Misinformation diffusion and spread

Misinformation diffusion research studies how false claims and misleading content spread through social networks. Key empirical findings include: false information spreads faster and wider than truth, emotional content amplifies spread, and early amplification by influential accounts shapes cascade trajectory.

Empirical patterns

• Speed and breadth: False news cascades grow faster, reach larger populations, and are less likely to be retracted than true news (Vosoughi et al., 2017)
• Emotional drivers: Content with high moral-emotional language (anger, moral outrage) spreads more readily than neutral factual claims (Brady et al., 2017)
• Social amplification: Influencers and bots disproportionately amplify low-credibility content in early stages, creating visibility that triggers shared attention (Shao et al., 2017)
• Community structure: Misinformation and fact-checking form segregated communities with minimal interaction; fact-checking cannot compete with misinformation in highly-connected cores (Shao et al., 2018)
• Correction delay: Corrections spread slower than original claims, and reach fewer people (Starbird, 2017; Nyhan & Reifler, 2016)

Theoretical mechanisms

• Algorithmic amplification: Platform algorithms prioritize engagement; emotional, novel, surprising content (including false claims) receive higher visibility
• Homophily and polarization: Users follow similar others; misinformation thrives in ideologically-homogeneous networks where critical voices are absent
• Confirmation bias: Users preferentially consume information confirming pre-existing beliefs, reducing exposure to corrections
• Social proof: Seeing many shares/likes of a claim increases perceived credibility (even if shares are from bots or coordinated campaigns)

Diffusion measurement

• Cascade structure: Retweet chains, share chains, and mention networks reveal who amplifies what
• Cascade shape: Misinformation cascades tend to be "bushy" (many independent shares) while true news cascades are "narrow" (retweeted via few influencers)
• Temporal signature: Viral false claims show early rapid growth, then plateau; true news grows more steadily
• Reach vs. depth: Wide reach (many people see the claim) vs. deep reach (many layers of retweets)

Key papers in this wiki

• Shao et al. (2018) — Anatomy of an online misinformation network — Network analysis of competing misinformation and fact-checking spread during 2016 election; characterizes core-periphery structure and diffusion dynamics
• Vosoughi et al. (2017) — The spread of true and false news online — Foundational empirical work: false news spreads faster, wider, and deeper than true news; studies cascade properties across 10 years and 126,000 cascades
• Brady et al. (2017) — Moral-emotional language amplifies information diffusion — Shows that content with high moral-emotional language spreads more readily; effect independent of political ideology
• Shao et al. (2017) — The spread of low-credibility content by social bots — Bots amplify misinformation via early high-volume tweeting and targeting of influential users; constitute only 6% of accounts but drive 31% of tweet volume spreading low-credibility content

Connections

• Network analysis of misinformation — diffusion occurs on social networks; network structure shapes cascade properties
• Bot detection — bots drive amplification in early stages, shaping cascade trajectory
• Social media polarization and echo chambers — polarized networks constrain diffusion and reduce fact-checking effectiveness
• Fact-checking and corrections — corrections spread slower and less widely than original false claims