Transforming Science: How ROR² and NanoBubbles are Enhancing Transparency, Trust, and Integrity in Research

Two groundbreaking research projects, ROR² and NanoBubbles, are paving the way for a more transparent and trustworthy scientific ecosystem. 

  • ROR² examines the systemic dynamics of publication practices and platforms, addressing challenges like “Open Washing” and promoting responsible research.
  • NanoBubbles dives deep into error correction processes, uncovering why science sometimes struggles to self-correct and how this can improve across disciplines.
Frédérique Bordignon
Frédérique Bordignon

Both projects are committed to interdisciplinary approaches, aiming to enhance transparency, trust, and efficiency in conducting and communicating science. Frédérique Bordignon is a researcher at École des Ponts ParisTech and a member of the LISIS laboratory (Interdisciplinary Laboratory of Sciences, Innovations, and Societies). With a background in linguistics and expertise in scientometrics, research integrity, and scholarly communication, she plays a pivotal role in both projects. 

In this interview, she will share insights into her contributions, discuss the challenges these initiatives face, and explain how her expertise helps advance the ambitious objectives of NanoBubbles and ROR² (disclaimer: I am also member of the ROR² project) . 

Ulrich Herb: The ROR² and NanoBubbles projects share a focus on improving scientific practices, with ROR² exploring the interplay between Open Science and integrity, and NanoBubbles addressing the challenges of correcting scientific errors. Both use multidisciplinary approaches and aim to influence policy and practice to enhance trust and accountability in research. That is at least the perception—could you tell us more about the projects, your role in both, and your specific contribution to them considering your professional and academic background?

Frédérique Bordignon: The NanoBubbles project began before the ROR² project, so the two are at different stages of development. NanoBubbles started in June 2021 and is set to conclude in May 2026, whereas ROR² only launched in November 2024 and will run for three years. NanoBubbles is a Synergy project funded by the ERC with a budget of 8 million euros, while ROR² is funded by the French National Research Agency (ANR) with a budget of 276,000 euros, primarily allocated for hiring two postdoctoral researchers.

ROR² leverages a theoretical framework to tackle the challenges of a political economy of scholarly communication, shaped by digital regulations and internationalization. This draws on Information and Communication Sciences, an area I know well since I have spent over 15 years promoting Open Science and addressing various scholarly communication issues as an academic librarian. I co-developed the ROR² project with Chérifa Boukacem-Zeghmouri, professor at the University Claude Bernard Lyon (FR), who officially leads the project. Together, we will supervise the postdocs’ work in collaboration with two other researchers.

The NanoBubbles project, by contrast, spans an even broader range of disciplines, combining approaches from the natural sciences, engineering (natural language processing), and the humanities and social sciences (linguistics, sociology, philosophy, and history of science).

Since its inception, I have been a permanent member of the NanoBubbles project, collaborating with many team members, particularly on the creation and analysis of corpora. I bring my expertise in scientometrics and natural language processing. As a trained linguist, I am especially interested in academic writing. I also co-supervise two PhD students, one based in Grenoble (FR) and the other in Nijmegen (NL).

UH: Let’s talk in further detail about NanoBubbles, as this project has been running for some time. What are the institutions and people behind the project, and what are their contributions to it?

FB: The NanoBubbles project is led by four researchers from the Universities of Sorbonne Paris Nord (FR), Maastricht (NL), Grenoble-Alpes (FR) and Radboud (NL), in collaboration with 4 researchers from the CNRS (FR), the University of Twente (NL), IRIT (FR) and Ecole nationale des ponts et chaussées (FR), which is my employer.

With the recruitment of our project manager, PhD students, postdocs, and engineers, the project has grown to 25 members — it would take far too long to describe each person’s contribution! I invite you to visit our website for more detailed outputs. Here, I will try to summarize others’ research and focus more on my own work, as I am best positioned to describe it.

Work Package 1 focuses on building and analyzing corpora. This involves developing a technical infrastructure for annotation and text extraction, provided by the team of computer scientists from Grenoble. I contributed to this work package by assisting my colleagues in exploring bibliographic databases to construct corpora, whether for metadata or full-texts.

For example, we needed to build a corpus of citation contexts for a particular paper, to analyze how and why it is cited, I made up this corpus. I am also interested in critical citations — citations that express disagreement or challenge the study being referenced. After constructing a dedicated corpus and working on defining this phenomenon (which is very rare), I will continue this research with the aim of determining whether critical citations serve as an effective mechanism for correcting science.

Work Package 2 employs diverse methodologies to trace contested claims and errors. This includes studying how scientific articles present claims (particularly about nanoparticles) and analyzing their argumentative structure to better understand the propagation of claims and the consequences of citing (and relying on) retracted papers. It also involves examining future-oriented scientific predictions, investigating how scientific “bubbles” form and evolve, and developing automated systems for claim detection.

Work Package 3 addresses correction mechanisms through various approaches, such as historical case studies, automated screening tools to identify problematic papers (e.g., the Problematic Paper Screener developed by Guillaume Cabanac and Cyril Labbé), and post-publication peer review, which involves critical evaluations of published articles (resulting in over 40 comments posted on PubPeer). As for me, I wanted to learn more about researchers’ behavior when confronted with errors. I have conducted a survey among chemists. I found that 88% of them have already spotted an error in a publication, and 50% in their own papers. Results show also that researchers are quite willing to act to correct errors (their own as well). Nevertheless, the verbatims show unambiguously that good intentions are often hindered by reluctant publishers/editors rather than by shyness or fear of hurting peers, even if social relationships, and particularly the presence of a “hierarchical” link with a peer, are sometimes presented as an obstacle to taking action to correct an error (article to be published soon (hopefully…)).

Work Package 4 explores ways to enhance trust. The research examines how to communicate about scientific corrections without undermining public trust in science and improving standards for communicating retractions. With a case study, I investigated how information about a paper being retracted is disseminated across the various venues and databases. Results highlight challenges in correcting the scientific record and the need for collaborative efforts among publishers, researchers, librarians, and information providers. They also show very poor reporting in Open Access archives. I took this problem a step further by studying what role open archives could and should play in terms of correcting science and updating the scholarly record.

UH:  What motivated the focus on error correction within the field of nanobiology, and how do you see this influencing other scientific disciplines?

FB: The project focuses on nanobiology, which emerged around 2000 as an interdisciplinary scientific domain. Despite its relatively young age, the field has already witnessed several instances of claims that were later found to be overstated or incorrect. Through the NanoBubbles project, researchers are analyzing three specific cases of scientific “bubbles,” tracking the circulation of claims and counter-claims as they spread through laboratories, conferences, journals… These bubbles were suggested to us by Raphaël Lévy, one of the project’s PIs, whose unhappy experience in correcting scientific errors that he spotted as a physicist was at the origin of the NanoBubbles project. 

Our  objective  is  to  understand  how,  when  and  why  science  fails  to  correct  itself, taking  three controversies both as research fields (to which some members also contribute) and as objects of study which we think can be representative of other rapidly expanding interdisciplinary fields that attract substantial funding and attention, similar to artificial intelligence and synthetic biology.

UH: What role do interdisciplinary methods play in NanoBubbles, and how do they enhance the understanding of scientific error correction in general beyond any disciplinary focus?

FB: The NanoBubbles project has prioritized fostering interdisciplinary collaboration and communication as a core component of its mission. To achieve this, we have implemented several strategies to create meaningful cross-team synergy, including regular online meetings, biannual workshops, and an active Slack group that facilitates daily communication and exchange. Significant time during in-person meetings is dedicated to breakout groups, where participants re-organize into interdisciplinary subgroups to tackle specific projects collaboratively. This flexible approach encourages cross-fertilization of ideas between natural scientists, engineers, and researchers in the humanities and social sciences.

One standout example of interdisciplinary success is the workshop co-organized by Maha Said (a biologist) and Candida Sanchez Burmester (an ethnographer). This unique event brought together nanoscientists and scholars from Science and Technology Studies (STS) to engage in a discussion on the topic of replication and science correction.

UH: What was the most surprising, the most shocking, and the most hope-giving result from the project?

FB: Colleagues, both within and outside the NanoBubbles project, who work tirelessly to detect and report fraud, identify hundreds or even thousands of problematic papers. They file numerous reports on PubPeer and alert publishers. Yet, in the end, very few retractions or corrections result. When you consider that the cases identified are likely just the tip of the iceberg, it’s easy to feel pessimistic.

However, to take a more optimistic view, I believe research integrity holds significant meaning for the vast majority of researchers. For several years now, I have been leading integrity training sessions at Ecole des Ponts, and I’ve observed the participants’ concern and genuine interest in actions that promote more open research conducted in accordance with the principles of research integrity. Beyond this, I also see a growing awareness of the environmental impact of research and the pressing need to adapt research practices accordingly. 

UH: Let’s talk about ROR², as it focuses on Open Science, which is generally seen as an antidote to manipulation in science. How does ROR² define the relationship between Open Science and research integrity, and what key tensions have been identified?

FB: Research conducted on Open Science, both as a concept and as a practice, including research by members of the ROR² consortium, has not always connected the question of openness with that of research integrity, despite both being fundamental to scientific progress. However, the experience of consortium members and some of their work show that openness alone is not a sufficient condition for producing research that adheres to principles of integrity: Openness considered in isolation can be the starting point for misuse, misconduct, or even fraud.

Integrity issues in research and publication practices interact with both implicit principles of Open Science (like values and norms) and explicit principles (mainly policies, infrastructures, and also assessment). Thus, the crucial challenge in contemporary studies of Open Science practices is to understand whether these practices are founded on more virtuous values, constituting a culture of responsible research that can guide it (or even transform it, let’s hope so!), toward principles of honesty, rigor, and transparency that align with those of research integrity.

The consortium we have set up is a sort of dream team made up of both researchers and professionals in scholarly communication, some of whom are both, starting with you, Ulrich! And myself, who have this dual profile. We are fortunate to have several researchers in information and communication science, but also a sociologist with expertise in scientific publishing, a philosopher of science whose expertise is in ethical issues, a medical doctor involved in the fight against fraud, a geographer and an economist whose skills in scientometrics will be invaluable to us, and 2 professional experts in scholarly communication who have been working for many years to open up science and are familiar with researchers’ practices. Finally, a private partner, Datactivist, which is very committed to Open Data, will be helping us to process the data and design our DMP.

UH: Can you elaborate on the taxonomy of research platforms that ROR² is developing and its significance for understanding publishing practices?

FB: The aim is to examine a wide range of platforms (publishers, social networks, data repositories, etc.), collect all the necessary information about them and identify their strategies, their discourse, their economic development and partnerships. The use of natural language processing tools to identify patterns useful for classification may also be considered. WP3 aims to identify when Open Science and research integrity practices align and in which contexts they fail to do so. For example, we hypothesise that there is only partial alignment of practices when we detect a phenomenon of washing (derived from the greenwashing concept), in other words advertising effects that are distortions of the principles of Open Science (Open Washing), but also of the principles of research integrity (Integrity Washing) that take up the markers and flavours of integrity without being able to claim the values.

UH: What long-term impact would ROR² ideally achieve in transforming the culture and policies surrounding research integrity and Open Science?

FB: The long-term impact of ROR² can be envisioned through proactive recommendations for promoting research integrity rather than merely reacting to problems. The project will help illuminate strategies and policies that leverage Open Science to enhance integrity. Through its recommendations, ROR² will improve research practices and their implementation on platforms. This work will contribute to discussions around research evaluation reform, particularly in the French national context and through COARA. ROR²’s impact will be amplified through discussions with specialist networks (COAR, ENRIO, …) who can provide insights on implementing the recommendations in practice.

UH: Dear  Frédérique Bordignon, thanks so much for giving us this profound insight to the ROR² and NanoBubbles projects and for demonstrating how interdisciplinary approaches can enhance transparency and integrity in science. Let’s see how both initiatives will address critical challenges such as error correction and “open washing” while paving the way for sustainable and responsible research practices.

By Ulrich Herb

Graduate sociologist, information scientist (PhD degree), associate of scidecode science consulting – De Castro, Herb, Rothfritz GbR, working for the Saarland University and State Library (Germany)

Leave a Reply

Your email address will not be published. Required fields are marked *