Skip to content

Kimball Lecture: Political Misinformation and the Problem of Expertise

2018
Open on YouTube

In brief

Adam Berinsky examines declining trust in media and scientific institutions and argues that expertise alone does not guarantee credibility or persuasion. The central problem: experts and lay audiences often speak different languages. He identifies two key communication strategies for building credibility in a low-trust environment: (1) speak your audience's language—don't assume authority automatically transfers knowledge; (2) find unexpected messengers—credible messengers are often those arguing against their own partisan interests. He illustrates this with examples from climate science, vaccines, and medical authority, concluding that building trust requires active work to align expertise with audience understanding, not passive assertion of credentials.

Key themes

  • Trust in institutions is declining. Media trust has dropped dramatically since 1997; confidence in medical institutions has declined since the 1970s despite stable confidence in scientific leadership. This creates a crisis where experts cannot rely on their authority to be persuasive.

  • Authority is not automatically granted. Despite high expertise and good intentions (e.g., MIT climate scientists), experts often fail to convince lay audiences because they assume credentials confer credibility. Instead, credibility must be earned through communication that meets audiences where they are.

  • The language gap between experts and audiences is real. Experts have domain-specific background knowledge, training, and lingo (e.g., academic journal articles) that lay audiences lack. Handing a research paper to a patient and saying "this explains everything" is ineffective because the expert has decodified knowledge the audience does not possess.

  • Communication strategy matters as much as content. In a world where authority is contested, how and who delivers a message often matters more than the message itself. An MIT scientist saying "trust me" may fail, but a credible messenger arguing against their apparent partisan interests succeeds.

  • Three levers for building credibility. 1. Put yourself in your audience's shoes—understand their concerns and vocabulary. 2. Use language your audience understands, not expert jargon. 3. Fund authority through action and unexpected sources—find people with no partisan motive to deny a claim.

  • Implications for vaccine hesitancy, climate denial, and political rumors. The problem is not that audiences lack information—it's that they don't trust the people delivering it. A Republican senator denying "death panel" claims carries more weight than a nonpartisan fact-checker, because the senator has no motive to mislead the Republicans in his base.

  • Hardliners may be a lost cause. In any belief landscape, there are people so committed to a false belief that persuasion attempts fail. However, there are also "uncertain" and "moveable" audiences who can be shifted toward accuracy if the right messenger and framing are used.

  • One-off corrections don't stick. A single fact-check or correction may work immediately but decays quickly. Rumors, by contrast, persist in public consciousness. This asymmetry matters for platform design and policy.

Notes

This talk, delivered in 2018, captures Berinsky's thinking on the expert-credibility crisis before the COVID-19 pandemic amplified these tensions. His emphasis on the communication gap between experts and audiences—and his focus on earned rather than assumed authority—is foundational to understanding why expert credentials alone fail to persuade in polarized contexts. The advice to find "unexpected messengers" (people arguing against their partisan interest) has proven prescient as a mechanism for reaching audiences in asymmetric information environments. His candor about the absence of a simple solution and the limits of one-off corrections contrasts with many contemporary policy approaches that assume fact-checking or media literacy alone will solve misinformation problems.

How it connects

Similar papers

Summarized from the full text. Added 2026-11-10 from a submission by reza. Every statement on this page traces back to the source.