After a few days of peace and quiet, the folks in my lab are starting to trickle back from the Society for Neuroscience annual meeting. Today I had this conversation with my grad student after he returned from the meeting:
“OH MY GOD!!! We need to get my paper published NOW, there was a poster at the meeting describing the exact same experiment as mine, they’re breathing down our necks and… they’re German!!”
“Whoa now, Hobo Joe, tell me what the poster was about…”
“Well they are working on the same neurological condition we are and they also applied the treatment twice and the results were the same as mine, and they did the same thing, with the double treatment and we’re TOTALLY SCREWED!!!”
“Yes, but what was the point of their experiment, what was the question?”
“OK, let’s look at the abstract, what lab was this?”
“They were German.”
So we finally find the abstract and read it over, calmly.
“Dude, did you actually read the poster, or did you just read the title, the first panel and then freak out?”
“Because this has nothing to do with what you are working on, at least not more than in a very superficial way. Sure, they are working on the same condition like another three-thousand people are, and they did apply the treatment twice, but they’re not even close to doing anything that resembles your project.”
“Um… but they were… German…”
This led me to think about scooping. Not the kind that I do after I walk my little mutt every morning, but scientific scooping. The question is, how likely is it really that one’s project might get scooped by another lab, and more importantly, does it really matter? In certain types of scientific research where there is a single answer to a question, such as a crystal structure to a protein, or a specific receptor to a ligand or a gene sequence, I guess scooping would matter. Particularly if you want to be able to say, “Wooohooo! We were the first to describe the structure of blah, blah, blah…” By this point, the contribution of the lab coming in second place would not be that significant. But most science is not really like this. Usually, one is either describing a complex process, or testing a series of interrelated hypotheses, and there is lots of room for different approaches, methods and even answers. Unless it is a very low-hanging fruit, the chances that two labs are doing the same experiment in the exact same way is extremely low. And in fact this is a good thing, because (assuming both labs are competent and good at what they do) that means that the different labs will validate each other’s results, using slightly different approaches, making the discovery more likely to be significant and true. And if one lab does it better than the other, then even if that lab published second, that lab’s paper is likely to become the go-to citation for that finding.
I remember when I first joined PhD lab, my advisor showed me a paper with a very interesting finding, published in a very crappy (and I mean crappy) journal, with somewhat crappily-done experiments. He said, “to start, why don’t you repeat these experiments?” I said, “but why, they’ve already done it?” He replied that “yes, but it is a potentially very interesting finding which, if true, could change the way we think about our field, but I don’t really believe the results. If you are able to replicate them, and extend them, not only will you kick-ass thesis project, but will be able to get some kick-ass publications.” And he was right on both counts. And I was surprised that we were able to publish this in a high profile journal even though we explicitly stated we were not the first ones to discover this phenomenon.
So if you find out a competing lab is working on a similar project, should one be worried about being scooped? In most cases, if you are working on a complex enough problem, I think its worth making sure your science is as strong as possible, rather than rush to publish something that is suboptimal just to beat a competitor. Because in the end the best science will have the most lasting power, and nobody will care if it came out in 2010 or 2011. If you are chasing down a receptor or a really obvious experiment, or your competitors are German, then maybe you should panic.
As far as my student, he should get his paper out anyway. Maybe I should have let him panic a bit more.