Schematic of the experimental design (IMAGE)
Caption
Schematic of the experimental design
In Study 1, we recruit 500 residents of the United States and record their demographics and political leaning. Participants then indicate their prior beliefs on six topics.
For each topic, participants are then presented with 12 argumentative posts that they must either upvote, downvote, or pass (at least two upvotes and two downvotes per topic). The output of Study 1 is a content inventory — 72 posts across six topics that have been engaged with by liberals and conservatives — from which we can algorithmically sample and rank posts to generate feeds.
In Study 2, we recruit a new sample of 1,000 United States residents and record their demographics and political leaning. Participants then proceed through a belief updating task where, for each topic, they state their prior belief on a topic, then view a feed of three posts pertaining to said topic (selected and ranked by one of the five algorithms tested), and then revise their belief on said topic. Once the belief updating task has been completed for each of the six topics, participants indicate their subjective perceptions of the posts they saw along three dimensions: how {civil, insightful, emotional} the posts were.
Credit
Credit: Jason William Burton, University of Copenhagen
Usage Restrictions
Must be credited
License
Original content