The relationship between science and the wider public is built on trust, and that trust is shifting in ways that matter. In the United States, a pattern of broad bipartisan confidence in science that held steady for decades has fractured sharply since 2018, with political orientation now predicting a person’s trust in scientific institutions more strongly than at any point in the past half-century. That fracture sits at the center of a much larger web of questions about how scientific knowledge reaches people, shapes policy, and gets contested or embraced. Understanding how science and society interact today means grappling with communication failures, misinformation, funding incentives, and the difficult politics of turning evidence into action.
The Widening Trust Gap
For most of the past fifty years, Americans across the political spectrum reported broadly similar levels of confidence in science. That started to change in the 1990s, with a slow divergence between political orientations, but the gap accelerated dramatically after 2018. Research tracking long-term survey data found that the recent split is driven by two simultaneous forces: trust among conservatives has dropped, and trust among liberals has risen, pulling the two groups apart faster than at any previous point on record.1PubMed Central. Rapidly diverging public trust in science in the United States The result is a public landscape where a person’s political identity has become one of the strongest predictors of whether they consider scientific institutions credible.
This matters because trust in science is not just a polling curiosity. It affects whether people follow public health guidance, support funding for research, and accept the conclusions that inform regulations. The COVID-19 pandemic made this painfully visible: when epidemiological models projected different scenarios depending on assumptions, the way those projections were communicated had real consequences. One study found that emphasizing uncertainty in pandemic projections could erode public trust, while downplaying uncertainty raised short-term support but risked backlash when projections were later revised.2PubMed Central. Model uncertainty, political contestation, and public trust in science: Evidence from the COVID-19 pandemic Neither approach was risk-free, and both fed into the political dynamics already dividing public opinion.
There is, however, a more hopeful thread in the research on uncertainty. A systematic review of studies on how communicating scientific uncertainty affects trust found that, overall, being open about what scientists do and do not know tends to have a positive effect on trust, while downplaying uncertainty can foster distrust.3PubMed. How Communication of Scientific Uncertainty Affects Trust in Science-A Systematic Review A separate large experimental study found that when uncertainty was communicated, people did perceive greater uncertainty, but the actual decrease in trust in both the numbers and the source was small.4PubMed Central. The effects of communicating uncertainty on public trust in facts and numbers The takeaway for scientists and public health officials is counterintuitive: being upfront about the limits of knowledge tends to do less damage to credibility than trying to project false certainty.
Why the Old Way of Communicating Science Stopped Working
For much of the twentieth century, the dominant model for science communication was straightforward: experts know things, the public does not, and the job is to fill the gap. This “deficit model” assumes that skepticism toward science comes from a lack of information and that providing more facts will fix the problem. It sounds reasonable, but decades of research have shown it does not hold up. Giving people more information does not reliably change their attitudes toward science, because attitudes are shaped by values, identity, emotions, and social context, not just knowledge.5PubMed. The lure of rationality: Why does the deficit model persist in science communication?
Despite its poor track record, the deficit model persists. Scientists default to it partly because it matches how they think about their own work: they trade in evidence, so it feels natural to assume everyone else will respond to evidence the same way. Moving past it means adopting what communication researchers call a “dialogue model,” where scientists do more than deliver expert knowledge. The dialogue approach treats the public as participants with their own perspectives, not empty vessels waiting to be filled.6PubMed Central. From deficit to dialogue in science communication: The dialogue communication model requires additional roles from scientists A related framework, sometimes called the “ambassador approach,” encourages scientists to focus less on persuading people to accept a specific conclusion and more on fostering genuine, open-ended exchange.7BioScience. Beyond the Deficit Model: The Ambassador Approach to Public Engagement
This shift is not just theoretical. It changes how scientists are trained, how public events are designed, and what counts as successful communication. A lecture where an audience quietly absorbs slides is a deficit-model event. A community meeting where residents share observations that shape the direction of a research project is a dialogue event. The second is harder to organize and harder to measure, but it tends to produce more durable trust.
Misinformation and the Psychology Behind It
The challenge of science communication is made harder by the speed and scale at which misinformation spreads online. Understanding why people believe and share false claims is not simply a matter of blaming ignorance. A systematic review of the psychological factors behind fake news found that cognitive biases, emotional reactions, and social identity motivations all play roles in shaping how people interact with misinformation on social media.8PubMed Central. Psychological factors contributing to the creation and dissemination of fake news among social media users: a systematic review People do not share false stories because they cannot tell the difference; often they share them because the stories align with what they already believe or what their social group values.
A large meta-analysis examining who falls for misinformation found that, on average, people are actually reasonably good at distinguishing true headlines from false ones, correctly judging accuracy about two-thirds of the time for both categories. But accuracy varies with individual traits. Older age and stronger analytical thinking skills were both linked to a greater ability to discriminate between true and false headlines, along with a tendency to be more cautious and label news as false. Ideological alignment with a headline, on the other hand, made people more likely to accept it as true, regardless of its actual accuracy.9PubMed Central. Susceptibility to online misinformation: A systematic meta-analysis of demographic and psychological factors
Algorithmic personalization adds another layer. Search engines and social media feeds curate content based on past behavior, creating environments where users encounter information that reinforces their existing views while limiting exposure to alternatives.10PubMed Central. The filter bubble and its effect on online personal health information For health and science topics, this can mean that someone who begins searching for skeptical views about a medical intervention quickly finds themselves in an information environment where skepticism is the norm and supporting evidence is invisible.
One of the more promising strategies for fighting misinformation borrows from medicine: inoculation. The idea is to preemptively expose people to weakened versions of the manipulation techniques used in misinformation, helping them recognize and resist those techniques when they encounter real examples. A series of randomized experiments has shown that this approach can meaningfully reduce susceptibility to false claims.11The ANNALS of the American Academy of Political and Social Science. Psychological Inoculation against Misinformation: Current Evidence and Future Directions The technique works even for contested science topics like climate change, where inoculation messages can help protect accurate beliefs against misleading counter-narratives, and in some cases shift people who already hold inaccurate views.12Social and Personality Psychology Compass. Inoculation theory in the post‐truth era: Extant findings and new frontiers for contested science, misinformation, and conspiracy theories
Turning Evidence Into Policy
Even when the science is solid and the public is receptive, getting research findings into actual policy is its own struggle. The phrase “evidence-based policymaking” implies a clean process where decision-makers consult the best available research and act accordingly, but the reality is messier. Policymakers operate under time pressure, political constraints, and institutional incentives that do not reward careful review of systematic evidence. Research on the gap between evidence and policy argues that successful engagement requires researchers to learn how policymaking actually works: to recognize that policymakers often rely on beliefs, emotional shortcuts, and familiar narratives, and to translate complex evidence into simpler stories that fit those decision-making habits.13PubMed Central. Evidence-based policymaking is not like evidence-based medicine, so how far should you go to bridge the divide between evidence and policy?
This is not just about communication skills. How expertise is organized within government agencies matters too. Whether scientists are embedded in policy offices, consulted at arm’s length, or sidelined entirely shapes how much their work influences decisions. Growing awareness of this organizational dimension has led scholars to argue that strengthening the role of evidence in policymaking depends not only on producing rigorous research or communicating it clearly, but on how expertise itself is structured within public organizations.14Public Administration Review. Evidence and Policy‐Making: An Organizational Approach
Governing Technologies That Move Faster Than Regulation
Some of the most consequential interactions between science and society involve technologies that develop faster than the rules meant to govern them. Gene editing, artificial intelligence, and pathogen research all raise questions that cannot wait for decades of regulatory precedent to accumulate. The governance challenge is anticipatory: how do you set guardrails for something whose full capabilities and risks you cannot yet see?
Human genome editing is a case study. After a Chinese researcher created gene-edited babies in 2018, sparking international alarm, calls intensified for meaningful public engagement on the technology. Researchers have since developed anticipatory governance approaches that solicit expert input on future scenarios and then use those scenarios to surface the values, hopes, and concerns of non-scientist publics.15PubMed Central. Governing with public engagement: an anticipatory approach to human genome editing The goal is to move away from expert-driven models that treat public engagement as an exercise in educating laypeople and toward genuine deliberation about what kind of future people want.
Dual-use research, where the same knowledge can be used both beneficially and harmfully, presents a different governance headache. Updated U.S. policy on dual-use research of concern and pathogens with enhanced pandemic potential has strengthened oversight requirements, contributing to an emerging international consensus that funders and institutions should implement enhanced scrutiny for high-risk research.16PubMed Central. What Does the Updated U.S. Dual-Use Research of Concern and Pathogens with Enhanced Pandemic Potential Policy Mean for Asia? But a review of biosafety and biosecurity governance frameworks found that policies in this area have historically been reactive, triggered by specific incidents rather than proactive risk management, and that gaps in transparency and international coordination remain.17PubMed Central. Balancing Innovation and Safety: Frameworks and Considerations for the Governance of Dual-Use Research of Concern and Potential Pandemic Pathogens
The rise of AI adds a new dimension to these concerns. Biological AI models can potentially be misused to design dangerous pathogens or identify novel routes to harm. Researchers have argued that evaluations of these dual-use capabilities should happen before models are deployed, so that biosafety measures can be put in place before, rather than after, harmful applications emerge.18PubMed Central. Dual-use capabilities of concern of biological AI models
The Publish-or-Perish Problem
The relationship between science and society is not shaped only by external forces like politics and misinformation. Internal incentives within academia also distort what gets researched, published, and ultimately communicated. The publish-or-perish culture rewards novelty and positive results, which means that studies confirming a null hypothesis or replicating previous work are harder to publish and less career-enhancing. This creates a systematic bias in the published literature: the file drawer fills with negative and inconclusive results while journals fill with findings that are more likely to be false positives.19PubMed Central. Scientific Utopia: II. Restructuring Incentives and Practices to Promote Truth Over Publishability
Modeling work on the dynamics of scientific publishing under these conditions has shown that when careers depend primarily on publication output, careless and even fraudulent research can gain a competitive advantage, because cutting corners produces more papers faster. The predisposition of top-tier journals toward positive and novel findings compounds the problem, contributing to reproducibility crises across several fields and, by extension, risking public trust in scientific findings broadly.20PubMed Central. Modelling science trustworthiness under publish or perish pressure Reform efforts like registered reports, open data mandates, and revised tenure criteria are underway, but they are fighting entrenched institutional habits.
Diversity and the Quality of Science
Who does the science also shapes what gets discovered. Research on the composition of scientific teams has found that gender-diverse teams produce work that is substantially more novel and higher-impact than same-gender teams of equivalent size. The effect scales with balance: the greater the gender diversity on a team, the better its publications score on measures of novelty and impact, even after accounting for individual researchers’ track records and network positioning.21PubMed Central. Gender-diverse teams produce more novel and higher-impact scientific ideas
But there is a painful paradox in this data. Research on scientific careers has found that underrepresented groups produce higher rates of novel contributions. Yet their novel work is taken up by other scholars at lower rates than equivalent contributions from majority groups, and equally impactful publications by gender and racial minorities are less likely to lead to successful careers.22PubMed Central. The Diversity-Innovation Paradox in Science The scientific enterprise benefits from diversity in its outputs but systematically undervalues the people producing that diversity. This is not just an equity concern; it represents a direct loss of scientific potential.
Opening Research Up
One practical lever for improving the relationship between science and the public is access. Most scientific research is published behind paywalls, meaning that the public, journalists, policymakers, and researchers at underfunded institutions often cannot read the work their taxes funded. Open access publishing changes this calculus, and the evidence suggests it also amplifies scientific reach. A study comparing open-access and paywalled publications in medical education found that open-access articles received roughly fifty percent more citations.23PubMed Central. Is it worth publishing Open Access? – the scientific impact of Open Access publications in the field of medical education A separate analysis of global health research reached a similar conclusion, finding that open access augments citation rates and arguing that self-archiving paywalled papers is a cost-efficient way for institutions to broaden their research impact.24PubMed Central. Knowledge sharing in global health research – the impact, uptake and cost of open access to scholarly literature
Citizen science represents another form of opening up. Projects that enlist volunteers to collect ecological or environmental data have long faced skepticism about data quality, but the evidence has matured considerably. Reviews of citizen science projects have found that well-designed programs, with proper training, expert validation, and statistical methods for handling bias, can produce data with accuracy equal to or surpassing that of professionals.25Frontiers in Ecology and the Environment. Assessing data quality in citizen science An eleven-year biodiversity monitoring study in the Mediterranean confirmed that recreational citizen science can produce generally reliable data when projects are tailored to volunteer capabilities.26Environmental and Sustainability Indicators. Citizen science data reliability enhancing scientific research: insights from an 11-year study in the Mediterranean Sea The quality varies with project design, though: programs with structured sampling protocols and trained volunteers produce data suited for serious research, while opportunistic data collection with minimal training is better suited for education and exploration.27PubMed Central. The potential for citizen science to produce reliable and useful information in ecology
Indigenous Knowledge and Western Science
The conversation about who contributes to scientific understanding extends beyond professional researchers and citizen volunteers to communities whose knowledge systems predate modern science by centuries or millennia. Indigenous knowledge, accumulated through generations of close observation of specific ecosystems, offers ecological insights that formal scientific methods sometimes miss. Researchers increasingly incorporate this knowledge into ecology and evolutionary biology, where it has enhanced understanding and contributed genuinely novel findings about species behavior, landscape management, and ecological relationships.28Frontiers in Ecology and the Environment. Contributions of Indigenous Knowledge to ecological and evolutionary understanding
Integrating these knowledge systems is not simple. A systematic review of efforts to combine indigenous and scientific knowledge found that the approach holds real promise for addressing global environmental challenges but faces obstacles including power disparities between indigenous communities and research institutions, different ways of validating knowledge, and the risk that indigenous contributions get extracted without proper reciprocity. Successful integration requires treating indigenous communities as equal partners, protecting their intellectual property, and developing frameworks sensitive to the cultural and spiritual dimensions of their knowledge.29Environmental Science & Policy. Integration of indigenous knowledge with scientific knowledge: A systematic review
Science Across Borders
Science has long served as a bridge between nations, sometimes holding channels of communication open even when political relations deteriorated. For much of the post-Cold War era, science diplomacy was framed optimistically as a way to build trust, find common ground through shared research, and maintain international cooperation in the face of geopolitical tension. That aspirational model helped connect researchers across divides and was treated as a cornerstone of internationalism.30Science. Rewiring science diplomacy
But the current moment is testing those assumptions. Increasing restrictions on international research collaboration, concerns about intellectual property theft, and the politicization of scientific exchange with certain countries have made science diplomacy more fraught. Researchers who once moved freely between institutions in different countries now navigate export controls, visa restrictions, and institutional compliance offices. The idealistic vision of science as a neutral space above politics was always somewhat naive, and the tensions of the 2020s have made that clearer. What remains is a more grounded question: under what conditions can scientific cooperation actually survive and serve its diplomatic function, and when is it being used as cover for other agendas? The field is still working that out.
What Science Journalists Think Their Job Is
The people who translate scientific findings into stories the public encounters also have their own evolving understanding of their role. A qualitative study among German science journalists found that they see their primary function as informing the public about scientific developments and contextualizing findings within current societal challenges. They also consider themselves critics of scientific institutions and practices, not just conduits for scientific results. When asked to describe the main attributes of science, journalists frequently cited independence and objectivity, along with terms like “empirically verifiable” and “plausible.”31PubMed Central. The journalistic understanding of science as process and social system: A qualitative exploration in the German science journalism community This self-conception as both translator and watchdog is relatively healthy, though the economic collapse of many newsrooms means fewer journalists are doing this work, and those who remain face pressure to produce content quickly, often without the time to scrutinize the studies they cover.