Spreading preprints in population biology

Ewen Callaway reports on the increasing use of the arXiv preprint server by geneticists and biologists: "Geneticists eye the potential of arXiv". With the near-arrival of the PeerJ system, which promises to seamlessly integrate preprints and pre-publication review with ultimate publication, this is a very timely story. Last week I pointed to the new paper on arXiv by Joseph Pickrell and colleagues, and there have been a few other notable ones recently.

But Ginsparg says that pre-publication is more likely to stop scientists from being scooped. In many physics fields, publication on arXiv is what counts for claiming priority, and journal reviewers can use the server to check that discoveries are correctly attributed. An authoring history that accompanies all arXiv papers also allows scientists to arbitrate disputes over priority. In the 21 years since arXiv began, Ginsparg has seen astrophysicists, computer scientists and others go from sceptics to devotees. Once a community adopts arXiv, it never seems to relinquish it, he says.

What readers probably don't know is that I have been experimenting quietly with preprints for the last year -- not front-paging, but putting up to make available and allow me to use the bibliographic system. Several of our in-progress manuscripts are online here on the blog and discoverable by Google, as are preprints of some of my published work. I've been motivated to publish preprints of published papers because copyright agreements generally do not allow authors to post final PDF versions, but do allow posting either pre-review or pre-publication manuscripts. The most frequently read preprint here is my 2008 book chapter, "From genes to numbers: effective population sizes in human evolution"

More to the point, I posted one of my own preprints on arXiv last year, regarding shrinking brains: "Selection for smaller brains in Holocene human evolution". I wanted to know how widely a biology preprint would be read, without promoting it myself which would skew the numbers. The results have been interesting. The paper got an initial mention on the physics arXiv blog but little attention otherwise.

That is, until the last few months. I've had a dozen requests from colleagues to cite the paper (which anyone is welcome to do by using the arXiv number). I also had two great interactions with colleagues who had comments and suggestions on the preprint, which I am now incorporating into a revision. So presubmission review actually does work, when the paper comes to the attention of the right people. But without promoting the preprint, that feedback won't happen for a while.

Recent events suggest that many population biologists may be ready to go to the arXiv. I think we should do everything we can to encourage this trend.