Friday, August 05, 2011

Subparallel Research

Let's say you heard a rumor that another group of researchers was working on a possibly identical, or at least very similar, project to your own research. You had both been working on this project for about a year, and had nothing published yet, not even a conference paper or abstract.

Would you:

- Contact the other group, seeking to open lines of communication? Possible motivation for this approach includes a desire to minimize overlap, share resources, and avoid negative consequences for students involved in the research.

- Do nothing and continue to work separately, waiting for a publication by you (ideally) or the other group (alas) to indicate results? Possible motivation for this approach includes a distrust of others and a wish to keep ideas and results confidential until it is time to submit something for publication.

In a recent experience with this situation, a researcher heard that we were working on something similar to his research project. He told mutual colleagues to ask my group to contact him. OK, so that was a bit indirect, but it was a way of opening communication without officially taking the first step: a sort of testing the waters without committing too much.

So we contacted him, and subsequent communication has been very friendly and interesting, with a bit of territory marking, but nothing too extreme. In the end, it will be particularly important for our students that our groups are now in communication and discussing complementary vs. overlapping research efforts.

In other cases, however, I have not been as interested in communicating information, although I typically don't mind giving general outlines of what I am doing.

For me, a key factor in my enthusiasm level re. communication is what I think of the other group -- that is, whether I think we are likely to have open, sincere, constructive discussions about our subparallel research.. or not. Sometimes you can't predict that if you don't know someone well, but sometimes there are clues (or prior negative experience) as a guide.

If you have heard rumors of possible or definite identical/similar research by others, what have you done?

And what influenced your approach? Whether/how well you know the other researcher(s)? Paranoia level? Desire to get the scoop? Other? Or does your research group (or field) have a particular philosophy of non-communication from which you do not stray (until you publish) no matter how nice the 'competing' researchers?


Sandlin said...

In my subfield, researchers seems pretty open and aggressive about doing overlapping working. When I was worried about getting scooped as a second year, I was told that was a real possibility, and the best solution is to work harder. I know the PIs usually share (brag?) about what they are working on, so we seem to know when there are imminent competitors, that just does change anyone's research directions.

Anonymous said...

In bioinformatics, many of our papers are analyses of other people's data. It can be quite frustrating when collaborators want to hold their data secret until they have the time or funding do to some more experiments, as the bioinformatics publications require that the data be made public.

Working with small research groups can be particularly hard, as they are often worried about being scooped by bigger, better funded labs that can redirect existing funds rather than having to wait the 2 years it usually takes to get grant funds for a new idea.

Anonymous said...

This topic raises the main issue I have with my field. It's intensely competitive, and every time I hear that someone else is working on an identical or similar problem, I usually try to hone in on exactly what questions we're each addressing, and I usually offer some nugget that I've learned that might help the other person. If I see any potential for collaboration or synergy, I mention it. I'm adamant that this is how science should be done; it avoids wasted effort and races to publish sloppy results, and open communication helps people move faster and more creatively. (This isn't to say we shouldn't check each other's work, think independently, and attempt different approaches.)

But I find that most of the time I share and try to coordinate, it's like talking to a wall--or worse, I get scooped. I've realized in my postdoc that a large fraction of researchers seem to be quite driven by their egos and potential personal glory. This makes me really sad and frustrated. I feel like I'm alone in thinking this way, and it makes me think I don't belong here. Obviously, we're all worried about funding and jobs (for those of us pre-TT or pre-tenure), but that's all the more reason to get the science done right.

The only time I've acted like a wall myself was when I was second author on a very high-impact project. Another researcher explained his idea, and I realized it was phenomenologically identical to our idea, though his invoked a different explanation. I said, "I'm pretty sure your idea will work, and ours is extremely similar--you should talk to [First Author] about it." That was the best I could do. I tried to be supportive and open.

Anonymous said...

I'm pretty new to this (2 yrs out of phd), but I've contacted. So far, this has led me to discover that the other groups were working on similar, but not the same, research projects. One of these contacts led to a unique publication with our shared data!
The way I see it, my field is too small for me to be getting a reputation for being sneaky or secretive so early in my career.

Andrea said...

What I do - panic, cry , eat chocolate.

Anonymous said...

I have a friend who presented a poster at a meeting only to find that the guy at the poster next to his was presenting the ~same project! The two presenters freaked out. The PI's (who had never met previously) talked and found that each group had a key thing the other lacked - one group had confirmed the finding in *human* disease (not just an animal model), and the other group had a clear mechanism (the first group had no clue about mechanism). They ended up publishing the work with the two grad students as co-first authors, and it was actually worked out well.

Anonymous said...

Very timely post! Here's a variation on the question of contacting other groups: what can/should you do if you learn of the overlapping work while reviewing a manuscript? ie. the project is referred to as related "unpublished results" and is clearly work in progress that may be a manuscript in preparation.

We would love to know how close the group is to submission and the major questions they are asking (since we're heavily invested and have been working on this project 3 yrs), but don't want to reveal we were reviewers of the related manuscript.

I don't think we can contact the other group, so we're just pushing hard to wrap up experiments (very stressful!) Suggestions?

GMP said...

I have had it happen recently. The group is fairly friendly, and we did share some basic information regarding where we're going, but since we have similar expertise it makes no sense from my standpoint to collaborate (we can do everything they can). The efforts are similar but not identical, and we keep communication channels semi-permeable, kind of keeping tabs on one another and citing each other's work. So far no major scoopage or feuds.

EscapedWestOfTheBigMuddy said...

In experimental particle physics, it isn't possible to get very far in without knowing who else is doing something similar or without them knowing what you are up to.

I'm currently involved in a effort with two major competitors. We were the first to get some data--in a limited way--but both competitors have the expectation of more precision in the final analysis, and one of them is just on the verge of starting data taking.

The interesting thing is that the early stages of funding processes mean that experimental design ideas are mostly explored in the open.

Anonymous said...

Work separately. In my case the two groups are composed almost entirely of senior faculty, but each is lead by a postdoc. I.e., someone who really needs to PI proposals and be first author on papers to have a chance at advancing. (I'm one of the postdocs.)

If we combined forces, it's guaranteed that at least one of us would lose, because only one person can be PI or first author (someone above mentioned "co-first authors" - this doesn't exist in my field). Working separately, we've actually both managed to have successful proposals in this subfield, and first-author papers, and of course we cite each other.

And having two independent groups working on it means better science, in my opinion, since we don't use exactly the same analysis.

Which isn't to say I don't have sleepless moments wondering when they're going to publish, or that I don't wait like a hawk for their data to become public :)

Doctor Pion said...

I would have done exactly what you did.

Why? Because in one past instance I ignored warnings that a well-known professor was a crook, and he ripped me off. But in other instances, the result was a complimentary set of projects that had the side benefit of opening doors for students to post doc in the other group and grow from the result (because the projects were not identical).

Science Professor Mum said...

Hmm. In general I would say I gulp, sigh, throw something at my office wall and then pick up the phone or email and get in touch. I am particularly keen to do this if it involves students or early career postdocs who may lose out. However, I have had 2 experiences with the same senior scientist where a student of mine has presented at a conference some results of their PhD (still a way off from publication at the time) and a notable senior figure in the field has talked to them about it and the next month has managed to drag up some simulations very similar and publish quickly. It may be that the questions we are addressing are obvious and have been hanging around for a while, but the timing is just too much of a co-incidence for it to happen twice with the same person. I am now much more cautious about what I and my students present at conferences, even though I feel that this is counter to the point of conferences...

Digbijoy Nath said...

Depends on what type of project, it's probable impact, honesty of the other group and many other factors.

I would, in general, prefer to work as separate groups. In research, you have to be either the best, or the first. If you are neither, there's no point doing research !!

Anonymous said...

I've heard many a horror story of two competing groups of researchers racing to be The First To Publish. Team A would submit their results--the first results--for publication, and, wouldn't you know, one of the folks from Team B would be asked to referee Team A's paper!

What happens then? The referee at Team B stalls the publication process while racing to submit their own results to another journal. Once Team B's paper has been published, the referee at Team B finally gives the journal the go-ahead to publish Team A's work.

And who gets credit for being The First To Publish? Team B.