30 June 2014

Claim for unknown human species based on unknown data

It’s weird how social media works sometimes. This appeared in two of my social media feed this morning, Facebook and Quora:

Seeing this in two places simultaneously, I assumed this was news. I Fucking Love Science links out to a longer article from November 2013. So right away, something that seemed to be new was not. As far as I can tell, I Fucking Love Science posted the picture yesterday / today, so they are responsible for the news necromancy.

Then, the I Fucking Love Science article reveals that this is based on research presented at a conference. Conferences are great, but the research presented at conferences is rarely peer reviewed. So people make all sorts of claims at conferences that may – or may not – withstand the deep scrutiny that extraordinary claims get. This is why publishing in peer reviewed journals or depositing data is valuable.

Similarly, the data presented at conferences tends to be work in progress. That preliminary analyses may not pan out. I’ve presented a lot of stuff at conferences that I wasn’t able to convert into papers.

Worse, people who were not at the conference don’t have any way to evaluate the claims.

That we haven’t heard anything on this since November of last year suggests that the evidence might be a bit preliminary.

The I Fucking Love Science article doesn’t do us any favours by not even mentioning the authors of this research. One link out (misleading linked to text saying “Royal Society”) leads to a Nature News piece. Based on that, part of the team seems to be David Reich. He posts PDFs of his papers, and none seem to list anything about the unknown human.

But I Fucking Love Science presents this conference research as an established fact on the (very simple and easily shared) image, with no link or date or context on the image. This can make the rounds on social media for a long time, even if it never makes it into a proper scientific journal.

There is a lot to learn from the successful formula of I Fucking Love Science. Pictures get shared; see the data from Google Plus below:

People interested in spreading their science shouldn’t just work on their sound bites. They should work on their social media meme images.

Incidentally, this claim might not be that wild. We already have a candidate for that “third, unknown species”: the lineage that includes Homo florisiensis, a.k.a. “the hobbits.” They are a third human species. But we don’t have any DNA sequenced from H. florisiensis, so if there was interbreeding between them and modern humans, we wouldn’t have a genetic signature to compare it to.

Update: On Quora, Ian York found the peer reviewed paper that I couldn’t. It’s “The complete genome sequence of a Neanderthal from the Altai Mountains,” published back in January.

External links

Interbreeding Among Early Hominins
Mystery humans spiced up ancients’ sex lives

No comments: