1

Introduction

Science is embedded in almost all aspects of modern life, and the process of science (the investigation of phenomena through observation, measurement, and analysis) has long been used to understand the world and advance knowledge and technological innovation. From the discovery of cures for life-threatening diseases to the development of crops that can adapt to environmental threats, to the construction of vehicles and devices for exploring the ocean floor and outer space, the benefits of science to individuals, communities, and society are well documented. Additionally, information from science is often used to inform personal and policy decisions related to medical care, food supply and safety, environmental health, and national security, among others. Given there are many contexts in which this information may be leveraged to advance specific interests, the reliability of these scientific findings is critically important. Science information is typically communicated and disseminated by individuals and institutions (e.g., scientists, healthcare professionals, journalists, philanthropists, universities, science associations, non-profit organizations, governments agencies, citizens, etc.), to achieve one or more goals identified in the National Academies of Sciences, Engineering, and Medicine (National Academies) report, Communicating Science Effectively: A Research Agenda (2017, p. 2):

• to share the findings and excitement of science,
• to increase the appreciation for science as a useful way of understanding and navigating the world,

• to increase knowledge and understanding of the science related to a specific issue,
• to influence people’s opinions, behavior, and policy preferences, and
• to engage with diverse groups so that their perspectives about science related to important social issues can be considered in seeking solutions to societal problems.

While distinct approaches may be required to accomplish each of these goals, they all reflect ways to support better integration of scientific knowledge with personal values and other considerations for decision making. Thus, misinformation related to science can greatly influence this process. For example, the spread of misinformation about science can plausibly lead to ill-informed personal choices about disease treatment, lack of planning for natural disasters, higher rates of death from vaccine-preventable diseases, and limitations on productive debate about addressing climate change and other environmental hazards (e.g., water pollution). Additionally, communities that are already experiencing risks to their well-being—due to a variety of factors including health inequities, limited access to affordable and nutritious food, environmental degradation, poverty, and structural and systemic racism—may be further harmed by the uptake of misinformation about science. For these reasons, concern about the spread of misinformation, and the overall role of scientific expertise in civic life, democracy, and policy has grown significantly in recent years (Kavanagh & Rich, 2018; Scheufele et al., 2021; Southwell et al., 2018; Watts et al., 2021). The topic of misinformation about science has garnered significant public attention not only in the news media but also from policymakers who are interested in mitigating the associated negative impacts.

Misinformation about science, however, is not a new phenomenon. For example, misinformation about vaccines dates back to the invention of the smallpox vaccine in the late 18th century, and even more recent narratives predate the era of social media and the coronavirus (COVID-19) pandemic (Colgrove & Samuel, 2022; Schwartz, 2012). Additionally, the U.S. Food and Drug Administration (FDA) was created in 1906 to enforce the Pure Food and Drug Act in response to widespread misinformation about the efficacy and safety of drugs, food additives, and biological substances (Denham, 2020; Jaafar et al., 2021). To this end, there is a long-standing body of research related to misinformation about science across diverse disciplines, including science, health, and risk communication, computational social science, history, political science, information science, journalism, law, media studies, psychology, sociology, agriculture, and engineering. Several initiatives have been launched to leverage this evidence to address misinformation about science (Cacciatore, 2021; Lazer et al., 2018). Think

tanks, government agencies, non-governmental organizations, and civil society organizations, among many others, have released reports and/or have funded initiatives to do more research and make policy recommendations in this area. But to date, these efforts have not yielded a clear understanding of the state of knowledge of the problem of misinformation about science and ways to address it.

STUDY CHARGE

With support from the National Science Foundation (NSF), the Board on Science Education of the National Academies initiated this consensus study to characterize the nature and scope of misinformation about science and its differential impacts; identify solutions to limit its spread; and provide guidance on interventions, policies, and future research to reduce associated harms (see Box 1-1). A 15-member expert committee representing multidisciplinary expertise across the social, biological, and applied sciences—psychology, sociology, political science, science and health

communication, journalism, computational social science, information science, engineering, technology, and agricultural sciences—was appointed to examine the extant literature in science communication and misinformation and develop this consensus report (Appendix B includes brief biographies of the committee members and staff). Committee members also had expertise regarding public understanding of science, the nature of misinformation within different communities and groups in the United States and abroad, and the nature of information spread through social networks, including, but not limited to, social media platforms. Finally, the committee included community engagement experts and health practitioners, as well as those with expertise in developing and implementing responsible and ethical innovations.

STUDY APPROACH

Over the course of 15 months, the committee held several fact-finding meetings, gathering evidence from expert presentations as well as from the existing literature, which included peer-reviewed journal articles, book chapters, policy reports, editorials, white papers, and previous National Academies reports. The committee also benefited from discussions and presentations from a variety of experts who participated in four fact-finding meetings. At the first and second meetings, the committee heard presentations on the scope and composition of the science information landscape, frameworks, and considerations for defining misinformation and disinformation, different classes of interventions that have been employed to address misinformation in both the online and offline environments, and differential strategies for addressing misinformation versus disinformation.

In conjunction with the third meeting, the committee hosted a one-day public workshop that brought together researchers, practitioners, philanthropists, and policymakers, among others. The workshop featured a series of discussions on the nature, mechanisms, and differential impacts of misinformation about science and on select interventions for addressing misinformation as it relates to their respective theory of change, target audience(s), intended and unintended outcomes, and effectiveness. At the final public information-gathering session, the committee invited two discussions of topics relevant to the statement of task: understanding misinformation in the context of the history and nature of science, and implications for addressing misinformation given advancements in information technologies (e.g., artificial intelligence, machine learning, etc.).

As part of its information-gathering efforts, the committee also commissioned three papers in additional areas that were identified as important for inclusion in the report. Joseph Polman (University of Colorado Boulder) authored a paper that examined how science learning in formal

and informal learning contexts can support the development of relevant competencies for navigating a complex information environment (e.g., identifying credible information, managing misinformation and disinformation, assessing, and weighing evidence, science-informed decision making, etc.). Rachel Kuo (University of Illinois Urbana-Champaign) and Sarah Nguyễn (University of Washington) authored a paper that reviewed the extant evidence on the origin(s), diffusion, and effects of misinformation about science within information networks primarily composed of nonnative speakers of English and immigrant populations in the United States. Lastly, Nicole Buckley (private law firm) and Ryan Calo (University of Washington) authored a paper that described the range of possibilities for addressing misinformation through regulatory mechanisms in the United States.

STUDY SCOPE

In interpreting its charge, the committee made several decisions that shaped its review of the evidence and the conclusions and recommendations that resulted from it. These include defining key terms—science information, misinformation about science, and disinformation about science—and using these definitions as boundaries for its analysis; adopting a systems perspective in seeking to holistically understanding the spread, mechanisms, impacts, and solutions for addressing misinformation about science; and determining which documented impacts of misinformation are most consequential toward prioritizing recommended actions for intervention.

Defining Key Terms

As a mode of inquiry, science provides an important way to understand and engage with the natural and material world, and as a discipline is constituted by a set of practices, values, and concepts that scientists have established and adhere to—such as expectations of appeal to empirical evidence and acting with integrity (see Chapter 2). Science as a discipline spans the physical, biological, social, health, and applied sciences, and as a way of knowing is often leveraged by individuals, communities, and societies to make important decisions regarding human and environmental health and well-being, among other reasons. Building on this, the committee adopted a definition of “information” from Wanless & Berk (2021), defined as “anything that is processed to provide meaning of the world.” Scientific knowledge at its broadest can therefore be viewed as any information that is generated through the process of science, and as such, the committee defines science information as any claim about a phenomenon within any of the science disciplines. In this consensus report, science and health

misinformation is the focus, both within the institution of science and outside the institution of science—both issues of concern (West & Bergstrom, 2021). This focus includes misinformation in science (e.g., fraud, the reproducibility crisis, hype) and about science topics (e.g., climate change, genetically modified organisms, vaccines, management of pain, smoking, COVID-19) as defined by researchers themselves.

As a fundamental component of the study charge, the committee was asked to define misinformation and disinformation about science. Currently, there is general agreement among researchers who study misinformation that false and misleading information both fall into the category of misinformation (Søe, 2021). At the same time, researchers have not yet agreed on a single definition of misinformation across disciplines and methods, and there is also disagreement in the field about the importance of intentionality within the definition (Altay et al., 2023; Vraga & Bode, 2020). This lack of an explicit and consistent definition can lead to interdisciplinary misunderstandings and artificially create contradictory findings. This confusion in turn, can lead to unwarranted policy responses to misinformation that reflect either an underestimation or overestimation of the problem. Thus, a clear definition of misinformation about science is essential for advancing scientific understanding of the phenomenon and determining when and to what extent a specific intervention is needed.

In this chapter, we briefly present definitions of misinformation and disinformation about science that were developed by the committee and guided our work, and Chapter 2 of this report provides a more detailed discussion of the committee’s process for developing these definitions. Importantly, the complexities in establishing widely-shared definitions are not trivial. Scientific knowledge is not static, and therefore the nature of scientific consensus is inherently contingent on current evidence. Debates arise within science as new information emerges and leads to the revision of what may have previously been understood, and at times, value judgments may largely shape scientific agreement at a given time (e.g., regarding the risks and benefits of new technologies). In addition, science can simply be poorly communicated, or it can be miscommunicated, hyped, or prone to publishing biases (Phillips et al., 2005; Southwell et al., 2019; West & Bergstrom, 2021), and currently there are no bright lines that distinguish between scientific uncertainty, science done poorly, and misinformation about science. Nevertheless, clearly defining these key terms was a fundamental aspect of the study charge, both to guide the committee’s work and to function as a possible guidepost for the broader research community. The committee’s definition states:

Misinformation about science is information that asserts or implies claims that are inconsistent with the weight of accepted scientific evidence at the time (reflecting both quality and quantity of evidence).

Which claims are determined to be misinformation about science can evolve over time as new evidence accumulates and scientific knowledge regarding those claims advances.

In the case of disinformation—a subset of misinformation—the notion of intent is often emphasized as a distinguishing feature of this particular information type (Freelon & Wells, 2020). However, it is difficult to measure the intent or beliefs of an agent who is sharing information, which also presents an operational challenge to research. Indeed, some agents are clearly self-interested and purely tactical in promulgating falsehoods (see Chapter 4); nonetheless, the committee determined that the intent of the sharer is immaterial to the potential harm of that information to the receiver(s) and to the way it influences and shapes their sense of what decisions or actions are possible. For these reasons, misinformation is considered to be agnostic with respect to intentionality, and in the view of this committee, misinformation is a phenomenon that encompasses disinformation. Thus, the committee also states:

Disinformation about science is a sub-category of misinformation that is circulated by agents who are aware that the science information they are circulating is false.

Here, false science information is defined as a mischaracterization of the “weight” of evidence as found in the literature at a particular moment in time and that underpins the consensus position (see Chapter 2).

Adopting a Systems Perspective

The committee also recognizes that people are differently situated with respect to exposure to information. The existence of the alternative press is a reflection of this reality (e.g., immigrant press, Black press, feminist press, military press, etc.). Hence, in seeking to understand the dynamics of misinformation about science (i.e., nature, scope, spread, and impact), the committee determined it was important to consider how the broader historical and contemporary contexts of people’s lived experiences shape their relationships to and with information. For example, systemic factors such as social stratification, structural and systemic racism, bias, and discrimination create conditions in which access to power, resources, and opportunities are constrained on the basis of identity, and as they relate to this report, can result in disparate access to and trust in high-quality science information across groups (see Chapters 3, 5, and 6). Taking a systems perspective afforded the examination of the intersections between misinformation about science and existing risk factors and inequities, and the potential

impacts these have on well-being. Globalization, including shifts in technology, labor, economy, migration patterns, and geopolitical relationships between states, also influences the relationships between people and the relationships between people and information. To that end, the committee also considered the role that history, values, culture, language, and identity play in influencing differential exposure, engagement, and impacts of misinformation about science across different communities.

Characterizing the Impacts of Misinformation About Science

Finally, the committee placed an emphasis on the misinformation about science that is most consequential. While people may be exposed to varying degrees of misinformation about science, it is possible that only a subset of this information might impair decisions that individuals or communities make with consequences for their health and well-being. Similarly, scope and scale of the misinformation provided another lens for the committee’s determination about potential for greatest negative impacts. Misinformation about science that can spread to millions of people through television, radio, social media, or a statement by prominent public figures has the potential to negatively impact more people than the misinformation that a single individual might encounter in a conversation. Undoubtedly, any misinformation can pose a risk to health and well-being to some extent (Krause et al., 2020; Swire-Thompson & Lazer, 2020); however, the committee was cognizant of the importance of representing the problem as accurately as possible. Building on its review of the evidence and conclusions related to the dynamics of misinformation about science, the committee also chose to prioritize recommendations with the greatest potential for mitigating its negative impacts.

STANDARDS OF EVIDENCE

As previously mentioned, this report reflects a range of sources consulted during the course of the study process. The committee gave the most weight to empirical evidence appearing in peer-reviewed publication outlets. As with other consensus reports published by the National Academies, this committee did not focus exclusively on literature associated with any one method for information gathering. This committee similarly draws on a National Research Council (NRC) report (2002, p. 6), which has informed subsequent consensus studies, to adopt the view that “A wide variety of legitimate scientific designs are available for […] research.” This stance meant that the committee considered qualitative and quantitative evidence as well as evidence generated by experimental studies, survey research, case studies, and observational data. Across these sources of evidence, the

committee prioritized information with relevant implications for understanding the production, transmission, consumption, and consequences of misinformation, and with a specific focus on science contexts. At times, the committee drew from the broader literature on misinformation, communications, mass media, learning sciences, cognitive psychology, and law and technology, and as such gave careful consideration to the strength of the evidence and degree of informativeness for reconciling existing knowledge gaps about the nature and scope of the problem of misinformation about science.

The absence of clear definitions of misinformation and related concepts, as mentioned above, complicates the assessment of published evidence. In the wake of prominent news coverage regarding the general challenges of false and inaccurate information, literature that features the keyword “misinformation” and “science” (along with other related terms) has increased considerably in recent years. For example, a search for the keywords misinformation and science on Google Scholar yields approximately 16,200 articles for the period 1990–1999; 26,800 for the period 2000–2009; 34,400 for the period 2010–2019; and 203,000 for the period 2020–2023.¹ Indeed, this body of work is rapidly advancing, and with new literature being consistently added, an exhaustive analysis of the issues addressed in this report and synthesis of the extant literature is not feasible.

It is also important to note that the extant literature includes myriad examples of manuscripts in which misinformation is not explicitly defined as a concept or in which the operationalization or measurement of misinformation is not clear. Additionally, the evidence base on the topic of misinformation also reflects studies that focus on other distinct, but related concepts to describe the various ways that information can be distorted within the contemporary information ecosystem more broadly. Such concepts include but are not limited to information disorder² (Wardle & Derakhshan, 2017), information integrity³ (National Science and Technology Council, 2022), and infodemic⁴ (World Health Organization [WHO], 2020). Thus, in carrying out its charge, the committee prioritized

___________________

¹ The committee notes that these data are not based on a systematic search of the extant literature.

² Wardle & Derakhshan (2017) define information disorder as the combined spectrum of misinformation, disinformation, and malinformation (truthful information used to harm).

³ According to the National Science and Technology Council, information integrity is defined as the spectrum of information and associated patterns of creation, exchange, and consumption in society, where high-integrity information is trustworthy; distinguishes fact from fiction, opinion, and inference; acknowledges uncertainties; and is transparent about its level of vetting (National Institute of Standards and Technology, 2024, p. 9).

⁴ The WHO defines infodemic as too much information, including false or misleading information in the digital and physical environments during a disease outbreak (WHO, n.d.).

publications that included clear definitions and operationalization of misinformation about science as a stimulus external to a person, and a subset of the wider universe of all publications that mention misinformation about science in some way. Because common encounters with misinformation about science involve both people as well as informational content and information environments that develop across time, the committee included not only studies of human participants but also studies in which a unit of analysis other than an individual person (e.g., a unit of media content or a community or time) constituted the primary unit of analysis. Research on misinformation about science has also largely focused on a narrow set of topics such as vaccines, COVID-19, genetically modified organisms, and climate change. Hence, throughout the report, these topics are frequently referenced as part of the committee’s analysis of the evidence base and the examples used to illustrate its findings. Likewise, online platforms are a prominent source of current data about the nature of misinformation. Indeed, misinformation about science exists beyond these topics and contexts, and the committee highlights the need for more scholarly attention to a broader range of science topics and media types in the Research Agenda.

Importantly, the committee also prioritized studies of the United States in its review and analysis of the evidence, including those that consider the United States as one among multiple national cases. This focus was essential to render the committee’s work feasible, as an understanding of the state of misinformation about science across the entire world lies beyond the scope of this report, and more importantly, the recommendations of this committee are most consequential for the United States. Thus, two types of studies of misinformation about science were mostly excluded from consideration: studies primarily about countries other than the United States, and studies written in languages other than English.

Finally, the committee employed a rubric to characterize the strength of the diverse research that exists on the topic of misinformation (Box 1-2). Where applicable throughout the report, the committee articulates the type of evidence being reviewed and its strength and adopted similar phrase definitions from a previous NRC report (2012) for this purpose. When reaching consensus on conclusions and developing recommendations, the committee took care not to present findings based solely on oral testimony or limited evidence as adequately supported, but instead framed them as potential areas for further investigation without recommending a specific course of action. As a result of these deliberative processes, all conclusions and recommendations outlined in this report reflect the full consensus of the Committee on Understanding and Addressing Misinformation About Science.

REPORT ORGANIZATION

The report that follows this introduction illuminates a comprehensive view of the nature, scope, and impact of misinformation about science through broad examination of the extant literature across diverse disciplines, sources, and topics. This begins with further discussion of the committee’s rationale and process for defining key terms in Chapter 2, which clarifies the phenomena that are the focus of this report, situates misinformation in the context of other distinct, but related informational phenomena, and discusses important caveats that are associated with applying definitions. In Chapter 3, the committee describes the confluence of historical and contemporary systemic factors that intersect with a rapidly changing and complex information ecosystem to influence the nature of misinformation about science and people’s relationship to information, including misinformation about science. Chapter 4 examines the evidence related to the sources of misinformation about science and in Chapter 5, the committee discusses key factors and mechanisms that influence differential reach and spread.

In Chapter 6, the committee reviews the existing evidence on the impacts of misinformation about science, articulating which are most consequential and how such impacts may be similar or different across levels (i.e., individual, community, institutional, societal) and social groups (e.g., race/ethnicity, socio-economic status, geography). Chapter 7 summarizes what is currently known about the effectiveness of the range of interventions that have been employed by different sectors (i.e., government, industry, academia, civil society, and community organizations, etc.) to address misinformation about science either before or after it has been encountered. This chapter also identifies important considerations for both current and promising future interventions.

As previously noted, there is a rapidly growing body of research on misinformation about science across diverse disciplines, each bringing different theoretical and methodological lenses to study the phenomenon. To this end, Chapter 8 characterizes and discusses the state of the scholarship on misinformation about science, including some of the challenges associated with studying this topic. Finally, in Chapter 9 the committee presents its conceptual understanding of the misinformation about science landscape, which includes report conclusions and recommendations for multi-stakeholder action, as well as prioritized directions for future research.