Policy, Preprints, and Peer-Review

Since winter break, I’ve toyed with these two questions as a pair:

1. What if research funders prioritized the deposit of preprints (and let go of immediate open access for peer-reviewed articles) as suggested by Plan-U?

2. What if publishers (like ACS) steered authors with policy-driven OA statements toward depositing their preprint in an approved server (like the ACS-supported ChemRxiv)?

This happens in pockets (CZI, eLife), but what if were the norm? In the following post, I’ve written out where asking these questions has taken me.

Enjoy!

Proposal & Response

The Institute of Physics (IOP) Publishing, the American Institute of Physics (AIP) Publishing, and the American Physical Society (APS) wrote a joint response to the cOAlition S’s proposal “Towards responsible publishing” for the Scholarly Kitchen in mid-January. The proposal and response offer a lot to consider, but there was one particular section that felt very notable

“The new iteration of Plan S contends that the current publishing system causes unnecessary delays in sharing research outputs,” wrote the publishers. “However, preprint platforms such as arXiv have been enabling free, informal dissemination of research since 1991.”

The cOAlition-S proposal describes the current state of publishing as not yet meeting the full potential of peer review and that peer review takes a needlessly long time, yet many funders insist that the version of the paper they want to see made publicly available is the one that’s undergone this slow and suboptimal process.

In their response, the publishers tout the importance of their peer review and editorial checks. And for better or for worse, organizing a system of manuscript feedback for authors is a service that publishers provide. There is justifiable debate over who should pay and how much, and there is also a reasonable claim by publishers who wish to be paid for the services they provide.

So far, in response to policy pushes for immediate open access to papers that have undergone peer review (very often organized by a journal), we have seen an increase in paid publishing options like direct author fees and system-wide contracts which might be helping readers, but this state of affairs does little to increase equity for the authors of these works. Even the route in which authors deposit their accepted manuscripts has become a paid fee option. These charges are a barrier for authors from low-income backgrounds as well as from high incomes.

Meanwhile, it is free for authors to deposit a preprint on a server as immediately as they want to make it available for readers. IOP, AIP, and APS explicitly point this out to in their Scholarly Kitchen post. There is usually no cost-barrier to depositing a preprint or any practical time barrier. Once vetted by a server, a preprint is posted, without further delay by reviewer feedback or cash deposit.

Many researchers, especially those in the early career stages, have expressed a desire to share preprints ahead of peer-review, to speed dissemination, but reportedly don’t always do so for fear of harming their chance at eventual publication in a peer-reviewed outlet. This is not the case. Many desirable publishers advertise fairly liberal policies that permit preprinting manuscripts ahead of submission.

There is conflict between some current funder policies and some publisher business models. There is also alignment between what researchers want and what publishers allow. At the risk of naivety, is there not room here for compromise?

Remember Plan-U?

In 2019, Sever, Eisen & Inglis proposed Plan U, which is the idea that universal access to scientific and medical research could be brought about by funder preprint mandates.

An excerpt from the abstract:

If all funding agencies were to mandate posting of preprints by grantees—an approach we term Plan U (for “universal”)—free access to the world’s scientific output for everyone would be achieved with minimal effort. Moreover, the existence of all articles as preprints would create a fertile environment for experimentation with new peer review and research evaluation initiatives, which would benefit from a reduced barrier to entry because hosting and archiving costs were already covered.

As cOAlition-S puts it, the “coupling of editorial gatekeeping with academic career incentives is damaging science.” Funders (like cOAlition-s) have the opportunity to help decouple editorial gatekeeping from career incentives by shifting priority to preprints. Through their current priority on the peer-reviewed versions of papers, funders are part of the reason there is a chokepoint on peer review and mounting fees for publishing.

The WHOST wrote that “using preprints as a demonstration of productivity may ease the pressure to publish in journals.” If funders only asked for authors to preprint instead, it could lessen pressures on those seeking peer-review publication by providing an alternate recognized route to establishing primacy of ideas as well as an alternate currency toward future funding requests. (Fewer papers in the peer-review pipeline could also shorten the queue for other papers that choose that pipeline.)

Shifting policy away from outputs that represent publisher revenues and putting those policies toward outputs that publishers have indicated do not pose a financial threat could help cap the current spiraling costs for authors, libraries, institutions, and funders themselves.

Peer-Review Opportunity

If “the full potential of peer review is not realized”, as suggested, then it is doubly in the cOAlition’s interest to prioritize preprints. If we want peer review innovations to succeed, we shouldn’t deprive would-be innovators of their stock: unreviewed papers.

When we ask researchers for peer-reviewed papers, we get more peer-reviewed papers; if we ask for preprints, we may just find that we have more preprints. With more papers appearing first or only as preprints, the opportunity increases for open competition among reviewing individuals and organizations.

It’s possible to review journal-published papers after they appear, but that’s a lot less rewarding or valuable than getting the chance to comment or review preprints. Preprints offer the opportunity for individuals to be the first to pluck exciting new findings out from the crowd and share them with a following. Post-print review most often favors finding the bad that’s slipped through the established filters, a practice which may have more of a limited appeal than assumed.

Winners & Losers

In a report from the Brief, Clarke & Esposito react to the cOAlition-S proposal that “researchers are expected to release their outputs whenever they choose to do so [and] ...at some point in the process... choose to send those outputs to (as of yet, largely non-existent) peer review services, and then maybe, later, if they feel like it, submit them to curation services that will choose their favorite articles in the form of overlay journals.”

If researchers release a preprint to fulfill a policy requirement, it's possible they may not ever submit that manuscript elsewhere if they feel assured that this will not hinder future funding.

As Clark & Esposito write, most researchers "think of publication as a necessary distraction that takes time and effort away from doing actual research." Preprinting requires less time and effort than traditional peer-review publishing, so maybe researchers would choose to leave it at a preprint and get back to researching.

On the other hand, their employing institutions may not be so up to speed and still look for external certifiers.

While alternative review and curation services may not yet have any declared winners, the traditional review and curations services--academic & scholarly journals--will still (presumably) be there. That may be bad news if your wish was to defeat journals, but pretty good news if your wish was only to remove some of the more notable harms that arise from the journal system.

No, implementing something like Plan-U won’t necessarily defeat journals. In fact, there are probably benefits for publishers. As Johnson and Chiarelli concluded, “the biggest threat to academic publishers will come, not from preprint servers, but from other publishers that do a better job of addressing authors’ desire for accelerated dissemination, feedback and scholarly credit.”

Publishers Relationship with Preprints

And so, publishers are making investments in preprinting and integrating preprint sharing into their submission process. Motivations may include the chance to offer novelty over competitors, lock in intellectual investment, and, well, the chance to commodify preprints (while simultaneously discouraging the deposit of author’s accepted manuscripts).

Whether or not the motivations are self-serving (which they most assuredly are), the headline is that publishers have dipped their toes and may be willing to plunge deeper.

You can imagine other ways this would help publishers, and eventually the public, including time savings and transparent assessment. If a paper were found to be truly abominable, a publisher could elect to note that permanently on the preprint page, which could save precious editorial resources for sister journals, and the public. While under review, appropriate public reviewers and commentators might also become more quickly identified. Or, maybe they monitor public comment for red flags that assigned reviewers might miss.

Once published, the journal and preprint server can communicate to better connect the versions. And in doing so, a journal could also note to preprint readers not only if/when there is a newer version, but create an alert if there was a substantial change that occurred between the two versions that make it imperative to read the newer version.

That alone would also create a signal for the value of peer review, or lack thereof.

What About OA?

So what, then, about open access to journal’s peer-reviewed versions?

As an open access advocate, speaking for myself solely, I would be satisfied not having immediate open access to (otherwise paywalled) works under two conditions.

The first condition is if preprints were assured to be deposited, making them available months before the public would ordinarily get to see the results of research.

The second is if the (paywalled) journal made their articles voluntarily open access after 12 months; this would not bar other routes to immediate free reading or OA, just that no later than one year, all papers would be free.

Whether or not funders would agree with my perspective is unknown. I hope a few will read this and consider. I, like some, am not aware of any business models put forth by cOAlition-S (or OSTP) for how non-Diamond OA publishers should change their business models. But that may not ever be necessary.

If funders went forward on a Plan-U type policy model that made its focus on preprint deposit, it seems like there is at least some chance of a workable compromise between the goals of funders (increased openness), researchers (speed of communication), and publishers (continued solvency).

We’ll see!

_________________________________________

This post appeared originally on LinkedIn at https://www.linkedin.com/pulse/policy-preprints-peer-review-arthur-aj-boston-o0qwc/

***************************

Like this content?

Consider supporting: https://buymeacoffee.com/arthur.boston

Or leaving feedback: https://forms.gle/eDiiYfeEMNciYdJr5