It is an exciting time in scientific publishing. Initiatives such as digital identifiers for authors through ORCID, more granular recognition of collaborative work with standardized language for specific roles with CRedIT, and more competition in the Open Access publishing world benefit researchers and move the scientific endeavor toward a more transparent and accountable future.
Yet, the write up and publication of results is one of the most challenging aspects of the endeavor, with peer review and reproducibility at the heart of this stage of the research lifecycle. We have previously acknowledged on The Official PLOS Blog that the public
“relies on the belief that content published in peer-reviewed journals is trustworthy, despite the fact that this is too often not the case.”
We have also acknowledged that we must do better: all stakeholders, including publishers, are accountable. Although the overall concept of peer review is an accepted form of quality control and valued by the scientific community, in practice it suffers from imperfections that prevent it from achieving that one great thing: advancing research communication.
In a thoughtful consideration of Truth in Science Publishing: A Personal Perspective, Thomas Sudhof eloquently describes peer review and reproducibility as flawed checkpoints that impair the “validity of published scientific results” and impede trust in science.
As a recipient of both the Nobel Prize and the Lasker Award for his work on synaptic transmission, Sudhof brings perspective and integrity to his thought leadership. Highlighting hidden conflicts of interest, too little accountability for journals and reviewers, and lack of competition between journals as three problems with peer review that have “corrupted the process, decreasing its value,” Sudhof endorses more transparency in the peer review process to reduce bias.
At PLOS there are a range of ways to improve the process without diminishing those aspects that the community values. Current tools and systems that address these limitations include the posting of research to preprint servers before formal publication, to enable researchers to improve their work and share it earlier. There is an opportunity to improve review forms that may be cumbersome or insufficient to provide thoughtful and constructive feedback to authors. Appropriate and rigorous reviewer and editor training can help to mitigate potential reviewer bias and mentor early career researchers. With improved technologies and processes, publishers have an opportunity to improve efficiencies, quality, trustworthiness and authenticity of the process.
As for reproducibility, Sudhof outlines increasingly complex experiments that are impossible to reproduce, “tweaked or selected” results that do not hold up with repetition, lack of validation of reagents and methods, and the “near impossibility” of publishing negative results as contributors to the problem.
Providing opportunity to showcase peer-reviewed articles that address the reproducibility issue is an important value of PLOS and PLOS ONE; the journal welcomes submission of negative, null and inconclusive results. PLOS Biology’s Meta-Research section welcomes experimental, observational, modeling and meta-analyses that address research design, methods, reporting, verification or evaluation.
PLOS Biology and PLOS Genetics authors can contribute to the reproducibility effort by identifying model organisms, antibodies or tools with a unique Research Resource Identifier (RRID). PLOS is a part of the Research Resource Identification Initiative, a cross-publisher effort to promote reproducibility in science and enable effective tracking of the use of particular research resources across the biomedical literature.
PLOS works toward a future where research is published without unnecessary delays, and continual assessment and commentary is provided by a robust and ethical system of visible, engaged pre- and post-publication peer review. We strive to engage a global editorial and reviewer contributor community, appropriately trained, recognized and incentivized. With regard to journal-facilitated peer review, rigorous input from experts in a relevant field of research is highly valued by both authors and readers, and contributes to trust of research results for working scientists, clinicians, patient advocates, policymakers and educators.
Addressing the issues and challenges that perversely incentivize unreliable research or prevent peer review from achieving its scholarly ideal will not be easy or quick. The challenges are substantial and the solutions must be as well, and satisfy a diverse researcher and stakeholder community. Broader adoption of reproducibility efforts and better recognition for the range of contributions made by researchers and reviewers will not be enough without the engagement of early career researchers, junior investigators and senior leadership with the power to influence change.
Image credit: one-vibe, pixabay.com