A feed-ranking algorithm that inserts news posts from a set of credible, ideologically balanced news outlets into users’ feeds, increasing their exposure to professional reporting on public affairs.

Partisan animosity thrives with a consistent supply of biased and emotionally charged information. The logic of the intervention is that, by increasing the presence of more balanced and professional editorial framing in users’ feeds, the impact of biased and emotionally charged content will be mitigated.

Stray et al. (2026) tested this approach as one arm of a large independent field experiment across Facebook, Reddit, and X using a browser extension. Over a 6-month period during the 2024 U.S. electoral cycle, the algorithm inserted personally relevant news (based on users’ prior histories) into users’ feeds drawn from a list of credible outlets weighted across the American political spectrum. The results indicate that adding balanced news content to users’ feeds significantly reduced partisan animosity (around 0.04 standard deviations), the largest reduction of any algorithm tested in the study (Note: all effects we include are statistically significant, unless otherwise stated. We report effect sizes using the metrics in the authors’ paper.)

A long-standing concern about social media is that it highlights and amplifies sensationalized, biased, and unverified information. This, in turn, is associated with distorted perceptions of political opponents and issues, as well as affectively charged discourse that can fuel, among other negative outcomes, polarization and hostility. Increasing the footprint of professional news organizations from different segments of the ideological spectrum in users’ feeds appears to be an effective way of mitigating political animosity among social media users.