Abstract
Events such as the climate crisis, developments in artificial intelligence (AI) and the COVID-19 pandemic, compel interaction between researchers and journalists to inform citizens and increase public understanding of science. This literature review provides an in-depth analysis of publications about the changing relationship between researchers and journalists. The analysis is based on two main trends which impact science-media interactions, changing media landscapes and medialisation of science. We find that the roles of both researchers and journalists have diversified in response to increased competition and increased time pressure for the science journalism profession. The combination of the two trends may cause journalists to increasingly depend on strategic communication from researchers and public relations departments of scientific organizations, which could reduce the quality of science journalism. Additional research is recommended, in particular regarding how journalists and researchers can ensure science journalism of good quality with an emphasis on analysing understudied countries and recent developments in generative AI.
Keywords
Introduction
Science journalism plays a vital role in bringing people into contact with science news (Dunwoody, 2014, 2021). Global challenges, such as natural disasters caused by climate change, the COVID-19 pandemic, and developments in artificial intelligence (AI) highlight the impact that science news may have on people's daily lives. For example, people with high trust in science were more likely to comply with COVID-19 interventions and are more likely to engage in climate change action (Cologna et al., 2025). In contrast, minorities who distrust science may affect public opinion and the consideration of scientific evidence in policymaking (Cologna et al., 2025). Nevertheless, Cologna et al. (2025) found that scientists and scientific methods are trusted in most countries and that science communication efforts are desired. In addition, Denham (2021) found that U.S. citizens who had higher confidence in the press also had more confidence in the scientific community, which may indicate that science journalism can impact audiences’ trust in science.
Meanwhile, a growing number of people use the internet and social media as primary sources of science news, instead of traditional news sources (Keng & Cheng, 2023; Verstappen & Opgenhaffen, 2024). The increasing reliance on digital media platforms to access news has sparked concerns about the impact of these platforms on trust in news, which has been declining in many countries (Ross Arguedas et al., 2022). Ross Arguedas et al. (2022) found that journalists are mostly critical of the impact of digital media platforms, because these disrupt journalistic practices and contribute to the circulation of damaging discourses about journalism. In addition, the rise of online spreading of mis- and disinformation puts additional pressure on journalists and researchers who need to fact-check, identify and spread reliable information (Scheufele & Krause, 2019; Westlund et al., 2024). Together, these developments highlight the need to better understand the science-media relationship.
Despite its current relevance, very few literature reviews on the science-media relationship are available. Notable exceptions include a narrative literature review on how the science-media relationship evolved (Dudo, 2015) and a scoping review on co-production between researchers and journalists in science communication (MacGregor et al., 2020). Regarding science communication research in general, Wu et al. (2019) conducted an automatic scientometric analysis of 393 articles. They identified seven topic categories in science communication research and found that topics related to media and science received much attention, but scientists’ relationship with media and the public was among the smallest topic categories (Wu et al., 2019). Guenther and Joubert (2017) conducted a bibliometric analysis of research papers published in the three main science communication journals. They concluded that research in science communication has increased and became more institutionalized and diversified. They called for additional research on the content of science communication research, including topics and methods (Guenther & Joubert, 2017). In response to the existing reviews, our literature review focuses on the content of publications about science journalism and science communication that studied interactions between journalists and researchers.
Following Bos and Nuijens (2020), we define science journalism as a type of journalism primarily dealing with scientific achievements, breakthroughs, processes, and quests. That includes research in the social sciences and the humanities and news about science covered in environmental, health, and technology journalism. Our literature review aims to provide an in-depth overview of changes in the science-media relationship, and how these changes impact the roles of journalists and researchers as well as the quality of science journalism. Our analysis is based on two main trends, the changing media landscapes and the process of medialisation of science. These two trends have received considerable attention in research on science journalism, but, to our knowledge, no comprehensive overview exists of their combined impact.
The first trend concerns how media landscapes are changing. During the past few decades, the development of internet and social media technologies coupled with a significant reduction of funding for news media has caused widespread changes in media landscapes worldwide (cf. Dunwoody, 2014, 2021; Ross Arguedas et al., 2022). Simultaneously, the amount of misinformation and disinformation has been rising, with new developments in generative AI creating risks of furthering this problem (Schäfer, 2023; Scheufele & Krause, 2019). Even though scientists remain among the most trusted sources, trust in journalism has decreased, affecting trust in science news as well (Cologna et al., 2025; Denham, 2021; Pew Research Centre, 2020). These developments have challenged the position of science journalists and changed media ecosystems in many countries.
The second trend concerns the process of medialisation of science. Medialisation, sometimes also labelled mediatisation or mediazation, is a broad concept that can be applied to a rich variety of processes, e.g., politics, religion and health, as well as science, which is the focus in our review (Väliverronen et al., 2020). Välliverronen et al. (2020) discussed medialisation as an ambiguous concept in media studies, arguing that media logics (including rules, regulations and customs) are intertwined with many other logics in society. Despite the entanglement, however, in our analysis, we focused solely on the narrower concept of medialisation of science to understand the relationship between scientists, their institutions, and the media. Medialisation of science, thus, refers to the process through which, from the 1990s onwards, scientists have increasingly taken up an active role in reaching out to audiences, supported by their research institutes (e.g., Franzen et al., 2012). Franzen et al. (2012) described an increasing media orientation of universities and research institutions, which led to active management of visibility in media and messages that can include ‘over-optimistic claims’ of scientific outcomes. Thus, the increasing media orientation of science may affect the science-media relationship and the quality of science journalism.
The main research question this literature review aims to answer is: How do the processes of changing media landscapes and the medialisation of science impact the science-media relationship? In the remainder, we present our results and identify changes in science-media interactions based on the two trends. We discuss how the two trends impacted and continue to impact the roles of journalists and researchers and the quality of science journalism. Finally, we highlight understudied questions about the science-media relationship and provide suggestions for further research.
Methodology
Sample
To select the sample for our review study, we searched for literature on the databases Scopus and Web of Science, using the search string: ‘(“science journalism” OR “science communication” OR “science-media”) AND (relationship OR collaboration OR interaction)’. In addition, a snowball search was conducted by screening the citations of relevant articles. Studies written in English and published between 1 January 2000 and 31 October 2024 in scientific journals, conference papers or books were selected. Additionally, the publications had to include information about direct interactions between science and media, on an individual or institutional level. This criterion excluded, for example, studies on the framing of scientific information in news sources and content analyses of social media messages written by individual scientists or science communicators. Both empirical studies and literature reviews were included. Although, as described above, medialisation of science started in the 1990s, we decided to begin the period of analysis in 2000, since online interaction and user participation, which drastically changed interactions between science, media and society, started to appear from the 2000s onwards (Almeida, 2017).
The database search initially generated 2244 results in total. After removing duplicates and adding 86 additional publications through citation searching, 1614 publications remained to be screened. The first author screened all publications by reading the titles and abstracts of the articles and assessing their relevance. The screening process resulted in 112 publications relevant for retrieval, of which five could not be retrieved. The remaining 107 publications were assessed for eligibility by the first two authors, based on the inclusion criteria mentioned above. In case of doubt, both researchers read the article and discussed whether to include it in the sample. In this step, another 41 publications were excluded which did not concern direct science-media interactions at the individual or institutional level. Therefore, in total, 66 publications were included in the review (see Figure 1 for an overview of these steps).
Flow diagram of steps to select literature.
Data Analysis
For the analysis of the literature, a review form was developed to qualitatively assess the included publications. To enhance the validity of the review form, eleven publications were independently reviewed by the two authors and differences were discussed until agreement was reached. From the remaining sample, 47 publications were assessed and reviewed by the first author and eight publications by the second author.
The review form consisted of an Excel template including multiple-choice and open questions. First, the reviewers filled in information about the authors, year of publication, title, country, source, type of literature (journal article or book chapter), topic and abstract of each publication. Next, the reviewers filled in information about the content of the publications. For the questions about included participants, methods, roles of researchers, roles of journalists and reasons for interactions, multiple-choice options were included, as shown in Table 1, with an additional column for notes on these questions. The list of the roles of researchers and journalists was taken from Dijkstra et al. (2024) and the categorization of reasons for interactions was based on Davies (2021). If a publication mentioned multiple roles or reasons for interactions, “other” was selected and the roles and reasons were described in the comment box and added to the total mentions of each option later. Next to these multiple-choice questions, open questions were included to collect what concepts and processes were mentioned, and it was possible to add notes on other mentioned actors, notes on the science-media relationship, a summary for the literature review, notes on references and other general notes.
Overview of Multiple-Choice Questions and Answer Categories.
Results
Results of the literature review are informed by all 66 publications included in our sample. We have added references to exemplify findings, while the complete reference list of all included publications can be found in Appendix 1. We do not refer explicitly to each study that has informed the aggregated results in the text. To provide an overview, where relevant, we have added the number of times a topic was mentioned. Below, we first provide an overview of general findings, followed by an analysis of the two trends affecting science-media relationships and ending with an analysis of the roles of journalists and researchers.
Overview of the Sample
To provide an overview of the included studies, Table 2 shows, for each publication, the authors, publication year and source, the countries where the studies were conducted, the studied topics, and the methods used with the number of included participants. The total sample included four book chapters and 62 journal articles. Analysis of regional distribution shows that data for 38 studies were collected in one country, whereas 28 publications included data from multiple countries, with seven studies including data collection in more than ten countries. These data were most frequently collected in European countries (72 times), followed by North American (32), Asian (11), African (8), Australian (6), and South American (3) countries, with the USA, Germany, UK, and Canada as the single countries where data was collected most frequently (each over 10 times).
Overview of Publications Included in the Sample.
Interviews were the most used research method (in twenty studies), while in twelve studies multiple methods were applied, eleven studies were based on survey data, eight studies conducted literature reviews, and four studies conducted content analyses. Eleven publications reported using other methods, such as philosophical analyses, Q-methodology, focus groups, training workshops, and ethnographic studies. In sum, 38 publications used qualitative methods, 16 publications used quantitative methods, and 12 publications used mixed methods. Finally, shedding light on the types of participants in the studies, 22 studies included multiple types of participants, all including researchers and journalists, with some also including other participants such as PR officers, editors, or science communicators. Of the studies that included only one type of participant, nineteen included researchers, nine journalists, and two institutes. Fourteen publications did not directly include any participants.
Changing Media Landscapes
The first trend of changing media landscapes encompasses several technological and societal developments that have changed the position of news media. The studies show that throughout the past few decades, media landscapes have changed drastically, with both positive and negative consequences for science journalism. Dudo (2015) provides an overview of how changes in media landscapes, especially in North America and Europe, have affected the science-media relationship. He argues that the increasing popularity of digital media, a reduction of funding for news media, and a decreasing interest in news about science and technology among citizens have weakened the traditional position of media as the primary source through which most citizens encounter science (Dudo, 2015).
Larsson et al. (2019) describes how changes in the media landscape affect the working conditions of news media production. Since digital media platforms provide increased opportunities to publish news more regularly, journalists must deal with more deadlines. Pitrelli (2017) adds that science journalists also need to acquire multimedia and digital skills and deal with competition from researchers and universities who communicate directly with their audiences. Furthermore, science journalists have fewer opportunities to join newsrooms, work under more stressful conditions, and get paid less than before (Pitrelli, 2017). Several studies echo the challenges journalists face in writing in-depth articles due to time and financial pressures (Appiah et al., 2020; Dijkstra et al., 2015).
Schäfer and Painter (2021) emphasize in their conclusions that changes in the media landscape and how they affect science journalism differ by country and depend on varying politics and media cultures. Dijkstra et al. (2024) identify large differences between the media landscapes of Belgium, Italy, Portugal, and Spain. Nguyen and Tran (2019) conducted a systematic literature review of challenges for science journalists in the Global South. In their findings, they emphasize that problems that exist in the North can be exacerbated in the South because they interact with other challenges distinctive to the Global South, and because there are fewer institutional and civil mechanisms to solve problems within professional journalism. Several publications mention that journalists in Africa and the Arab world especially have difficulty accessing local researchers and that their relationship with researchers is strained due to a lack of resources and a lack of skills for both parties (Appiah et al., 2020; Kaye et al., 2011; Lublinski et al., 2014; Nguyen & Tran, 2019).
Changes in media landscapes may negatively impact the quality of science journalism. Several authors argue that the increasing pressure to publish news quickly may cause journalists to depend more on press release material and to become less critical of their sources (Ashwell, 2016; Schäfer & Painter, 2021). In their view, the reliance on science public relations (PR) sources increases the risk of churnalism, when news media reproduce press release material verbatim (McKinnon et al., 2019). Franks et al. (2023) add that, as a result of under-resourced newsrooms, churnalism threatens science journalism in South Africa, even though the journalists they interviewed emphasized that PR content should always be investigated further.
Even when journalists interact with researchers directly, the time pressure journalists must deal with may cause less critical reporting. Journalists struggle to get uninvolved expert opinions, because of time constraints and because scientists are reluctant to criticize each other (Dijkstra et al., 2024; Geller et al., 2005). In addition, Albaek (2011) found that journalists usually initiate contact with scientists they already know when they need a source for science news. McKinnon et al. (2019) call this reliance on a select group of trusted sources journalists’ ‘trust portfolio’. Franks et al. (2023) elaborate on this topic and argue that journalists need to balance their relationship with researchers, so that researchers trust them sufficiently to join interviews, but do not feel betrayed if a report is critical of their findings. In contrast, Geller et al. (2005) report journalists had more opportunities to explore sensitive and controversial topics if they had a trusting relationship with researchers.
Changing media landscapes may also benefit science-society relationships. Peters’ (2013) findings indicate opportunities to use online communities and social media for more interactive and dialogic forms of science journalism, with increased possibilities for citizens to engage in debates about scientific topics and their impact on society. Lutz et al. (2018) also highlight the potential of social media to enable more democratic and participatory discussions between researchers and citizens. In their study, Martin Neira et al. (2022) specifically focus on the benefits of disseminating science news on TikTok as part of transmedia narratives that take advantage of the narrative possibilities of different media platforms. Results from Dijkstra et al. (2024) suggest that developments in artificial intelligence, while they might enhance negative impacts of changing media landscapes as well, can provide opportunities to increase the accessibility, impact, and understandability of science news.
In the studies, it was noted that science journalists have found various ways to adapt their roles to changing media landscapes. Several studies describe how changes in settled routines and assumptions in journalism are changing the roles of science journalists and blurring the lines between journalism and other forms of knowledge production (Secko et al. 2011; Van Witsen & Takahashi, 2018). Brüggeman et al. (2020) argue that changing media landscapes, polarizing societies, and the proliferation of science in society leads to post-normal science communication which values pro-active communication, advocacy, interpretation, and transparency over objectivity. Brennen (2018) demonstrates how science journalists continue the (scientific) process of producing meanings by reducing technical complexity and by constructing and justifying truth claims in their coverage of science.
Next to the changes in media landscapes described above, the impact of the COVID-19 pandemic on science-media interactions was a recurring theme in recent studies. In their study during the beginning of the COVID-19 pandemic, Metcalfe et al. (2020) conclude that traditional media outlets remain an important source of information for many people, because of their perceived credibility. Several studies described that researchers and journalists experienced their interactions with each other during the pandemic as predominantly positive and noticed a rising public interest in science (De Jong et al., 2024; Marín-Gonzalez et al., 2023; Massarani et al., 2021). In his in-depth analysis of how questions were asked and answered in podcast episodes about COVID-19 throughout the pandemic, Nohl (2023) shows an increasing mutual understanding between researchers, journalists, and their audience over time, as the reporting style of the podcast became more educational.
The COVID-19 pandemic also results in new digitization opportunities in newsrooms that can help to generate attractive content and reach new audiences (Martin Neira et al., 2022). During the COVID-19 pandemic, it became more common for journalists to use pre-print articles. More than half of the participants in the study by Massarani et al. (2021) used preprint articles during the pandemic, and 59% of them treated preprints differently in their outputs by stating that the research had not been peer-reviewed. Fleerackers et al. (2022) conclude that news messages about COVID-19 pre-prints often did not mention the research was published as a pre-print and in 40% of the messages pre-print studies were not framed as uncertain. When pre-prints were framed as uncertain this was usually done by stating that the research was not peer-reviewed (Fleerackers et al., 2022).
Medialisation of Science
The second trend in the science-media relationship is the process of medialisation of science. Included in the sample are studies that discuss the concept of medialisation of science as introduced by Peter Weingart (1998). For example, Rödder (2011) describes the concept in her study analysing communication processes between science and mass media. Bauer and Entradas (2022) and Rödder (2011) also discuss two core elements of the medialisation of science: that media attention on scientific issues increases and that, in turn, science increasingly adapts to media logic and values. The process of medialisation of science was initially expected to occur predominantly on an organisational level (Rödder, 2011), but the concept has since been applied to various aspects of the science-media relationship.
Multiple studies describe how medialised specific cases or topics in science are. For example, Rödder (2009) found that the Human Genome project exploited all four of the studied medialisation indicators of press conferences, visibility of scientists, media publications before scientific publications, and an intertwining of the scientific and political discourse. In addition, Rödder and Schäfer (2010) compared the medialization of human genome research and particle physics research, concluding that the latter field is less medialized and that indicators of medialisation differ somewhat per field.
Differences in the medialisation of scientific fields and topics indicate that the process of medialisation of science is context-dependent, and subject to cultural differences as well. Most studies focus on perspectives from the Global North, though there are some exceptions. Lo and Peters (2015) compare medialisation among German and Taiwanese life scientists. They conclude that Taiwanese scientists are less medialised since they are less likely to prioritise interactions with journalists, adapt to journalistic styles, and discuss the context, limitations, and uncertainties of their research than their German colleagues (Lo & Peters, 2015). Koso (2021) studied how Japanese research organisations adapt to the unique media landscape in her country and conclude that these research organisations are somewhat medialised. She did not observe the Western indicators of professionalisation and institutionalisation of medialisation of science. Instead, she found that research organisations adapt to local press clubs, which are important in the Japanese media landscape (Koso, 2021).
Public relations (PR) departments at research organisations and independent science PR organisations play an important role in the medialisation of science. For example, Williams and Gajevic (2013) describe the institutionalisation of science communication about hybrid embryo research via the UK Science Media Centre as an important indicator of medialisation. Similarly, Ivanova et al. (2013) emphasize the role of universities’ PR departments in the medialisation of German climate scientists. Bauer and Entradas (2022) demonstrate that scientific institutions increase their PR efforts in response to increasing competition.
Several studies have analysed the medialisation of individual researchers, by looking at changes in their interactions with media, including how skills, motivations, and roles changed (Lo & Peters, 2015; Moorhead et al., 2023; Olesk, 2021). Olesk (2021) categorizes visible scientists into adopters, who are more medialized and proactively manage media interactions to achieve their goals, and adapters, who are less medialized and reactively use interviews to explain their work accurately. Moorhead et al. (2023) add a third category of affiliation, for researchers who enthusiastically engage in public outreach. They also emphasize that researchers’ career stage, institutional context, and pressure from journals may influence how researchers engage with journalists (Moorhead et al., 2023).
Underlying the discussions about the medialisation of science and the increasing orientation of researchers towards the media is the worry that scientists may increasingly adhere to media norms and values to make decisions about their research, while these decisions should be governed by scientific values (Allgaier et al., 2013; Ivanova et al., 2013). Ivanova et al. (2013) found that for 82% of the German climate scientists who participated in their survey, media interest played some role in their scientific decision-making, but this role was fairly unimportant in most cases. Similarly, Allgaier et al. (2013) found little evidence that scientists feel constrained or guided by public visibility in their research. Focusing on media coverage instead of researcher perspectives, Lehmkuhl et al. (2023) found that journalism rarely succeeds in focusing public attention on specific scientists and scientific events. While emphasizing that media coverage is only one part of the medialisation process, they conclude that it is unlikely that journalism plays an important role in scientific decision-making (Lehmkuhl et al., 2023).
Despite the limited impact on scientific decision-making, several studies observe a growing media orientation among researchers and scientific organizations. Researchers’ knowledge about journalism, and their strategic use of this knowledge has increased, which can improve their relationship with journalists (Allgaier et al., 2013; Moorhead et al., 2023; Olesk, 2021). However, the increasing importance of science PR departments and organizations (cf. Bauer & Entradas, 2022; Ivanova et al., 2013) may also harm the quality of science journalism (Williams & Gajevic, 2013). Williams and Gajevic (2013) argue that an increase in science PR can lead to simplification, hype, and scientific self-censorship, which weakens the position of both researchers and journalists as sceptical interrogators.
The combination of a growing availability of science PR material and a higher dependence of journalists on a limited number of sources increases the risk of churnalism (Franks et al., 2023; Williams & Gajevic, 2013). Franks et al. 2023, therefore, propose that a better understanding among researchers of journalists’ duty to take a critical stance may help to prevent this. Meyer (2006) argues that in response to extensive science PR, journalists may report on researchers as mere powerbrokers in a “power vs. the people” frame, undermining trust in science. To avoid this, she proposes that researchers should have a public discussion on how to put the ideal of scientific disinterestedness into practice whilst integrating science in society (Meyer, 2006).
Roles of Journalists and Researchers
In total, 47 publications describe one or more roles of journalists, as shown in Table 3. Critical roles for journalists, such as gatekeeper and watchdog roles, are the most common. The gatekeeper role is a traditional role for journalists, which Dudo (2015, p. 762) explains as follows: “Journalists have traditionally been the gatekeepers of scientific information, charged with surveilling and disseminating science news in ways that are (ideally) accessible, instructive, and interesting”. In deciding what science news to cover, journalists also take on a role of agenda setter, although Lehmkuhl et al. (2023) and Secko et al. (2011) indicate that science journalists have limited success as agenda setters. Several publications highlight that journalists continue preferring critical roles such as that of gatekeeper, watchdog, independent reporter, and trustworthy source (Dijkstra et al., 2015; Franks et al., 2023; Waddell et al., 2005).
Overview of the Roles of Journalists and Researchers and How Often They Were Mentioned in an Included Publication.
Note. In case a publication mentioned multiple roles of journalists or researchers, “other” was selected as the multiple-choice option and a comment was made of which roles were mentioned. The number of mentions in this table includes all publications in which each role was mentioned.
In contrast, researchers may expect journalists to help them promote the interests and goals of science (Ashwell, 2016; Peters, 2005). Some studies describe roles for journalists that fit these expectations of researchers more closely, including roles of cheerleader/science populariser and civic educator. As cheerleaders or popularisers, journalists can help to legitimise science (Peters et al., 2008; Schiele, 2008). As civic educators, journalists focus more on informing their audience and contributing to public understanding of relevant scientific topics (Appiah, 2020; MacGregor et al., 2020). Over the past few decades, journalists have also started to act as public intellectuals, synthesizing complex information about a specific scientific topic they specialize in (Fahy & Nisbet, 2011; Giannoulis et al., 2010; Møller Hartley, 2017).
In total, as shown in Table 3, 53 publications discuss one or more roles for researchers, with expert roles as the most common. Several studies describe how the majority of researchers used to act as autonomous scientists, who prefer not to interact with media and the public (Brüggeman et al., 2020; Schiele, 2008). Even though some researchers still adhere to autonomous scientist roles (Kaye et al., 2011; Carr Kelman, 2023; Martin Neira et al., 2022), over time, researchers have taken on increasingly interactive roles in public debates (Besley et al., 2018; Larsson et al., 2019; Lutz et al., 2018). Initially, the main role of scientists is described as a passive expert, with journalists fulfilling a role as intermediaries, fitting with the deficit model of science communication (Gerhards & Schäfer, 2009; Mogendorff et al., 2012; Schiele, 2008). Several studies focusing on journalists’ perspectives still find roles of researchers as passive experts (Brennen, 2018; Giannoulis et al., 2010; Massarani et al., 2021).
In the shift towards more participatory models of science communication with open dialogues, more active roles for researchers in public debates emerge (Brüggeman et al., 2020; Schiele, 2008). Fifteen studies found roles of researchers as public communicators, who actively and strategically seek out science communication opportunities, for example by contacting the media themselves to achieve their goals (Besley et al., 2020; Pinto & Matias, 2023). Other researchers act as promotors of science, providing a positive image of their own research and science in general. The role of promotor of science is closely related to that of researchers as educators. Journalists also appreciate researchers who take on an educator role to help them with factchecking and ensuring accuracy (Dijkstra et al., 2024; MacGregor et al., 2020).
In several studies, authors identify more critical and political roles for researchers, for example by commenting on the work of colleagues or policy claims (Dijkstra et al., 2015; Olesk, 2021). Albaek (2011) notes that the role of scientists has shifted from focusing on their own work towards providing a critical expert opinion by commenting on issues related to their area of expertise. Several studies highlight that researchers disagree on the acceptability of taking on roles of advocates or policy advisers (de Jong et al., 2024; Swenson & Marson, 2024). Metcalfe et al. (2020) add that researchers may be reluctant to participate in controversial discussions because it can jeopardise their reputations.
Discussion
In our literature review, we aimed to gain an in-depth insight into the impact of changing media landscapes and the medialisation of science on the science-media relationship. The findings show that within the science-media relationship, both journalists and researchers take on a plurality of roles. For science journalists, critical roles as watchdog or gatekeeper, which match with expected roles for journalists in general, remain dominant. This finding corresponds with the observation that news coverage depends mainly on its production infrastructure and not on its content (Dunwoody, 2021). Roles for researchers have shifted from reluctant or passive participants as part of autonomous scientist and expert roles, to increasingly interactive roles, fitting with the emergence of increasingly interactive (social) media platforms (Keng & Cheng, 2023). Nevertheless, the expert role, which can be considered as more passive than active, is still the most common for researchers.
Changing media landscapes have drastically influenced the roles of journalists in several ways. The increasing popularity of digital and social media offers new opportunities to increase interaction and reach new audiences (Lutz et al., 2018; Martin Neira et al., 2023). Thus, journalists are encouraged to develop digital competences while continuing to adhere to the fundamental principles of the profession (Marta-Lazo et al., 2020). However, combined with a reduction in funding for science news, these developments also put pressure on journalists to publish more quickly and compete with more actors, including researchers and science PR departments (Dudo, 2015; Pitrelli, 2017). The resulting changes in settled routines and assumptions in journalism have blurred the lines between journalism and science, leading to new roles for journalists as public intellectuals who pro-actively synthesize and justify information (Brennen, 2018; Møller Hartley, 2017).
The process of medialisation of science has predominantly impacted the roles of researchers. The medialisation of science has led to higher levels of professionalisation and institutionalisation of science-media relationships (Allgaier et al., 2013; Olesk, 2021). Researchers have an increased understanding of journalism (Allgaier et al., 2013) and are more likely to see public communication as an important part of their responsibilities (Marín-González et al., 2023). As public communicators, researchers may act reactively, proactively to achieve their own goals, or proactively out of enthusiasm for public outreach (Moorhead et al., 2023; Olesk, 2021). Additionally, Dempster and Sutherland (2024) argued that the medialisation of science may have helped scientists to act as public experts during the COVID-19 pandemic. This offers a potential explanation for the relatively positive experiences of scientists with media during the COVID-19 pandemic reported in studies in our literature review (de Jong et al., 2024; Marín-González et al., 2023).
The trends of changing media landscapes and the medialisation of science may positively impact the quality of science journalism. On the one hand, digital and social media provide increased opportunities for interaction between researchers, journalists and their audiences, which can potentially be used to democratize discussions about science (Keng & Cheng, 2023; Lutz et al., 2018). Moreover, Nelson (2025) argued that research institutes could use these opportunities to internalize public responsiveness in science, helping to rebuild and merit public trust in science. In addition, researchers’ increased knowledge about journalism and their strategic use of this knowledge may improve their relationship with journalists (Allgaier et al., 2013; Olesk, 2021). Trusting relationships between researchers and journalists could also increase opportunities to explore controversial topics in science news (Geller et al., 2005).
On the other hand, the negative impacts of changing media landscapes on the roles of journalists may harm the quality of science journalism. For example, Ross Arguedas et al. (2022) reported that journalists described how digital media platforms strain traditional norms which underlie trust in journalism through their disruption of journalistic practices. In addition, Román San Miguel et al. (2022) found that journalists believed that the increase in misinformation in media during the COVID-19 pandemic was partially caused by the excess of information on social media and the lack of time for journalists to check and contrast information. Our literature review confirmed that, due to increased time pressure and competition, journalists have less time to write and ensure the quality of their work (Appiah et al., 2020; Dijkstra et al., 2024). Consequently, science journalists can be more dependent on press releases or on specific familiar sources, which Dunwoody (2021) described as reporter-source intimacy. Several studies in our sample described similar phenomena and argued that this might lead to less critical coverage of science (Albaek, 2011; Ashwell, 2016; McKinnon et al., 2018; Schäfer & Painter, 2020).
As Williams and Gajevic (2013) argued, the increase in science PR as part of the medialisation of science can harm the quality of science journalism through simplification, hype, and scientific self-censorship. The medialisation of science may also lead to media norms influencing scientists’ decisions, which are expected to be based on scientific values, affecting the quality of science (Allgaier et al., 2013; Franzen et al., 2012). However, so far, no conclusive evidence seems available that this happens in practice (Allgaier et al., 2013; Ivanova et al., 2013; Lehmkuhl et al., 2023). In contrast, new findings from Fleerackers et al. (2025) show that academic debates influence journalists’ perception of research studies as well as the norms and practices journalists use for reporting, which indicates that scientisation of journalism is part of the medialisation of science process.
The combination of the two trends of changing media landscapes and the medialisation of science extends the negative impact of each trend separately. The increasing availability of science PR material and the increasing pressure on journalists, as pointed out in several studies we reviewed, may result in increased levels of churnalism, the verbatim reproduction of PR material in news media (McKinnon et al., 2018). Several studies have indeed found high similarities between press releases and science news articles, indicating the presence of churnalism (Comfort et al., 2024; Vonk et al., 2024). Our literature review also highlights that, alternatively, in response to extensive science PR, journalists may report on researchers as mere powerbrokers in a “power vs. the people” frame, undermining trust in science (Meyer, 2006).
Our results also indicate that the trends of changing media landscapes and the medialisation of science are context-dependent, with differences per country, culture and research field. Negative impacts of changing media landscapes that affect science journalism in the Global North are likely to cause greater problems in the Global South (Nguyen & Tran, 2019). Koso (2021) demonstrated how medialisation of science looks different in Japan than in Western countries, with Japanese research institutes adapting to the unique media landscape in their country.
The context-dependency of the trends is important to note regarding the observation that English-language publications on science journalism in the Global South are limited, as both our sample as well as previous literature reviews show (e.g., Guenther & Joubert, 2017). Notably, even though our sample included 28 publications with data from multiple countries, only eight publications compared perspectives from the Global North and South. Therefore, we echo the recommendation from Guenther and Joubert (2017) to stimulate collaboration between researchers from different countries, especially between the Global North and South. In addition, we recommend further research on how the two trends affect science journalism and roles of researchers and journalists in understudied countries, especially in the regions of Asia, Africa and South America.
Another important research gap to be addressed concerns the impact on the science-media relationship of new developments in artificial intelligence (AI), and especially of applications using generative language models such as ChatGPT. Our sample included only one publication (Dijkstra et al., 2024) that focused specifically on this topic, despite the large impact these developments are expected to have on science journalism (Møller et al., 2025; Nelson, 2025; Schäfer, 2023). The application of generative AI in newsrooms is likely to further increase existing pressures on science journalists and decrease their job security (Schäfer, 2023). Møller et al. (2025) found that while journalists worry about job displacement, they are more concerned about the impact of AI on their professional identity. As a result, developments in AI prompt journalists to reflect on their changing roles and responsibilities, emphasizing their human qualities (Møller et al., 2025).
Our results indicate that positions for science journalists in newsrooms have decreased, confirming findings from Schäfer (2017). The limited number of specialist science journalists means that generalist journalists are more likely to cover scientific topics as well. Therefore, a relevant area for further research would be to compare the coverage of scientific topics by specialized science journalists and generalist journalists who occasionally cover science news. Considering the impact of the two trends on science journalism, it would be especially interesting to compare how science journalists and generalist journalists view their roles and what quality criteria they apply to ensure the quality of science journalism.
Limitations
Some limitations of this literature review should be acknowledged. Our sample of 66 publications is limited and contains some biases. Despite the use of several broad search strings to find relevant studies on the science-media relationship, the search was limited to two databases with scientific publications, supplemented by a snowball search. The sample predominantly contains studies that analysed the situation in countries in the Global North, underrepresenting insights from other parts of the world. Partly, this is due to the inclusion of publications in the English language only. We recommend literature reviews of publications in languages other than English to get a more comprehensive overview. In addition, the studies predominantly included perspectives of researchers on the science-media relationship. More research can add to understanding the roles and perspectives of journalists.
The focus of this literature review on changes in the science-media relationship is too narrow to provide a complete and comprehensive overview of all content in science journalism studies, as Guenther and Joubert (2017) called for. In addition, our emphasis on the science-media relationship highlights the roles of researchers, journalists and, to a lesser extent, science PR departments. Consequently, the roles of other stakeholders that affect science journalism, such as political and industrial actors and audiences, received little attention in this review. Nevertheless, this narrower perspective allowed us to synthesize in-depth information about how changing media landscapes and the medialisation of science impact science-media relationships and the quality of science journalism.
Conclusion
Global challenges that are informed by science and scientific processes highlight the importance of good quality science journalism. The growing spread of misinformation and disinformation makes this especially relevant. Therefore, we conducted an in-depth analysis of scientific literature on the science-media relationship based on two main trends, changing media landscapes and medialisation of science. Changing media landscapes result in limited funding, increased competition, and increased time pressure for journalists. Medialisation of science leads to researchers taking on increasingly active roles in public debates but also leads to more strategic communication from researchers and science PR departments. The combination of these two trends may cause journalists to increasingly depend on press releases and a limited number of sources, which may increase churnalism and lead to less critical coverage of science. We suggest further research in different cultural contexts on how journalists and researchers can ensure quality science journalism in response to these trends and new developments in (generative) AI.
Supplemental Material
sj-docx-1-bst-10.1177_02704676261452144 - Supplemental material for Current Changes in the Relationship Between Scientists and Journalists. A Systematic Literature Review
Supplemental material, sj-docx-1-bst-10.1177_02704676261452144 for Current Changes in the Relationship Between Scientists and Journalists. A Systematic Literature Review by Anouk de Jong and Anne M. Dijkstra in Bulletin of Science, Technology & Society
Footnotes
Acknowledgements
The authors thank Menno de Jong and Miles Macleod for their helpful suggestions during our discussions of this literature review.
Ethical Considerations
Not applicable
Consent to Participate
Not applicable
Consent for Publication
Not applicable
Author Contributions
Anouk de Jong: Conceptualization, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing.
Anne M. Dijkstra: Conceptualization, Formal analysis, Funding acquisition, Methodology, Supervision, Writing – review & editing
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This manuscript is based on a literature review conducted as part of the ENJOI project (De Jong & Dijkstra, 2022), which received funding from the European Union’s Horizon 2020 Research and Innovation program under Grant Agreement no.101006407.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability
Not applicable
Supplemental Material
Supplemental material for this article is available online.
Author Biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
