Public Trust in Science Is Fragile and Complicated

My mother-in-law is a smart woman. Raised four kids, ran a small business for twenty years, reads obsessively. She also spent a solid stretch of the pandemic convinced that certain public health guidance was designed specifically to control people rather than protect them. Not every piece of it—she's not a flat-earther—but enough that conversations at our dinner table got genuinely uncomfortable. I remember sitting there thinking: I don't know how to do this. Not because she was stupid, and not because I had some ironclad counter-argument ready to go. But because I could see that her skepticism wasn't really about the science. It was about something older and harder to name.

I've thought about that dinner table a lot while working through the recent research on public trust in science. Because the data keeps doing something I find genuinely interesting: it refuses to confirm the catastrophe narrative.

The headline that flatters both sides

The dominant story in public discourse is that we are in a trust crisis—that science denial has gone mainstream, that misinformation has broken the social contract between researchers and the public. It's a story that's easy to tell and politically convenient for a range of people across the spectrum.

But Nature recently reported something that complicates that narrative significantly: from a global perspective, researchers studying public trust have concluded that trust in science and scientists is actually high. Not eroding-but-still-okay high. Actually, genuinely high.

Sit with that for a second, because it cuts against a lot of what we've been telling ourselves.

And yet. Nature also reports that trust is important precisely because scientific knowledge cannot influence decisions or improve lives unless citizens and policymakers consider it trustworthy—and that confidence in scientific processes remains uneven. So we have two things that are both true: globally, people still trust science at relatively high levels, and that trust is also unevenly distributed, politically inflected, and genuinely fragile in ways that matter for policy.

The question isn't "is there a trust crisis?" The question is: where, among whom, and about what?

The fracture lines underneath

Here's where the aggregate data gets less reassuring. Trust in science isn't distributed evenly across society, and the fault lines tend to run along exactly the lines you'd expect: identity, ideology, social relationships, economic interest.

Psychology Today cites research by Sutton et al. on how rooted attitudes—stemming from identity, ideologies, social relationships, and fears—shape whether people accept or reject scientific findings. Their example is pointed: if scientific research suggests your occupation might disappear due to AI, you might be motivated to question the credibility of that research. Which is not irrational, exactly. It's human. The problem is that "human" and "epistemically sound" are not always the same thing.

This is the part that tends to get lost in the trust-crisis discourse. People frame science skepticism as a knowledge deficit—if people just understood the science better, they'd trust it more. But that's not what the research consistently shows. What shapes trust isn't primarily information. It's the relationship between the person and the institutions delivering that information.

Psychology Today's piece on the quiet beginning of trust gets at this directly. Trust doesn't begin with data. It begins in something quieter and harder to engineer—a sense that the person or institution communicating with you is doing so for you rather than to you. That distinction is doing a lot of work.

Think about my mother-in-law. She wasn't rejecting the virology. She was rejecting a felt sense that the institutions communicating public health guidance were leveling with her, that they had her particular situation in mind, that they weren't operating with interests and incentives that didn't include her. Whether she was right about any of that is almost beside the point. The feeling was the data she was working from.

The chain of trust and where it breaks

The Lindau Nobel Laureate Meetings write about this elegantly: most of us can't evaluate primary scientific research ourselves. We trust the chain—the researchers, the reviewers, the journals, the institutions, the regulators. For most of modern history, that chain held. It worked as an operating assumption even if few people could have articulated it.

What's changed isn't that people have become stupider or more irrational. What's changed is that several links in that chain have visibly failed, in public, in ways people remember. And once you've seen one link fail—once you've watched an institution you trusted turn out to have interests that weren't yours—the whole chain looks different.

This is not an argument for false equivalence. Some scientific consensus is robust and the skepticism directed at it is not. But understanding why that skepticism persists requires engaging with the legitimate grievances underneath it rather than diagnosing people as simply credulous or tribal.

Nature argues that trust in science risks becoming ideologically siloed—that it can't be sustained by the support of narrow constituencies at one end of the political spectrum. That's a real concern. If science is perceived as belonging to one political tribe, then skepticism of science becomes a way of marking tribal loyalty for the other—and at that point, no amount of better communication fixes the problem, because the problem isn't really about communication anymore.

What actually helps—and what doesn't

The brief that started this piece gestures toward "increasing accessibility and engagement" as the path forward, and that's not wrong exactly, but it's also pretty thin as a prescription. The research suggests something more specific.

Psychology Today points to the importance of reducing perceived distance between scientific communities and the public. Not dumbing things down. Not better infographics. Actual proximity—scientists who are legible as human beings with stakes and uncertainties and blind spots, rather than as oracles descending from institutional heights.

The trust-building that works, according to the healthcare trust research, happens in small, accumulated moments. It is not a campaign. It's not a messaging strategy. It's the slow accumulation of experiences in which institutions behave in ways that make their stated values credible. Which is, for the record, extremely hard to manufacture and very easy to destroy.

I keep coming back to my mother-in-law and what would have actually moved her. Not a better explainer video. Not a celebrity endorsement. Something closer to: an honest accounting of what we knew and didn't know, delivered by someone who seemed to understand her specific life and wasn't treating her skepticism as a pathology to be corrected.

She came around, mostly. Not because she was persuaded by an argument. Because she watched people she trusted—in her actual life, not on television—make decisions based on that guidance and be okay. Trust, in the end, traveled through relationships, not institutions.

Which is maybe the most important and least scalable finding in all of this research: the institutions most invested in rebuilding public trust in science are probably not the ones best positioned to do it. The work happens closer to the ground than that. The question is whether the specific humans running those institutions understand that, and whether they're willing to act as if they do—or whether they'll keep designing better charts and wondering why nothing changes.

---

Marcus Obi covers parenting and family dynamics for Buzzrag.