To me, the two biggest questions, what is the nature of the universe? I'm not going to do that. You have to be a theoretical physicist to get there. But the other one is, what is it that makes us human? It really became clear to me that an understanding of the brain was the way to most address that second question. The human brain is the most complex structure in the known universe. And we are in the middle of a scientific revolution to understand its inner workings.
Join us for a conversation with world-renowned neuroscientists as they visit Rochester. I am Dr. John Foxe, Director of the Del Monte Institute for Neuroscience at the University of Rochester. And you are listening to Neuroscience Prospectives. We are almost 30 years into the functional neuroimaging revolution. And if you asked anybody in our field to name the top 10 people who have been influential in the field of functional neuroimaging, Steve Peterson's name would be on everybody's list.
So Steve, it's really great to have you here today. Well, thanks for the introduction. I'm happy to be here. That's great. So I have a few hard questions for you today. OK. All right, you ready? I'm ready. I'm very interested, first of all, actually, in how did you end up as a neuroscientist? What's the story? Well, I started out as an anthropology major as an undergraduate. It was a long time ago. And there were a lot of people talking about the evolution of the brain.
But it was very obvious that they didn't know anything about the brain. So I decided I was going to be the anthropologist that goes and learns about the brain and brings it back to anthropology. And I knew two names. A small goal. A small goal. And built of great naivety, I knew two names in neuroscience. Roger Sperry, who did the split brain work, and Jim Olds, who's been lost to the sands of time a little bit.
But he was the first guy that began self-stimulation experiments where the rats will push a key because they're stimulating their brain and getting reinforcement. And they were both at Caltech. So I applied to a bunch of anthropology programs. And I applied to Caltech, the most selective place in the country. I got into no anthropology programs. And I got into Caltech. So that's how I got into neuroscience. I went to Caltech.
And there was a guy, John Olman, who also had an anthropology degree and was doing monkey vision physiology. And so I went into his lab for my graduate work. Could you imagine yourself doing anything else? Was it the right decision? I didn't know that it was right. But it seemed very cool at the time. I mean, I didn't know much about neurophysiology or about vision. But I learned a tremendous amount there over the five years that I was there.
Now, you've watched neuroimaging from its genesis, really, to the explosion that it is today. I was going to say cottage industry. But it's a military industrial complex, essentially. If you had to pick out the top one or two things that we've learned from neuroimaging, is there something you could point to? Or is it much broader than that? I think it's much broader than that. I think the idea that you could take pictures of the human brain at work, that's just totally amazing.
How long ago I got into this is PET didn't even exist at the time I started graduate school. It developed over the years I was in graduate school. And functional MRI was not even a twinkle in people's eyes at that point. But no, it's covered everything. It's changed neurology. It's changed psychiatry. And it's changed cognitive neuroscience, our understanding of how people think. So it's hard to go back and pick out one or two.
I had a little birdie tell me that one of your favorite things to do if you're not doing neuroscience involves cards. Yes. Are you going to tell us a little bit about that? And we'll get back to science afterwards. All right. Well, I love poker. Being able to see what the people have and being able to play out their thought process, I got really enamored of it. I played in graduate school a little bit, drinking and playing for Nicodemus Quarter.
But going to the casino and playing against other people, it's really fun. And you get to be mathematical. And you get to try to read other people's minds. So it's really a fun game. So it's leveraging some of the skill set that a neuroscientist has. I think also it's deeply competitive. And I picked it up in my 50s. And so things that were really competitive for me, like sports, they were on the big time ebb. And so it also gave me a place to put the competitiveness in.
Let's go back to science. I think one of the things that scientists often struggle with, apart from just generally communicating what can be complex thoughts to folks, the folks that pay for our work, the taxpayer. One difficulty can be sometimes to explain to folks why you're not necessarily working on a specific disease model. You're not working on autism or schizophrenia or stroke. You're just working on the basics of how the brain works.
Do you have a way, for example, to think about that, to explain to folks why it's really important to be able to do really basic stuff? To me, the two biggest questions that scientists, philosophers, and intellectual people in general, the two big things are, what is the nature of the universe? And I'm not going to do that. You have to be a theoretical physicist to get there. But the other one is, what is it that makes us human? And to me, what makes us human is our ability to think.
And that's what, as I was doing anthropology, it's why I studied anthropology, it really became clear to me that an understanding of the brain was the way to most address that second question, is what is the nature of humanity? And we get way down in the weeds and sometimes very far from that. But I always hope that ultimately we're trying to ask and answer questions about our basic humanity. So I don't feel very apologetic that I'm not studying a disease.
I also think that many of the ways that we've come at solving diseases is by having really good basic science. So it just flows out of our understanding, our basic understanding. And I think that's true about the brain as well. And I think some of our recent failures probably are a little bit losing sight of the human aspects of some of these things. Right, right. So you've watched the system for a few decades. A while. A while. And you've seen how it operates.
You've seen, for example, how the National Institute of Health works, how the National Science Foundation works, essentially the economics of the scientific enterprise. And I don't think there's too many people would argue that in the US we have the best science engine. But if you were given the magic wand, or I gave you the scepter in the morning, said, Steve, what would you do to reorganize? You're the new CEO of the whole scientific enterprise. I know that's a big question.
Can I just do biomedical research? Biomedical research. I meant that. Sorry. So I think the NIH has gone down some bad roads. I would change things in probably three-ish ways. So the first thing is, it's hard for somebody to get into the system. So the first thing I would do is make it much easier for people to get into the system. Now, I don't know what the age is this year, but it keeps going up. Time to your first R01, people are in mid-40s. And that's totally ridiculous. Absolutely.
I wasn't the fastest person through graduate school and post-doc, but I got my first grant at 35. And it changes your life. And that should not be another decade for everybody on average. So the system should be much more permissive to people at the front end. And I think that's merely a matter of will. You get good people who've had good training, and you make damn sure that half of them get funded as fast as possible. The second thing I would do is not make the funding one-zero.
So now you get your first grant, you go along, you try to get it renewed. If you don't get it renewed, you have no funding. And the lab that you've built up can disappear overnight. It should be more docile to the person so that you look and go, well, you're doing pretty well. We're going to maintain you at this level. You're not doing so well. We're going to cut you back. This is your warning. If you're doing better, we're going to give you more.
And that presupposes the third thing I would do, which is make the funding more retrospective. The best predictor of future success is past success. So the idea that your grant goes into a study section, and it may get turned down because somebody doesn't like the control on your 10th experiment is just crazy. You've been incredibly productive.
The first grant that I did not get, I had just probably the most productive period of my life, and I had to sit and wait for funding because one of the reviewers didn't like a couple of the things I said on a couple of experiments. And to me, that was one, it was really irritating, but it's debilitating. And so people have worked for years to build a lab group and technicians and administrative help and a series of students and post-docs. And now all of a sudden, you're just without.
Because of the serendipity or lack of serendipity in a review, I think it's just a really bad system. It's crazy to me because the NIH pretty much has the system I just described for their intramural program. And I've served as a reviewer and a counselor for the intramural program. It's much easier to do reviews when what you're deciding about is how productive the person has been.
And the reviewers pretty much all agree there's never a huge disagreements about things because it's not so focused on such small issues. The other worst thing about the NIH is the number, because it's difficult to get through the system, people are throwing more and more grants at less and less money. And so the funding percentages go down. You could solve all of those, all of that with the ideas I just portrayed.
I'd love to say that these are only my ideas, but lots of people have very similar ideas about this. Good practical solutions. Now, there's the funding aspect of it, but then the unit of productivity for a scientist is the paper, a published paper. And you've served as a senior editor on one of our great publications in the field, Cervo Cortex, for many years now. I know you've edited some of my own papers. Thank you. Hopefully, you were handled well.
Well, they were always handled well, I have to say. But I know you have strong thoughts as well about where we are with the output and how we publish our papers. And there's a revolution now in the field with regard to open access. And there are more than 300 journals in the neuroscience space. What's going to happen here? Where is it going? What do you see? What do you like? What don't you like? Well, boy, this is a much, to me, NIH is simple compared to scientific publishing.
I think the difficulty comes from a similar problem, is there's more and more papers chasing publishing utility. So people tend to want their paper to be in Nature or Science, the absolute top journals. And then they step down and they want to be in Nature Neuroscience. And then each phase that they go through till they get to the journal that they like, there are reviews and reviewers that have to be obtained.
So I know you're also editing and see hundreds of papers, new papers come in a year that you have to handle. And it often takes six or eight requests for review before you can get two reviewers. The sheer number of people that you have to get to do this overwhelms the system. There aren't enough good reviewers to review papers well. I don't see an answer to this. I think this will just continue to be a real bottleneck for useful publication.
I thought you were going to go the other way and say, what are the metrics for good publications? Well, how about that? What are the metrics for good publications? Well, I have one that I like that other people just really seem to not like. So I think the H index for somebody's scientific productivity is pretty good. Absolutely, yeah. So maybe just for our audience, a quick explanation of the H index. So the H index, it sounds more complicated than it is.
It's the number of papers that you have that are cited greater than the number of your publications. So if you have 10 papers with 10 or more citations, then your H index is 10. And if you look within a field, the H index actually varies quite a bit by the field that you're in. So imaging, people tend to have higher H indexes, so you can't indices. You can't judge people sort of outside their field.
So somebody in single unit physiology, they just don't publish as much because it's so much more demanding to reach a publishable unit. But if you look within a field and you pick out the people that you think intuitively are the best people, in a great number of cases, they will have the higher H index. And so I think it's intuitively a good one. Right. The only argument there is it works very well when you get to mid-career and above because it's an index of your history.
And of course, the more history you have, the better chance you have to have this thing represent your real productivity. But for a postdoc who's had a couple of papers and is just coming out, the H index hasn't really kicked in as well. And I don't think there is a good metric. At the postdoctoral level, I've had truly awesome postdocs who don't have very many publications and not many citations. At the time when they're looking for their job, and then it's the man, the fan, and the plan.
So that's where the mentors are incredibly important. And that's where the person having a clear idea of where they want to go and being able to articulate it to the people who may or may not hire them. The man, the fan, and the plan. I like that. That's very good. Or it could be the woman, the fan, and the plan. Offline, we were talking about a couple of papers. And you mentioned a paper that you did, one I remember quite well, looking at perceptual closure.
And we won't bore people with what perceptual closure is. But you were on the hunt for the seat of consciousness. I was. That really brings up a question. Is it possible? Can we find it with our tools? I don't know. I thought the experiment was really good. I just don't trust the outcome. I have to say where this idea came from is I was at the cognitive neuroscience meeting with two people that I knew. And we were closing a bar.
And we were having the deep intellectual conversation that you have after you've had too many drinks. Drawing pictures on napkins. And basically, every time we've done this, you go up to your room. You have the napkins with you. You get up in the morning and go, either I have no idea what this is about, or that's the stupidest idea I've ever had. And in this case, we saw each other the next day and go, I don't think this is a bad experiment to try. And it was basically a hangman experiment.
So you basically have people play a game of hangman where you have blanks and you fill in serially. And all of a sudden, you go, the word. So it's blank H, blank T, blank something. And you get along and you go, oh, rhythm, which is a great hangman word. And so that moment in which you have real recognition, the aha moment and the magical moment of recognition, and we thought we could take a picture of that with imaging. So you get a picture of what your brain does at the moment of aha.
And we ran the experiment. And we got stuff. I'm just not convinced that we got the magical moment of recognition. The problem, of course, with consciousness is it's totally internal and it's mine and mine alone. But I thought experiments like that. Getting out of innocence. Yeah, a little bit close to it. I suppose there are philosophical arguments about the difference between awareness and consciousness.
And maybe this sounds like an awareness of the relevance of something or a piece of information as it enters. So my inspiration for the experiment was actually a bunch of experiments that were done by John Duncan. And it's called, it's the moment of the past. So you can have a bunch of stimuli on the screen and be monitoring them for something to happen. And then at the moment that there is a target at one of the locations, you get this bottleneck of attention.
And the argument that Duncan made, and I think it's a good one, is when that target comes, it is passed into consciousness. And that's your total perception of what's going on, because that T all of a sudden is there out of nowhere and it captures your attention. So the idea of this perceptual closer experiment was to try to take advantage of that. And the interesting thing about that moment is other stuff can't get in.
It blocks out all the other things that could be going on so that if something else happened at one of the other locations, you wouldn't see it. You wouldn't have consciousness of that or awareness of that. So I think it gets close to what I think of as sort of phenomenal consciousness. And so that's what got me going down this magical moment kind of experimental line. Now that sounds like it. That's a great experiment. I love that John Duncan paper. It's a fantastic paper.
Do you have a favorite paper? You have a massive CV of papers. Is there one that just makes you happy to think about? Oh, boy. I have probably a small number of them for different reasons. So the very first big imaging paper was a nature paper done using PET in, I think, 1988. So it's a while back. I love that paper. One, because it was like a breakthrough paper. It was written up by John Marshall at the front of Nature that time. And it has kind of an interesting history.
So Nature actually asked us for that paper. Somebody at a cocktail party had said, I saw this preprint of this paper, which we didn't know where we were going to send because it was so huge. And Nature said, well, have them contact us. We can't do a nature paper. And they said, yes, you can. And so we put the paper together and send it to Nature. And I love that paper partly because it's a breakthrough cognitive imaging paper.
People think that's among the early ones, depending on how you counted. And also, that paper, I started giving talks everywhere because of that paper. So I got a lot of pub. I got on newspapers. So I was exposed to the press for the first time and learned some kind of bad things you should do and not do.
And it was just, and I think it's fair to say, that paper got me the Society for Neuroscience Young Investigator Award a couple of years later, which was great because I got in a time in my life where we had no money, I got $5,000. Excellent. Paid off cars. So I love that paper for that.
I love the paper Maurizio Corbetta and several of us did where we showed that paying attention to different aspects of visual information, so whether if you're trying to make a color decision or a movement decision or a shape decision, it actually changed the pictures in your brain. So it wasn't what you were seeing. Or how you were responding. It was what you were thinking about. And I thought that was really cool. That paper launched many careers. The dorsal ventral potential activation paper.
Yes. I love that paper. And I like this new stuff. I know it hasn't really quite captured people's fancy in the sense that it's people's fancy in the same kind of romantic way. But I love this network stuff. I think it's really fun. I was getting tired of doing the next task, FMRI. What does this kind of stimulus do to your brain? And so it kind of revitalized my energy because it was so different from all the stuff that we'd done before. So that, I don't know which paper I'd pick.
Maybe the first Jonathan Power Network one. That sort of started that phase off. So that was really fun. It's really fun. Excellent. Question for you from an electrophysiologist. So the brain's an electrical device. And those of us who, yes, you've heard. There's a rumor going around. So those of us who measure the primary signal sometimes get a little bit suspicious of you plumbers measuring the hemodynamics in there.
How much should we worry about the direct coupling of blood flow and the blood flow measures, perfusion measures to the electrical activity? And where are we on solving that coupling issue or understanding? So I think it's very easy to overdo the bold signal as being information carrying. But I think it does certain things. I think you can say this.
I think the bold signal, if it's handled properly, is monotonically related to the bulk amount of synaptic activity that takes place in one of the little image elements. I think going further than that is overstating it. And I think particularly in the cortex, where you have all this complex circuitry, excitation, inhibition, I think it gives you at least a probably monotonically related signal of the amount of information processing that goes on in that volume element.
I think you can go a long way with that. And I think we and other people have shown that you can play temporal games like the Hangman experiment to try to slow down all this incredibly fast, very malleable electrical activity. You can play some games to get some temporal access to what's going on. But it's very easy to over-represent how deeply we can get a representation of that. But I think just with what I said, I think you can learn a lot.
But I think we have to be humble that there was a whole universe of time scales and spatial scales that we're just not addressing with imaging. Steve, I know I speak for a lot of people in the field to say that it is a really good thing that this anthropologist became a neuroscientist. And your contributions to the field have been enormous. And you've had a massive impact on many of us. Thank you for being in Rochester with us. Thank you for having me. Thank you.