A “bottoms-up” approach to scientific publishing

A few days ago, a piece was posted at ASAPbio which reminded me of an idea I had a long time ago about a new model for publishing and peer review in the life sciences.

I described my idea on Twitter, but I decided to provide a little more detail about it.

Was reminded today of an idea I had some time ago about having a bottoms-up approach to cascading, instead of the current top-down one. Key point is the focus on scientific soundness only, and ease of publishing. Check thread below! https://t.co/OJXN02Kkc9

— Alejandro Montenegro (@aemonten) January 21, 2018

My idea was based on submitting to a suite of journals, one that would include both a sound-science title (like PLoS One, Health Science Reports, Biology Open, PeerJ, and others) and journals using ‘significance’ or ‘impact’ as a criteria for publication (or any other type of journals; the more, the merrier). These journals may differ on a number of issues besides acceptance criteria.

This idea is based on a “bottoms-up” approach, which differs from the current “top-down” cascading strategy normally employed by authors and publishers. By top-down, I mean submitting to a high profile journal and then having to ‘cascade’ down to other journals until you find a home. In many cases, the last part of that ‘journal-shopping’ journey is a sound-science title. Here, I propose the opposite, where you would start there.

In my proposal, you would submit your manuscript to the suite. There, it would be evaluated, through standard peer review (which could include some nice modifications), for scientific and technical soundness only. In a way, this means you would be submitting to “the bottom” title. If the manuscript, after revisions when necessary, is determined to be scientifically and technically sound, it will be accepted for publication “in the suite”.

What does that mean? It means that after acceptance, your manuscript will be published, by one of the journals in the suite. After acceptance based on objective criteria (technical soundness), the editors of the other journals in the suite can take a look and see if they are interested in the manuscript for their journals. The more selective titles in the suite -i.e. those interested in featuring groundbreaking reports and studies that they think “significantly advance a field- will evaluate the already accepted manuscript, and if the topic is of interest to them, they can offer the author to publish it in their journal. The author will be able to choose where to publish if many offers are received. If none of the other titles in the suite are interested, or if the authors decline every offer, then the manuscript is published in the sound-science journal.

A crucial aspect of this proposal, is that after acceptance based on technical soundness, the other journals in the suite can no longer request for additional experiments: the manuscript has already been reviewed and accepted. This would eliminate one of the complains authors usually have: “they only asked for this experiment because I submitted to [insert glam title]; if I had submitted to [insert lower tier title], they wouldn’t have asked for this“.

Now, the suite of titles can have any numbers of origins. It could be ‘publisher-based’, ‘society-based’ or just a regular centralized system (connecting authors to many titles  from different publishers), as long as the peer review is done with the criteria outlined above.

I think that uncoupling the required revisions from the journal the manuscript is submitted to, is an important step to take. By this, I mean that reviewers should not be requesting experiments based on the journal the manuscript was submitted to; it either is scientifically and technically sound or it isn’t. If it is, then editors of specific journals can then offer to publish it based on subjective criteria like ‘significance’ or ‘impact’. That’s fine, that’s their prerogative: they can choose what they want to publish, but this should not delay communication of technically sound manuscripts. Preprints are a way to try to avoid the latter though, and I’m all for preprints, as you can see in the pic on the right. In my proposal, manuscripts could also be deposited as preprints; disclosure and “initial validation” are different things.

Point is, a fair, ‘sound-science’-based peer review, independent of the journal, is probably a step in the right direction. People might argue though, that journal-independent platforms (e.g. Axios) have not worked, as journals nevertheless asked for additional reviews, from reviewers they liked and trusted. Here, that wouldn’t exist: journals in the suite would not ask for additional revisions after the initial peer review. The suites would be created among journals and editors who trust each other. In some of the other ideas that have been proposed (e.g. the “Peer Feedback” proposal), however, the aforementioned problem of independent review is not fully addressed, although they mention that ” reviewers will be selected through partnerships with several scientific societies”; if the journals in the ‘Peer Feedback’ project would take the reviewed manuscripts without asking for additional experiments, then both of our proposals would be quite similar. Indeed, in the “Peer Feedback” proposal they state that “journals may indicate whether they will accept [the revised manuscript] in its revised form or request additional experiments or peer review”.

I was excited to read something similar to what I’ve been thinking of; as I said, ideas of this kind are, in my opinion, a step in the right direction.