102: Soledad Gonzalo Cogno: Sloooow oscillations in entorhinal cortex, mentoring, and the physics approach to neuroscience

Soledad Gonzalo Cogno is a group leader at the Kavli Institute for Science Neuroscience in Trondheim. We talk about how she went from studying physics in Argentina to working on the brain in Norway, the importance of interdisciplinary approaches to neuroscience, why researchers should give their research animals a nice life, mentorship, and discuss her recent Nature paper on ultraslow oscillatory sequences in medial entorhinal cortex.

BJKS Podcast is a podcast about neuroscience, psychology, and anything vaguely related, hosted by Benjamin James Kuper-Smith.

Support the show: https://geni.us/bjks-patreon

Timestamps
0:00:00: Studying physics in Argentina
0:12:30: The advantages of a physics background - interdisciplinarity in neuroscience
0:27:31: How Soledad ended up in Trondheim
0:32:46: Rodent heaven in Norway
0:36:19: Start discussing Soledad's paper on ultraslow oscillatory sequences
1:03:12: So what do those ultraslow oscillatory sequences do?
1:16:18: A book or paper more people should read
1:22:30: Something Soledad wishes she'd learnt sooner
1:30:51: Advice for PhD students/postdocs

Podcast links

• Website: https://geni.us/bjks-pod
• Twitter: https://geni.us/bjks-pod-twt

Soledad's links

• Website: https://geni.us/gonzalo_cogno-web
• Twitter: https://geni.us/gonzalo_cogno-twt

Ben's links

• Website: https://geni.us/bjks-web
• Google Scholar: https://geni.us/bjks-scholar
• Twitter: https://geni.us/bjks-twt

References

Episode about Ramon y Cajal: https://geni.us/bjks-ehrlich

Brun, Solstad, Kjelstrup, Fyhn, Witter, Moser & Moser (2008). Progressive increase in grid scale from dorsal to ventral medial entorhinal cortex. Hippocampus.
Constantinou, Gonzalo Cogno, Elijah, Kropff, Gigg, Samengo & Montemurro (2016). Bursting neurons in the hippocampal formation encode features of LFP rhythms. Frontiers in computational neuroscience.
Dayan & Abbott (2005). Theoretical neuroscience: computational and mathematical modeling of neural systems.
Gonzalo Cogno, Obenhaus, Lautrup, Jacobsen, Clopath, Andersson, ... & Moser (2024). Minute-scale oscillatory sequences in medial entorhinal cortex. Nature.
Hastie, Tibshirani & Friedman (2009). The elements of statistical learning: data mining, inference, and prediction.
Kropff, Carmichael, Moser & Moser (2015). Speed cells in the medial entorhinal cortex. Nature.
MacKay (2003). Information theory, inference and learning algorithms.

Chapters

• 0:00 Studying physics in Argentina
• 12:30 The advantages of a physics background - interdisciplinarity in neuroscience
• 27:31 How Soledad ended up in Trondheim
• 32:46 Rodent heaven in Norway
• 36:19 Start discussing Soledad's paper on ultraslow oscillatory sequences
• 1:03:12 So what do those ultraslow oscillatory sequences do?
• 1:16:18 A book or paper more people should read
• 1:22:30 Something Soledad wishes she'd learnt sooner
• 1:30:51 Advice for PhD students/postdocs

Transcript

[This is an automated transcript that contains many errors]


Benjamin James Kuper-Smith: [00:00:00] So before we started recording, just as a sound check, you mentioned basically being a very non stereotypical Argentinian person.

Is there anything as stereotypical that you do or

Soledad Gonzalo Cogno: It's not, no, there's not something like very stereotypical that I would do, but I think my it's very funny because I became less and less Argentinian as I lived more and more abroad, but at the same time, in some regards and some aspects of my personality became. Very much more strengthened.

So some parts of my personality became more and more Argentinian and are more Argentinian than ever. But other parts of my personality kind of faded away. For example, I never drank mate and it was kind of an issue here in Argentina. It's like, Oh, you didn't really don't like it. And now it's like, okay, I can live my life without drinking mate at all.

No one will question me because of that. But but I do interact with people sometimes in a very Argentinian way. If [00:01:00] the context allows for it. Right. So.

Benjamin James Kuper-Smith: What would that mean?

Soledad Gonzalo Cogno: I can be very friendly. Argentinians are very friendly, like they will hug you five seconds after they talk to you for the first time and they will, you know, invite you over to their houses to have dinner and to hang out. That's something, that's very Argentinian. It's a very friendly

Benjamin James Kuper-Smith: not very Norwegian, yeah.

Soledad Gonzalo Cogno: And it's not very Norwegian. I can be very much as a Norwegian would expect me to be because I've been living there for such a long time now that I know the culture, I know the habits, and I can adjust very well to that. But the moment I see the opportunity to relax that a little bit, then I can enter the Argentinian mode and be, you know, more friendly, more open and all of that.

Benjamin James Kuper-Smith: Yeah, it's fine. I mean, basically the reason I thought I'd just ask that is because, we'll be talking about your research mainly today. But also kind of your career that kind of led you up to where you are right now. And you know, one place I I wanted to start [00:02:00] was I mean, I assumed you were Argentinian because you have a degree from Argentina, which typically,

Soledad Gonzalo Cogno: Yeah.

Benjamin James Kuper-Smith: Yeah, it means you're from there.

So I was just curious, you're doing kind of computational systems neuroscience, I guess, something like that but you started off doing physics from what I understand. So, maybe, yeah, when you were in, in school or whenever you made the decision to study physics, what was going on?

Why did you decide to do physics?

Soledad Gonzalo Cogno: Oh, yeah. That was a very interesting moment in my life because I really liked math. That was the one thing I really liked. And the one thing that I could be studying math all day, every single day. I didn't mind at all. I really liked it. And I wanted to be good at it. I remember that feeling, the feeling of enjoying it, and at the same time, wanting to be very good at it.

And it became very clear to me that even though there were many things that were within my control, umbrella of interest. Math was really the one thing that, that I felt passionate about. So then I started thinking, okay what is the right path for [00:03:00] me? And then I was, okay, perhaps this is engineering, but I knew that was more applied and I was not, that was not really resonating with my personality.

And I was like, okay, perhaps I should study like, I should just get a degree in math. And then physics became part of the equation. And I It's very interesting because I went to a high school where I didn't have any physics course at all. So all my education in physics started at the university level.

But there were, I found here and there I found some physics books and I started reading those on my own in my free time and I became in love with it. I fell in love with it. I really like it. And it was more like a gut feeling. I just went for it feeling that, okay this is something that I feel I would be happy if I started this.

And that was the case. I mean, it was just a gut feeling, but it was very, it was the right choice for me at that time. And that's how I started. So that's why I started physics. It was really a gut feeling and something that I really, it was not about the job that I would get or [00:04:00] the doors that it might open, it was really about what I enjoyed and what I liked. And then I studied physics. And then, you know, one thing takes to another neuroscience became an option. 

Benjamin James Kuper-Smith: I was a little surprised when you said you didn't have physics at all in school. Was that, I mean, not even, so I don't know what the school system is like in Argentina. I went to school in Germany and you always have to have some sort of physics and chemistry and biology throughout school.

Maybe not the last few years of your high school but certainly, you know, when you're like 12 to 16 or something like that, you always have something like that. Did you actually have none or just not in the latter stages?

Soledad Gonzalo Cogno: No, I didn't have any physics at all. So first of all, because I went to a high school, which was focused on humanities. So I had philosophy, I had psychology, I have tons of courses that were more humanities oriented. And then there was this one course on that. It was a combination of physics and chemistry.

But then the physics teacher she had a health [00:05:00] issue and therefore she couldn't teach. And then there was another person, a substitute that was supposed to teach both physics and chemistry, but she was only trained in chemistry. So I had some basic chemistry in high school, but I didn't have any physics at all.

Benjamin James Kuper-Smith: That's really interesting. Yeah, it's just because like, I mean, I chose psychology as my undergraduate in large, but I think because it was, Because I never had it in school. I mean, I also didn't like school. So I think it seems like you actually enjoyed some of the stuff you did in school which would be a contrast to me, but I think, you know, one of the reasons I chose psychology is because I just had, I had no experience in school with it so far.

So everything I did was kind of on my own, like finding out stuff on my own, which made it special. I'm curious, was that also kind of part of the appeal? This kind of like something you've. I mean, it sounds silly to say, like, you found psychology, like, I found psychology, you found physics, but was that also part of it, this kind of, like, doing something that's a bit unusual, or was it just a coincidence?

Soledad Gonzalo Cogno: perhaps there was a little bit of that because even [00:06:00] though I didn't have a bad time at high school, it was not something I didn't feel I fit it, you know, I didn't feel like, okay. I didn't feel part of that community in a way I felt that I was different. And I, and at that time, I think I even interpreted as some part of me, perhaps failing, because it was not a perfect match with the rest and with the interest and with the topics were being taught at that time, so it was very clear to me that I didn't fit there.

And perhaps that feeling is what triggered this need that I had to learn more math and to read physics books. And I did that and it was, you know, it was a happy place in a way, let's put it like that. I had fun doing that. And then everything kind of made sense when I started college and I went to the university and I properly started studying physics and everything was like, okay, yeah, this is how it was supposed to be.

Now it makes [00:07:00] sense.

Benjamin James Kuper-Smith: So it actually was you know, often, I guess, when you do something that you had little exposure to, you create this, like, false image in your head, and then it's very different, but in your case, it actually was what you were hoping for.

Soledad Gonzalo Cogno: Yes, it was even, it was, it went beyond my expectation. It was really. Yeah, it was really good. It's like, even today, like many of my closest friends ever, I met them during those first years of university and we were studying together. We were preparing exams together and they are, most of them are physicists and they're working on the most different topics that you can imagine.

Some of them are working as data scientists. Others are developing lasers. They're doing many different things. I'm the only one doing neuroscience at the moment. But it was really, it was a fantastic time because it's, it was, I remember, you know, this feeling of there are things that I like and things that I want to do, but nothing seems to be, you know, the normal path until it became the normal path.

But I just have to [00:08:00] let myself understand and acknowledge the fact that sometimes what doesn't seem to be normal, it's not bad, it's just different and you just have to go for it.

Benjamin James Kuper-Smith: And then you got validated, kind of, by

Soledad Gonzalo Cogno: Yes.

Benjamin James Kuper-Smith: Yeah, by the degree and the experience. So, but then, but why am I talking to a neuroscientist now and not a physicist? Or,

Soledad Gonzalo Cogno: approach neuroscience the way a physicist does. 

Benjamin James Kuper-Smith: Okay, but then why I guess, how did you, I guess you must've had some sort of contact to topics that are out. I'm assuming computational neuroscience isn't part of a mainstream physics curriculum. Or. So basically, how did that happen?

Soledad Gonzalo Cogno: Yeah. So when I started studying physics, you know, I wanted to do what you usually identify as physics, right? Like relativity or quantum mechanics or string theory. That's what I wanted to do. But as time went by, I [00:09:00] realized that Perhaps that was not really what I wanted to do, but I still remember that the moment I decided I enrolled in the master program and I had to choose one project and work on it on a different thesis, I did consider string theory at that time.

Benjamin James Kuper-Smith: So the master's was also in physics, or was it one program, or?

Soledad Gonzalo Cogno: No, so my degree, my master's and my PhD, all of them are in physics. And that's because as part of those programs, I had to take courses that are physics oriented. So, despite the fact that my, the work that I did wasn't computation on your science, both during my master's and my PhD. And but then at some point, you know, by the time I had to.

Choose a topic for my master's I was hesitating and I did consider strength theory I remember I even have an interview with a person that could supervise me, but I don't know It was again a gut feeling it was not nothing more [00:10:00] sophisticated on that. I was just feeling that was super interesting, a very cool problem to work on, but not the problem that would make me happy.

I just wanted to have a good time. I wanted to do it. I always did science and I do science nowadays because it's fun because I like it. I enjoy it. And the day I stop enjoying it, I would just get another job. I will not just continue doing it because it's what I have to do. No, that will never happen. But at that time, that was my feeling that I would have more fun, you know, working on neuroscience.

And that's what ended up happening. I had fun during my post master. I really liked it. I enjoyed it. And what's starting being more like a game I'm saying, okay what would it be like to be a physicist doing neuroscience? It became more and more of a real thing for me. And I don't regret it. I think it was the best decision I could have ever made.

Benjamin James Kuper-Smith: How did you choose your PhD topic?

Or first, maybe, was it clear that you wanted to do PhD? 

Soledad Gonzalo Cogno: Yes, I wanted to do a PhD. I was very sure [00:11:00] about that. I didn't know if at that time I knew I wanted to do a PhD, but I didn't know if in the long term I would want to have my own lab or if I would want to go to the industry that I didn't know at that time. But I did know that doing a PhD would put me in a good place for whatever option I wanted to pursue in the end. So that was always part of the plan. And yeah, so then I finished my master's and then I started applying for PhD positions like everybody else does. And then I had some personal issues family related issues that forced me in a way to not force me really, it was a choice that I made, but that very much encouraged me to stay in Argentina in my home country.

So that's why I decided to do my PhD here. And and for me it was a very natural thing to stay in the same place where I was, where I did my master's, which was interdisciplinary physics group with a focus [00:12:00] on statistical physics. So that's where I did my PhD. So it was really nice because even though I was doing neuroscience, I was still in contact with people doing statistical physics, and I took a lot of courses that were on the physics side.

I'm not so much on the biology side. So that gave me a very good training. Then there was a challenge during my postdoc in which I transitioned from being surrounded by physicists to being surrounded by biologists, but that's a different story. But that's a little bit how things were.

Benjamin James Kuper-Smith: One thing I wanted to ask, and I guess you Kind of, alluded to a little bit just now is I was just curious kind of what the what your studies in physics kind of background you have, what that maybe gives you that, you know, people with other backgrounds maybe don't have. And I guess that's always difficult to answer because, you know, you dunno what you would, how you would be thinking if you did something else.

But maybe we can approach it from the angle of. people you're surrounded by. So I [00:13:00] guess you went from a environment where you were mainly surrounded by physicists to one where, I don't know exactly what it was like in, in Tom time, but I'm assuming it was more biologists, I guess, in that sense. So maybe what's what differences have you noticed in the way that let's say for example, these two groups will choose whatever contrast you want think about the brain or science or whatever.

Soledad Gonzalo Cogno: Yes. So I think there may be differences. And first of all, I think neuroscience is an interdisciplinary effort. So we will not understand the brain. only with biologists or only with physicists or only with medical doctors. I think it's the combined activity of all of us in a synergistic manner that will really us to understand something about the brain.

I'm very much convinced about that. That said, I think there's something very unique about being trained in physics. When you want to understand the brain, which is the fact that you have a very formal way and [00:14:00] an abstract way of thinking, right? So you can, first of all, there's something about the formality of the way you think and the formality of the way you introduce the concepts, you define the concepts and you build on those concepts.

For me, sometimes it's very easy to identify when a paper was written by a physicist and when a paper was not written by a physicist, because when it's not written by a physicist, you know, you start with a definition and that definition starts changing along the paper. And that's something a little bit confusing for me.

And I was like, okay, but we were saying that at the beginning in the abstract, this word meant this thing. And then it turns out that it means something different. I feel that's not so much the case when you Have some experience with math and you were training math or physics. 

Benjamin James Kuper-Smith: You mean because basically your definitions are mathematical?

Soledad Gonzalo Cogno: yes, you have very well defined

concepts. 

Benjamin James Kuper-Smith: per se, but

Soledad Gonzalo Cogno: Exactly. Something has a very specific meaning and a very exact definition and [00:15:00] you build from those definitions. And that is something that that I didn't find so much in biology, sometimes. It's not always like that, of course. And then sometimes the meaning of understanding something is also a little bit different. The way of answering questions is sometimes a little bit different. So I want something that is quantitative, that you can put a number on it and that number is not subjective to anything that relies on hyper, pardon me. Or it could rely on hyper parameters, but that is something that is subjective, you know?

And, you know, those are some differences. 

Benjamin James Kuper-Smith: So what do you mean by hyperparameters?

Soledad Gonzalo Cogno: For example, I want a definition of let's put it like this. Okay. Like, if we are thinking about now, navigation and the meat and trying to cortex. Okay. What is it? What does it? What is the field? Right? So you have a neuron firing and it's firing in a field. What is the field? How do you define a field?

So I want a definition that is Okay, this means that this [00:16:00] neuron is firing here, and it's not firing here, and it's firing within, I don't know, one given radio, something like that. This is an

example. So you have an animal moving in a box, and then you have a cell that is selected to one specific location in that box, and it's going to fire in a bump within the location of that box. And okay, that's a field, right? Okay, what does it mean to be a field?

I want a definition that tells me, okay, this means that the animal is trying is that the cell has a high fragment right here, but not somewhere else, right? Okay. Something that, and that does not depend on how you calculated the fighting rate, how much is smoother defining rate whether the day that you were recording the animal was feeling like this or like that.

This is perhaps not the best example, but, 

Benjamin James Kuper-Smith: yeah, do you mean basically there's like different ways in which you can calculate things and you know some might be better for this or that but in the end you want an answer that's true for all of those approaches or?

Soledad Gonzalo Cogno: Yes, let's put it like that. Yes, exactly. And

Benjamin James Kuper-Smith: So I guess like a yeah like a [00:17:00] finding that's robust not yeah that you can

Soledad Gonzalo Cogno: that's one thing. Yes. And so one thing is how you define the quantities you want to work with. And the other thing is, what do you mean by understanding something and by quantifying something? And I can give you an example of that. Because. When I was doing my PhD, for example, I took several courses on information theory.

I calculated a lot of information, mutual information myself between different quantities. I don't know, for example, the number of spikes and features of the local free potential, the number of spikes and features of the space where the animal is running. I did that a lot. I had tons of experience about that. I start my postdoc and then there was this quantity which is it's based on information theory. And I have never heard of that. And then I started asking, but what is this quantity? What does it measure? And no one could write the equation. It's a very simple equation, right? So [00:18:00] for me, understanding what that measure was just writing it, assuming that the neuron is firing like like a Poisson neuron.

And then assuming that, then you have a very simple derivation and then you reach this This information measure, right? But my colleagues were so used to working with that and had a very developed intuition and much, much better intuition of what this quantity could tell you about. You're on a firing in the middle and trying to cortex than I did at that time.

Yeah, they couldn't tell me specifically what this measure was. So that's when you start understanding that other people think about things in different ways. And sometimes the way you think about it can contribute with something and the way another person thinks about it might contribute with something else.

So at that time, I did have the math, I didn't have the intuition about the brain. And I feel that only having the intuition is not enough, only having the math is not enough, then you have to, you know, bridge the gap. These two scenarios and these two situations in order [00:19:00] to have a better understanding and can and use that quantity to actually measure Something that is relevant in the brain.

Benjamin James Kuper-Smith: Yeah, I mean So what can, for example, intuition or something a bit fuzzier give you that formal mechanism or the kind of precision that you expect? What can that give you? Or maybe what are some of the potential let's call them problems. This may be a big word, but maybe blind spots of, you know, of having formal models. Maybe what are some things that, where it doesn't work quite as well and where you are maybe glad that you gained? Yeah, the kind of intuition that the biologists had.

Soledad Gonzalo Cogno: Yeah, so first of all Before I started my postdoc because during my postdoc I did experiments myself Which means that I did the searcher. I did the histology. I Made recordings myself And that was an eye opener for me because I realized how ignorant [00:20:00] I was about the brain and about the behavior of the animals and about the problem itself before I started doing experiments myself.

And I don't know what will happen in the future. I'm now, even though I did experiments in my postdoc now in my lab, I'm returning more and more to computational neuroscience. So I'm not sure how much more experiments I will do in the future, but regardless of how much I do. Having the experience of doing experiments was very transforming for me.

And I think. Every computational and theoretical neuroscientist that has the opportunity to not do experiments themselves, but to, you know, go to the lab every now and then and see how things are done or have a chat, you know, a constant dialogue with experimentalists. I think that's very important because you learn a lot about the system that you're investigating by just looking at it. I remember that before I did experiments myself, I used to assume that the animal, [00:21:00] you know, in the typical experiments of navigation where you have an animal which is running in a box, I would assume that the animal has a very homogeneous coverage of the box and that the speed is more or less always the same.

And the animals sometimes do not like running in the box. And always prefer the corners or to walk along the walls. And sometimes they don't like being in the center, meaning that sometimes the coverage is not homogeneous. Sometimes it is sometimes it's not. It really depends on the animal. And that's the key point.

It does depend on the animal. All the animals have different brains, the surgery is not always the same, the behavior of the animal is not always the same. The experimental conditions are not always the same. So things change and being aware of this variability is very important. And I think it makes you a better neuroscientist.

Thanks. So I very much treasure the time that I spend doing experiments myself, and I very much treasure the interactions that I have had with experimentalists. And I learn a lot discussing with my colleagues that are more [00:22:00] on the biology side, right? I remember that we were discussing oscillations very much in the early days of my postdocs.

of my postdoc, and we were discussing about oscillations. And then this colleague of mine used to tell me, but I want to see an oscillation. I want to see the figure of an oscillation. And I was like, but I'm showing you the Fourier transform of the activity of this neuron. It has a sharp peak. This is an oscillation. And they were like, no, but I'm not convinced about this. I want to see the oscillation. I want to see something that is going up and down. And for me, that was very transformative because it really made me understand that what is crystal clear for me might not be crystal clear for my colleague. And that there's something that I have to understand in how I communicate science and how I communicate with my colleagues, if I want to work in an interdisciplinary environment.

So perhaps that was really one of the. Most deep and meaningful things that I learned [00:23:00] during my poster, which was really how to interact with people from different disciplines and from different with different backgrounds. And I really, I think that was a fantastic experience for me.

Benjamin James Kuper-Smith: Yeah, I guess someone should offer a course like Psy what would be the word? Yeah, basically like a translation service from physicist to biologist and the other way around. But we'll start talking about your actual research soon, but I'm just curious because you're starting your lab now and or you have started your lab now.

I'm just curious now from this kind of, whilst we're talking about interdisciplinarity and, you know, this kinda stuff what kind of people do you want in your lab, like as PhD students or postdocs? Like, do you want, you know, the people who have. All the other experience who could, you know, a postdoc who did the entire lab in, you know, doing electrophysiology or whatever it might be.

Like, do you really want the kind of other side of neuroscience or do you want people who have the same [00:24:00] background as you and then you collaborate with someone else or what's your thinking there? And I guess your lab also isn't in a vacuum. I mean, you're still at the Cavalier in

Soledad Gonzalo Cogno: Yes, exactly.

Benjamin James Kuper-Smith: yeah.

Soledad Gonzalo Cogno: So, at the moment, I'm looking for, I mean, if I can choose, then I would prefer people who have a physics background because then I feel very comfortable because then I know the courses that they took and the things that they know, and we have a language that is shared, you know, it's like, my mother language is Spanish and I can speak English and we can communicate as we're doing now in English.

But there's something some, there's always something more comfortable when you can express yourself in the language in which you grew up. And for me, that's the language of math and physics, right? So I can adjust the way I communicate myself, but there's always something simpler when I exchange ideas with physicists.

So that's what I'm aiming [00:25:00] for. But at the same time, because. At Kavli, and this is more like a strategic decision, right? But at the Kavli City for Systems Neuroscience, we have, It's really an institute that is at the forefront of experimental neuroscience. So we have all the facilities for training people in electrophysiology, for training people in calcium imaging.

We have people that will be there and that their job is to train you and they will do it with the highest standards and with the most positive attitude, we don't have the infrastructure for training people. In math and data analysis, and as a matter of fact, I developed a PhD courses at Kavli on algebra and calculus and neural data analysis, but that's still at the very early stage.

So that's why. Just to summarize, because it's simpler for me and because we can train them in the experimental component of neuroscience, I prefer to aim for physicists at [00:26:00] the moment. And also because the lab is growing in the computational direction, right? So the lab myself is not gathering any data at the moment.

All the data that we're analyzing is in collaboration with other groups. So we need people to develop computational models and we need people to analyze data. Yeah, so that's why I'm also aiming for But that said, I'm also of the idea That you can study something and then you can learn whatever you want to learn as long as you are committed to it and you are willing to put the time and the effort in it.

So, yeah, I'm open to, you know, what, at the end of the day, I really want people that are passionate about neuroscience because I'm very passionate about it and I could not work with people that are not really passionate about it. So that's really the most important thing for me.

Benjamin James Kuper-Smith: Yeah. Yeah. I had something like that recently where someone said something like you know, a kind of passion and agency or drive is kind of more important than anything else because like the people who would just like figure stuff out, they figure it out kind of no [00:27:00]matter what their background is.

Soledad Gonzalo Cogno: exactly. Exactly.

Benjamin James Kuper-Smith: yeah, if someone has the perfect background you want, but like you have to feed them, hand feed, that's the word, not feed them, hand feed them or hold their hand or whatever. I'm getting my metaphors mixed now. That's a lot more work than just someone who will just, you know, run with it, maybe make some elemental, elementary errors, but that can be fixed.

Soledad Gonzalo Cogno: Yeah, absolutely. I think a creativity and passion and willingness to put the hard work that is by far more important than anything else.

Benjamin James Kuper-Smith: Okay. So how did you end up in Trondheim? Because from what I understand your research before wasn't about spatial navigation, or the hippocampal formation or anything like that. So how did that happen?

Soledad Gonzalo Cogno: Yeah, so the first half of my PhD was really on computation and models that were not very much that were inspired by experimental data, but we're not really, you know, tied to experimental data. So, I did a lot of computational modeling at that time. [00:28:00] But then, at some point, we started collaborating with a person who actually was supposed to be in the Moser lab before I joined the Moser lab, who is Emilio Krupp.

And Emilio Krupp is the first author of the Speed Cells paper. So, Emilio worked they discovered the speed cells in Trondheim, and then he's Argentinian, so he returned to Argentina, and by the time he returned, I was doing my PhD, and then we started collaborating. And then I started analyzing data collected in the medial intranal cortex and in the hippocampus with tetrodes at that time, and mainly local fin potential.

So I was mainly working with the theta rhythm with oscillations at that time. So, during the second half of my PhD, I did a lot of data analysis and I had this very nice project. About resettings of the theta rhythm in the median adrenal cortex and in the hippocampus at the potential link of those three settings to a reset of the path integrator, [00:29:00] which then I finished my PhD and I never wrap it up.

So, it's not published yet. Hopefully, we'll come at some point in the future, but that was, but that's what I was working on at the moment. So I was already working in navigation at that time and I learned how to analyze data or became familiar with how to do it already in my PhD. So that was what's happening, what was happening at that moment.

And then I applied to a summer school in Norway, because when I was a PhD, I was, you know, I was in Argentina, we were very isolated. So I was making, I was very proactive in trying to Collaborate not collaborate, but to interact with as many people as possible from abroad because that was the only way in which I could get a glimpse of what was going on in the world.

So I applied to this school and then I was selected. So then I was selected to come to Norway and to participate in the summer school, which was going to be held at the Kavli Institute for Systems Neuroscience. And then because we saw the opportunity, which would let Edvar [00:30:00] and Mybred know that I was working on this data set that was collected in their lab, and that I had this very nice finding, and they invited me to give a talk.

So that's how it all started. Came to Trondheim, I gave this talk on what I was doing during my PhD. And this was essentially the day before the summer school started. And the talk went well, and then I got offered a postdoc position. So that's how I ended up in Trondheim. And that was already like seven years ago.

Benjamin James Kuper-Smith: Okay. This is a question that won't come up often in my interviews. But if I get my timeline correctly, I think the Moses had won the Nobel prize like three years before you started there. Something like that. Was that a reason? Pro or against going there? Because in some reasons it seems like an obvious pro, but in other reasons, I mean, senior professors already get invited to lots of stuff and don't have a lot of time.

And I would imagine that you know, if you win a Nobel Prize, you get invited to basically everything that, that [00:31:00] happens. Um, yeah, I was just curious, like, was that was that anything you considered or was it you know, they do great research. The lab is great. And yeah,

Soledad Gonzalo Cogno: No, well, first of all, I started analyzing data from their lab before they got the Nobel Prize. So I was already working with that data before they got the Nobel Prize. And I remember very well the day that I found out that they got it, that they won the Nobel Prize and I was in shock. And then, of course, I mean, it's flashy, right, to have your supervisors that have this amazing discovery and that they are awarded the Nobel Prize for that. Of course that's something nice, but that was not really the reason why I decided to join the lab and as a matter of fact, I They receive applications from people every single day.

I never applied to the lab. I just went there, I saw the place, I discussed with Edward, I discussed with Myra, and I felt comfortable. I liked the place, I liked the people. I thought the two of them were absolutely amazing. [00:32:00] Not only because of their science, but Because they were so kind to me and they were discussing science with me as if we were equals and we were not equals.

They were these gigantic people in neuroscience and I was this person doing her PhD in this remote country in the world and they were still Willing to discuss with me, putting time in discussing with me, discussing science, hearing my thoughts about what I was saying and their papers and everything.

And I thought, okay, this, they are amazing. science is amazing. That, of course, there's no doubt about that, but, you know, people wise, they are awesome. And I just thought, okay, this is an amazing place. And then I visited the animal facility. And trust me it's amazing. It's like wonderland for mice and rats.

It's like the animals are so happy there and we have the highest standards and the animals so happy [00:33:00] and are so well taken care of. And, It's really good. So all of, when I saw all of that, I realized that's the place where I wanted to do science and it was really not about what they accomplish, but it was about what I felt I could accomplish by being in a place like that.

Benjamin James Kuper-Smith: Yeah, just out of curiosity What does it mean? Like, you know, you just mentioned the facilities being amazing. I mean, I've never worked with animals what does that can you yeah, just a bit more detail what that means because I guess those are the kind of reasons why I have this podcast so I can kind of learn about these things that I would never find in a paper or casually briefly discussing stuff with people

Soledad Gonzalo Cogno: yes. So, we have very big cages for all of our animals. We have tons of toys in all the cages. We have an amazing group of people that take care of our animals, that check that the animals are in good shape. The moment the animals are in good shape, not in good shape. It doesn't matter if it is Saturday, Sunday, or Christmas, they will give you a phone call and they will [00:34:00]ask you to please come to the lab because this animal is in pain.

And then you have to do something about it. You have to take care of your animal. So really we have the highest standards for animal welfare. And I'm very proud of that because the moment I started working there, I realized that it could never, ever work. with experimentalists that do not take good care of their animals. They're not only our instruments for us to perform our research. Those are, they are alive. They experience happiness. They experience pain and we have to make sure that they have a good life. So, for me, that was amazing. And, you know, it's a little bit of a culture that we have at Kavli.

When I was trained on how to do surgery, I was trained not only on how to implant, you know, the electrodes and how to check that the animal is breathing correctly, but it's also about how to take care of the animal, how to take care of the animal while the animal is recovering, how to play with the animal, how to allow the animal to, to know you, you know, and to get used to you so that the animal is [00:35:00] not in stress.

We are trained on how to recognize stress signs in an animal and to avoid that. We never pick the animals from the tail, for example, because that's a stressor for them. So we put our hands like in a cap so that they can jump on it. And our animals really get to know us very well because then you just approach your hand and the animal jumps to your hand directly.

And that's really nice. So when I saw all of that and that the, you know, the values when it comes to animal welfare aligned with the values that I had, I was like, okay, this is amazing. And if there's one place in the world where I will try to learn, at least try to learn how to perform experiments and how to work with animals, I want these, the place where I will develop myself in that direction.

And that's what I wanted at that time. So it worked out very well.

Benjamin James Kuper-Smith: mean presumably it also apart from an ethical perspective just from a You know scientific [00:36:00] practical perspective. I imagine it also helps with science if your animals are generally happy and you know, not stressed the entire time and 

Soledad Gonzalo Cogno: Absolutely. It's really it's about happy animals and happy animals enabling you to do excellent science. So yes, of course.

Benjamin James Kuper-Smith: Okay, so, yeah, the main kind of paper I guess we want to talk about is the paper about the what do you call it? Ultra slow oscillatory sequences. Yeah, maybe how did kind of, what was the starting point of that? How did that start? What was the initial, I don't know. Yeah. What kind of sparked that project?

Soledad Gonzalo Cogno: Yeah. So, um, what sparked that project was essentially this project is very tied to the one thing that I have always been very interested in when it comes to neuroscience. which is how the activity of many neurons is coordinated as a function of time. So I was always very passionate about that and [00:37:00] understanding not what one individual neuron is doing, and how one individual neuron is responding to this specific stimulus.

But more generally speaking, now you have a population of neurons, a neural circuit, the neural circuit, these neurons are doing different things. How is their activity modulated as a function of time? So that's always, that's the question I brought with me to Norway. Even before this project started, that was already the question that I was, that I wanted to somehow tackle.

But then this question, of course, has been addressed in the context of oscillations, right? Because people have studied oscillations and how oscillations can modulate activity of individual neurons and also of ensembles of neurons as a function of time. For many years, right? And this is the entire field of oscillations and you have so many different frequency bounds like theta, gamma, slow gamma, fast gamma, delta, you have so many frequency [00:38:00] bounds.

And of course, these oscillations at these different frequency bounds are known to be associated to different computations and brain functions. So I was very interested about that and of course, as I mentioned before, I was also, you know, trained in the oscillation field. But then all of that was always in the millisecond timescale, right?

So, meaning that you have several cycles within a second. But we also know that there are many brain functions that unfold at much slower timescales. then the question is, are there coordination mechanisms that coordinate the activity of the population as a whole at these lower timescales? So that was a little bit the driving question that we had.

And we decided to work on the internal cortex because As we know, the internal cortex plays a crucial role in navigation, but also in epistatic memory formation, which are, you know, brain functions that unfold in this very slow [00:39:00] timescale. So that's why we decided to look into the MST. And that's how it all started in a way.

Benjamin James Kuper-Smith: Yeah. Could you give like a few examples of just like what you mean by like events that happen on a slower timescale, just to make it a little bit more concrete to kind of, , just to kind of, so we have a slightly more concrete idea of different behaviors and maybe also some that were like the very fast timescales do make sense.

Soledad Gonzalo Cogno: Yes, for example let's think about different examples, right? But if we think about birds the process in which a bird learns a song and performs a song that, that happens over a timescale of seconds, right? So that's one very specific example. But there's also in general, you know, going from A to B, you know, describing a trajectory in space, taking you from one point to another one, that is also something that unfolds on very slow timescales.

And that might require activity to also unfold in [00:40:00] very slow timescales. So, but episodic memory formation, right, which we know. The entorhinal cortex is very much associated with, or not associated with, but important for this brain function, in which you have a series of events. That you recollect as a memory as a whole.

Right. So these are things that can unfold in many seconds, even minutes, or sometimes more than that. But beyond this specific examples, everything that is behavioral. Or behaviorally related, that's also slow, right? So if you have an animal that in a box, right? And the animal is climbing or the animal is grooming, those things always happen in the time scale of seconds or more.

It does not happen in fractions of seconds. So, we have studied neural activity in a time scale of milliseconds. And as a matter of fact, when you receive, for [00:41:00] example, like this. Pi times of one neuron in your data, and then you want to create a spike train, and then you build the data. You usually use a few milliseconds as a beam size, right?

And that's because that's what we usually do. But that's already putting in a bit, not a constraint, but a prior on the timescale at which you will analyze your neural data. And we were just wondering whether there's something that goes beyond those timescales. And that's how it started in a way.

Benjamin James Kuper-Smith: so what you do then you just I mean, you know, the experiments you have in the paper, are those basically how you started? Or was there like a different version you tried before or kind of

Soledad Gonzalo Cogno: No,

no, those were

only, 

Benjamin James Kuper-Smith: go from that idea to the

Soledad Gonzalo Cogno: no those were the only experiments that we had. And the rationale for that experiment is that in that paper, the experiment was like this. We had an animal on a wheel. It was head restraint and the animal was performing a free [00:42:00] pace task on a wheel in what we call sensory minimized conditions.

Meaning that. The recording room was dark, there were no sounds and there was no visual stimulation or rewards associated, for example, with the position of the animal, nor anything that could indicate to the animal where it was positioned on the wheel. And the reason why we did that was because if we didn't have that, so let's say that you have someone go in a simulation that may reset these very slow oscillations, right?

That's the problem when you want to investigate things that start unfolding at behaviorally relevant timescales, that you might have features of behavior or stimulation that might start interfering or modulating or resetting the signal that you want to investigate. So, in order to get rid of all of that.

And have this thing now in its pure or most pure possible way, we decided to peel off all of these [00:43:00] stimuli or behavioral or reward and. Features and then only focus on what's happening in this free based talk. So that's the only experiment that we had. We didn't try anything else. We recorded many animals, of course, and we improve the tracking system of the animals while the experiments were developed.

And I should acknowledge here that. The people who collected the data, it's really amazing people. All the coauthors in that paper are really talented and I learned so much from all of them. So it was really a team effort. It was not only me. But that's how that's, was the only experiment that we had.

Yes. Oh,

Benjamin James Kuper-Smith: I understand correctly, there was basically in, you know, Dark, silent room. There was a mouse, right? And it was on top of some sort of ball that would move if the mouse moves. So, I mean, why wouldn't the mouse just, like, lie down and sleep or something? I mean, it was head fixated or something, right?

But, like, [00:44:00] basically, I'm just curious, like, in a sense, what is the mouse thinking

Soledad Gonzalo Cogno: that's a great question. I would love to, I would love to know, I would love to know. So, but to answer your question your intuition is correct, right? It's like, why would you care in running if you can just be there and just lay there and some animals do that and they're. Lazy and they just, you know, sit there and they don't want to move.

And then if they don't want to move, they don't move. That's totally fine. Then we recorded for half an hour or one hour. And if you see the behavior or the speed of the animal as a function of time, you will see essentially a flat line at zero. But then we had animals that were running all the time that were running like crazy and that they really enjoy running.

So we had other animals in that paper that were essentially running all the time. And that's a matter of fact. This combination this, by model distribution, if you want, in the behavior of the animal where there were animals that were running all the time or running or animals that were not running at all, made the analysis of the behavioral component [00:45:00] of the data very difficult because we didn't have enough data in between.

We didn't have alternations between running and non running. So now in the foot, we are working out on, we're working on follow up project. Now of that paper one of the things we want to achieve is a better sampling of behavior and running speed and onsets and offsets of running. But to answer your question, yeah, some animals were sitting still, but some animals were running a lot.

Benjamin James Kuper-Smith: Yeah, if I remember actually correctly, anything in the paper, the one you provide the running speed of one animal that is actually

Soledad Gonzalo Cogno: Exactly.

Benjamin James Kuper-Smith: quite a lot of the time, if I remember, yeah, okay. So, okay, I mean, we actually haven't talked about, like, what you actually found then so maybe, I mean, you found quite a lot of things, but yeah, can you kind of, step one of the results, basically, what did you first kind of look at?

Soledad Gonzalo Cogno: So essentially we started analyzing the activity of individual neurons. Kind of very typical approach when you're

Benjamin James Kuper-Smith: Oh, sorry, actually can I interrupt you just first? [00:46:00] You, one thing I wanted to ask that maybe it's better to ask this first. First rather than later, is you used calcium imaging and Neuropixels, right? I think whether,

Soledad Gonzalo Cogno: We started,

yeah. 

Benjamin James Kuper-Smith: So basically why those two modalities you know, what's better or worse for your context about either yeah.

Soledad Gonzalo Cogno: Yeah. So we started The main data set that we analyzed in that paper was with calcium imaging and calcium imaging is really great because when you implant the prism, then the prism gives you really optical access to layer 2 of the median intranal cortex, which is where grid cells are. So, it's really beautiful because you have access to layer 2 of the median intranal cortex.

And not too deep layers, for example. So we started with Calcium in Machine and most of that paper is devoted to Calcium in Machine recordings. For the follow up projects, we started some time ago recording with Neuropixels. And when [00:47:00] that project started a few years ago now, Neuropixels were not a thing at that time.

While we were working on that project, Neuropixels became an available technology. And by the time we were wrapping up the project, I was already wanting to switch to Neuropixels because of some scientific questions that I want to answer that I think Neuropixels and electrophysiology in general might be better suited for those questions.

So that's why we also started doing Neuropixels and we wanted to show in the paper that the results of Were robust to the technique that you were using to record and your own data, but your pixels is more challenging in the sense that when you implant your pixels in in, in the mouse brain, you don't.

Sample on the cells from layer 2, but also from deeper layers, because the way you access the brain is very different from the way you access the brain when you're doing cost and emotional recordings. So you are recording from different [00:48:00] anatomical locations. And we are working on that at the moment.

Benjamin James Kuper-Smith: And the other locations are just not that relevant for what you're doing, or what's the problem with

Soledad Gonzalo Cogno: No. It's not a problem. It's a still medial internal cortex, but now you're recording from deeper layers as well. So now we are recording not only from very extended medial lateral locations, but also not only from layer two, but also from deeper layers as well. So we have a collection from cells from, so there's nothing wrong or

Benjamin James Kuper-Smith: the reason I was asking, so I was just asking because I think you mentioned that grid cells are found in Layer 2, if I remember correctly, and, so do you, basically my question was basically whether you thought that maybe Layer 2 is the most relevant for what you're interested in, therefore recording the others maybe isn't that interesting, or,

Soledad Gonzalo Cogno: no, it's centrally, it's certainly interesting, but that's something I want to understand, right? Because. Connectivity wise, Layer 2 is not the same as Layer 5. And in terms of what the cells are doing, it's [00:49:00] not the same either, right? So for example, we found very beautiful great cells in Layer 2. And very beautiful head direction cells in layer in layer 5, for example, so the cells are doing different things.

Connectivity is different. So what I would like to understand is how these different connectivity patterns in these different layers are shaping the dynamics of the circuit. And for that, I want to tease apart. these different contributions and these different layers. So, but Neuropixels is great in that regards because now we can record simultaneously from different layers and we can put the data together or separate them according to, you know, a dorsal medial sorry, dorsal ventral axis or superficial to deep layers.

And we can start looking at the dynamics of the population as a whole and at the modulation by these ultra slow rhythms as a function of the anatomical location as well. Okay.

Benjamin James Kuper-Smith: as an outsider, the I haven't even thought about [00:50:00] layers. Cortex for me, that's just enter out of cortex. That's the precision I need right now. Well, like that, that I remember. But okay. So I interrupted you earlier uh, results.

Soledad Gonzalo Cogno: No, well, so then, all right, so we had data from individual neurons, many neurons recorded at the same time. So then we took like the easiest approach and the most traditional approach, which is to investigating whether the activity of this individual neurons was modulated by rhythms and by ultra slow rhythms. And that's figure one in the paper, actually. And essentially, we took a very simple approach in which we calculated the autocorrelation single cells. And if you were to do that in a cell that is theta modulated, you would see a very nice oscillation in the autocorrelation at a frequency of eight hertz.

So. So we did exactly the same and we found this oscillation in the other correlation and when we calculated the power spectral density, it turned out that this oscillation [00:51:00] was ultra slow because it had a frequency that was way below 0. 1 hertz, which is what in the community we identified with the ultra slow regime.

So that was the first result of that paper, that the activity of individual neurons in the median entorhinal cortex when the animals are running in this sensory minimized state condition, the activity of neurons is oscillating and is oscillating at ultra slow frequencies. So that was the main,

the first main result. had already actually I mean not for enter random cortex, but I have a link in the introduction already say like there was already evidence for these kind of very slow oscillations. So I think that wasn't the I think now kind of it gets interesting, right? When we get to the sequences.

Because that was not, that was a little bit surprising, but not entirely surprising because ultra slow oscillations or infraslow oscillations have been reported in the past, but it has been reporting, you know, like a bit in an anecdotal way, you know, a little bit here and there, but there's not a community of people investigating that.

So you would find [00:52:00] examples in this brain area and this species here and a few years later in this other brain area and this other species here, but there was not, you know, a big body of literature dedicated to those oscillations. And we found these oscillations in essentially all the neurons that we were recording from.

So that was very interesting. But then, the fact that we could record many neurons at the same time, that was really exciting, and that was the main difference with older papers. Where there were only a few neurons recorded at the same time. No, here we had 500 neurons or more recorded at the same time.

All of them were oscillating at roughly the same frequency. That was mind blowing. And that was amazing because then that triggered the question of how are these oscillations at the single cell level organized at the network level? What is the population dynamics? When you have these ultra slow oscillations at the single cell level and express so strikingly in an entire population that was recorded.

So that was the next question and that [00:53:00] took us to the second piece of results of that paper, which is the fact that these ultra slow oscillations at the single cell level are organized at the neural network level into these ultra slow oscillatory sequences. This means what? This means that you have now this bunch of oscillators and these oscillators are phase shifted one with respect to the other, meaning that cell one fires and a little bit later cell two fires and a little bit later cell three fires.

So it's a sequence. It could be synchronized activity, but it's not, it's a sequential. So those are the two main results of that paper, that the activity. of individual neurons in the entorhinal cortex in this sensory minimized behavioral setting. The activity of those neurons is oscillatory and oscillating at ultra slow frequencies, and that at the network level, these ultra slow oscillations are organized into ultra slow oscillatory sequences.[00:54:00]

Benjamin James Kuper-Smith: Just to to check that I understand something in the paper correctly you mentioned that they are, I'll get the mathematical terming correct now, but that they're organized in a circle basically. It's not a circle but I hope you know what I mean. By that you just mean that basically once neuron 10 fires it goes back to one again, right?

That's basically all that means.

Soledad Gonzalo Cogno: Exactly. That you

can 

Benjamin James Kuper-Smith: can you maybe explain a little bit more about that just to clarify that, that it's this kind of cycle, I guess, of.

Soledad Gonzalo Cogno: Yes. This means that, well, essentially you have all your neurons, these neurons will have one given position in the, on the cortex, you know, an X and Y position if you disregard the depth of the plane that we were imaging from. And they, these neurons are doing something, right? And that's something that they're doing, which is these oscillatory sequences.

Now you can, you could conceptualize as all these neurons now being arranged, not on the surface of the internal cortex, but instead of being arranged on this [00:55:00] abstract ring. And the ordering of the neurons on this ring is given by the position in which they're firing in the sequence. And because it is a ring, once the last neuron that was recorded fired, then the first one will record.

Meaning that it has periodic boundary conditions. And this is because, it has periodic boundary conditions because these sequences progress one after the other. So you have one sequence and then the next sequence comes, and that's why they are, we call them oscillatory, or you could think about them as periodic.

Those are periodic sequences. Okay. Thanks. Which is consistent with the activity traversing a ring shaped manifold.

Benjamin James Kuper-Smith: Okay, that's a more precise way of saying it. Thank you. Um, from what I understand, those are kind of the highlight findings. And then the remaining results you did were checking up exactly what this means and [00:56:00] what it might be, or what might not be.

Maybe so that you mentioned a term in the in the paper that I've heard but I've, I'm not particularly familiar with, which is a traveling wave. So what is that and why did, from what I understand, you not find

Soledad Gonzalo Cogno: Yes, exactly, that's correct. Okay, so let us start with a travelling wave. What is a travelling wave? A travelling wave is essentially what you would find, for example during development, that's, I think, one of the first papers were probably ways were reported in the visual cortex, for example, and this means that neurons that fire close in time. Are close in anatomical space. So this means if I fire first and a few milliseconds afterwards, you fire, that means that we are neighbors in a way. And if there's a time lag between me firing and you firing of, I don't know, several seconds, that means that you and I are far away within the cortex. if [00:57:00] that happens, then it's a traveling way. And essentially, how do you, I mean, it's very nice when it's calcium emission, because then you have the movie and then you can see. The field of view and when it's a traveling way, essentially what you see is the activity sweeping across the cortex and it can be in a linear way, meaning that the center of mass of the activity, if you want is moving as a line is describing a line in, in, in the cortex, or you can have a more complicated shape, or it can be, you know, like, like a circular way.

So it means that it starts and then it starts propagating such that the radius.

Benjamin James Kuper-Smith: drop a

Soledad Gonzalo Cogno: a stone in the water, exactly. And then the radius starts increasing as a function of time. So those are different types of traveling waves. And then the question was whether these sequences were topographically organized in the medial entorhinal cortex.

Because if they were topographically organized, then these sequences would manifest as traveling waves. We had reasons [00:58:00] to believe that this could be traveling waves, because the entorhinal cortex is topographically organized. Right. For example, grid cells fire with a very small scale in more dorsal locations and with a much larger scale in ventral locations.

So that's one example of a gradient in the medial and frontal cortex. So we thought, okay, perhaps this is a traveling way, but it turns out that it's not. And we showed that in many different ways. We've had tons of quantifications. Despite all the quantifications, I think that the most convincing result is when you just look at the field of view and you see that there's really nothing that resembles a traveling wave.

It's more of a salt on paper organization of the activity. So, no, we didn't find any traveling waves or hints of traveling waves there.

Benjamin James Kuper-Smith: So what does that tell you? What would, if you had found traveling waves versus now that you didn't, how does that change how you think about these oscillatory sequences?

Soledad Gonzalo Cogno: Well, [00:59:00] you know, oscillations and sometimes, for example, the theta rhythm are known to be topographically organized. So the fact that these sequences are not topographically organized has implications, for example, regarding connectivity and has implications of different sorts. But and essentially that's one of the things we want to understand now, right?

We want to now build on this phenomenon by understanding More mechanistically, how the internal cortex can generate these dynamics and whether there's something as specific and special of the median internal cortex in generating these dynamics that you will not find in other brain areas. So, that's a little bit what we are working on at the moment, but the fact that these are not traveling waves is telling us, it has implications for connectivity, but it also is telling us that traveling waves have been reported.

So many times in the brain, so I think it's it makes it more novel that these are not traveling ways. [01:00:00] It's more like food for thought.

Benjamin James Kuper-Smith: Is it in a sense that almost a traveling wave is easier to explain almost? Like you almost have like just, like it's just like activation that kind of spreads through rather than yours seems to be, I don't know, some sort of more complicated 

Soledad Gonzalo Cogno: Um,

Benjamin James Kuper-Smith: not mechanism, but

Soledad Gonzalo Cogno: no, I wouldn't say the traveling ways are easy to explain. I feel that when it comes to the rain, nothing is really easy, but but they were more thoroughly described. And we know, for example, that data can be topographically organized. So in a way, if this had been traveling waves. This would have been more in line with already reported reasons just that this would have been at the slower time scale, but this seems to be a little bit different in that regards. But for me, if this had been traveling ways, it would have been equally fascinating for me as the person working on this.

Benjamin James Kuper-Smith: Okay. You also recorded from two other brain regions. I took [01:01:00] many from that, that, you know, it is you know, that you didn't just find this everywhere. Is that, is there more to the results than just showing that, you know, it's somewhat specific to entorhinal cortex or how would you summarize that?

Okay.

Soledad Gonzalo Cogno: Yeah, no, exactly. That's what we wanted to check whether this is something that we would expect in other brain areas as well. Because. The thing is that ultrasound oscillations, as we were discussing before, were reported in many different brain areas. across the years and in the literature, of course, right?

But there's not a very good system and systematic over the years of, of ultra slow oscillations, but hopefully this will change from now on. I mean, there are amazing papers on the topic, but it's, there's, it's not, you know, as well established or as well characterized or investigated as the theta rhythm or the gamma rhythm.

So hopefully that will change in the future. That said, these ultra slow oscillations have been reported in different brain areas. we wanted to investigate whether these ultra slow oscillatory [01:02:00] sequences were also present in different brain areas, but it turns out that's not the case. So this is suggesting that the brain can have the mechanisms for generating ultra slow oscillations. But that the way those ultra slow oscillations are organized at the network level might depend on the brain area you're recording from. And in that regard, it is pointed to the fact that there might be something special about the connectivity in the adrenal cortex that is organizing these ultra slow oscillations into the ultra slow oscillatory sequences that we found.

Now, whether these ultra slow oscillatory sequences are internally generated in the MEC or inherited from a neighboring brain area, that is something that, We are investigating at the moment, our working hypothesis is that it is internally generated, but we do not rule out that it could be inherited from other brain areas.

And as a matter of fact, we're already recording in the lateral internal cortex. I know you said that for you, everything is the internal cortex, but the medial and the lateral are actually very different. So now we're [01:03:00] trying to see whether this the lateral entorhinal cortex could also play a role in this story.

But long story short, we didn't find this in the prosopiculum and we didn't find this in the visual cortex.

Benjamin James Kuper-Smith: So, I mean, what, basically one, one kind of question I'm left when I finished reading your paper was a kind of uncertainty of like, what exactly to make of this? Because in some sense, because it's not like, you know, you can say like, Oh, great. Now we understand how animals know where they are or whatever, you know, because you take all of that out of the context, right?

You basically, yeah, put an animal in a black box and see kind of, and then you find these long roots. So is it, yeah, basically, I mean, what is it? What are these sequences doing? Are they, I mean, are they doing anything? Or are they the consequence of something? Or

Soledad Gonzalo Cogno: no, I mean, I don't have a a definitive answer to that question. Because what the functional role of the sequences is that is not yet fully established. So we have a working hypothesis, and this working hypothesis is that these sequences could serve as a scaffold for enabling [01:04:00]new. Patterns of neural activity or more in general for computations that are unfolding at very slow timescales.

Meaning that if you have this scaffold, this sequence is going on in the median intronal cortex that might enable computations that otherwise would be harder to achieve. So this is scaffolding idea. It's like saying a little bit like now you have your, the scaffold of a house. And then you can build using that scaffold very different houses, right?

So with different windows, different doors, different colors. This is kind of the same idea. You could support different computations by having these ongoing sequences. And we are not the first ones to propose this scaffolding idea when it comes to sequences, right? Because. This has been very much for example, suggested in the field of song learning and song in, in, with birds meaning that having ongoing sequences facilitates learning songs for several finches, I [01:05:00] think.

And then, of course, you have different sequences that in the case of hippocampus, you can have, um, Sequences of play cells that are already there before the animal explores the environment and then those sequences recur. So the fact that you can have these backbones of sequences that can be used for something else.

This is not new. This is not novel. Many people have pushed this idea. And there's also the work by Dean Guanomano in that regards, which also I really like. So we're not the first one pushing for that idea, but in our case, it's not only sequences, but it's also very slow sequences. So this ultra slow component of the dynamics by may, might be enabling ultra slow computations.

And that's exactly what we wanted to illustrate with the toy model that we added in the last extended figure, where we are showing that if you have ongoing sequences, then you could support a [01:06:00]variety of finding patterns. And we're gonna, at the moment we are working on. Computational model that illustrates this as yes, further, but experimental wise, essentially, what you have to do is to have your sequences to disrupt the sequences and then show that you're disrupting the performance of the animal in one specific behavior or cognitive task, and we're not even close to that.

Yet, because, of course, it requires not only having the sequences, but disrupting the sequences and having one behavior associated to that, that we can estimate how good or bad is, but that will come. It's in the pipeline. So that's the idea on the working hypothesis that we have at the moment. But of course, we don't rule out that other things might be happening.

For example, that I don't think that this is the case, but I have discussed about this with other people, for example, that this could be associated to accumulation of error in the path integrator [01:07:00] and that the path integrator in the medial entorhinal cortex Might be accumulating error when you are in, in, in dark conditions and that this could be somehow related to that.

It is hard to believe that if you're accumulating error in your path integrator, it would be this stereotypic and this low and with this fixed frequency, but I don't really doubt either. So that's another hypothesis. And we are working on all of them at the moment. We have a bunch of projects that follow up on that nature paper.

Benjamin James Kuper-Smith: Yeah, One thing that you didn't mention, I think you didn't mention at all in the paper, that was kind of my initial, just, association when I thought of this, that it had something to do with like diurnal rhythms or something like that, is that you know, I guess rats just being able to tell the time more or less it could be anything like that, or because I'm, yeah, I'm just curious, that was like my initial thought when I You know, sort of basically a study where nothing really happens.

That is just

Soledad Gonzalo Cogno: I mean, with

Benjamin James Kuper-Smith: keeping track of 

Soledad Gonzalo Cogno: yes. So in that 1st [01:08:00] paper, we didn't do decoding of time, meaning that we could have, let's say, take our 1 hour recording and then use that data to train a decoder and see if we can, if we could identify perhaps a time epoch. We didn't do that. So I couldn't tell you whether. Those sequences could be used for identifying time epochs, for example, for me, if that's a little bit hard to believe, because it's so stereotypic.

And if you look, and as a matter of fact, if you do, for example, cross validation, and you take the 1st, 3rd of the data, and you sort your neurons, right? And then you apply the sorting to the rest. Essentially, what you see is, It's always the same, right? So meaning that what you find in the last third is essentially the same thing as you find in the first third.

And for me, that points against the idea of using that to keep track of time. Another thing that for me points a little bit,

Benjamin James Kuper-Smith: [01:09:00] why? I mean, isn't that exactly what a clock does? It just runs around under the same,

Soledad Gonzalo Cogno: yes, but then that means that you have a bunch of neurons that are doing exactly the same at time point one and at time point 3000, because the sequences are essentially. always the same because they're periodic. There is variation sequence to sequence. So, okay, let me be more precise. In that variation from sequence to sequence, there could be information encoded about time.

That's true. I agree with that. But at the first glance, that was not my first hypothesis because it is so stereotypic. And this, and you know, you have these oscillations and you have these sequences that for me, it was kind of hard to believe at the beginning, that if you have something that Always kind of looking that you could use that to tell time for me.

It's like, okay, there has to be better ways for telling time, but perhaps you can tell time from that. We didn't run an analysis on decoding time from [01:10:00] that. And for example, this is medial entorhinal cortex in the lateral entorhinal cortex. There's a work done in the muscle lab where they showed that the activity of the lateral adrenal cortex is drifting over time and this drift, this change in the population activity as a whole, as a function of time is enabling Okay. This is enabling a time readout because the activity is changing and because the activity in time point one is different from time point three thousand. Now, you can tell those two time points apart, but that's not so much what is happening with our sequences. So that's, for me, the first intuition. Why for me, this might not be the best mechanism for telling time.

And also,

Benjamin James Kuper-Smith: it maybe to summarize it, is it basically, so because at first I thought the, that it is so precisely periodic and stereotypical, I thought that would be exactly a good thing to tell time, but are you, is it, are you basically saying like it's like having, you know, the second or [01:11:00]minute hand on the clock that's running around without having the hour hand moving, so then you know where you are on the cycle but not like, yeah, you know what minute it is, but not what hour.

Soledad Gonzalo Cogno: Yes, exactly. So how can you tell apart 1 o'clock from 4 o'clock? Because the middle of the clock is in a

different 

position. Exactly. So it's changing over time. So, that's a little bit how I think about it. But that's my intuition. So the way I'm answering your question is with my intuition, but we haven't quantified yet how well we can tell time when you have these ongoing sequences.

We have another ongoing project, which is in collaboration with the Moser lab and is actually the person who is driving that project. Are actually to one former PhD student from the most 11 one PhD student who is co supervised by the three of us and they are working together on how you can keep track of time in the lateral internal cortex as compared to the median and trying to cortex and it's very different as a matter of fact, so [01:12:00] my gut feeling is that if you want to tell time, it's better that or episodic time.

Let's put it like that. It's better that you focus on the lateral internal cortex. And in the medial entorhinal cortex, that information is integrated with the spatial signals. That is coming, for example, from grid cells, head direction cells, pit cells, and other spatially modulated cells.

Benjamin James Kuper-Smith: Whilst you're just mentioning those different kind of cells I mean, so I don't know basically my question is whether you have any idea about whether the cells that you measured, whether those are grid cells or whatever Again I know there are grid cells in a Toronto cortex, but I don't know like how many, what percentage and all this kind of stuff.

And yeah, so basically do you know, like whether the cells you have also do other stuff or like, let me rephrase that. Do you know the other functions that the cells you measured are involved in, or is that just open because, you know, the animals [01:13:00] didn't do anything. Um,

Soledad Gonzalo Cogno: On that at the moment. So in that first nature paper, we didn't, we only recorded with the animal running on a wheel. And because we only recorded with the animal running on a wheel, we don't have open field recordings, meaning that we cannot tell whether this cell is a great cell, this cell is a head direction.

Cells. We don't have that distinction, but we had a very big field of view. And from where the field of view is located, it could be the case that we have different functional cell types, meaning head direction cells, grid cells, and more. But we don't have the identity of the cells in that data set. So that's actually what we want to do next.

We are now Do it using your pixels to record from the median internal cortex when the animal are running, the animals are running in this box and we are identifying the functionality of this. It's very, I mean, we just started. So it's too early to tell. But but that's coming. So, long story short. [01:14:00] It seems to be the case that these sequences are recruiting many different functional cell types based on the anatomical location of the person and the field of view and how big it was, but we don't have the final proof yet and that will come with a new set of experiments.

Benjamin James Kuper-Smith: Yeah. Yeah. Yeah. It seems like you're doing a lot of cool stuff based on this research. I mean, I think you've already mentioned like four projects or something that you're working on

Soledad Gonzalo Cogno: Yeah uh,

Benjamin James Kuper-Smith: sounds of it.

Soledad Gonzalo Cogno: that's kind of, that defines most of the research that we're doing in my lab. It's not all about that. But but yeah, we were following up on that paper.

Benjamin James Kuper-Smith: Yeah. I mean, I, usually I, you know, I asked like what people want to do next, but I guess we kind of already discussed that at length of the last half an hour or whatever. Unless there's like anything else you want to 

Soledad Gonzalo Cogno: No, so regarding what comes next, it's essentially we [01:15:00] are looking at different behavioral conditions now. We are recording not only on the wheel, but also during sleep and in open field recordings. And we want to see how, so that's what we're doing. But rather than telling you what we're doing, let me tell you the questions that we have.

So the questions that we have are these sequences expressing also in other behavioral conditions that are more naturalistic, for example, sleep or free foraging? Okay. Does the intranal cortex store only one sequence or can it store multiple sequences? Because if it can store multiple sequences, then you can ask about the capacity of the intranal cortex, similarly to what you would ask in the terms of attractors and the hippocampus. And then we want to understand how these sequences can couple cells with different functionality as you were pointing out before, right? Great cells, head direction cells. And finally, the big question that I also have is what, how is this generated? What are the [01:16:00] mechanisms that are enabling this ultra slow oscillations and the organization into this ultra slow sequences? And for that we're also working on computational model. So that's what's coming next.

Benjamin James Kuper-Smith: sounds pretty cool. Uh, So yeah, at the end of each interview, I asked my guests the same three questions. The first one is what's a book or paper more people should read or that you think what people should read which can be old or new or famous or completely unknown.

Yeah. As I said, something that you would recommend to people.

Soledad Gonzalo Cogno: Yeah I think what people should read is what people find interesting and would contribute to The research questions and because I don't know what that is for people I will not tell you what I think people should read but I will tell you what I like reading and There are three books that I like reading Or that I always go back to, you know, it's like those Places you tend to go back to because it's good and you like it and I also teach at the Kavli Institute So the first [01:17:00] book is the elements of a statistical learning which I use for teaching neural data analysis And this is a book by Hasti, Tifshirani, and Fridman. And it's a really nice book on statistical learning that I also use a lot for when I have to do data analysis myself, or when I have to teach data analysis. And then the other book that I really like is the well, actually, there's so many books.

I really like also information theory by Mackay. So that's a very book that I very much recommend because it's. Information theory is presented the concept of entropy. It's also everything is presented so nicely and so clearly and with, you know, a hint of humor. That is really good. Very enjoyable.

And there's also the book, which is a classic theoretical neuroscience by Peter Dayan and Larry Abbott. I always go back to that book too. So, yeah, let's go with those [01:18:00] 3. Yeah, theoretical science and elements of a statistical learning. 

Benjamin James Kuper-Smith: I think I've said this before. Well, I have said this before on the podcast, when about this question, that usually it's either books I've heard about a lot and that I have wanted to read but haven't gotten around to, or it's something I've never heard of. And those are three books that I've definitely heard about.

And I think I actually had the Theoretical Neuroscience book. I think I owned that once, but I think I was a bit intimidated by the maths because I have like no background in that. And then I was a little bit lost and ended up doing other things. But, uh, 

Soledad Gonzalo Cogno: For me, that's also, it's like one of the first book that I was recommended to read when I started doing neuroscience during my masters, because during my masters, when I I didn't know anything about the brain, I still feel I don't know much about the brain. But at that time, I didn't even know what a neuron was.

I took this first course on neural networks, and that was one of the first books that was listed in the recommended [01:19:00] bibliography. So I like the book, and I, you know, it's like, it brings back memories of when I was a student learning what how to model a neuron. So it brings back many nice memories.

And I like the work done by the two authors as well. So this is great,

Benjamin James Kuper-Smith: Yeah, I think I should probably revisit. I think back then I just had like no framework at all for how to, you know, You know, I remember like once I so my father studied mathematics. So I, you know, I had like this one form and I was like, dad, what does this mean? And he just explained, I was like, I don't know what to do with this information.

Like, you know, it just had like no concept of yeah. What to do with it. But hopefully I have a little bit more of that now.

Soledad Gonzalo Cogno: but you know what I feel, okay, you should perhaps give it a try because I feel that most people that were not trained in math, the biggest challenge is to feel it's not really about the understanding because once they do it, it's actually simple, but it's, you know, defeating that fear of. Oh, what if I don't understand it?

Or perhaps this is [01:20:00] overwhelming because when I, some, now I'm not doing it anymore, but I used to teach a course on linear algebra and now other people are taking care of that. But I remember that people used to come to me and tell me, I always struggled with math. Math was always very difficult for me.

I'm not sure I will be able to do this. And I was like, you know what, calm down, breathe, take a deep breath. And if you need help, I will help you. We will do it together. And they do it. You know, it's sometimes And for me, that was very, that was amazing because I like seeing the process of someone learning what a subspace is.

It's not, it's great, I like it, but observing the process of someone being able to challenged by something and ending up mastering that something. That was mind blowing for me. It was such an amazing experience. So, and I feel that book is, it's, things are presented in a very clear way. So perhaps you will enjoy it.

Hopefully. Yes.

Benjamin James Kuper-Smith: [01:21:00] Yeah, that's what I mean because I have done a little bit of math since and then it's really weird like I always found it I mean, as I said, I didn't do much in school. But, like, I found math. I remember it was one of the few subjects I practiced for the final exam, because I thought it would be stupid not to do very well in it.

So, like, in principle, I have Like, in literature analysis, I never knew what they wanted from me, but in math, it's always like, yeah, just, like, there is an answer and you can just figure it out. But somehow I don't know, it's, yeah, I had the same kind of like, I think when I first read that book, or like, tried to, I had the same kind of sense of like, what do I do with this, and this like, slight sense of panic of like, Ah, I don't get this.

But hopefully, so since I've had a little bit more exposure, so hopefully that's, um, Yeah.

Soledad Gonzalo Cogno: That's good.

Benjamin James Kuper-Smith: We'll see. Yeah, I mean, I also have to say, I mean, this is one of those things. I think I've asked this question. How many times? Like 30 times now, something like that. So I didn't do it from the beginning. I don't think I've read a single thing I've recommended. I mean, in part of it, I guess it is because, you know, [01:22:00] every on average two weeks or something, you get recommended this like series of papers or new books or whatever, but, um, yeah.

It is a cool list though. I'm really, I mean, and I think your books fit very well into it and added a nice other flavor that it needed a bit more from the let's just say computational side. Even that's maybe not the entirely correct word. But Yeah, one day I'll read all the things that people have recommended to me.

Uh, Anyway um, second question is, what's something you wish you learned sooner? This can be from your work life, from your private life, I don't really care. Just something that you're willing to share, that you learned, and that you think, if you'd learned that sooner, that might have helped you out a bit.

And then maybe also like, how you went about learning it, or what to do about it, or, yeah.

Soledad Gonzalo Cogno: Yes. Yeah, I was thinking about that after you sent the email and what my answer could be to that. And I think my answer would be that What I wish I had learned sooner is [01:23:00] that we all have different stories and we all have different paths and that's okay and that doesn't mean anything at all and that the outcome of what you do and of the things you achieve is not related to what your starting point was and that things can change and that you can change and that's okay. And I wish I had knew that when I had knew that when I was younger. Because sometimes, you know, you tend to measure yourself and your achievements compared to what other people do and other people achieve. But the thing is that we all have different circumstances. We all have different stories. We all have different wishes and different priorities, and we're all different people.

And there's no one specific way of doing things such that if you don't do it that way, then you're failing. And also what does fail mean, right? That's another question. So [01:24:00] I think if I had known that all paths are different and that you can still have a career that is solid, that is good, and that makes you happy, regardless of what your path looked like, that would have removed some of the tension for me and the stress. Also because in my specific situation, I did my PhD in Argentina. So I was all the time wondering, okay. What if I cannot, you know, do a postdoc abroad and what if I cannot learn the things that I want to learn or what if I cannot publish it if I want to publish all these things that I feel that at that time I was very much influenced by the past that the people that I consider successful was, I was measuring myself against those rules.

And then in the end, I managed to go abroad for a postdoc. I did a postdoc with two mentors that I really like, and I had an amazing experience. [01:25:00] And it was good. It worked out well for me. So that's what I would tell myself that everything's going to be fine. As long as you remain focused and you know what you want and you're willing to work for it.

Benjamin James Kuper-Smith: Can I ask a little bit based on what you said about the, you know, if

I understand you correctly, part of it was, you know, doing a PhD in a country that isn't, I don't know what the word, right, what would be part of like the mainstream scientific culture or something like that. I mean, if I look at the places of downloads for this podcast you know, obviously lots of them are from the mainstream countries, but lots of them also aren't.

So I'm just curious whether you can talk a little bit, yeah, it seems to me as if there, there was some questions and uncertainties or insecurities about that, if I understood you correctly.

So I was just curious whether you could speak a little bit more about that because that's probably something that I mean, that applies to many people, even, you know, even if you do your PhD in the U. S., you maybe don't do it at [01:26:00] a famous place or whatever, right? That's, by definition, the majority of people, basically.

Soledad Gonzalo Cogno: Yes, exactly. So, I mean, okay. So I don't know how. It is in other places, right? So I, I know my experience and experience of studying and doing a PhD in Argentina. There are some limitations, right? So, for example we didn't used to, we didn't have much money for traveling at that time. So being exposed to the community, you know, learning what people are doing, getting inspired by what people are doing.

And, you know, learning about the latest developments in terms of ideas, in terms of techniques, in terms of whatever you want. That is usually difficult, right? Also because Argentina is very far away from absolutely everything, pretty much. Essentially, all the countries that are in South America, so getting exposed to the community, [01:27:00] and that was always a challenge.

And also, you know, there's value in going to a conference and going to a summer school and discussing with people about what you're doing, getting feedback from your peers and your colleagues. It's amazing. It's inspiring. It's also, you know, helps you to raise your standards and do things that are more important in terms of asking relevant questions, in terms of, you know, having good methods and good ways for modeling your data, for analyzing your data, for collecting your data. It's a team effort. Science is a team effort. And when you are in an isolated country with not a lot of to travel that component becomes a little bit challenging in my case. Luckily, I was doing computational neuroscience, so all I needed was a computer, but if you want to do experimental neuroscience, now it becomes even more challenging, right?

Because it's very hard to get equipment. It's very hard to essentially, you cannot [01:28:00] compete with other countries. And I'm not saying this in the ugly sense of the word, you can really not work on similar questions or in in, like, for example, now we are recording with Neuropixels, and this is something that I'm doing in collaboration with the Molson Lab.

If I wanted to do this in Argentina, most likely I couldn't. So the fact that there are some limitations, unfortunately, also puts a constraint on the scientific questions that you can answer. And that for me, is not, I don't want to negotiate on that. I am doing science because I love science and because there are questions that I feel passionate about and that I want to answer.

And I don't want to have any constraints when it comes to that, as long as it's possible, right? So that can be that can be, those are different challenges. Yeah, but then we could talk hours about this, right? Because this also relates to different, to how each country prioritizes science and prioritizes [01:29:00]funding science and basic research, which is also very bad in Argentina at the moment likely it's in, in Norway.

I'm very well funded, so I'm very happy about that.

Benjamin James Kuper-Smith: yeah maybe as a, as an addition to that I did an episode with uh, Ben, I don't know how to pronounce his name, it's a German surname, but he's American. Uh, Erlich, I guess would be the English pronunciation. Yeah. He wrote the biography of Santiago Ramon y Cajal, and I'm sorry for my Spanish

Soledad Gonzalo Cogno: No, that was very good.

Benjamin James Kuper-Smith: Okay. I'll take it. And just to add to that, I mean, I guess he had the, in a sense, the most extreme story of what we talked about almost not exactly the most extreme, but he was someone who I mean, we talked about this also in the episode. who was someone who was very aware that there hadn't been a great Spanish scientist when he was around and he wanted to show that Spaniards could be scientists.

That was like part of a lot of his thinking. And he did, I think, his the research that he later on when ended up winning a Nobel Prize [01:30:00] with, I think he did that on like really cheap microscopes that were way worse than any of his, I think at the time, especially German Not colleagues, but like scientists in Germany, they had way better equipment than he had.

And you know, so even if it's not ideal, I guess you can still make it work. If you're one of the greatest neuroscientists of all time,

Soledad Gonzalo Cogno: Which is not easy, right? But no, but that said, I mean, it's if you ever visit like a lab in Argentina, you would be amazed by how amazing things they can do with essentially no funding. It's really creative. It's really good. The standard is very high, but it's very difficult, you know, when you have to do everything yourself. So yeah, but I agree with that.

Benjamin James Kuper-Smith: Yeah. Okay. I'll link to, I mean, I'll link to everything we talked about but also to the episode with about Cajal. In case people want to hear more about that. Final question is, so I started a PhD exactly two months ago today. Which is scary to think about that. It's already been like nine weeks since I started.

But yeah, any [01:31:00] advice for people in this kind of period where there may be like, you know, Advanced PhD students thinking about postdocs or people who just started a postdoc. Yeah. Any

Soledad Gonzalo Cogno: No, but you studied your postdoc, right? You didn't, it's

Benjamin James Kuper-Smith: I have started mine. 

Soledad Gonzalo Cogno: your postdoc. Yeah, 

Benjamin James Kuper-Smith: I have. 

Soledad Gonzalo Cogno: Yeah. 

Benjamin James Kuper-Smith: But just basically, I mean, pick whatever you want, but like in this kind of period of one's academic career any advice,

Soledad Gonzalo Cogno: Yeah. For me, the most important thing is to have good mentors and to be very clever when you choose your mentors. I think that's way more important than the specific question in which you're working, because having good mentors that will help you do good science. Yeah. And with good science, I mean addressing important questions about how the brain works with the highest standards in terms of methods and concepts for addressing those questions.

Having good mentors that help you with that, [01:32:00] that also have very high standards in the science that is conducted. And that would also help you to become the scientist that you want to become and to achieve the things that you want to achieve. For me, that's the most important thing by far. And I learned, for example, in my case, I did my postdoc in the group, and during my 6 years as a postdoc there, I learned a lot about internal cortex.

I learned a lot about navigation. I learned a lot about many things. This is part for me. By far, the best thing was to have them as mentors because they helped me grow as a scientist and they supported me in becoming the scientist that I always wanted to be. So yeah, good mentors are forever, you know?

So I think that's the only advice I can give. Be very careful and clever when you choose your mentors.

Benjamin James Kuper-Smith: but how do you choose a good mentor? [01:33:00] I mean, what are you looking for? Let's say someone is, let's say someone is a. You know, they've got like one and a half years left in a PhD or something. They want to do postdoc and you know, they've got like four labs or whatever that they're interested in.

They mail them. There's kind of some interest. They're about to meet them. What should they pay attention to 

Soledad Gonzalo Cogno: I think they should talk to the PIs. and ask as many questions as they want. But I think it's also important that they talk to the people in the lab, to the postdocs and the PhD students, because those are the ones that know how the daily life in the lab is. And, you know, let us be honest, some PIs are super nice when they're discussing among PIs, but they're not so nice when they're discussing with PhD students, for example, and that's not a nice thing to say, but we all know that sometimes happens.

Unfortunately. So, the only way in which you will find out about that is by talking to other people in the lab and people in the lab, hopefully, will be honest about it. [01:34:00] So, I think that's a good thing. And then. When you are interviewing with a PI and the PI is encouraging you to talk to people in the lab, I think that's a good sign because they have nothing to hide. all the times that I was offered that. It was always a good PI. And then that's also what you hear from other people. And now that I have my group and it's a very small group and I'm in the growing phase of the lab, but also that's something that I want to do. It's like people initially want to talk to you because they want to know about your research.

They want to know what they did, what they could do in the lab. But I think it's equally important that they talk to the people that work with you, because even if. You know, you're not a horrible person and you want to be a good mentor, sometimes, you know, the personalities simply do not match.

And that's also something that you can learn by discussing with lab members. So I think that's crucial. [01:35:00] And then you have to trust your gut feeling as well.

Benjamin James Kuper-Smith: when you talk to them and yeah meet everyone. And you just get a sense for like what it would be

Soledad Gonzalo Cogno: Yes, exactly. I was honestly, I was very lucky because During my PhD, I, my, my PhD advisors were amazing. Edvard and May-Britt are absolutely wonderful. And as a matter of fact I continue, I mean, now I'm not formally a member of their lab anymore because they have my own group, but whenever I have questions, I always ask for their advice.

So, so yeah, I think I really like that, you know, it's like this idea that there are people that are more senior than you are, and that can help you. I really like that. So I really hope that now that I have my group, I hope that in the long term I can become two people in my lab with these amazing people represented to me in, in, in the development of my career.

If I achieve that, I [01:36:00] feel things will be better. that I will have returned what I also received during my years as a trainee.

Benjamin James Kuper-Smith: Yeah. Just one final thought to that is that I mean, I always want to end it there because that's a nice thought to end on, but just to add to that is that also a mentor doesn't only mean the PI. I mean, you know, I guess a mentor is anyone who knows more than you do about something right now.

I'm being helped a lot by one postdoc in the lab who knows a lot more about what I want to do. So it's And another good reason for talking to the other people in the lab is that, yeah, mentorships are not just with people who are formally above you.

Soledad Gonzalo Cogno: exactly, it can be your peers, and it can also be people that are more senior than you are, but that you're not related to them in any specific way. For example, there's this person, I can actually, say this very openly. There's Sarah Sola, she's a computational neuroscientist. I have never published with her.

She has never been on my advisor, but she's amazing. And then whenever I need advice or I need [01:37:00] help, either, you know, in terms of career development or in terms of, you know, how to navigate in computational neuroscience, when you're a female researcher or when it comes to, okay, I have this idea, I have this project.

What do you think this is good? Is this solid enough? I always. You know, ask her because I trust her and I know that she's the person that is willing to help younger generations. So identifying these people that will help you because they care about you because they care about your research and teaming up with them.

I think that's one of the best things about being in academia.

Benjamin James Kuper-Smith: Nice, you managed, even with my interruption, to put it back on a nice ending. So thank you for that and thank you for the conversation.

Soledad Gonzalo Cogno: very much. I had a great time.