Getting Funded from the National Institute of Mental Health
(NIMH)
By Edward Fantino, Ph.D.
My first application to NIMH, back in 1964, was unsuccessful,
but the rejection was informative. I had put together what I
thought were proposals for three good sets of experiments. I
learned that the experiments were not all that good, that they were
too unrelated and that I was asking for far too much. Since that
first application I have submitted dozens more. In all I tried to
take into account the lessons learned from that first ill-fated
attempt. While several of these proposals were not funded, and
while I certainly bristled from time to time at the reviews, based
on my experiences with NIMH and the National Science Foundation
(NSF) and based on a term on the NSF panel, in large measure I
believe that the grant process is fair. Perhaps I have been asked
to write this piece because I have succeeded in obtaining
uninterrupted funding from NIMH and/or NSF since my first NSF grant
in 1965. Currently we are completing the second year of a five-year
NIMH grant. But it has never been easy (and a thick skin is
recommended). What have I learned? While there is a certain degree
of variability in the process, I would say that there generally is
a correlation between the perceived quality of the proposal and the
likelihood of funding. So the first principle involves putting in
the time and planning required for a strong proposal. The proposal
should have implications either for theory or application if it is
to compete successfully. At the very least, the proposal should
make clear why the outcomes of the proposed experiments are likely
to be of interest to others. Above all the proposal should be
clearly written. The application for NIMH funding features a
single-spaced proposal of up to 25 pages. If the application
proposes experiments studying non-human animals it goes to a panel
that includes several members sympathetic to behavior-analytic
research. The panel meets three times a year.
Unlike NSF panels, which rely on the reports of outside
reviewers as well as on the evaluations of the panel members, NIMH
relies solely on the opinions of the panelists. Thus, one's first
step should be to check out the membership of the appropriate panel
on the NIMH web site or by phoning the office of the relevant
program director (currently, Howard S. Kurtzman, PhD, Chief,
Cognitive Science Program). Obviously one takes special care not to
ignore research relevant to one's proposal done by any of the panel
members. Since our laboratory investigates choice in humans as well
as in pigeons this panel was appropriate. If our proposal had
consisted entirely of human research, however, it would have been
sent to another panel (I know not which), and very possibly would
have received a less favorable reception. There can be a downside
to this strategy however: if the human and non-human research is
not well integrated the proposal may be judged unfocused. Since I
have been repeatedly advised of the importance of coherent
well-focused proposals there is a definite tradeoff here. Indeed I
have had a proposal from our lab derided as "a potpourri proposal"
for this reason (one person's breadth is another's potpourri). In
my judgment steering the proposal to the appropriate panel is
probably even more important than submitting the best possible
proposal. In our most recent proposal that strategy was successful.
Although one or two of the three reviewers did comment on the lack
of focus, they nonetheless recommended funding.
While all of the panelists on an NIMH panel vote (giving a score
in a secret tally) three panelists are primarily responsible for
influencing the fate of the proposal (one "primary" and two
"secondary" reviewers). These panelists present their detailed
evaluations to the rest of the group and lead the discussion. Soon
after the panel meeting, the investigator receives the three
detailed reviews and a "priority score" and percentile rank. The
percentile rank is critical. The cutoff for funding will depend on
how well funded NIMH is at the time. In the past few years at least
the top 20% have been funded. If the proposal is not funded, the
investigator may revise and resubmit it twice (for a total of up to
three considerations). If the reviewers' concerns can be adequately
dealt with, chances for funding obviously improve.
What can we do to increase the number of behavior-analytic
proposals that are funded by NIMH (and NSF for that matter)? The
most obvious and effective strategy IS TO APPLY! The more
applications received from behavior-analytic researchers, the more
qualified panelists will be appointed in our area. And the more
panelists who are knowledgeable about, and sympathetic to, research
in behavior analysis, then the brighter the prospects for the
funding of behavior-analytic proposals.