December 19, 2024 • 28 mins
Can we predict brain health across a lifetime? Just as a knee injury at 20 might cause problems later, researchers are exploring how early brain experiences shape future brain health. In this episode of Neuroscience Perspectives, Dr. Randy McIntosh, Professor and Director at the Institute for Neuroscience and Neurotechnology at Simon Fraser University, discusses groundbreaking insights into brain health, neuroscience, and AI with Dr. John Foxe, Director of the Del Monte Institute for Neuroscience. Learn about the Virtual Brain, an open-source neuroscience platform that simulates human brain function, offering new ways to understand conditions like stroke, epilepsy, and neurodegenerative diseases. Discover how technology, artificial intelligence, and neuroscience are converging to unlock the secrets of the human brain.
Virtual Brain: https://www.thevirtualbrain.org/tvb/zwei
🧠Stay Connected
Subscribe: https://www.youtube.com/@URNeuroscience
X: https://x.com/URNeuroscience
BlueSky: https://bsky.app/profile/urneuroscience.bsky.social
🧠Experts in this Episode
John Foxe, PhD: https://www.urmc.rochester.edu/people/112360965-john-j-foxe
Randy McIntosh, PhD: https://www.sfu.ca/bpk/about/people/faculty/randy-mcintosh.html
🧠Labs Mentioned
Frederick J. and Marion A. Schindler Cognitive Neurophysiology Lab: https://urmc.info/1RG
McIntosh Lab: https://www.armcintosh.com/home
#brain #science #podcast
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
a lot of what we do in neuroscience just focuses on the patient. And even that's even probably an
(00:05):
overstatement. It's really focused on the person's brain. So we don't think about that the brain's
sitting in this person's body, the person's in this environment, the person has to talk to their
loved ones, they have to interact with them and so on and so forth. And it's like, well, if we
took that model and thought about how do we take them out of the MRI facility and put them into
having dinner with their loved ones, how do we actually study that in a way that's useful?
(00:38):
I'm John Fox, Director of the Dal Monte Institute for Neuroscience at the University of Rochester.
And I'm absolutely delighted to welcome you to another episode of Neuroscience Perspectives.
We've taken the podcast on the road to the Society for Neuroscience Conference here in lovely,
beautiful, sunny Chicago. And I'm thrilled to be joined by Dr. Randi McIntosh, Professor and
(00:59):
Director of the Institute for Neuroscience and Neurotechnology at Simon Fraser University,
one of Canada's preeminent academic institutions in Burnaby and Vancouver, Canada. He is also one
of the founders of the Virtual Brain, an open source neuroscience platform that simulates the
human brain and offers insight into how it works and how disorders like stroke, epilepsy, or
(01:21):
neurodegenerative diseases impact its function. Randi, thank you for taking the time to come to
join us today. Happy to be here. Let's dive into your research. We'll come back to a bit to the
computational stuff. You're really interested in the development of the brain really across
the lifespan and the sort of functional architecture of the brain. Where does that even come from? How
(01:42):
did you decide to study this? Well, that's a long story. I think the motivation really came from my
foray initially when I was a postdoc, was doing some work on aging at NIH. And then when I was
in Toronto, I was in a geriatric research facility called Baycrest. And there, obviously, the focus
(02:04):
is on aging. It occurred to me, it's not terribly surprising that a lot of what happens to you when
you get older is influenced by what happened to you when you were younger. Anybody who's an athlete
knows that if you got a busted knee when you were 20, you try and go skiing when you're 50, of course.
It's like, my knee's sore. What's going on? But certainly it's going to happen with the brain as
(02:24):
well. So a lot of the approaches that one uses to understand that trajectory requires you to really
have that sort of lifespan perspective. The challenge, of course, is that we really don't have
data sets that cover that whole span. It's usually longitudinal studies typically are short,
relatively speaking. So we had to find a way to kind of pull these data together. That's kind of
(02:46):
the motivation behind doing aging and development, but then also trying to find a computational
platform like Virtual Brain that allows you to sort of stitch these things together to make sense
of it all. So this idea of sort of building a computational representation of a biological
structure in silico, in a computer, where does that come from? Tell the lay viewer out there
(03:14):
why you would expect to be able to really model a human brain in a computer.
Yeah. I mean, it's not like we're trying to build a completely accurate representation of the brain.
That would be otherwise we'd just build another brain. What we're trying to do is understand
what are the critical pieces that we can simulate that allow us to understand something important
(03:37):
about it. I think the great analogy is something like, for example, flight simulators, where
the whole aeronautics industry really had a huge boon when they improve their ability to simulate
the consequences of certain design considerations for wings, for example, that sort. We can do
(03:58):
something similar in the brain. So we know that the areas are connected in a certain way. Does
that have any relevance? If we change the connections and then try and simulate the activity
that comes from that, does that look anything different in terms of what we see in real data
sets? And you can go back and forth between them and say, if I model the critical features that I
think are important for capturing dementia, for example, for capturing the evolution of language,
(04:23):
for instance, what are those critical features? And I can test my hypotheses in the simulation
itself. And the nice thing in simulation is that you can do the experiments without having to
actually paper scans and so on and so forth. So it's a very efficient way to test your
speculations and then try and incorporate the empirical validity of that validation.
(04:45):
The same way that you test wing designs and come up with a small set of candidates for
wings that actually work and say, well, this one's the one I'm going to put onto this particular
plane. And it's great because that way you can avoid having to actually test it and have the
planes crashing, which you don't want to have happen, of course. And all the work and effort
that goes into performing human experimentation too. But is it one of these cases then where you
(05:08):
can make some predictions from the virtual brain and then take it back into the field and say,
does this actually happen in reality? Precisely. I mean, there's some papers recently,
not from my group necessarily, but one in development looking at the change, for example,
of the balance of excitation inhibition, which is a big thing in the brain, and how that evolves as
(05:28):
we develop. And the model shows that there are certain things that should happen. It's never
been really shown, but those people went back and sort of doing some pharmacological experiments
to show that in fact, yeah, that's actually what we see empirically. So it's a great way of
validating and testing and generating hypotheses for that matter as well.
(05:48):
Yeah. You know, it's interesting because in the field, right, there are folks that come
to neuroscience from a computational background, a math background, and people who come to it from
biology and pharmacology. And then there's a very huge group of people who come to it from the
cognitive sciences, right? They want to understand the brain. And many of those people, and I'm going
(06:12):
to paint in a very broad stroke here, you know, might be a little mathaphobic. They didn't, they
came through the, you know, interest from a psychology perspective. Would you go so far as
to say that really there's no way forward in understanding the brain without having
computational models, or is that too extreme?
I don't think...
(06:33):
I'm not trying to get you in trouble with all the psychology.
No, no, no. I mean, my pitch isn't psychology, sorry. I come from that lineage. And I have
those battles a lot. Certainly like my, when I was in Toronto, the courses I taught were mainly
multivariate stats in the psychology department. So, you know, half the classes absolutely hated
me for how dare you put up equations. So I do think it's important to have a computational
(06:58):
perspective. And it's important to realize that these computational models are great tools.
Right.
I don't think it's fair to expect all of the psychologists to learn dynamical systems theory
and be able to derive the equations, but it is worthwhile to be open to what they show
and to at least be able to have a conversation about them.
(07:19):
Right. At an intuitive level.
Yeah. And what that means though, is that it's on the other side, the people who are
the maths that are doing this, the applied mathematicians, the physicists and so on,
have to learn how to converse with the psychologist and learn how to do the math.
With the psychologist and make analogies or just show the equations in a way that's
tractable so they can have a conversation and then build the models jointly, for example.
(07:44):
Yeah. Maybe even take that one step further. And the end of the day, right, our duty is to the
people who pay the taxes to pay for the, so we need to be able to communicate this
for sure.
Folks who aren't even in the field in a way that's understandable.
And you typically find that that way of trying to communicate actually works really well for
communicating to your colleagues as well. Because it's not going down the lowest common denominator,
(08:04):
but it's actually trying to make it tractable. And that exercise, that capacity is one that
needs to be developed better in science.
Right. Yeah. You know, I'm just thinking, you were certainly one of the pioneers of not the
person who really introduced structural equation modelling to what we do and that. And brought a
(08:26):
lot of very good mathematical tools into the space. There's right now we're in the middle of this AI
revolution. It's as if AI showed up a couple of years ago. And yet, you know, I think if people
in our field, you know, we've been talking about these same mathematical procedures for 20, 25
years. Are we at a discontinuity now? Did something happen two years ago or is it just that it's
(08:46):
finally emerged into the public sphere?
I think it's become easier to use the tools and there's now a critical corpus of students,
critical number of students and scientists that are using these tools that are interested in
applying it to the brain. Historically, it was interesting when I was in Toronto again, that a
(09:08):
lot of people who were developing these technologies, the algorithms for AI and machine
learning and so on, they did it a lot of times without thinking about the brain, without
actually talking to someone who was actually doing the wet science. So for them, it didn't really
matter if that's how the brain did it. This does, this learns face recognition really well. So we
(09:29):
really cared about, but now I think there's that linkage that's been formed. And that's really what
I think changed things is that all of a sudden the gates to open and I've got people going back and
forth who are really facile in the computer science underlying AI and people who are willing to take
that step to apply it to their own data sets. There is a course and you probably know this as
(09:49):
well as I do that, there's an assumption it's almost going to be a panacea for us. And I think
it's important to make sure that the realistic aspects of AI machine learning are conveyed to
people who think it's going to solve all the problems. And that's where the problems come in,
I think is that over promise, under deliver kind of thing. Yeah, exactly. Yeah. It's very interesting
(10:12):
just to think that the dialogue, the narrative in the media about AI is so extreme. Yeah, unfortunate.
Yeah. And yet again, like these mechanisms, convolution nets, all this stuff, it's really
been around for decades and the neurosciences trying to solve the problems that we think about
every day. And again, as you say, there's a real risk of expecting too much too quickly. Yeah.
(10:36):
Yeah. It's going to be an evolution, there's no question that all techniques do that.
Depending on how far you're removed from the actual work done at ML and AI is we need to be
careful of not putting too much of a spin in either direction on that. And that's where that
(10:57):
tempering, the enthusiasm, but kind of saying this is a great tool because it can do things that we
could not do 20 years ago, but it's not going to solve your problem. It's not going to write your thesis,
it's not going to solve your problem for you. You have to actually still think about what's going on.
You and I had a conversation about this over the last few weeks,
thinking of the traditional way in which experimentation is done in human beings.
(11:21):
We put them in an isolated, electrically shielded booth or we have them doing something
in a magnet and they're highly unnatural. And we do it for control, right? So we can really turn
the dials in a very specific way and an experiment to control everything as best we can. But you've
really started to think about getting out into the environment and measuring people in the real world,
(11:45):
doing what it is they do actually in reality. Tell us a little bit about that and where your
thinking is going. Yeah, the motivation for that really came from getting back to my
years in Toronto. And you probably know this as well, when you're in a hospital facility and you
kind of see what people are actually faced with when they're faced with different kinds of neurological
(12:08):
aspects, stroke, dementia, and so on and so forth. So there's them and then there's the families and
then there's the extended family and all having to deal with the challenges that are around that.
A lot of what we do in neuroscience just focuses on the patient and even that's even the patient
is probably an overstatement. It's really focused on the person's brain. So we don't think about
(12:29):
that the brain sitting in this person's body, the person's in this environment, the person has to
talk to their loved ones, they have to interact with them and so on and so forth. And it's like,
well, if we took that model and thought about how do we take them out of the MRI facility and put
them into having dinner with their loved ones, how do we actually study that in a way that's useful?
(12:50):
There's all this thing about collaborative cognition that's been out for quite a while where
all of a sudden the person with their families seems to do better than if they're doing the
regular cognitive assessment. They shouldn't be able to do that and all of a sudden they're
remembering things. And that's the kind of thing that's kind of puzzling because we don't have a
way of studying that because we want to do it in a scanner. But what if we did it in the person's
(13:10):
kitchen or went to their community? And that was the motivation. I didn't get a lot of traction
primarily because there's a lot of other things going on. In BC and especially in Simon Fraser
University, there is this notion that the university is part of the community. And if that's true,
if we really aspire to be part of the community, let's bring the community in. Let's work with them
(13:30):
to start designing our studies. So I said, let's actually put our money where our mouths are and
try and build a platform that we can take out into the different communities in and around Vancouver,
take it into the interior BC and talk to the communities about what they think is important
for brain health, for brain resilience. What do they think are the most pressing issues?
(13:51):
Work with them on trying to design some studies, have wearable devices that can measure not just
brain, but body and get those integrated in a ways that we can now model with AI and virtual brain,
but also use the communities as our collaborators. And I think that's going to really push
the neuroscience in a direction that we need to go because it does integrate the person in the
(14:14):
environment and the environmental issues that we deal with as well. But it also then allows the
community to feel like they're actually part of the enterprise. And that's the important thing for me.
Yeah, fantastic. And we're sort of at this historical nexus to where the sensors, high bandwidth,
easy, deployable, cheap sensors are now available to us. And I mean, I think neuroscience has always
been very good at adopting these technologies. So the ability is there now, maybe it wasn't there
(14:39):
20 years ago where everything was heavy and lugubrious and difficult and we didn't have the
bandwidth or computing power. Yeah. Yeah. And now it's you can have it in your back pocket.
You know, I love what you just said about to the listening to the community at Rochester.
Rochester is actually the home of what they call the biopsychosocial model of medicine. And this was
the thought that you stopped thinking about a person as the disease that they walk through
(15:03):
the door with, but as a person that was embedded in a family and a community and a culture and a
medical history. And so I think, you know, you've literally restated that very beautifully.
And we're in a position now to really measure the biopsychosocial context in which somebody lives.
Yeah. And it's a complicated problem. I use the word complicated in the sense of
(15:25):
getting back to some more discussion earlier about the math underlying these things,
that it's going to be an enormous challenge to find a way to integrate these data in the
way that can be used. Because you think about, again, when we're talking about brain, we kind
of, we can measure neurons, we can measure populations of neurons, we can make multi-scale
models, at least putatively. But now we're talking about bringing us in a body. So we have these
(15:45):
interactions between the visceral and the metabolomics, the motor system and how the
motor interacts. And that becomes another level of modeling that we really have to kind of think
about how to do well. Extraordinary complexity, but the essence of actually being a healthy
human being, a lot of interactivity in the other systems. You know, and I suppose another challenge
will be doing it equitably and making sure that we get it to people who are in less technologically
(16:13):
rich parts of the country. Definitely. This is certainly a big consideration for folks,
for example, with disabilities, movement disabilities, who can't come to the mothership,
you know, to the main university to get an MRI scan and that. So there's real hope there
that these sort of mobile technologies that you can push out into the community might have
(16:34):
much greater reach than our current model. I'm hoping so. And again, it's also like the
communities, like the indigenous communities, First Nations communities are obviously
skeptical of these kinds of things. But I think there's a great opportunity there to
work with them as partners on these things to really understand their perspective on things.
(16:56):
And that there's some, you know, some of the conversations I've had have been extremely
illuminating to really get a perspective that it's almost that the perspective they bring is
that perspective that there is the person is part of this broader community, broader environment.
So if we have a way of actually incorporating that into our, how we do our science, that can
really make a big difference, not just to us, but also to them. Fantastic. Fantastic. Let's talk a
(17:21):
bit about Randy McIntosh. Let's talk about the five year old. Oh, okay. You know, so what were
you like growing up? Where does the passion for science come from? I think I saw a story about
you and your brother coming up with experiments and being just very inquisitive young man.
He was motivated by Ivo Knievel. Ivo Knievel, that's what I saw. Ivo Knievel was a motorcycle
(17:44):
daredevil. He used to jump over incredible things. And one of the ones was the Snake River Canyon,
which was, I don't remember when exactly it was, but, and he had a motorcycle that was basically
a rocket. So my brother and I tried to recreate that and ended up putting my brother in a box and
then setting it on fire. Unfortunately, we got him out of it before it got too bad. But trying to
build a rocket. My brother and our version of that was to get the Ivo Knievel action, man. We didn't
(18:07):
go any further than that. Oh, that's wonderful. We should have prototyped it first, I guess.
There you go. So you almost set your brother on fire. That's how it started. Yeah. I think he's
forgiven me. I think he's forgiven me, right? But yeah, so I started reading quite early.
I don't, my mom says I was around four and I started reading psychopedias and I was just,
(18:31):
this thirst for knowledge was always there. What I was interested in was quite diverse,
everything from dinosaurs to climate to like different languages to history and so on. So the
science has always been something that's really been driven. It's been this, I guess, an innate
curiosity that was always there. And it never really changed even in the high school days when
(18:55):
I was doing the rebellious things, there was still a scientist that was there. Certainly when I
transitioned to university, there was a brief moment where I thought I wanted to be a lawyer,
a very brief moment after my first year pre-loss, like, yeah, that's not a good idea. So
I stayed with science after that and ended up being neuroscience. It was partly from
(19:18):
some work in high school, but also partly from taking some classes in my hometown on Lethbridge,
where it was really sort of inspired by the classes I had with people like Rob Sutherland and
Ian Wishaw in particular. So again, again, we hear of people who come in contact with very specific
individuals who really just float their boat, get the intellectual juices from it.
(19:43):
Yeah, yeah, yeah. And Rob was a new faculty at that point in time, so he was extremely
interactive as well because you could see the enthusiasm for what he was trying to do.
And it's that enthusiasm for that path of discovery that really kind of grabbed me. It's like, wow,
I could really see myself doing this.
(20:03):
I could really see myself doing this. And then you did your original undergraduate
university up in Canada in Calgary and then headed down to Texas for your PhD.
Right. What brought you to Texas?
It was really the person I wanted to work with. So in Calgary, I started working with a technology
called autoradiographic 2-deoxyglucose, which is basically a predecessor of the imaging stuff
(20:26):
that we're doing now with humans, but I was doing that with rats. And there was a guy in Texas
named Francisco Gonzalez Lima who was using this method 2-DG to make it easier to spay
in a behavioral sense. So when it was first used, it was kind of like what happens to
glucose metabolism if we give an animal ketamine, if we give it a stroke. But there were very
(20:52):
few people doing like what happens if an animal learns, for example, that an auditory stimulus is
an aversive signal, a classical conditioning kind of thing. And he was doing that and was like,
I want to do that. So I sent him a letter, handwritten letter, and I got a letter back
about two weeks later saying, yeah, come on down. So at that point he was in Texas,
(21:15):
saying I'm in university, which is at the medical school there. And then a couple of years later,
he got a position in Austin. And you moved to Austin with him.
I moved to University of Texas at Austin, which was...
Pretty good. Good place to land for you, I happen to know because of your passion and love for music.
Right? So music's another big component of your life.
Definitely. Definitely. Yeah. I was one of these...
I can't tell you the number of people who've sat in that spot. Music is highly,
(21:40):
highly overrepresented in the neurosciences or in the sciences. I think it's amazing.
Yeah. It's interesting because I don't know if people think about it as a separate thing.
I mean, certainly the process of creating music is different than analyzing data and thinking about
the brain, but I don't know that they're different enough so that they're considered
(22:03):
almost separate activities. Right.
That a lot of... And we can test this, of course. A lot of the processes,
the underlying biological process should be similar. It could be very much the case
that a lot of the fluidity in terms of the mental processes could be similar.
And this could be why music actually has such an appeal for people who are thinking about
high thoughts, like either would they be scientists or whether they be other people who are also in
(22:27):
that creative genre that use these other things to continue to engage broader networks for
their own enjoyment, but also potentially to also help think through problems.
I had somebody say to me one time, one of the things that I was thinking about was
somebody say to me one time, one of the things that she got out of music was that in the sciences
(22:51):
and what we do, that there's a tremendous delayed gratification component to it.
You start something today and you'll be very lucky if you see something useful or something
that will give you a little bit of a buzz for two or three years and often longer.
And she went home and she played some music and the product was right there.
(23:11):
And that was something that she needed to tie it through.
This is very true. This is very true. There's a gratification you get from
playing music or even listening to music that's harder to get from
getting your paper accepted in nature.
It's a long day.
Tell us, with your trainees, and well, I often ask this question,
(23:34):
if given the Gestalt, the Zeitgeist in 2024 for a youngster coming through graduate school,
do you have pearls of wisdom, advice about the world that they're entering into in the sciences?
Would you do it again today if you had it to do over?
Oh yeah, no question.
As I said, I mean, science is deeply ingrained in who I am.
(23:57):
So I don't think about my career choice as being a job. And that was something that I
really found peculiar in talking to some of my other colleagues who treated science as a job.
And this is true for other professions, whether it be art, you can think about other professions,
(24:19):
medical doctor, for example, that you don't shut things off at five o'clock.
And I think that's one of the things in science that it's going to sound smug and maybe a little
bit elitist, I guess, but it's really don't go into science if it's not a passion.
Yeah.
Don't think about it as like, my job is to be a scientist because it doesn't stop at five o'clock
(24:44):
on Friday. Some of my insights come to me on Sunday mornings, unfortunately.
So I gotta write that down. So I'm up at four o'clock in the morning on a Sunday trying to
get this thing written down. And that's the way, I mean, musicians do that as well, right?
Yeah, of course.
Musicians get inspiration or artists do the same kind of thing. And I think that's the
(25:06):
attraction of being in this field. And it's for me, I feel blessed that I actually get to do this.
Yeah.
I'm kind of stunned sometimes that I get paid to do this. Wow. And that's, I think, one of the
things that people going through the system now really have to think about is that, is this
something that really I think of as a passion or is it something I'm doing just because this is
(25:31):
what you do?
Yeah.
So going through bachelor's, going through your PhD, going through your postdoc, if you're just
doing this because those are the steps, it's probably the wrong reason for doing it. And stop
at some point and think about, is that really what I want to do? I've been quite deliberate
with my students to have those conversations a number of times and be very frank with them.
(25:54):
I'm not going to think less highly of you if you decide I want to go into industry, for example.
Right.
Because that's a good choice and maybe that's actually what you want to do and maybe be
successful there. Try it.
Yeah.
Why not? There's no penalty really. You may take a pay cut. Actually, most people go into industry,
get a pay raise.
Yeah, they pay raise.
But it's really allowing that opportunity to explore where they want to go and supporting
(26:21):
them for that. And that's my job is to really support their journey so that they're making
the decision when they finish their PhD that they're going to be doing something that they're
going to be passionate about when they're 60 or 70 years old as well.
Right. Right. No off switch. In the end of the day, it's a living breathing thing. It's part
and parcel of your very being to be on the whole time. And I appreciate that too. It's not a job.
(26:46):
It's a privilege to get paid to do it.
For sure. For sure. Yeah. Like when we moved to Vancouver, we have a place now on Vancouver
Island as well, which is a great retreat. But when I go there, it's not like I shut everything off.
It's a different perspective.
Yeah.
No question. But it's not like I...
That's an important point of being a scientist too, right? Which is it's one thing to be in the lab
(27:09):
all the time and doing experiments. But the quiet time and the thoughtful time is where
a lot of the work gets done. And that's the part maybe that doesn't feel like work.
Yeah.
Where you're riding your bike and...
Well, the virtual brain history actually came not because Victor, you're sitting in a lab,
you were sitting in a pub having a conversation about it. And all of a sudden inspiration came
(27:30):
from that. And he and I do a lot of running together. So when we are on runs, we have these
great conversations that are kind of meandering and all of a sudden, oh, we should stop and record
this.
Yeah. So you could have been talking about the football game or you're talking about this.
It's just...
Right. Yeah. Yeah. Yeah. For sure.
Superb. Randy, thanks so much for taking the time with us.
Pleasure. I enjoyed it.
It's really great to have you here. Looking forward to seeing your work.
Yeah. Thank you.
(27:50):
Over the years to come.
Take care.
a lot of what we do in neuroscience just focuses on the patient. And even that's even probably an
(00:05):
overstatement. It's really focused on the person's brain. So we don't think about that the brain's
sitting in this person's body, the person's in this environment, the person has to talk to their
loved ones, they have to interact with them and so on and so forth. And it's like, well, if we
took that model and thought about how do we take them out of the MRI facility and put them into
having dinner with their loved ones, how do we actually study that in a way that's useful?
(00:38):
I'm John Fox, Director of the Dal Monte Institute for Neuroscience at the University of Rochester.
And I'm absolutely delighted to welcome you to another episode of Neuroscience Perspectives.
We've taken the podcast on the road to the Society for Neuroscience Conference here in lovely,
beautiful, sunny Chicago. And I'm thrilled to be joined by Dr. Randi McIntosh, Professor and
(00:59):
Director of the Institute for Neuroscience and Neurotechnology at Simon Fraser University,
one of Canada's preeminent academic institutions in Burnaby and Vancouver, Canada. He is also one
of the founders of the Virtual Brain, an open source neuroscience platform that simulates the
human brain and offers insight into how it works and how disorders like stroke, epilepsy, or
(01:21):
neurodegenerative diseases impact its function. Randi, thank you for taking the time to come to
join us today. Happy to be here. Let's dive into your research. We'll come back to a bit to the
computational stuff. You're really interested in the development of the brain really across
the lifespan and the sort of functional architecture of the brain. Where does that even come from? How
(01:42):
did you decide to study this? Well, that's a long story. I think the motivation really came from my
foray initially when I was a postdoc, was doing some work on aging at NIH. And then when I was
in Toronto, I was in a geriatric research facility called Baycrest. And there, obviously, the focus
(02:04):
is on aging. It occurred to me, it's not terribly surprising that a lot of what happens to you when
you get older is influenced by what happened to you when you were younger. Anybody who's an athlete
knows that if you got a busted knee when you were 20, you try and go skiing when you're 50, of course.
It's like, my knee's sore. What's going on? But certainly it's going to happen with the brain as
(02:24):
well. So a lot of the approaches that one uses to understand that trajectory requires you to really
have that sort of lifespan perspective. The challenge, of course, is that we really don't have
data sets that cover that whole span. It's usually longitudinal studies typically are short,
relatively speaking. So we had to find a way to kind of pull these data together. That's kind of
(02:46):
the motivation behind doing aging and development, but then also trying to find a computational
platform like Virtual Brain that allows you to sort of stitch these things together to make sense
of it all. So this idea of sort of building a computational representation of a biological
structure in silico, in a computer, where does that come from? Tell the lay viewer out there
(03:14):
why you would expect to be able to really model a human brain in a computer.
Yeah. I mean, it's not like we're trying to build a completely accurate representation of the brain.
That would be otherwise we'd just build another brain. What we're trying to do is understand
what are the critical pieces that we can simulate that allow us to understand something important
(03:37):
about it. I think the great analogy is something like, for example, flight simulators, where
the whole aeronautics industry really had a huge boon when they improve their ability to simulate
the consequences of certain design considerations for wings, for example, that sort. We can do
(03:58):
something similar in the brain. So we know that the areas are connected in a certain way. Does
that have any relevance? If we change the connections and then try and simulate the activity
that comes from that, does that look anything different in terms of what we see in real data
sets? And you can go back and forth between them and say, if I model the critical features that I
think are important for capturing dementia, for example, for capturing the evolution of language,
(04:23):
for instance, what are those critical features? And I can test my hypotheses in the simulation
itself. And the nice thing in simulation is that you can do the experiments without having to
actually paper scans and so on and so forth. So it's a very efficient way to test your
speculations and then try and incorporate the empirical validity of that validation.
(04:45):
The same way that you test wing designs and come up with a small set of candidates for
wings that actually work and say, well, this one's the one I'm going to put onto this particular
plane. And it's great because that way you can avoid having to actually test it and have the
planes crashing, which you don't want to have happen, of course. And all the work and effort
that goes into performing human experimentation too. But is it one of these cases then where you
(05:08):
can make some predictions from the virtual brain and then take it back into the field and say,
does this actually happen in reality? Precisely. I mean, there's some papers recently,
not from my group necessarily, but one in development looking at the change, for example,
of the balance of excitation inhibition, which is a big thing in the brain, and how that evolves as
(05:28):
we develop. And the model shows that there are certain things that should happen. It's never
been really shown, but those people went back and sort of doing some pharmacological experiments
to show that in fact, yeah, that's actually what we see empirically. So it's a great way of
validating and testing and generating hypotheses for that matter as well.
(05:48):
Yeah. You know, it's interesting because in the field, right, there are folks that come
to neuroscience from a computational background, a math background, and people who come to it from
biology and pharmacology. And then there's a very huge group of people who come to it from the
cognitive sciences, right? They want to understand the brain. And many of those people, and I'm going
(06:12):
to paint in a very broad stroke here, you know, might be a little mathaphobic. They didn't, they
came through the, you know, interest from a psychology perspective. Would you go so far as
to say that really there's no way forward in understanding the brain without having
computational models, or is that too extreme?
I don't think...
(06:33):
I'm not trying to get you in trouble with all the psychology.
No, no, no. I mean, my pitch isn't psychology, sorry. I come from that lineage. And I have
those battles a lot. Certainly like my, when I was in Toronto, the courses I taught were mainly
multivariate stats in the psychology department. So, you know, half the classes absolutely hated
me for how dare you put up equations. So I do think it's important to have a computational
(06:58):
perspective. And it's important to realize that these computational models are great tools.
Right.
I don't think it's fair to expect all of the psychologists to learn dynamical systems theory
and be able to derive the equations, but it is worthwhile to be open to what they show
and to at least be able to have a conversation about them.
(07:19):
Right. At an intuitive level.
Yeah. And what that means though, is that it's on the other side, the people who are
the maths that are doing this, the applied mathematicians, the physicists and so on,
have to learn how to converse with the psychologist and learn how to do the math.
With the psychologist and make analogies or just show the equations in a way that's
tractable so they can have a conversation and then build the models jointly, for example.
(07:44):
Yeah. Maybe even take that one step further. And the end of the day, right, our duty is to the
people who pay the taxes to pay for the, so we need to be able to communicate this
for sure.
Folks who aren't even in the field in a way that's understandable.
And you typically find that that way of trying to communicate actually works really well for
communicating to your colleagues as well. Because it's not going down the lowest common denominator,
(08:04):
but it's actually trying to make it tractable. And that exercise, that capacity is one that
needs to be developed better in science.
Right. Yeah. You know, I'm just thinking, you were certainly one of the pioneers of not the
person who really introduced structural equation modelling to what we do and that. And brought a
(08:26):
lot of very good mathematical tools into the space. There's right now we're in the middle of this AI
revolution. It's as if AI showed up a couple of years ago. And yet, you know, I think if people
in our field, you know, we've been talking about these same mathematical procedures for 20, 25
years. Are we at a discontinuity now? Did something happen two years ago or is it just that it's
(08:46):
finally emerged into the public sphere?
I think it's become easier to use the tools and there's now a critical corpus of students,
critical number of students and scientists that are using these tools that are interested in
applying it to the brain. Historically, it was interesting when I was in Toronto again, that a
(09:08):
lot of people who were developing these technologies, the algorithms for AI and machine
learning and so on, they did it a lot of times without thinking about the brain, without
actually talking to someone who was actually doing the wet science. So for them, it didn't really
matter if that's how the brain did it. This does, this learns face recognition really well. So we
(09:29):
really cared about, but now I think there's that linkage that's been formed. And that's really what
I think changed things is that all of a sudden the gates to open and I've got people going back and
forth who are really facile in the computer science underlying AI and people who are willing to take
that step to apply it to their own data sets. There is a course and you probably know this as
(09:49):
well as I do that, there's an assumption it's almost going to be a panacea for us. And I think
it's important to make sure that the realistic aspects of AI machine learning are conveyed to
people who think it's going to solve all the problems. And that's where the problems come in,
I think is that over promise, under deliver kind of thing. Yeah, exactly. Yeah. It's very interesting
(10:12):
just to think that the dialogue, the narrative in the media about AI is so extreme. Yeah, unfortunate.
Yeah. And yet again, like these mechanisms, convolution nets, all this stuff, it's really
been around for decades and the neurosciences trying to solve the problems that we think about
every day. And again, as you say, there's a real risk of expecting too much too quickly. Yeah.
(10:36):
Yeah. It's going to be an evolution, there's no question that all techniques do that.
Depending on how far you're removed from the actual work done at ML and AI is we need to be
careful of not putting too much of a spin in either direction on that. And that's where that
(10:57):
tempering, the enthusiasm, but kind of saying this is a great tool because it can do things that we
could not do 20 years ago, but it's not going to solve your problem. It's not going to write your thesis,
it's not going to solve your problem for you. You have to actually still think about what's going on.
You and I had a conversation about this over the last few weeks,
thinking of the traditional way in which experimentation is done in human beings.
(11:21):
We put them in an isolated, electrically shielded booth or we have them doing something
in a magnet and they're highly unnatural. And we do it for control, right? So we can really turn
the dials in a very specific way and an experiment to control everything as best we can. But you've
really started to think about getting out into the environment and measuring people in the real world,
(11:45):
doing what it is they do actually in reality. Tell us a little bit about that and where your
thinking is going. Yeah, the motivation for that really came from getting back to my
years in Toronto. And you probably know this as well, when you're in a hospital facility and you
kind of see what people are actually faced with when they're faced with different kinds of neurological
(12:08):
aspects, stroke, dementia, and so on and so forth. So there's them and then there's the families and
then there's the extended family and all having to deal with the challenges that are around that.
A lot of what we do in neuroscience just focuses on the patient and even that's even the patient
is probably an overstatement. It's really focused on the person's brain. So we don't think about
(12:29):
that the brain sitting in this person's body, the person's in this environment, the person has to
talk to their loved ones, they have to interact with them and so on and so forth. And it's like,
well, if we took that model and thought about how do we take them out of the MRI facility and put
them into having dinner with their loved ones, how do we actually study that in a way that's useful?
(12:50):
There's all this thing about collaborative cognition that's been out for quite a while where
all of a sudden the person with their families seems to do better than if they're doing the
regular cognitive assessment. They shouldn't be able to do that and all of a sudden they're
remembering things. And that's the kind of thing that's kind of puzzling because we don't have a
way of studying that because we want to do it in a scanner. But what if we did it in the person's
(13:10):
kitchen or went to their community? And that was the motivation. I didn't get a lot of traction
primarily because there's a lot of other things going on. In BC and especially in Simon Fraser
University, there is this notion that the university is part of the community. And if that's true,
if we really aspire to be part of the community, let's bring the community in. Let's work with them
(13:30):
to start designing our studies. So I said, let's actually put our money where our mouths are and
try and build a platform that we can take out into the different communities in and around Vancouver,
take it into the interior BC and talk to the communities about what they think is important
for brain health, for brain resilience. What do they think are the most pressing issues?
(13:51):
Work with them on trying to design some studies, have wearable devices that can measure not just
brain, but body and get those integrated in a ways that we can now model with AI and virtual brain,
but also use the communities as our collaborators. And I think that's going to really push
the neuroscience in a direction that we need to go because it does integrate the person in the
(14:14):
environment and the environmental issues that we deal with as well. But it also then allows the
community to feel like they're actually part of the enterprise. And that's the important thing for me.
Yeah, fantastic. And we're sort of at this historical nexus to where the sensors, high bandwidth,
easy, deployable, cheap sensors are now available to us. And I mean, I think neuroscience has always
been very good at adopting these technologies. So the ability is there now, maybe it wasn't there
(14:39):
20 years ago where everything was heavy and lugubrious and difficult and we didn't have the
bandwidth or computing power. Yeah. Yeah. And now it's you can have it in your back pocket.
You know, I love what you just said about to the listening to the community at Rochester.
Rochester is actually the home of what they call the biopsychosocial model of medicine. And this was
the thought that you stopped thinking about a person as the disease that they walk through
(15:03):
the door with, but as a person that was embedded in a family and a community and a culture and a
medical history. And so I think, you know, you've literally restated that very beautifully.
And we're in a position now to really measure the biopsychosocial context in which somebody lives.
Yeah. And it's a complicated problem. I use the word complicated in the sense of
(15:25):
getting back to some more discussion earlier about the math underlying these things,
that it's going to be an enormous challenge to find a way to integrate these data in the
way that can be used. Because you think about, again, when we're talking about brain, we kind
of, we can measure neurons, we can measure populations of neurons, we can make multi-scale
models, at least putatively. But now we're talking about bringing us in a body. So we have these
(15:45):
interactions between the visceral and the metabolomics, the motor system and how the
motor interacts. And that becomes another level of modeling that we really have to kind of think
about how to do well. Extraordinary complexity, but the essence of actually being a healthy
human being, a lot of interactivity in the other systems. You know, and I suppose another challenge
will be doing it equitably and making sure that we get it to people who are in less technologically
(16:13):
rich parts of the country. Definitely. This is certainly a big consideration for folks,
for example, with disabilities, movement disabilities, who can't come to the mothership,
you know, to the main university to get an MRI scan and that. So there's real hope there
that these sort of mobile technologies that you can push out into the community might have
(16:34):
much greater reach than our current model. I'm hoping so. And again, it's also like the
communities, like the indigenous communities, First Nations communities are obviously
skeptical of these kinds of things. But I think there's a great opportunity there to
work with them as partners on these things to really understand their perspective on things.
(16:56):
And that there's some, you know, some of the conversations I've had have been extremely
illuminating to really get a perspective that it's almost that the perspective they bring is
that perspective that there is the person is part of this broader community, broader environment.
So if we have a way of actually incorporating that into our, how we do our science, that can
really make a big difference, not just to us, but also to them. Fantastic. Fantastic. Let's talk a
(17:21):
bit about Randy McIntosh. Let's talk about the five year old. Oh, okay. You know, so what were
you like growing up? Where does the passion for science come from? I think I saw a story about
you and your brother coming up with experiments and being just very inquisitive young man.
He was motivated by Ivo Knievel. Ivo Knievel, that's what I saw. Ivo Knievel was a motorcycle
(17:44):
daredevil. He used to jump over incredible things. And one of the ones was the Snake River Canyon,
which was, I don't remember when exactly it was, but, and he had a motorcycle that was basically
a rocket. So my brother and I tried to recreate that and ended up putting my brother in a box and
then setting it on fire. Unfortunately, we got him out of it before it got too bad. But trying to
build a rocket. My brother and our version of that was to get the Ivo Knievel action, man. We didn't
(18:07):
go any further than that. Oh, that's wonderful. We should have prototyped it first, I guess.
There you go. So you almost set your brother on fire. That's how it started. Yeah. I think he's
forgiven me. I think he's forgiven me, right? But yeah, so I started reading quite early.
I don't, my mom says I was around four and I started reading psychopedias and I was just,
(18:31):
this thirst for knowledge was always there. What I was interested in was quite diverse,
everything from dinosaurs to climate to like different languages to history and so on. So the
science has always been something that's really been driven. It's been this, I guess, an innate
curiosity that was always there. And it never really changed even in the high school days when
(18:55):
I was doing the rebellious things, there was still a scientist that was there. Certainly when I
transitioned to university, there was a brief moment where I thought I wanted to be a lawyer,
a very brief moment after my first year pre-loss, like, yeah, that's not a good idea. So
I stayed with science after that and ended up being neuroscience. It was partly from
(19:18):
some work in high school, but also partly from taking some classes in my hometown on Lethbridge,
where it was really sort of inspired by the classes I had with people like Rob Sutherland and
Ian Wishaw in particular. So again, again, we hear of people who come in contact with very specific
individuals who really just float their boat, get the intellectual juices from it.
(19:43):
Yeah, yeah, yeah. And Rob was a new faculty at that point in time, so he was extremely
interactive as well because you could see the enthusiasm for what he was trying to do.
And it's that enthusiasm for that path of discovery that really kind of grabbed me. It's like, wow,
I could really see myself doing this.
(20:03):
I could really see myself doing this. And then you did your original undergraduate
university up in Canada in Calgary and then headed down to Texas for your PhD.
Right. What brought you to Texas?
It was really the person I wanted to work with. So in Calgary, I started working with a technology
called autoradiographic 2-deoxyglucose, which is basically a predecessor of the imaging stuff
(20:26):
that we're doing now with humans, but I was doing that with rats. And there was a guy in Texas
named Francisco Gonzalez Lima who was using this method 2-DG to make it easier to spay
in a behavioral sense. So when it was first used, it was kind of like what happens to
glucose metabolism if we give an animal ketamine, if we give it a stroke. But there were very
(20:52):
few people doing like what happens if an animal learns, for example, that an auditory stimulus is
an aversive signal, a classical conditioning kind of thing. And he was doing that and was like,
I want to do that. So I sent him a letter, handwritten letter, and I got a letter back
about two weeks later saying, yeah, come on down. So at that point he was in Texas,
(21:15):
saying I'm in university, which is at the medical school there. And then a couple of years later,
he got a position in Austin. And you moved to Austin with him.
I moved to University of Texas at Austin, which was...
Pretty good. Good place to land for you, I happen to know because of your passion and love for music.
Right? So music's another big component of your life.
Definitely. Definitely. Yeah. I was one of these...
I can't tell you the number of people who've sat in that spot. Music is highly,
(21:40):
highly overrepresented in the neurosciences or in the sciences. I think it's amazing.
Yeah. It's interesting because I don't know if people think about it as a separate thing.
I mean, certainly the process of creating music is different than analyzing data and thinking about
the brain, but I don't know that they're different enough so that they're considered
(22:03):
almost separate activities. Right.
That a lot of... And we can test this, of course. A lot of the processes,
the underlying biological process should be similar. It could be very much the case
that a lot of the fluidity in terms of the mental processes could be similar.
And this could be why music actually has such an appeal for people who are thinking about
high thoughts, like either would they be scientists or whether they be other people who are also in
(22:27):
that creative genre that use these other things to continue to engage broader networks for
their own enjoyment, but also potentially to also help think through problems.
I had somebody say to me one time, one of the things that I was thinking about was
somebody say to me one time, one of the things that she got out of music was that in the sciences
(22:51):
and what we do, that there's a tremendous delayed gratification component to it.
You start something today and you'll be very lucky if you see something useful or something
that will give you a little bit of a buzz for two or three years and often longer.
And she went home and she played some music and the product was right there.
(23:11):
And that was something that she needed to tie it through.
This is very true. This is very true. There's a gratification you get from
playing music or even listening to music that's harder to get from
getting your paper accepted in nature.
It's a long day.
Tell us, with your trainees, and well, I often ask this question,
(23:34):
if given the Gestalt, the Zeitgeist in 2024 for a youngster coming through graduate school,
do you have pearls of wisdom, advice about the world that they're entering into in the sciences?
Would you do it again today if you had it to do over?
Oh yeah, no question.
As I said, I mean, science is deeply ingrained in who I am.
(23:57):
So I don't think about my career choice as being a job. And that was something that I
really found peculiar in talking to some of my other colleagues who treated science as a job.
And this is true for other professions, whether it be art, you can think about other professions,
(24:19):
medical doctor, for example, that you don't shut things off at five o'clock.
And I think that's one of the things in science that it's going to sound smug and maybe a little
bit elitist, I guess, but it's really don't go into science if it's not a passion.
Yeah.
Don't think about it as like, my job is to be a scientist because it doesn't stop at five o'clock
(24:44):
on Friday. Some of my insights come to me on Sunday mornings, unfortunately.
So I gotta write that down. So I'm up at four o'clock in the morning on a Sunday trying to
get this thing written down. And that's the way, I mean, musicians do that as well, right?
Yeah, of course.
Musicians get inspiration or artists do the same kind of thing. And I think that's the
(25:06):
attraction of being in this field. And it's for me, I feel blessed that I actually get to do this.
Yeah.
I'm kind of stunned sometimes that I get paid to do this. Wow. And that's, I think, one of the
things that people going through the system now really have to think about is that, is this
something that really I think of as a passion or is it something I'm doing just because this is
(25:31):
what you do?
Yeah.
So going through bachelor's, going through your PhD, going through your postdoc, if you're just
doing this because those are the steps, it's probably the wrong reason for doing it. And stop
at some point and think about, is that really what I want to do? I've been quite deliberate
with my students to have those conversations a number of times and be very frank with them.
(25:54):
I'm not going to think less highly of you if you decide I want to go into industry, for example.
Right.
Because that's a good choice and maybe that's actually what you want to do and maybe be
successful there. Try it.
Yeah.
Why not? There's no penalty really. You may take a pay cut. Actually, most people go into industry,
get a pay raise.
Yeah, they pay raise.
But it's really allowing that opportunity to explore where they want to go and supporting
(26:21):
them for that. And that's my job is to really support their journey so that they're making
the decision when they finish their PhD that they're going to be doing something that they're
going to be passionate about when they're 60 or 70 years old as well.
Right. Right. No off switch. In the end of the day, it's a living breathing thing. It's part
and parcel of your very being to be on the whole time. And I appreciate that too. It's not a job.
(26:46):
It's a privilege to get paid to do it.
For sure. For sure. Yeah. Like when we moved to Vancouver, we have a place now on Vancouver
Island as well, which is a great retreat. But when I go there, it's not like I shut everything off.
It's a different perspective.
Yeah.
No question. But it's not like I...
That's an important point of being a scientist too, right? Which is it's one thing to be in the lab
(27:09):
all the time and doing experiments. But the quiet time and the thoughtful time is where
a lot of the work gets done. And that's the part maybe that doesn't feel like work.
Yeah.
Where you're riding your bike and...
Well, the virtual brain history actually came not because Victor, you're sitting in a lab,
you were sitting in a pub having a conversation about it. And all of a sudden inspiration came
(27:30):
from that. And he and I do a lot of running together. So when we are on runs, we have these
great conversations that are kind of meandering and all of a sudden, oh, we should stop and record
this.
Yeah. So you could have been talking about the football game or you're talking about this.
It's just...
Right. Yeah. Yeah. Yeah. For sure.
Superb. Randy, thanks so much for taking the time with us.
Pleasure. I enjoyed it.
It's really great to have you here. Looking forward to seeing your work.
Yeah. Thank you.
(27:50):
Over the years to come.
Take care.
Popular Podcasts
United States of Kennedy
United States of Kennedy is a podcast about our cultural fascination with the Kennedy dynasty. Every week, hosts Lyra Smith and George Civeris go into one aspect of the Kennedy story.
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com