Independent Investigators

Much of what I write in this blog is about how to modernize the research enterprise. One aspect that I would like to discuss is the concept of the “independent investigator”. I can't deny that the independent investigator has proven to be a very successful way to do research. It is the cornerstone of biomedical research in the United States, and the independent investigator is frequently described as the reason that the United Stated is a worldwide leader in biomedical research. That being said, I believe that our focus on the independent investigator is now hindering our research productivity in two ways: it is creating a tortuous path to being an investigator, and it is hindering collaboration.

There are basically two groups that participate in the research enterprise; labrats (i.e. students, post-docs, and technicians) and independent investigators. For the career of a labrat to progress he/she must become an independent investigator, and one of the most important qualifications for this job is the demonstration of independence. I find this to be a very strange job qualification, especially when one of the other major qualifications is to have a successful publication record. Let me clarify this a bit. For a post-doc to get a faculty position they must, first, become an expert in a field and develop a quality publication record in that field. Second, they have to find a new area that is clearly independent from their first area and independent of the investigator for whom they worked. And, lastly, they have to find funding for their new work, which requires preliminary data and a publication record in that new field. The requirement for independence in an investigator results in a very difficult career path that is constantly getting longer. Are these hurdles part of the reason that so many scientists leave academics? Are these hurdles necessary to select for the best scientists? Should there be a position for a career scientist who does not want to be independent? I don't have the answers to these questions and would love to get your opinions.

The second problem with focusing on independent investigators is that it hinders collaboration. A scientist must continuously be concerned with how they are demonstrating their independence. In a collaboration a scientist must be a part of a larger team, or contribute part of a larger project. It is only beneficial for an independent investigator to lead a collaboration, it is not beneficial to just contribute. In fact, it is detrimental for an investigator to just contribute to a collaboration. Being part of a collaboration uses resources but does not demonstrate independence and thus has a cost. This should reduce the number of investigators who want to participate in collaborations. It is also difficult to have a collaboration where every investigator wants to lead. Because of these problems, the NIH has set up a variety of grants to try and foster collaboration, i.e. SPORE's, P01's, etc. However, the groups that form for these grants often lack a true collaboration. It is easy for a number of independent investigators to put together their independent projects and call it a multi-departmental collaboration. Each investigator can lead their part of the collaboration and benefit without having to truly collaborate, but this defeats the purpose of collaboration.

The solution to this problem is both simple and incredibly difficult. The system must find a way to reward collaboration. Would it be possible to focus on an investigator's contribution instead of their independence? I believe it is possible, although difficult. Some journals now require that each investigator's contribution be listed on the publication, and this should make it easier to determine someone's contribution. Rewarding collaboration is not a new concept. It's only new to the biological sciences. Physics often has publications with hundreds of authors. How do they reward each investigator's contribution? Maybe biology needs to learn a few things from physics?

I don't believe that it is wrong to have independent investigators. It has proven to be very successful. However, I believe that requiring investigators to be independent can be harmful to the research enterprise. Geography and technology no longer require investigators to be independent and we need to modernize the system to reward collaborative investigators. Maybe we should just create a position called “collaborative investigator”.