactivity, making it “interpassive,” which lies between completely passive autoplay and manual initiation of each video. We empirically compare these three modes of video viewing in a user study (N = 394), which exposed participants to either extreme or non-extreme content under conditions of manual play, interpassive autoplay, or completely passive autoplay. Results show that interpassive autoplay is favored over the other two. It triggers the control heuristic compared to passive autoplay, but leads to higher inattentiveness compared to manual play. Both the invoked control heuristic and inattentiveness result in higher rabbit hole perception. These findings have implications for socially responsible design of the autoplay feature.

Original languageEnglish (US)
Article number103303
JournalInternational Journal of Human Computer Studies
Volume190
DOIs
StatePublished - Oct 2024

All Science Journal Classification (ASJC) codes

  • Human Factors and Ergonomics
  • Software
  • Education
  • General Engineering
  • Human-Computer Interaction
  • Hardware and Architecture

Cite this