Trust in governments and expert institutions is declining across democracies. Scientists and policymakers have largely responded by trying to communicate more clearly, more accessibly, more frequently. A new paper in Nature, co-authored by twelve researchers across four continents, argues that this response misunderstands the problem. The issue isn’t explanation. It’s exclusion.
The Deficit Model and Why It Fails
For most of the 20th century, science communication operated on what researchers call the deficit model: the public distrusts or misunderstands science because they lack information, and the solution is to provide it. Experts produce knowledge; communicators translate it; citizens receive it. The flow runs in one direction.
That model has a specific failure mode. When people feel that decisions affecting their lives are made by others, communicated to them after the fact, and presented as settled conclusions they should accept, providing more information doesn’t build trust. It reinforces the sense that the system is not theirs to participate in.
The Nature paper, led by Chris Tyler of University College London and colleagues from Aalborg University, George Mason University, Stony Brook University, and elsewhere, identifies this as the central challenge. Trust in elite institutions is falling across many democracies. Populists routinely frame scientists and academics as untrustworthy elites. Yet the response from scientific institutions has mainly been to double down on outreach — producing more publications, more public events, more accessible summaries. The channel changes. The structure stays the same.
What the Evidence Actually Shows
The paper draws on cases where genuine participation has produced better outcomes. Between 2015 and 2018, the EU invited citizens across 30 countries to co-create research priorities for its Horizon 2020 funding agenda. The results were striking. Sixteen expert-led foresight reports recommended prioritising technological advances. Citizens, given real input, placed higher value on strong communities, health and wellbeing, education, and local economies. The preferences were different — not because citizens were less informed, but because they were optimising for different things.
The implication cuts deep. If science policy is set by experts without public input, it will reliably reflect expert values rather than public ones. That’s not a failure of communication. It’s a structural feature of how priorities get set. More clarity about expert preferences doesn’t resolve the mismatch — it just makes it more visible.
The Danish case offers a longer track record. Since the mid-1980s, Democracy x (formerly the Danish Board of Technology) has run consensus conferences — deliberative processes in which randomly selected citizens work through complex science policy questions over several days and produce recommendations. These have shaped Danish legislative agendas on gene technology, the marine environment, and other contested issues. Citizens in these processes don’t just receive information. They define the values that should guide decisions, which is precisely what scientific expertise alone cannot do.
A NASA case study makes the same point in a different domain. Following a public consultation in 2014, the agency pursued a planetary defence mission judged by the public to be more useful for deflecting hazardous asteroids. Public involvement didn’t override scientific judgement. It clarified what the scientific capability should actually be directed toward.
Transparency as a Structural Requirement
The paper is direct about what transparency means in practice. During the COVID-19 pandemic, the UK government’s SAGE committee damaged its own credibility by initially withholding meeting minutes. By contrast, the Joint Committee on Vaccination and Immunisation published its reasoning throughout — including for contested decisions like not initially recommending universal COVID-19 vaccination for healthy 12-to-15-year-olds. Publishing the rationale didn’t end disagreement. But it gave journalists, clinicians, and the public something to scrutinise, which is a fundamentally different relationship than being handed a conclusion.
This is what the paper means by transparency. Not press releases. Not FAQ pages. The actual reasoning, including the uncertainties and the values embedded in the conclusions. Scientists, the paper argues, should acknowledge the values shaping their recommendations — something that almost never happens in current science advice. A precautionary approach to safety and a preference for minimising regulation are both value positions, not neutral scientific facts. Presenting them as neutral facts is itself a choice that erodes trust when people notice.
The Structural Changes Required
The paper’s six recommendations amount to a redesign of how science and policy interact, not a refinement of how they communicate. Among the concrete proposals: universities should establish units dedicated to participatory research methods, not just science communication. Funders should make public involvement an expected condition of grants — from agenda-setting through to dissemination — and finance it accordingly. Promotion criteria at universities should reward co-produced research, which currently almost none do.
The University of Victoria in Canada offers one working example: research involving Indigenous communities must follow protocols defined by those communities, who own and control the resulting data. The paper cites a community-based mangrove restoration project in the Philippines, in Kalibo, where lasting success came only once a local people’s organisation received long-term rights to manage the forest. Scientists’ knowledge of species mix combined with the community’s knowledge of the coastline. Neither was sufficient alone.
These examples don’t resolve the hardest tensions. The paper acknowledges that participation favours those with time, education, and existing civic engagement. It’s expensive and slow. And it can blur the distinction between scientific evidence and value judgements in ways that create their own problems. Not every question benefits from public deliberation — but the default, the paper argues, should shift. Exclusion should require justification, not inclusion.
Science as a Shared Project, Not a Finished Product
The deeper argument is about the nature of legitimacy itself. People support policies they helped to shape at a rate substantially higher than those handed down to them, even when the content is identical. This isn’t irrational. It reflects something true about the relationship between knowledge, authority, and democratic life. Science doesn’t generate its own legitimacy — it borrows it from the institutions and processes that house it.
This connects to a broader pattern across contemporary governance. As explored in Europe Is Running Out of Volunteers, civil society depends on participation that institutions tend to take for granted until it diminishes. The crisis in science trust follows a related logic: institutions that assumed deference as a given are now navigating a world where it has to be actively earned.
The Nature paper doesn’t argue that scientists should surrender their expertise to popular opinion. It argues that expertise without participatory structures produces decisions that lack the legitimacy to actually work. Getting the science right is necessary. It turns out it isn’t sufficient.
Key Sources
- Nature – Six Ways to Put the Public at the Heart of Science and Policy
- Nature – Have People Stopped Trusting Science? The Data Tell a Surprising Story
- OECD – Innovative Citizen Participation and New Democratic Institutions
Subscribe to EuroLuminant for independent European journalism.
