Authors and institutions need to acknowledge problems with published studies when they are found.Credit: Getty
Scientists rightly want to be and should be credited for discoveries and inventions. But in addition to receiving credit, scientific authorship also means being accountable if issues arise. It is a matter of responsibility and transparency.
Retractions are part of science, but misconduct isn’t — lessons from a superconductivity lab
Authors need to acknowledge errors in their studies that have come to light after publication, for instance, or if their published findings could not be verified subsequently. Institutions and journals have clear responsibilities, too, to correct the publication record when needed. As we have argued previously, this process works best when all parties cooperate.
The process of correcting the scientific record often raises difficult questions, particularly when it comes to retractions. For example, what are the respective roles of institutions and journals in the process of retractions? How should authors distribute, and accept, responsibility when a study has been retracted if data have been manipulated or if there are concerns that they might have been manipulated?
Updated guidelines
The Committee on Publication Ethics (COPE), a non-profit organization based in the United Kingdom that advises on standards in scholarly publishing, has been considering these questions and updated its guidelines for retractions in August 2025 (see go.nature.com/496corh). The new guidelines reiterate the previous guidance in highlighting that “authorship requires joint responsibility for the integrity of the reported research”.
They also restate that “if retraction is due to the actions of some but not all of the authors of a publication, the notice of retraction should state this if possible”. But the updated guidance now also states: “This approach would only be appropriate if an institutional investigation concluded that a specific author or authors were responsible for the errors or actions. The retraction notice should reference the institutional investigation.”
Nature has followed these updated principles for a study on lung cancer immunotherapy that the authors have now retracted formally (K. W. Ng et al. Nature 616, 563–573 (2023); retraction https://doi.org/10.1038/s41586-026-10104-7; 2026).
Science communication will benefit from research integrity standards
The retraction follows an investigation by the authors’ institution, the Francis Crick Institute in London. The investigation found that in some instances, the study data had been manipulated, could not be reproduced or had been included or excluded selectively (see go.nature.com/4jd1nz6). The authors have apologized to the research community and say that they are committed to revising the study.
The Crick investigation identified one author, who has been named in the institute’s report, as being responsible for these analyses. This author is also named in the retraction. We are doing this in line with the revised COPE guidelines following the conclusions of the institutional investigation.
We have also done so because, in science, credit comes with responsibility. Researchers agree to follow a set of behaviours that should amount to the highest standards of integrity. This is what trust in science is built on. If an investigation shows that those standards have fallen, or if that trust has been betrayed, then it is only right that all those responsible should be held accountable by the community.
Joint responsibility
In the current era of team science, researchers are right to be concerned about what the impact on their careers and reputations might be if one or more group members compromise a study’s research integrity.
The Crick also deserves recognition for acting swiftly with its institutional investigation and communicating the results in confidence to Nature’s editors. It is regrettable when there is insufficient cooperation between institutions and journals leading to protracted, parallel investigations.
Usually, when journals perform their own investigations they do so without access to the results of an institutional investigation, if there has been one. We have argued before that institutional investigations are necessary to assess cases of possible misconduct. Improved communication between institutions and journals is essential, too.
A ten-year drive to credit authors for their work — and why there’s still more to do
Sometimes, universities can be reluctant to investigate. Such reluctance or delays in investigating cases can stem from concerns about reputational damage to institutions or to their research groups if wrongdoing is uncovered. But the damage to reputations from inaction can be just as bad, if not worse. It is much better to be transparent and acknowledge any mistakes that were made. We hope that institutions are encouraged to act when necessary, and to do so quickly, according to the COPE guidelines and with journals committed to working with them.
Scientific authorship is a two-sided coin. Credit goes hand in hand with responsibility, which is essential for transparency and maintaining trust in science. That responsibility includes acknowledging when things don’t go right.
Retractions are part of the scientific process, and we will continue to follow the updated COPE guidelines if and when the need arises in future.
