ClearFactAll topics

Social media drives anger and polarization?

8 peer-reviewed studies for, 2 against

Key takeaways

Stronger evidence: For

Verdict

The best evidence suggests social media can contribute to polarization, but effects are heterogeneous and often modest on average. On the “for” side, randomized evidence shows that certain forms of exposure on social platforms can backfire: in a Twitter field experiment, exposure to opposing-party messages increased polarization for some users, indicating plausible causal pathways through motivated reasoning and antagonistic engagement. A systematic review of the broader literature similarly concludes that social media can foster polarization via selective exposure, platform incentives that reward outrage, and affective reactions.\n\nOn the “against” side, high-quality interventions often find smaller or null average effects. A field experiment finds social-media-driven changes in news exposure can reduce negative feelings toward the other party and has little effect on political opinions. Large-scale feed interventions on Facebook that reduce like-minded content exposure also show minimal changes in attitudes and polarization measures.\n\nNet: social media is not the sole driver of polarization, and changing feeds alone may not depolarize the public. Still, credible causal evidence indicates that social media can increase polarization for meaningful subgroups and under common engagement dynamics, so the stronger side is for.

Strongest study on each side

Ranked by citation count.

Strongest study for

Strongest study against

Open the interactive page →

The evidence

For (8)

Overperception of Moral Outrage in Online Social Networks Inflates Beliefs About Intergroup Hostility (Nature Human Behavior, 2023)

William J. Brady, Killian L. McLoughlin et al. · 67 citations

Observers systematically overperceive moral outrage in authors, inferring more intense moral outrage experiences from messages than the authors of those messages actually reported. Overperception of individuals' moral outrage causes overperception of collective moral outrage and inflates beliefs about hostile communication norms, group affective polarization and ideological extremity.

The effects of social media criticism against public health institutions on trust, emotions, and social media engagement (Proceedings of the National Academy of Sciences, 2025)

Jonathan Y. Lee

Two online survey experiments (N=6,800) examining how social media criticism against public health institutions (CDC, AHRQ) generates anger and polarizes institutional trust, with experimental evidence that integrity-based attacks produce dramatically more anger and polarizing social media engagement preferences.

Reranking Partisan Animosity in Algorithmic Social Media Feeds Alters Affective Polarization (Science, 2025)

Tiziano Piccardi, Martin Saveski et al.

Preregistered field experiment with 1,256 participants on X/Twitter during 2024 election found that exposure to antidemocratic attitudes and partisan animosity (AAPA) content directly causes increased affective polarization and negative emotions (anger, sadness). Decreasing AAPA exposure led to warmer out-party feelings; increasing it led to colder feelings, providing causal evidence that algorithmic amplification of polarizing content drives polarization and anger.

Against (2)

Frequently asked questions

Do social media algorithms amplify anger?

An audit of Twitter's engagement-based ranking found it amplifies emotionally charged, out-group hostile content, with anger the predominant emotion among political tweets. Research also shows social reinforcement can amplify moral outrage expression online.

Does seeing opposing views on social media reduce polarization?

Not necessarily. A Twitter field experiment found that exposure to opposing-party messages increased polarization among Republicans, with different or limited effects for Democrats.

Would changing social media feeds reduce polarization?

Results are mixed. Reducing partisan animosity content on X led to warmer feelings toward the other party, but a large Facebook intervention reducing like-minded content showed minimal effects on attitudes and polarization.

Do people overestimate how angry others are online?

Yes, according to one study. Observers tended to infer more outrage from messages than authors reported feeling, which inflated beliefs about group hostility and polarization.

Related questions