Interaction

February 15, 2006

Who's in charge?

The catalysts for research are many. But without money most research doesn't happen.

The money, for the most part, is in Washington. Research and money engage in a mutually transformative relationship: Researchers at Stanford, like researchers anywhere else, go where the money is (one scientist called it a "command economy"), and, as a result, science and funding officials are working to make their grants more suitable and more stimulating for multidisciplinary research.

"All the meetings in Washington these days are about multidisciplinary research, but they haven't figured it out," said Anne Hannigan, associate vice president for research administration. "Everyone's struggling to reduce barriers."

In March 2004, a subcommittee of the White House Office of Science and Technology Policy recommended that federal agencies adjust their policies to help improve research. Encouraging collaborative and interdisciplinary work was high on the committee's list. In January 2005 the administration announced it would allow research projects to have more than one principal investigator (http://rbm.nih.gov/PI_memo_050104.pdf), and in July it requested comments from the scientific community on how to implement the new policy.

The National Institutes of Health (NIH) at present acknowledges only one principal investigator, or PI, per project. "One of the drivers behind wanting to recognize multiple PIs was the strong effort by NIH to recognize team science and interdisciplinary and multidisciplinary science," said Marcia Hahn, director of the division of grants policy at the agency's Office of Policy for Extramural Research Administration. "These are strong focuses of NIH, and the multiple PI is part of the changing way in which science is matching reality."

"It's a complex, vigorous discussion," Hahn said of the debate among scientists over principal investigators. "Apportionment is a whole new kettle of fish."

With a single principal investigator, only one person and that person's institution get maximum glory, even if (as is frequently the case) there are subsidiary collaborators on the grant. For some scientists, that system makes the most sense, but for many others—especially those whose universities use PI status as a criterion for tenure—it is both unfair and an unreliable way of assessing research. For them, credit and funding must be reapportioned.

If indeed the NIH opts for the multiple PI system, Hahn said, it probably will introduce what amounts to a prenuptial agreement in case "researchers fall out of love." It also will ask the participating research institutions to appoint one person as the principal NIH contact, someone who would get more paperwork than glory, though they're not mutually exclusive. Chemist W. E. Moerner, for example, principal investigator on an ongoing four-year, $2.9 million grant, said he spent three weeks in spring writing the annual progress report and a detailed budget for the next cycle.

The transformation would be substantial for the NIH, which has a far more hands-on approach to grant-giving than its sister institution, the National Science Foundation (NSF). Every step of the process for the NIH is predicated upon there being one principal investigator, which means that databases, finances and legal arrangements all will undergo an overhaul, Hahn said.

At the NSF, meanwhile, multiple principal investigators are old hat. Essentially, the agency lets the applicant institutions (or departments) sort out the hierarchy themselves by submitting collaborative proposals, each one a subset of one large project, each one with its own principal investigator and budget.

"The NSF will have a far easier road than the NIH because we've already solved a lot of these problems," said Hahn's counterpart at the NSF, Jean Feldman. "And we'll continue down that road." What the agency now calls sub-principal investigators or co-principal investigators will become multiple principal investigators, for example. "But in terms of substance and in terms of how it allows us to support multidisciplinary research, we're already doing a lot."

One Stanford researcher who fully agrees with Feldman is Jeff Koseff, a professor of civil and environmental engineering, director of the Stanford Institute for the Environment, and longtime champion and protagonist of interdisciplinary scientific research.

"The NSF has made it very easy," he said. "I've got to hand it to them."

Senior Associate Dean Eleanor Antonakos, also at the Engineering School, explained that adjusting a grant's management to the facts of divided research responsibility is not difficult. The lump sum is divided into subcontracts, with credit apportioned to each one. If one subcontractor fails to deliver the goods, the corresponding department pays the price. The NSF, meanwhile, stays out of the way.

The obvious problem, Antonakos pointed out, is that making departments ultimately responsible for what is supposed to be interdisciplinary in some way undermines the whole point. But bureaucratic adjustments can move only so quickly.

"The nature of science has changed," the NIH's Hahn said. "It's time for us to catch up."