The preprint is increasingly recognised as an effective vehicle for the early communication of research findings. It offers a solution to the long delays and suppression of research associated with the traditional scholarly publishing model. It is not an alternative to peer-reviewed publication, but complementary to it, and can combine effectively with a more transparent and equitable process of peer review.
A preprint is a version of a scholarly paper before submission to a journal and peer review, made publicly accessible online via a preprint server or a preprint publisher's platform. The preprint is usually assigned a unique identifier, such as a DOI, so that it can be cited, and may be indexed in some search databases (for example, Google Scholar). The preprint platform may offer a public comment function, enabling a kind of informal open peer review. There is no cost to the author to publish a preprint.
The preprint, as its name suggests, is typically an early version of a paper destined for formal publication in a journal. The main functions of the preprint are:
The majority of preprints are published using preprint servers. These are typically run on a not-for-profit basis by academic stakeholder organisations. The original preprint server, arXiv, established in 1991 and widely used by the physical and mathematical sciences communities, is owned and run by Cornell University; the immensely popular bioArXiv is run by Cold Spring Harbor Laboratory. Other servers gaining traction include the subject-specific ChemRxiv, EarthArXiv and PsyArXiv. There are also and generalist services, including PeerJ Preprints (owned and run by the publisher PeerJ) and OSF Preprints. No doubt more preprint servers will become established in the scholarly communications landscape over the next few years.
Publishing a preprint is in the majority of cases a stage on the journey to peer-reviewed publication. Many journals will happily consider articles that have been previously posted as preprints. This includes all journals published by Springer-Nature, most of those published by Elsevier, all the PLOS journals, and many others. Authors can very easily check whether their journal of choice allows preprint publication by looking up the journal’s policy in SHERPA/RoMEO.
In fact, making an early version of a paper available for public comment provides an informal peer review function: it can help authors to identify weaknesses and errors in their research, and improve the quality of the research output. This in turn can increase the chances the paper will be accepted by the authors’ journal of choice and that it will have a smoother journey through peer review and editorial processes. So preprint servers can actually help to raise the quality of material submitted to and published by journals.
An interesting take on the preprint as a vehicle for improving research quality is the concept of the preprint journal club.
Journal clubs, in which participants discuss published papers, offer a well-established mechanism for students and early career researchers to learn how to construct and evaluate research papers. By extending this model to preprints and using online peer review platforms such as PREreview and The Winnower, trainee researchers can peer-review live research papers, publish their reviews, enter into dialogue with the authors, and see their comments inform papers as they evolve towards a final version.
This turns the journal club from a discussion group into a practical real-world activity, making the process more engaging to the participants and giving hands-on training in the art of constructing a good peer review. Because the reviews published on sites like PREreview and The Winnower are assigned DOIs, they are also citable outputs that can be added to CVs.