Loren Frank is a professor of physiology at the University of California, San Francisco, and has been an HHMI Investigator since 2015.
1. In a layperson’s terms, what do you work on?
I’m interested in understanding how the brain makes it possible to think. For example, I’d like to know how an animal can use its past experiences to mentally simulate possible futures and then make sensible decisions based on those simulations? And how do those memories get created in the first place? We’re also trying to understand how the brain balances the need to pay attention to what’s going on now (reality) with the need to make predictions about what might come next.
2. Based on your experience with funders (private, NIH, others), what do funders do right vs. what could they improve? For example, if you could change anything about federal funders (NIH/NSF/etc.), such as their approach to peer review or to “high-risk” research, what would you change?
Let’s start with what they do right.
I’m tremendously fortunate to be an HHMI investigator. One of the many things that HHMI does right, in my opinion, is that they support people, not specific projects, and they do so for 7-year terms. The resulting intellectual freedom makes it possible to go after the fundamental questions in the field, even when those questions require effort spanning many years.
Similarly, I’ve been very lucky to be part of the Simons Foundation Collaboration for the Global Brain, which is now in its last year but provided funding for 10 years and, in my estimation, had a huge effect on our community. I’ve also been involved in a number of NIH BRAIN Initiative grants, which did a great job of supporting team science and new technologies.
But there are a number of things that could be improved:
The vast majority of funding in my field comes from the NIH, and while I try to remind myself that, historically speaking, public funding for science is a rare thing, the current approach leaves a lot to be desired.
First, science takes time, and there are very few awards that are long enough to actually see a project through from start to finish. Seven years is reasonable here, and while short grants can get things started, longer timescales would benefit both the science and the scientists. I’d like to see both government and private funders move to a longer cycle for all of their grants.
Second, while inflation has diminished the value of a dollar by about 55% since 2000, the size of NIH grants hasn’t kept up. A “modular” grant from NIH is still $250,000 a year, which was reasonable when it was introduced in 1998, but now a lab would need 2-4 of those to be viable. Add to that flat NIH budgets, stiff competition, and, typically, cuts of 10-20% of proposed budgets when grants are funded, and the result is that most scientists spend a huge fraction of their time writing and rewriting grants rather than actually doing the science. Doubling the amount for each grant to $500,000 would put us back where we were. It would also lower the funding rates, at least temporarily, but in the long run could allow scientists to focus on the science.
Third, we lack a mechanism to support professional scientists who don’t happen to be running their own labs. The work we all do is increasingly complex, and often requires teams of people who can work together. There are many postdocs who do not end up getting a job and leave science. If instead there was a dedicated career track, with salary-supporting grants, we could keep more of these people in science. It might cost twice as much as hiring a postdoc, but you’d have more “culture carriers”—people who maintain continuity from year to year rather than constantly moving on. Also, longer and larger grants would help here as well, as that would provide the funding stability needed to make these positions plausible. This would also help make science a more appealing career; right now there are increasing numbers of students and postdocs who look at what the life of a lab head is like and decide, entirely reasonably, to leave science and do something else.
3. Surveys show scientists spend upwards of 44% of their time on proposals, reports, IRBs, budgets, etc.–that is, administrative and regulatory requirements. Is that consistent with your experience? Is there anything that could be streamlined?
The administrative responsibilities of someone in my position are huge and growing. I’ve been able to hire people to help me —I have a half-time person who helps with progress reports, etc; a lab manager who does all of the lab management tasks; and also senior people who help with software development, grant applications and the like.
These people are critical to our enterprise, but most lab heads are not in a position to support individuals in these roles. As a result, and as I mentioned above, the large majority of lab heads spend a small fraction of their time actually doing science.
Streamlining here would require a commitment to reduce regulatory burden. This is complicated, as each regulation is introduced to minimize the chance of a particular harm. As someone once told me, whenever someone does something foolish or unethical, universities (and, I think, the government) introduce a new rule to forbid that action. I have some sympathy for this given the proclivity some people have to sue, but the impact on those of us trying to get science done is large.
Alternatively, we could get more support for filling out all the forms. This seems like something that a large language model could perhaps do, but it would require a lot of infrastructure work to make the data available in a form where this would be possible.
4. If you had no constraints in terms of funding or the need to publish, is there anything that would be different about your research?
In the context of funding, sometimes it is useful to write a grant to get into a certain mode of thinking. But for the most part, the time spent writing and preparing a grant is an exhausting and frustrating exercise. In my case, I’m involved in several long-term collaborations that all require funding from large, collaborative grants. I spend a lot of time working with my colleagues on these grants, and that is time that we could be spending building new technology and answering central questions in the field.
In the context of publishing, we need to distinguish between sharing our work and publishing in general, which is critical for science, and trying to get papers into high profile journals, which is not,. As scientists we share our work in publications, and while there are things we could do to make papers better, such as including the ability to interact with the data, the sharing part is essential. The part that is not as helpful is the need to publish in particular places, where somewhat arbitrary editorial decisions make the difference between success and failure.
If we felt no pressure to publish in particular places, we would work to write up our results, share them, and move to the next discovery. Instead, we write and then rewrite each paper multiple times before it’s published. Sometimes the paper gets a bit better in this process, but often the improvements do not make up for the huge additional time and effort required.
5. Just for fun, are there any recent papers in your field that you wish you could have written?
There’s a paper from some colleagues of mine at Cornell, where they developed a method that allowed them to disrupt activity in the hippocampus related specifically to sequences that link events over time. They then showed that these sequences are critical for flexible spatial behavior. It was very nice work.