Preprints and Open Access Research: Trends and Implications

The scientific record has never moved faster — or been more accessible to anyone with an internet connection. Preprints and open access publishing have reshaped how findings travel from a researcher's desk to the broader world, compressing timelines that once stretched to years and removing paywalls that once locked out most of the planet. This page examines what these models are, how they operate in practice, where they fit into the broader landscape of scientific publishing and journals, and the real trade-offs researchers, institutions, and readers face when navigating them.

Definition and scope

A preprint is a version of a research manuscript posted to a public server before — and sometimes instead of — formal peer review. The author submits the paper, a basic screening checks for obvious problems (spam, offensive content, off-topic material), and the manuscript becomes publicly visible, typically within 24 to 72 hours. No journal editor has weighed in. No anonymous reviewers have pushed back on the methods. The work is, in the authors' own framing, complete enough to share.

Open access (OA) is a related but distinct concept. It refers to research that is freely readable without a subscription — but it says nothing about whether that research was peer reviewed. Open access can describe a preprint on arXiv, a polished article in PLOS ONE, or a paper in a traditional journal whose authors paid an article processing charge (APC) to make their accepted manuscript publicly available. The Provider Network of Open Access Journals (DOAJ) indexes more than 20,000 peer-reviewed OA journals as of its publicly maintained registry, giving a sense of the field's scale.

The two concepts intersect constantly, which is precisely where confusion tends to breed.

How it works

The major preprint servers are domain-specific and operated by nonprofit or institutional entities:

1. arXiv — Operated by Cornell University; covers physics, mathematics, computer science, quantitative biology, and related fields; hosts more than 2.3 million papers (arXiv.org).
2. bioRxiv and medRxiv — Run by Cold Spring Harbor Laboratory; bioRxiv focuses on life sciences broadly; medRxiv covers clinical and public health research and applies additional screening given the potential for direct health impact.
3. SSRN — Originally social sciences, now broader; owned by Elsevier but free to post and read.
4. Zenodo — Operated by CERN under the European OpenAIRE program; accepts research outputs across disciplines including datasets and software.

On the funding side, the US federal government has been a major driver of open access mandates. The Office of Science and Technology Policy (OSTP) issued a memorandum in August 2022 directing all federal agencies to update their public access policies so that peer-reviewed publications resulting from federal funding are made freely available immediately upon publication — eliminating the previous 12-month embargo (OSTP Public Access Memo, August 2022). Agencies including NIH, NSF, and the Department of Energy are implementing this directive through their own updated policies.

For researchers, the typical open access route involves either publishing in a fully OA journal (gold OA) or depositing an accepted manuscript in an institutional repository after peer review (green OA). Gold OA often involves APCs ranging from a few hundred dollars to more than $11,000 at some high-impact journals — a cost that falls unevenly on researchers without institutional support.

Common scenarios

Pandemic-era preprints became the most visible illustration of both the power and peril of the model. During 2020 and 2021, thousands of COVID-19 preprints circulated on medRxiv and bioRxiv before peer review. Some shaped public health guidance. Others contained significant errors that reviewers would have caught. The National Institutes of Health acknowledged using preprint findings in its guidance development while simultaneously cautioning the public about their preliminary status.

Funding agency compliance is another common scenario. A researcher at a university receiving NIH funding must deposit accepted manuscripts in PubMed Central within the timeline specified by NIH's updated Public Access Policy (NIH Public Access Policy). Failure to comply can affect future funding eligibility — making open access less a philosophical choice and more an administrative requirement.

Journal embargo conflicts arise when a researcher posts a preprint and the target journal's policy restricts prior public availability. Most major journals, including those in the Nature and Science families, permit preprint posting under specified conditions, but policies vary enough that researchers working across research design and methodology fields must verify each journal's stance before submitting.

Decision boundaries

The central tension in preprint culture is speed versus validation. Posting early maximizes visibility and establishes priority — in competitive fields, being first matters. But unvetted findings that attract media attention before peer review has identified flaws can mislead the public, policymakers, and other researchers. The replication crisis already demonstrated how peer-reviewed work can fail at scale; preprints add another layer of complexity to that picture.

Three practical distinctions help clarify when each model is appropriate:

• High-stakes clinical claims carry a different risk profile than theoretical physics derivations. medRxiv applies additional screening; arXiv applies minimal filtering. The domain sets the floor.
• Preprint plus eventual journal publication is now the majority pathway in life sciences and physics — the preprint is not a substitute but an early broadcast.
• Gold OA versus green OA involves a cost-versus-access trade-off. Gold OA maximizes immediate public reach but often at significant author expense. Green OA is free to authors but may involve embargo periods, depending on the publisher.

The broader infrastructure supporting open science — including research data management standards and federal research funding agencies mandates — continues to push the field toward openness as a baseline expectation rather than an exception. Whether preprints settle into a permanent role as the first public version of every paper, or remain a contested middle ground, depends largely on how institutions and funders calibrate their incentives over the next decade. That calibration is already underway. For a broader orientation to the scientific enterprise, the National Science Authority index covers the full range of topics from methodology to research careers.