#93. Can Video Games Predict Cognitive Decline? - Dr. Emre Yavuz

WEBVTT
00:00:03.903 --> 00:00:07.524
Hello, hello, hello. And welcome to the Smooth Brain Society.
00:00:07.524 --> 00:00:26.356
So quick question. Have you ever played a mobile phone game and thoughts you were just wasting time? Well, what if I told you that over 4 million people played a game called See Hero Quest and without even realizing it, They helped build one of the biggest neuroscience data sets ever created.
00:00:26.356 --> 00:00:29.518
One that could change how we detect Alzheimer's disease.
00:00:29.518 --> 00:00:32.999
Today we are joined by Dr.
00:00:32.999 --> 00:00:40.500
Emre Yavus, a computational neuroscientist who just finished his PhD at UCL under Professor Hugo Spears.
00:00:40.500 --> 00:00:51.383
Emre's research is all about how humans navigate space, why some of us are better at than others, and how understanding that could help us spot the earliest signs of cognitive decline.
00:00:51.860 --> 00:00:53.612
long before they become a problem.
00:00:53.612 --> 00:01:12.365
He has used data from millions of players to understand how we find our way around the world and he has also done one of the first ever studies looking at two people's brains at the same time while they collaborate in Minecraft using virtual reality, which to be is just pretty damn cool.
00:01:12.365 --> 00:01:18.650
And on top of all of that, Emre has been winning AI hackathon after AI hackathon.
00:01:19.043 --> 00:01:21.683
building health tech tools in 24 hours.
00:01:21.683 --> 00:01:30.608
And he also happens to be a singer with the London Philharmonic Choir, who has performed at the BBC Proms and a cellist with the Kensington Symphony Orchestra.
00:01:30.608 --> 00:01:34.129
You are damn right, we have a lot to cover.
00:01:34.129 --> 00:01:35.480
Welcome, Emre.
00:01:37.007 --> 00:01:42.861
Hello, thank you so much for having me on~ here. Yeah, it's it's yeah, it's an amazing to be on here.
00:01:42.861 --> 00:01:44.427
So thank you. Excited No.
00:01:44.091 --> 00:01:47.385
to discuss yeah, everything we have to discuss.
00:01:47.084 --> 00:01:48.084
Yeah.
00:01:47.936 --> 00:01:49.427
Well, it's an absolute pleasure to have you on.
00:01:49.427 --> 00:01:50.977
I'm very excited for this episode.
00:01:50.977 --> 00:01:52.801
So let's Exactly.
00:01:52.418 --> 00:01:54.180
start off how we always start off.
00:01:54.219 --> 00:01:59.061
Can you give us a little bit of an origin story of what made you want to study the brain?
00:02:00.034 --> 00:02:22.503
For sure. ~ so I think it was when I was in sixth form I saw a talk by Professor Richard Morris, ~ who invented the Morris Water Maze, which was this ~ where it's a paradigm used very often to look at spatial memory,~ so how we remember where objects are in our environment and ~ most commonly done on in rodents, you know, mice and rats.
00:02:22.503 --> 00:02:29.127
And and I think I saw that talk and I it was all about why why humans remember, why humans forget.
00:02:29.953 --> 00:02:46.387
and I was so interested in ~ from that what it is that drives individual in ~ cognitive function, you know, why there's such a spectrum. Some people, you know, have such a good memory, I have quite a good memory, and some people they forget things, you know, so often, so easily.
00:02:46.387 --> 00:02:54.409
And so I thought ~ going into neuroscience was a great way to understand the real sort of core mechanisms driving those behavioral differences.
00:02:54.409 --> 00:03:07.917
And importantly, I wanted to translate it into something clinical or something that would actually be used to guide therapies, you know, ~ preventative medicine,~ what have you. So I wanted something with a that could impact lots of people.
00:03:07.917 --> 00:03:20.681
And by understanding, you know, these individual individual differences, hopefully one day we'll get to the level where we can look up, you know, sort of a precision med medicine approach to understanding why somebody might have a particular cognitive trajectory and not someone else.
00:03:20.681 --> 00:03:34.473
So I think from there I d I was just fascinated in that sort of that area of research. And then I thought neuroscience~ was the way to go. ~ and yeah, and then it kind of went things went from there.
00:03:35.811 --> 00:03:43.168
Yeah. So you did your MSc at Imperial beforehand, is that right? So Yes.
00:03:42.591 --> 00:03:54.455
you ended up in kind of navigation and yeah, kind of almost ~ testing, seeing if you can test Alzheimer's disease before it's like usually detectable.
00:03:54.455 --> 00:03:57.896
But just to very quickly touch on what you did before that.
00:03:57.896 --> 00:04:01.457
I believe you did, was it adolescent self-harm?
00:04:01.457 --> 00:04:04.299
And then what made you decide to go from that?
00:04:04.399 --> 00:04:06.144
and then move on to navigation.
00:04:07.191 --> 00:04:12.335
~ so yeah, so I my my ~ my master's was very much in sort of clinical psychiatry.
00:04:12.335 --> 00:04:16.379
And I think at that point I very much wanted to explore that space.
00:04:16.379 --> 00:04:35.274
I was very interested in, as I mentioned, the clinical applications of research and how we can develop better prognostics and ways of ~ understanding why people engage in the behaviors they do, the individual differences that come with that, and how that could be used by clinicians, you know, and like how we can spot if somebody's going to go on to have a particular trajectory. And I thought.
00:04:35.747 --> 00:04:52.074
go going into that area was a very good window into understanding how behavioural interventions can be applied to clinical populations who need them most. So it gave me experience in actually working with populations, which I always wanted, and actually interviewing them on very sensitive material video.
00:04:52.074 --> 00:05:04.439
So I it was a it was a a sort of a way for me to to develop my expertise in that area, ~ but also have the ~ experience in actually dealing with~ subject data clinical matters.
00:05:04.937 --> 00:05:09.067
~ and being able to apply behavioral psychology to that importantly.
00:05:09.067 --> 00:05:24.221
~ so yeah, and and secondly, my supervisor, she'd been working on a mental health-related app, digital tool for looking at what how we can in sort of enhance mental imagery in people as a mechanism for overcoming these negative emotions.
00:05:24.221 --> 00:05:34.485
So self-harm is a population I worked with, and they often have negative urges to injure themselves as a way, often propose the way to release their emotions,~ and the different theories surrounding that.
00:05:34.607 --> 00:05:41.670
But I she'd worked on an app. ~ so Martina di Sapliccio, who was my MSE supervisor, she was a psychiatrist.
00:05:41.670 --> 00:05:54.295
And so she'd been working on this app to deploy in patients where ~ we were trying to use mental imagery as a mechanism to overcome these negative emotions and therefore hopefully reduce self harm engagement.
00:05:54.295 --> 00:06:00.278
And I wanted to very much delve into the digital health side of things and see how that was deployed in clinical settings.
00:06:00.278 --> 00:06:04.598
So I think those things, so digital health is a big interest of mine as well.
00:06:04.598 --> 00:06:10.870
I think those two things coming together was made it seem like the right fit for a project. So yeah, so I went into that.
00:06:10.870 --> 00:06:15.952
And then I I really, you know, learnt a lot from it and that's now been published.
00:06:15.952 --> 00:06:24.225
So yeah, so I so I'm glad I've had that that insight 'cause really helped me now when I reframe other areas of research in a more clinical lens.
00:06:24.225 --> 00:06:30.206
I can look back on that and think of the challenges I overcame in that and how that could be relevant to future clinical research.
00:06:30.206 --> 00:06:31.206
Yeah.
00:06:32.067 --> 00:06:34.247
Yeah, that's really interesting.
00:06:34.247 --> 00:06:35.593
you Anyways.
00:06:35.249 --> 00:06:42.351
almost like just my next question was going to be how you went from that to, you know, how you found Hugo Spears lab.
00:06:42.351 --> 00:06:47.423
But you've just literally said clinical applications, trajectories and behavioral interventions.
00:06:47.423 --> 00:06:49.115
You've kind of summed that all up.
00:06:49.115 --> 00:06:56.598
But how did you find Hugo Spears lab and what made you decide to move over there and specifically more into navigation?
00:06:57.416 --> 00:07:03.680
Yeah, I know for sure. I mean, I I did my undergrad thesis actually on navigation related sort of research.
00:07:03.680 --> 00:07:11.656
So ~ VR environment and people had to navigate down this virtual corridor and stop ~ at where they thought the goal was.
00:07:11.656 --> 00:07:20.742
So it we were looking at how people estimate the distance they've traveled in virtual environments and how the mechanisms they do they use to do that.
00:07:20.742 --> 00:07:43.829
And this was really to develop with the idea of developing better boundaries for cognitive decline in the future, like Could one person sort of signal as such be tell tell us something meaningful about, you know, if they should ~ they're deteriorating faster, should go to a particular, you know, drug stratification, clinical trials, all sorts of things. So I became very interested in that idea of using as a behavioral biomarker, as a metric.
00:07:43.829 --> 00:07:46.271
So navigation is something we do every day.
00:07:46.271 --> 00:07:47.613
So I understood the importance of it.
00:07:47.613 --> 00:07:49.194
It's something I've always been good at.
00:07:49.194 --> 00:07:54.038
I've always had a good cognitive map, and I've always been the one sort of showing other people around when we're in a new place.
00:07:54.105 --> 00:07:58.857
So I was always fascinated as to why some people ~ seem to get lost so easily and not other people.
00:07:58.857 --> 00:08:01.889
My own personal experience, it's clinically relevant.
00:08:01.889 --> 00:08:10.103
So Alzheimer's disease, one of the earliest ~ signs or we believe from the literature is that people get lost, especially in their own homes and familiar environments.
00:08:10.103 --> 00:08:22.785
~ so that was a sort of an urge to sort of want well understand that to the fullest to be able to you know, help help as many people in understanding why they might be deteriorating quicker than other people.
00:08:22.785 --> 00:08:25.045
So I saw I saw the clinical relevance of it.
00:08:25.045 --> 00:08:36.389
And navigation is something we can measure virtually very easily on phones and digital tools. So it lends itself very well as something that can be measured because it's, you know, trajectory data we can use looking at their positions.
00:08:36.389 --> 00:08:59.472
That's something you can easily collect on a phone, unlike maybe other more complex things or nuanced things where it's not necessarily as can't necessarily be put down to some few numbers ~ and quantified in an easy way and looked at trajectories in that sense. So I thought the its ease of collection digitally also made it gave me another reason to to to go into that area.
00:08:59.472 --> 00:09:17.044
And then Hugo Spears specifically, so he developed h him and the team and working with engineers had developed this PCH request game which you mentioned, so a mobile app video based game that ~ people basically ~ navigate a virtual So they see a map, the map goes away and they have to navigate to a target.
00:09:17.640 --> 00:09:27.738
~ so it's a a flag in this virtual environment there on a boat, tends to see her request, and and that game had been developed by the lab that I was going to join, a Hugo Spears lab.
00:09:27.738 --> 00:09:47.299
And so I'd always from through my undergrad even had that interest in this and this is really exciting, not because of the not just because of the big wealth of data, but because of this potential to sort of lay the land for future gamification and gamification for therapeutic use for something I was I was I could see this field going in.
00:09:47.299 --> 00:09:50.091
So I thought, you know, this is really interesting.
00:09:50.091 --> 00:09:59.434
If I work with this, ~ with Hugo Spears, I'll get great insight into understanding how we analyze data from games and how that can be used to inform therapeutic approaches.
00:09:59.434 --> 00:10:05.488
So and ~ he done ~ lots of eminent work with taxi drivers, so the Big London taxi driver studies.
00:10:05.488 --> 00:10:08.429
He was involved in those with Eleanor McGuire.
00:10:08.429 --> 00:10:13.714
and so I So I thought, you know, this is real world wayfinding.
00:10:13.714 --> 00:10:16.885
This is the real world application of navigation.
00:10:16.885 --> 00:10:38.138
And I I know I wanted to be on the interface of real world clinical sort of mobile research, because therefore you could see its effects in everyday life and have a better understanding of how actual everyday behaviors can inform, you know, treatments ~ and how we can get that sort of data, the most sort of represented real world valid data for that.
00:10:38.138 --> 00:10:57.298
So I ~ So I think, yeah, though that combination of things made me want to come to Hugo Spears' lab and it was at UCL, which was a big hub for cognitive neuroscience, especially. Imperial didn't have as big a psychology department, it had lots of psychiatry. So I thought, you know, UCL is ~ particularly strong for cognitive neuroscience.
00:10:57.298 --> 00:11:08.225
And I thought you know, with the amount of collaboration in different departments, you know, engineering, that we could, you know, leverage tools to study neuroscience in real world settings from sense that it was the right.
00:11:08.376 --> 00:11:22.043
thing to do. And he was a supervisor on the ecological brain program, which is all about real world neuroscience, so mobile, you know, brain imaging,~ digital health, you know, so the program was very fitting with my general aims.
00:11:22.043 --> 00:11:25.485
So and then he and he was one of the supervisors on that programme.
00:11:25.485 --> 00:11:31.778
So so those th two things coming together just made it seem like a natural sort of segue from what I'd done previously.
00:11:33.523 --> 00:12:01.221
Really fascinating. I guess of you said a lot, and one thing which stuck to my mind is you said that you were quite good at navigation yourself compared to sort of your friends, peers, whatever. And and you mentioned that Alzheimer's sort of some one of the earliest sort of symptoms you see is people kind of getting lost in familiar locations. ~ is there when we talk about that as a symptom?
00:12:01.485 --> 00:12:11.988
What about would somebody who is not very good at navigation, so I would consider myself not very good at sort of knowing navigation in place, are they at like more risk in that way of Alzheimer's?
00:12:11.988 --> 00:12:20.889
Or is this more of like are you talking about symptoms in terms of like the risk of decline from what they were to what they become sort of thing?
00:12:21.923 --> 00:12:46.764
~ yeah, so so specifically for my PhD I didn't look at necessarily longitudinally or the change over time, how like how it necessarily i impinges their deterioration deterioration. But ~ we have looked at so with the C Heroquest game, we've shown that we can stratify ~ genetic allele carriers,~ for Alzheimer's apo E4 carriers from people who are non-carriers.
00:12:46.764 --> 00:12:50.486
So there is some sort of validity in discriminating those.
00:12:50.678 --> 00:12:52.659
groups based on performance in the game.
00:12:52.659 --> 00:13:20.296
~ and we haven't ~ well at least not definitely not in my case, we haven't looked at over time how sort of cognitive trajectories vary with game performance. So that would definitely be be a next step ~ for for ~ yeah for for research. So I think at the moment it's more on the sort of can we stratify sort of people side for~ future other other sorts of interventions, maybe for drug trials, lifestyle factors.
00:13:20.296 --> 00:13:38.721
~ how how can we understand? I think first question, which my PhD was a lot of about, was how can we understand the lifestyle factors that affect navigation ability, how those come together to shape navigation ability, and then how by understanding how those come together, can we better understand why someone has a particular cognitive trajectory at at the sort of Mm, yeah.
00:13:38.451 --> 00:13:41.370
cross sectional level, and then later for the interventional level.
00:13:41.370 --> 00:13:43.912
So we ha we haven't done any causal interventional work.
00:13:44.160 --> 00:13:49.703
necessarily with this tool, but it's been mostly sort of cross-sectional observational studies.
00:13:49.703 --> 00:13:57.158
So like I looked at video gaming and we showed that video gaming more~ when you're younger is associated better nav navigation later.
00:13:57.158 --> 00:14:00.759
So people were better at navigating if they video games more.
00:14:00.759 --> 00:14:06.263
So we and we understand that sort of cause effect relationship, but it's not really cause and effect until we've proven it is.
00:14:06.263 --> 00:14:10.115
So there's a suggestion that it might be ~ there's an association there.
00:14:10.115 --> 00:14:20.086
But yeah as you say, we need to do more of these~ sort of longitudinal, ~ sort of actually giving someone an intervention to see if that then actually changes their navigation.
00:14:21.191 --> 00:14:24.302
See parents, it could be beneficial to let your kids play video games.
00:14:25.376 --> 00:14:28.599
Yeah, yeah, no definitely. I mean, Welcome! I feel this place is shut up.
00:14:27.947 --> 00:14:30.028
yeah. Yeah.
00:14:28.599 --> 00:14:29.962
Sorry, carry on.
00:14:30.509 --> 00:14:55.038
No, for sure, no, definitely no, you're right. Definitely. Like particularly aspects of video games, like it's been proposed that, you know, flow and this sort of immersive experience of video games can be beneficial for sort of cognitive ~ deteriorating groups,~ but also like ~ video games, particularly like first person shooter games,~ where we're performing mental maps, where we have to sort of constantly look at what's around us, we have to sort of be engaged in our environment.
00:14:55.414 --> 00:15:01.777
It's been proposed that those sorts of games might be particularly important for navigation, although we didn't see a particular genre effect in my study.
00:15:01.777 --> 00:15:17.461
But there have been propositions that by training at least some certain components of video games, we can actually ~ yeah, improve people's navigation, hopefully improve their ~ their ~ their ability to overcome cognitive decline and cognitive resilience.
00:15:17.461 --> 00:15:24.082
But that remains untested. But that then these are definitely sort of some strong consensus sort of ideas from the field that.
00:15:24.503 --> 00:15:28.024
we can leverage these tools. So I h really hope so in the future.
00:15:28.139 --> 00:15:29.586
guess one Yeah.
00:15:29.129 --> 00:15:43.038
one more quick question then Beth can jump back in, sorry. ~ was ~ was you mentioned a little bit 'cause you're looking at correlational data and observational data, but you said that you you also mentioned the APO E genes and stuff like that.
00:15:43.038 --> 00:15:46.350
But if you were so I was just wondering how that played a role.
00:15:46.350 --> 00:15:57.360
So were you collecting sort of like genetic data as well from people who for participating people who played the video your video games and things like that. Like how did the study kind of work is Yeah, yeah.
00:15:57.139 --> 00:15:58.139
my question.
00:15:58.311 --> 00:16:02.212
So so so a lot of that so that particular study wasn't mine with the genetics.
00:16:02.212 --> 00:16:15.130
~ so we collected data. so there was so there's that sort of big bank of data of millions of people that we'd had ~ collected through Michael Hornberger at ~ at the University of East Anglia and our collaborators.
00:16:15.130 --> 00:16:23.645
So we had that big data set, sort of benchmark data set, but we also and in addition to that had about nine hundred people's worth of US data from a previous study we'd done.
00:16:24.090 --> 00:16:28.390
And so that's what I was looking at for the video gaming, driving, sleep studies that I had done.
00:16:28.390 --> 00:17:24.401
So particularly in in that US sort of young US sample, average age about maybe twenty-two, twenty-three, looking at that sample and seeing these associations with video and navigation. So the genetics part was something, yeah, we data data that we'd had been previously collected by our collaborators, including ~ people in ~ in America and and~ England and so we'd had that ~ previous data where they'd been where they'd had genetically tested and that was part of the the sort of the bank of data we'd yeah we'd leveraged but in my case there were there wasn't any ~ we didn't have genetics status as as a sort of variable we were looking at ~ for the US participants it was rather lifestyle factors we didn't actually look at genetics but it would have been very interesting to see the interaction between genetics and these lifestyle factors in in future work definitely ~ yeah so I think that's definitely one few one way this Work could be ~ could be going.
00:17:24.513 --> 00:17:34.003
Yeah, nice. I just want to very quickly go back just because I think you mentioned a study that I think is quite important and might kind of explain Mm-hmm.
00:17:33.402 --> 00:17:52.461
lot of people kind of why we're doing this research and it's this the taxi driver study. And as far as I'm aware, the taxi driver study was they looked at kind of the hippocampus in taxi drivers compared to people who were not taxi drivers.
00:17:52.461 --> 00:17:54.922
Is it that study you're talking about or is it another one?
00:17:54.753 --> 00:18:01.390
Yeah, yeah, yeah. So so that that's part of the yeah, research done by our team and collaborators.
00:18:01.390 --> 00:18:06.713
But that that specifically yeah, I didn't specifically work with taxi drivers for my research.
00:18:05.881 --> 00:18:08.203
No, absolutely. Yeah. And so But yeah, yeah, yeah, yeah.
00:18:09.303 --> 00:18:14.476
basically with the Taxi Driver Study, I'm assuming you could only maybe have a smaller sample sizes and it was specifically taxi drivers.
00:18:14.048 --> 00:18:15.048
Yeah, yeah.
00:18:14.476 --> 00:18:19.907
So with this See Hero quest, is it you're able to capture so much more?
00:18:19.907 --> 00:18:21.759
And basically, sorry, for those Mm-hmm.
00:18:21.209 --> 00:18:26.030
who don't have a taxi drive, basically the taxi drivers had larger hippocampuses than those who are not taxi drivers.
00:18:26.030 --> 00:18:28.292
And they think it's because they had to learn maps.
00:18:28.292 --> 00:18:34.625
~ And so I'm assuming then with the See Hero quest, then this is able to look at so many more people.
00:18:35.132 --> 00:18:36.423
and kind of understand Hm, yeah.
00:18:36.434 --> 00:18:42.868
kind of what's happening. And I'm assuming you scanned people as well in a MRI machine. Is that correct?
00:18:42.576 --> 00:18:44.211
so yeah, so for my PhD I Yeah.
00:18:44.267 --> 00:18:50.236
looked at mobile ~ FNIS, so actually mobile brain imaging, Yes! ~ nice.
00:18:47.978 --> 00:18:54.300
not fMRI, but our lab, our lab, our lab have a track record of fMRI. So we've done a lot of those fMRI studies.
00:18:54.300 --> 00:19:04.074
And yeah, you're you are right, like fMRI, you can get maybe 20 to 30 people a good sample once you've, you know, removed sort of the messy data and mobile Yeah.
00:19:03.765 --> 00:19:08.557
imaging. This yeah, mobile mobile, sorry, mobile ~ technology allows you to collect so much more.
00:19:08.557 --> 00:19:10.847
It's you know a 10-minute game, easy to administer.
00:19:11.223 --> 00:19:13.585
sort of you know cheap can do it anywhere around the world.
00:19:13.585 --> 00:19:27.280
So we have collected you know data globally on that from all sorts of And yes, like it just vastly, you can ask vastly more, many, many more questions because you have such a big data set to leverage with you know more power and sort of a robot bus data set.
00:19:27.280 --> 00:19:35.013
But yeah, FMRI, I did try to use that for one of my studies at the Minecraft study I did, I tried to use FMRI for that, but ~ participants would get very nauseous.
00:19:35.013 --> 00:19:48.067
So that's another sort of lesson I learned about how how about how how immersive environments fit in scanning settings and actually when you're lying down, you know, that was very challenging for people, it seemed, to actually carry out.
00:19:48.067 --> 00:20:00.161
So it it taught me a lot about, you know, what's doable in these fMRI scanners, how you design very, very simple naturalistic tasks and where the sort of the boundary is or the gray area between how naturalistic is too naturalistic and too involved.
00:20:00.161 --> 00:20:01.942
~ so that was a very interesting lesson.
00:20:01.942 --> 00:20:17.270
And so when we got to that stage, we moved to the mobile brain imaging, which was the FNIS, which is like fMRI in that it measures the sort of ~ oxygenated bold response, but uses light absorption to do that rather than sort of FMRI, which is the NMR resonance.
00:20:17.270 --> 00:20:24.114
So it's it's rather how much light is absorbed by hemoglobin and then that gives you a sort of readout of the brain activity.
00:20:24.114 --> 00:20:30.457
And so are they more active when they're hunting versus following so when they hunt a target, are they more active than when they simply follow one?
00:20:30.457 --> 00:20:31.707
So that was one question I'd asked.
00:20:31.707 --> 00:20:37.990
But yeah, we use mobile brain imaging because that allows for the ecological validity. People can actually move around when they're wearing that technology.
00:20:38.173 --> 00:20:42.786
And we can scan two brains at the same time, importantly. We couldn't do that in the fMRI scanner.
00:20:42.786 --> 00:20:51.329
I have a collaborator who in Japan in Kyoto there's a dual fMI scanner where you can scan two brains at the the same time, but we didn't have that technology.
00:20:51.329 --> 00:21:06.769
~ a UCL where you could do dual brain fMRI hyper scanning would have very interesting. But ~ but yeah, FNI is being more conductive to that because if it's a mobile, you know, cat wearing and people can just move around and logistically ~ sort of simpler than that would have been.
00:21:06.769 --> 00:21:25.435
So we thought we'd use the technology and we had collaborators who were in the FNAS department. They've got an FNAS established department at UCL as bioengineering team. And we'd so as I was saying before about leveraging other disciplines, other teams at UCL, we thought secondary supervisor Antonia Hamilton, because she's an expert in ethnic and social sort of brain studies.
00:21:25.435 --> 00:21:41.118
So we leveraged that and then yeah, it came to use FNIS. And then people, yeah, people wore caps and played the game with sort of one one or two people have reported a bit of nauseousness, ~ and we took longer breaks for those participants, but it wasn't nearly as drastic as FMRI.
00:21:42.039 --> 00:21:53.640
Yeah. So let's pretend that I'm going to play See Hero Quest now and you've got my data. What would that, how could you tell, what could you tell about someone's brain health?
00:21:53.640 --> 00:21:56.140
So what would be good? What would be bad?
00:21:56.140 --> 00:22:00.839
And how is there any way, you, I know you said it's, think you only did one time point.
00:22:00.839 --> 00:22:07.539
What would you be looking for to possibly say somebody is struggling and possibly could be a biomarker for Alzheimer's disease?
00:22:08.268 --> 00:22:18.541
Yes, for sure. so with that data, so we have this massive data set, often comes out as a raw JSON file,~ where we have the X ~ and sort of the w y coordinates in space.
00:22:18.541 --> 00:22:37.440
So imagine they played the game and you see the top-down view, you can map out the trajectory they've taken ~ every second of the game, because you have yeah, you have those X and Y coordinates in that plane, and we can basically take that, take those those coordinates and plot their trajectories, but also calculate a distance measure.
00:22:38.167 --> 00:22:44.131
which basically overall from sort of start to finish shows how total distance they've traveled.
00:22:44.131 --> 00:22:56.397
So ~ so a shorter distance would mean more efficient because they'd navigate to the goal ~ more directly, whereas a longer wayfinding distance would be less efficient, navigate to the goal less directly.
00:22:56.397 --> 00:23:04.332
And so we use that as the output marker of navigation efficiency, and then we can relate that to tons of variables as you say.
00:23:04.332 --> 00:23:07.554
So, you know, video gaming, driving, GPS, sleep.
00:23:07.766 --> 00:23:11.377
So we'd have those sort of variables matched by participant sort of ID.
00:23:11.377 --> 00:23:17.101
So we could have, you know, for example, which is even folks of basic morphometry, but in my case, it was these lifestyle factors.
00:23:17.101 --> 00:23:23.505
And so we can see basically, we can plot out like a, you know, a percent of distribution and we can see where someone fits.
00:23:23.505 --> 00:23:30.499
So in relation to the average person for their group, they are this is less efficient because they have a much longer distance.
00:23:30.499 --> 00:23:52.855
So we can do that and we can we as well as correlating it with these lifestyle variables, we can see ~ I guess the question you could ask is, you know, somebody has this this, you know, this level of gaming has been associated with ~ a pre-risk of cognitive decline. Therefore, where does this person fit fit on this spectrum?
00:23:52.855 --> 00:24:05.811
So if we know that, you know, doing X number of gaming is associated with better performance and that's linked to cognitive decline risk, in the future we ideally should be able to then look at that another person and say, well, where do they fit in relation to that?
00:24:06.016 --> 00:24:12.980
metric and how what does that tell us about how at risk they are of then going on to develop something cognitive decline or some adversity.
00:24:12.980 --> 00:24:38.700
So yeah, so we have because we didn't have that longitudinal data, we we did pull up these patients, we couldn't directly do that. But the idea is you'd have that a sort of a score for every person and then you could ~ sort of calibrate that to the population norm, to the sort of ideal value based on any kind of question you were asking. So it allows us to have to sort of do that sort of matching up of individual metric to sort of ~ population data.
00:24:38.700 --> 00:24:40.281
Yeah.
00:24:38.902 --> 00:24:41.119
Yeah, so it's basically normalizing the data.
00:24:39.718 --> 00:24:43.615
So Mm-hmm. Yeah, yeah, yeah, exactly.
00:24:43.615 --> 00:24:55.107
So lots of pre processing involved normalization and yeah, Yeah, as in like, of course, you know outlier remember you almost want to have like a population who's normal and then you could plot Yeah, yeah. Y you want a normal yeah, in the future and see where they play on that.
00:24:54.625 --> 00:25:09.155
yeah, exactly. A normal benchmark and then you can create all sorts of scores for like this is ~ what a you know, what a data set looks like for people who are particularly less risk of cognitive decline, you know, where where does the percentile fit in relation to those people?
00:25:09.155 --> 00:25:16.821
So exactly you can sort of ~ you can create different sort of distributions on different populations, I guess, and then relate different metrics to those.
00:25:16.821 --> 00:25:19.403
Yeah. Yeah.
00:25:20.059 --> 00:25:29.849
So I guess the follow up question for me is ~ w what were the sort of lifestyle factors which you found which came which were more relevant, I guess? Because I assume we were looking at quite a few.
00:25:31.109 --> 00:25:48.663
For sure. Yeah, yeah. so well, video gaming is ~ video gaming frequency was one of the key ones ~ because of the sort of rise in technological use, and we were very interested in ~ people, you know, people like using GPS every day now with Google Maps and and you know, our technological world, our mobile phone usage.
00:25:48.663 --> 00:25:51.676
We were very interested in sort of GPS and video gaming in particular.
00:25:51.676 --> 00:25:56.300
So we looked at video game frequency, how they rely how much they rely on GPS.
00:25:56.300 --> 00:25:57.662
So we found interestingly that.
00:25:57.781 --> 00:26:05.378
So video game frequency was associated with better navigation ability, but that GPS reliance wasn't associated with better navigation ability.
00:26:05.378 --> 00:26:15.407
~ and that's interesting because there was always this phenomenon in the literature that if we use GPS more, we, you know, use our brains less, supposedly, and therefore worse at navigation.
00:26:15.407 --> 00:26:30.434
But actually, we found there wasn't an association in this big sample, which puts together or I guess it set sets the scene for an interesting ~ idea that perhaps it's how we interact with GPS use that's important, with GPS devices that's important.
00:26:30.434 --> 00:26:50.645
So the reasons for using them, how how engaged we are in them, when we use them, sort of that sort of thing, our two-way relationship with these GPS tools might actually be ~ deciding how the effect it has on our our navigation ability and our health rather whether we simply use them or not, whether we simply rely on them more than someone else.
00:26:50.645 --> 00:26:53.768
So yeah, that was an interesting sort of takeaway from that.
00:26:53.867 --> 00:27:13.425
And I think that kind of leads into the question of developing AI assisted tools, understanding the hu relationship between human beings and AI, and how ~ how much sort of how we can develop GPS devices that adapt themselves to the user's preferences, the user's behaviors, and how we model that sort of two-way fluidity.
00:27:13.425 --> 00:27:38.355
~ so yeah, so that was ~ a t a very sort of interesting for me personally take home from that in particular. And also we looked at ~ Other lifestyle variables too, so like driving ability. So driving is something that had been perceived to be associated with smaller scale spatial cognition, so mental rotation, perspective taking, these sort of supposedly smaller scale abilities compared to the larger scale, navigating the real world, wayfinding.
00:27:38.355 --> 00:28:01.923
~ and it all it had been proposed that those were associated with these smaller scale abilities and driving, poorer driving, so more lane de a deviation, speeding, that sort of thing had been associated with ~ poorer cognitive health, so ~ increased ~ risk of cognitive decline and ~ increased likelihood of progression to dementia with you know failed~ t car tests, car crashes, that sort of thing.
00:28:01.923 --> 00:28:09.825
~ so there was an e a body of evidence there saying that cognitive function could be associated with driving or driving could be a biomarker for that.
00:28:09.825 --> 00:28:24.415
No one had ever looked at wayfinding and not a and not and not at not at scale, not this many people see how does driving associate itself with wayfind is it associates with wayfinding and What can that maybe tell us and you know be useful for in future research?
00:28:24.415 --> 00:28:27.148
So we thought, how does wayfinding relate to driving ability?
00:28:27.148 --> 00:28:37.105
And we found interestingly that driving independently at a younger age, rather than just simply driving or not, or simply how much driving was associated with with better navigation.
00:28:37.105 --> 00:28:41.638
So if you drove by yourself before the age of eighteen years old, you're a better navigator later.
00:28:41.638 --> 00:28:48.804
And that wasn't just if you drove more ~ in general or if when you said learn to drive, but if you drove by yourself.
00:28:48.804 --> 00:29:09.357
So There's something about the sort of independent ~ training of those cognitive perhaps that are involved in driving that~ something about it could be something to do with, you know, people having to take initiative when they're alone, people having to rely on the environment around them, people having to learn the scenes in their environment, people having to learn to make associations and form that cognitive map.
00:29:09.357 --> 00:29:16.539
There are many mechanisms by which that could have occurred that we don't know, we haven't explored. But it's yeah, it's interesting that.
00:29:16.503 --> 00:29:19.685
Describing solo seems to be an important predictor of wayfinding.
00:29:19.685 --> 00:29:37.159
And then we looked at sleep as a sort of final variable of that cluster because of associations, of course, with sleep, the sort of six to eight hours idea of it basically with cognition and a U-shaped curve. So more sleep, worse, less sleep, worse, sort of mid-range, better, supposedly from the literature.
00:29:37.159 --> 00:29:50.107
We'd shown that that was true in navigation in a previous study with CHERO but ~ We also know that there are several sleep variables, not just sleep duration. And that previous study had looked at sleep duration.
00:29:50.107 --> 00:30:13.788
And we wanted to see how many sleep factors, when considered holistically, could shape navigation and also how it interacts with gender. So there's differences in sleep quality, sleep length in males and females, and we wanted to see how those gender differences actually interact with and not just looking at a U-shaped quadratic curve, but also modeling it as a linear relationship to see if, in different ways to see how that relationship evolves, might be different in different.
00:30:13.981 --> 00:30:21.778
And different interactions. So we found that men in particular, more sleep was more important for their navigation.
00:30:21.778 --> 00:30:30.644
So men who had more sleep were better at navigating, but that what wasn't found in women, which is very interesting that there seems to be sort of sleep gender.
00:30:28.400 --> 00:30:29.518
That is fascinating.
00:30:30.384 --> 00:30:31.433
That's very interesting.
00:30:30.644 --> 00:30:37.729
And it's the first time we first time we'd ever showed that association with sleep and gender and So it was very interesting that there was something there.
00:30:37.729 --> 00:30:41.612
~ and it was this was sleep duration when we'd considered other sleep variables.
00:30:42.484 --> 00:30:44.796
within that sleep duration still survived.
00:30:44.796 --> 00:30:47.105
So something about sleep duration in particular.
00:30:47.105 --> 00:30:55.598
It was the first time we'd shown that sleep duration was particularly important compared to just sleepiness, difficulty falling asleep, you know,~ nap time, that sort of thing.
00:30:55.598 --> 00:31:15.278
So it seems to be something about sort of how long, which is interesting. ~ but yeah, the the gender as fact was particularly interesting because genders, you know, males have always been thought ~ across many studies, be supposed to be better in females at navigating across different environments, although studies have shown that that's not necessarily true.
00:31:15.278 --> 00:31:31.226
And you know, motivation levels, you know, context, task difficulty, a type of environment, type of task, that can vary. But but yeah, that was very interesting for me to see that that sleep is one of the metrics in which gender differences might be apparent.
00:31:31.226 --> 00:31:34.708
And ~ that in turn can shape navigation.
00:31:34.708 --> 00:31:42.000
So it yeah, there's definitely more exploring to do that could give us a c maybe a better answer as to why males and females perform differently.
00:31:42.000 --> 00:32:07.873
Because if we understand maybe there's something metabolic ~ metabolism in sleep, something to do with hormonal, you know, something to do with life, so sociocultural patterns, you know, women staying at home to look after the kids, all those sorts of stereotyped roles, you know, differences in work patterns, work distributions and It gives us a sort of a platform to look at those things that could give us a closer answer as to why males and females perform differently.
00:32:07.873 --> 00:32:15.066
So it nudges us in a direction to want to actually explore those things that we never thought maybe might have been important before in navigation.
00:32:15.066 --> 00:32:18.307
So yeah, so I think that's the the summary.
00:32:18.307 --> 00:32:25.950
~ that's that's a summary ~ yeah, s of where the sort of we are with the lifestyle factors.
00:32:25.950 --> 00:32:29.070
so yeah, I hope that's ~ interesting insight.
00:32:29.070 --> 00:32:31.637
~ yeah, we'll incredibly Incredible.
00:32:30.699 --> 00:32:55.464
interesting and really interesting because as well of course in Alzheimer's disease it's more in women than men so the fact that you found sex differences within navigation and sleep is particularly important so you looked at sleep driving kind of lone driving and kind of sex is there any other ~ ways that people can protect their spatial navigation abilities as they age or is it just luck?
00:32:57.296 --> 00:33:02.823
yes, I mean we're that's sort of again a big big question when you say about ~ Mm-hmm.
00:33:02.759 --> 00:33:53.624
protecting we I mean we're we're seeing across various different populations about navigation ability being compromised and traumatic brain injury,~ amnesia. ~ and we're ~ we're seeing that the sort of gap in between males and females navigation ability could perhaps be explained by~ differences in economic sort of value and so so GDP per capita could be explained ~ by by those sorts of economic socioeconomic factors as well as ~ we're seeing people in different environments~ so people in you know the rainf rainforests in Congo for example or you know, tribes when they learn to sort of navigate, they adapt to their environment very much and learn partic specific strategies which help them navigate in that specific environment.
00:33:53.624 --> 00:34:02.277
So in the Marshall Islands, people use stick charts to measure the pattern of the waves when they're on the boat to find their way to goals because they don't use GPS or signage really there.
00:34:02.277 --> 00:34:18.135
And so I think I think ~ it seems to be that, you know, getting getting off GPS and getting out there and actually learning to sort of explore without necessarily these sort of mobile tools.
00:34:18.135 --> 00:34:51.695
So learning to sort of ~ take take hostage of an environment and learn to exploit that environment for the better~ seems to be associated with better navigation ~ abilities. There was a study in taxi drivers that shown showed that sort of exploiting features of an environment was associated with better abilities. So it's how they exploit boundaries and that sort of led to the idea that it's maybe us being able to successfully exploit an environment to the fullest that might be associated with the best, maybe the best we can do to protect ourselves.
00:34:51.695 --> 00:35:05.440
~ so the more and like an orienteering sort of training in a lot of schools in Scandinavia, they have that orienteering training and Nordic populations have been shown to be associated with their navigation ability in Sea Hero Quest, ~ better wayfinding.
00:35:05.440 --> 00:35:16.333
And so I think early in life, definitely leveraging the environment we have, the the environments and the tools in that environment available to us ~ earlier on as well.
00:35:16.333 --> 00:35:29.380
So gaming, you know, sleep, driving, for example, ~ perhaps we could facilitate the use of these everyday tools that we use and actually, you know, use them to train our ourselves to be better at navigating.
00:35:29.380 --> 00:35:38.545
So I think it definitely seems to be that it's very much sort of me and the environment, ~ how do I utilize my environment well?
00:35:38.545 --> 00:35:41.867
~ how do I sort of use the environment available to me?
00:35:42.460 --> 00:36:09.210
~ it seems to be that sort of the way it's heading in terms of what can actually protect us or make us pr protect us from navigation decline as well we've seen recently that so ~ socially so socializing, engaging in social activity has been associated with better navigation,~ richer sort of ~ yeah so ri rich more engaging more with people that the things that have been associated with you know loneliness and dementia.
00:36:09.210 --> 00:36:23.757
So I think definitely definitely all of those things I mentioned, how we how we underst understanding the relationship between us and the environment and how that changes over time and how that how recipe the reciprocal nature of that.
00:36:23.757 --> 00:36:26.378
Like when is it more when is it more us?
00:36:26.378 --> 00:36:27.557
When is it more the environment?
00:36:27.557 --> 00:37:12.068
When is it how does the individual, how do the attributes of the individual fit into the wider environment that in? I think if we're understanding that relationship, that will give us the best ~ idea of ~ what can actually really protect our navigation skills, like the when, how often, the how, the random monitoring, ecological assessment, you know, passive monitoring, active monitoring, those things come together and understanding how individual attributes personality, things like that recently have been shown to shape the gender effect on navigation. So how do those things, how do our how does our desire, our tendency to want to exploit an environment actually affect our navigation ability? So how do those individual traits affect the environment itself are our are how we behave in in a given environment.
00:37:12.068 --> 00:37:14.349
So it's a very, very complicated picture.
00:37:14.349 --> 00:37:37.018
but I think definitely getting out there and you know, exploring as much as you can on your own terms, and it seems to be being curious and being, you know, willing to want to explore new environments seems to somewhat play a role in ~ better navigation ability as well.
00:37:37.402 --> 00:38:02.494
in terms of, you know, traits, and yeah, and definitely learning to adapt, you know, and adapt to different environments and learning how to use an environment to your advantage and exploit that environment, I think, is what I can say from the literature seems to be ~ the important the important things that are shaping our navigation ability.
00:38:02.494 --> 00:38:20.148
~ but I can't ~ we're not quite at the stage yet where we have concrete metric that f that predicts ten years later who go ~ goes on to develop dementia and who doesn't necessarily ~ a clear sort of composite ~ measure. So we're not not quite at that stage yet.
00:38:20.148 --> 00:38:27.034
But I think yeah, from the sort of complex literate body of evidence, that's what I can say is ~ the most promising thing.
00:38:28.083 --> 00:38:35.746
Yeah, well, it sounds to me like,~ think pirates had lower amounts of Alzheimer's disease then they were exploring. So.
00:38:35.215 --> 00:38:46.128
Yeah, I mean I I I I I predict so and I I I predict that they, you know, they they are constant they have less fear as r less of spatial anxiety as a result of that experience.
00:38:45.139 --> 00:38:47.509
Yeah, treasure hunting?
00:38:46.128 --> 00:38:53.543
And that yeah, spatial anxiety in itself has been associated with navigation ability, on the seas. Yeah.
00:38:52.061 --> 00:39:04.213
of course. Yeah, so like maybe people maybe maybe their differences in their tenacity and their willingness to ~ take on new challenges in general, i it i in itself, aside the spatial component, just having that.
00:39:04.389 --> 00:39:12.563
desire to have to ~ you know to to have yeah to to to take on novelty like that in itself I would predict with social navigation.
00:39:12.563 --> 00:39:14.335
So yeah, we're looking at different things there.
00:39:14.335 --> 00:39:17.373
We're looking at the novelty element, the actual spatial map element.
00:39:17.373 --> 00:39:18.938
There's so much in that as well.
00:39:18.938 --> 00:39:28.222
But it's very and they wouldn't have to rely have around, you know, ~ their positioning system and learning associations because they use GPS. And so so so yeah. Yeah.
00:39:26.206 --> 00:39:30.534
Yeah, exactly. Very social. Cockroach.
00:39:27.264 --> 00:39:34.485
I mean I I'm I mean So all of those. Yeah, yeah. Social, exactly. Yeah, yeah. So all of those things I think, yeah.
00:39:33.943 --> 00:39:36.552
I I mean you don't even need to go as far as pirates.
00:39:36.552 --> 00:39:39.385
I feel I am better at navigating when I'm touristing.
00:39:39.385 --> 00:39:45.201
So if I go to a new city, then you're trying to like find landmarks, trying to learn the space. I feel you do better.
00:39:44.913 --> 00:39:45.913
Yeah.
00:39:45.360 --> 00:39:50.976
At least personally, I feel like I'm more willing to learn the locations and try to figure out what's a bit down the road from me.
00:39:50.976 --> 00:39:54.668
And Exactly, Yeah, well, yeah. We shouldn't exactly. Like you know yeah, yeah.
00:39:54.668 --> 00:39:56.047
That rich life, yeah.
00:39:56.269 --> 00:39:58.061
use Google Maps basically is what you're saying.
00:39:58.061 --> 00:39:59.791
Or look at it. Like Mm-hmm.
00:40:00.143 --> 00:40:02.259
see here request, look at it and then try and figure out.
00:40:02.259 --> 00:40:08.967
So we've kind of been on a little bit of one part of your PhD, but I appreciate you did some stuff with Minecraft as well.
00:40:08.967 --> 00:40:10.362
~ So ~ Yeah, yes, yes.
00:40:11.268 --> 00:40:13.751
VR, made them play Minecraft together.
00:40:13.751 --> 00:40:17.813
And this is, think, what you're talking about with the minds together and seeing kind of like at the same time.
00:40:17.813 --> 00:40:19.934
Could we learn a bit more about that one as well?
00:40:18.896 --> 00:40:22.407
Yeah. Of course. Yeah.
00:40:19.934 --> 00:40:21.235
Sounds fascinating.
00:40:22.407 --> 00:40:34.708
So mine so that was sort of inspired by a lot of the navigation research we've done has been looking at an individual and how they navigate in an environment, their sort of performance, like how do I get to a train station or a landmark?
00:40:34.708 --> 00:40:57.199
~ we hadn't really explored how people relate to other people ~ when they and then we're recently sort of emerging evidence that there's this kind of social brain for navigation where we track other moving human beings, other moving goals, ~ It wasn't clear how and like hunting was the sort of the task we used for this was hunting as a sort of proxy of looking at this sort of navigation in social environments.
00:40:57.199 --> 00:41:04.907
Because when you hunt, ~ like a navigation, you have to plan, but you also have to work together to get to a share to sort of get to the shared goal.
00:41:04.907 --> 00:41:08.010
So we thought hunting was a good sort of mechanism to explore this.
00:41:08.010 --> 00:41:15.878
~ and we hadn't, no one had really done any sort of How do we navigate when we're with other people in a virtual environment, only when we were by ourselves?
00:41:15.878 --> 00:41:34.577
So we wanted to sort of explore and to leverage that unknown knowledge and and also understand with the EFNIAs how do your sort of brain sit how does our brain track other people in our in our environment when we navigate with them? And ~ and we hadn't really, yeah, so we no one had really looked at that.
00:41:36.007 --> 00:41:42.320
When sort of you have two people sort of moving together continuously, dynamically. And so we thought, you know, let's let's explore that.
00:41:42.320 --> 00:41:46.393
So we Minecraft is a great environment to do that because of its be a sandbox game.
00:41:46.393 --> 00:42:05.523
You can create anything, in other words,~ reliably key building blocks,~ cubes that can you can easily create environments with because of its sort of it can be stripped down to a simple block, and then you could sort of quantify where everything is and quantify positions in the game. So using Java, you can get the game data out, quantify the X, Y, and Z coordinates.
00:42:05.918 --> 00:42:11.099
quantify where people are in the game, and then so basically you can get get their trajectories out.
00:42:11.099 --> 00:42:34.690
And we can look at sort of the speed synchrony, we looked at distance direction, because those are the things in the navigation literature that we were in in, but also, you know, how they synchronize their paths, the two hunters. So we had two, two ~ participants who took on the roles of the hunters or the predators, and then one person who was a member of our research team who took on the role of the prey. And they were in the environment at the same time.
00:42:35.134 --> 00:42:40.516
And the pre they had to catch the two predators had to catch the prey as ~ quickly as possible within the one minute.
00:42:40.516 --> 00:42:44.588
~ and ~ so that was their task during the hunt trials.
00:42:44.588 --> 00:42:57.585
And there were also follow trials where two predators had to follow the path of the prey as closely as possible, but not hunt it. So they weren't hunting, but they were just following that visual stimulus.
00:42:57.585 --> 00:43:01.307
It was like a baseline control where they saw the same thing, but they weren't doing any of the catching.
00:43:01.307 --> 00:43:04.349
So we could compare the hunt and follow trials together.
00:43:05.045 --> 00:43:07.467
and the two predators had ethnia's caps on.
00:43:07.467 --> 00:43:11.409
The two predators both had brain activity recorded, so we could see their brain synchrony.
00:43:11.409 --> 00:43:28.838
And we found that there was greater brain-to-brain synchrony in the prefrontal cortex, both medial and lateral, during hunting, ~ compared with chance levels or compared with a random sort of distribution for Monte Carlo assimilations, and which was interesting because it showed that.
00:43:29.309 --> 00:43:44.762
The prefrontal cortex, which has previously at the individual level been with versatile navigation. So detours, shortcuts, you know, ~ changes on the go, was actually important for linking up the two brains of these hunters in this virtual Minecraft environment when they were doing this navigation task.
00:43:44.762 --> 00:44:20.896
So it was the first time we'd shown prefrontal synchrony during navigation two people in a virtual environment, and it extended the idea that the prefrontal cortex might have a social role in navigation and not just be important to the individual level, which opens the door towards ~ understanding social connection, loneliness, like the social factors associated with cognitive decline, and whether brain mechanisms might be important for mediating those effects and whether and if navigation is associated with that, is you know, social navigation relying on a different brain mechanism to individual navigation. And how do those talk to each other?
00:44:20.896 --> 00:44:21.896
How are they different?
00:44:22.077 --> 00:44:27.201
How might the mechanisms that are common or different to both be associated with cognitive decline?
00:44:27.201 --> 00:44:34.697
So is it that navigating socially can protect our brains ~ from the navigation decline at the individual level?
00:44:34.697 --> 00:44:44.773
You know, there's all sorts of questions you could ask about how to enhance navigate how to enhance our cognitive health through understanding why the brain is important in mediating social relationships.
00:44:44.773 --> 00:44:47.445
And there's lots of questions that you know come from that.
00:44:47.445 --> 00:44:49.516
So I think that's the direction this is going.
00:44:49.516 --> 00:45:04.764
And how can we leverage synchronization tools in the workplace, how could we develop tools that better help recruiters or better help team based on what we understand about brain synchronization ~ so an entrepreneurial sort of pursuit that I'm also interested in.
00:45:04.764 --> 00:45:08.985
So I think that it's it opens up sort of many doors in many different directions.
00:45:08.985 --> 00:45:12.447
But it ~ it was just a p a pilot study that was at the very beginning.
00:45:12.447 --> 00:45:19.250
~ but but yeah like it's the you see the layer of the land is very vast and multidisciplinary.
00:45:20.940 --> 00:45:23.771
So by pilot study, how many pairs of people did you do it with?
00:45:24.635 --> 00:45:46.378
Yeah, so ~ originally we had thirty-two sub subjects, sixteen dyads, but then with all the noise and you know,'cause you have light not penetrating through because hair gets in the way with ethners or too much light absorption because of, you know, the setting, the environment, the person's anatomy, we ended up with about nine dyads, which reviewer flagged as being very small.
00:45:46.378 --> 00:45:52.911
And we we know that too, but it's like so classifying as a pilot's preliminary study is probably the thoroughest assessment of that.
00:45:53.032 --> 00:45:55.023
So we to be careful and cautious of the findings.
00:45:55.023 --> 00:46:07.385
But ~ yeah, so you you know with it mobile with mobile data, you have to often get the problem of motion artifacts, people moving around a lot, or people, you know, naturalistic sort of things getting in the way of the data.
00:46:07.385 --> 00:46:13.688
So that's another sort of lesson I learned about the trade-off between spatial resolution and ecological validity.
00:46:13.688 --> 00:46:19.789
So you have to sort of think what's more important and when, what do I care more about for my scientific question?
00:46:19.789 --> 00:46:26.958
What's going to be more useful for this to actually know and What are the and so therefore what should I trade off with what sort of thing.
00:46:26.958 --> 00:46:33.344
So I think that's another maybe takeaway from that is, you know, how you go about deciding what tool to use when.
00:46:33.344 --> 00:46:34.539
Yeah.
00:46:34.759 --> 00:46:36.681
Need to get a few more bald friends involved.
00:46:36.681 --> 00:46:45.005
In your Yeah, yeah, exact yeah, no, you know, yeah, exactly. You know, yeah,'cause 'cause if ~ obviously you have a bold participant, amazing, you know, no hair to worry about.
00:46:45.005 --> 00:46:49.518
But of course that that ne never happened in our study, unfortunately. Yeah.
00:46:50.726 --> 00:47:03.762
I I d I mean so I I guess pilot study, so cautiously speaking, what did you see in terms of differences between like sort of navigation versus like the individual navigation?
00:47:03.762 --> 00:47:05.885
Wha do you see many differences?
00:47:05.885 --> 00:47:07.036
Do you see some differences?
00:47:08.248 --> 00:47:19.742
so we didn't actually compare like a social condition to an individual So that was what was missing, and a follow-up study should directly look at that because then I'd be able to, you know, answer what exactly was different.
00:47:19.742 --> 00:47:41.014
~ social, I mean, ~ in the past we've seen or a couple of recent studies have shown that when we track a moving goal or moving avatar, there's some ~ we get the same sort of regions like medial temporal lobe, for example, that been commonly implied in navigation, hippocampus, you interrogano cortic surrounding regions, but also parts of the frontal cortex as well and even in the posterior cortex too.
00:47:41.014 --> 00:47:45.489
So there's this kind of this interaction actually between different regions.
00:47:45.489 --> 00:47:56.197
But we sh we ~ they showed in this recent study that sort of lateral prefrontal connectivity with the sort of medial temporal lobe was important in tracking ~ the path of a moving avatar.
00:47:56.197 --> 00:48:26.057
So it seems to be there is some crosstalk between different regions implied in navigation. So it's very much a sort of a network that's recruited, it seems, at the individual level from what we've seen so far that might be important for goal tracking, then we need to do these days studies where we validate individual versus together versus to see actually how much overlap is there. Is it a c is it, you know, it what are the nuances and and in people where there isn't a significant difference where there is, why in those people is there a difference and not in other people?
00:48:26.057 --> 00:48:31.550
So it could be that that is me the differences are mediated by lifestyle factors that we know are associated with navigation.
00:48:31.550 --> 00:48:53.231
So is is it that people who drive independently are those are Are their social networks more distinct from people who don't for when they individual levels? So is it that the difference between the individual networks and the social networks recruited, is that gap bigger ~ the more one engages in these lifestyle factors like driving?
00:48:53.231 --> 00:49:03.289
Or that would be an interesting question to see whether differences can be explained. And if not, if they can't be explained by lifestyle factors, that's very interesting because it shows there's something inherently there.
00:49:03.666 --> 00:49:11.201
in the brain's neural wiring, that isn't explained by how they interact with environment. That could that it's almost like nature versus nurture.
00:49:11.201 --> 00:49:18.757
Are they predisposed to navigating differently because of the ~ fundamentally because of their difference in neural networks?
00:49:18.757 --> 00:49:24.851
That's, you know, not not explained by how they they evolve how they grew up and the environments they were exposed to throughout life.
00:49:24.851 --> 00:49:40.362
So that would be another sort of bigger question to look at sort of from childhood to adults, ~ to through maturity, how how do those yeah, how does that come to be? But I think that's a yeah, question for future research.
00:49:41.282 --> 00:49:54.983
Yeah, I think you spoke to us before, the big thing is like it's when you play Minecraft, it's socializing, it's planning, it's navigating. So when you see in those kind of different areas, the prefrontal cortex, is that all three of them working together?
00:49:54.983 --> 00:49:57.143
Or is it one of them? Or is it maybe two of them?
00:49:57.143 --> 00:49:58.726
Is that something Yeah.
00:49:58.202 --> 00:50:01.702
that can pull apart or what's your thoughts on that?
00:50:02.480 --> 00:50:10.114
Yeah, like I so I mean, FNIRs in my case is very sp I mean, the cellular did is very spatially limited.
00:50:10.114 --> 00:51:05.992
So you get about three centimetres penetration depth, very much a scalp recording. And I think like first thing would be even within the frontal cortex, like there's areas of the frontal ~ sort of lobe, like more dorsal anterior singular cortex that have been shown to be associated with tracking moving goals in monkeys, for example, or primates, where people have actually done intracranial recordings and got those deeper~ so I think the first thing would be maybe to fully explore the parts of the frontal lobe that from animal literature that have been associated with of tracking moving goals during navigation, see whether that carries through to humans, understand, you know, fully, because the frontal, especially the orbital frontal, the sort of ~ frontipolar cortex, I mean, has been sort of more recently proposed as a an area in itself based on the sort of monkey literature,~ looking at how sort of~ dissections and that sort of thing affect one's the ability to sort of keep two goals in mind, sort of thing.
00:51:05.992 --> 00:51:13.967
So I think it's only a recently emerging and accepted maybe region ~ to be important cognitively in in humans.
00:51:13.967 --> 00:51:28.329
And I think we need to do more digging and exploiting of the different components of the in a bigger pop population of the different parts, lateral versus medial, is it really distinct or is it really a broad range of ~ routines, even within the frontal cortex itself?
00:51:28.838 --> 00:51:38.581
And then adding on to that, we have separately the hippocampus and those~ to look at in in their in their own sort of right, because we haven't looked at them necessarily in this setting.
00:51:38.581 --> 00:51:56.039
We couldn't get hippocampus because it was too deep, for example. So we need to ful you know fully exploit ~ those different regions and then add on to that and see their connectivity and crosstalk as a sort of second step, and then ~ you know, from that see how those networks evolve over time and longitudinally.
00:51:56.039 --> 00:51:57.460
So I think I think.
00:51:57.670 --> 00:52:10.215
it's very hard to say right now if the functional prefrontal cortex can stand alone as its own entity or whether it d it actually needs that And of course in humans you can't, you know, knock a path out, you can't, you know, put a g genetic manipulation.
00:52:10.215 --> 00:52:14.067
You can't sort of physically manipulate that connection to see if it's necessary.
00:52:14.067 --> 00:52:25.041
We have, of course, case studies from people like frontal lobe dissection participants in, you know, hip patient HM with hippocampal abnormalities.
00:52:25.041 --> 00:53:04.231
So we have ~ case studies to go off, but in order to, you know, really manipulate that we need a more interventional studies where we might where I think the future's going, where we put people into this dual game and we assign other people to not play this dual game or to do a a very simple two D task that doesn't involve them necessarily ~ collaborating in this 3D environment and then see after a few sessions of training, ~ what are the mechanism that seem to be important in driving the this ability and and you know, is there is it in a large enough sample, is there a concrete enough effect?
00:53:04.231 --> 00:53:44.545
If we see it's very robust and there's lots of frontal cortex to posterior cross talk across the population in a big sample, then we can be more confident and certain that there seem that crosstalk seems to be necessary. ~ if it seems to be that, you know, it's the mirror frontal activity, but very weak effect size from the hippocampal or the functional connectivity, then yeah, so that sort of thing, com comparing the sort of looking at the readouts of the study with interventions, I think is going to be the the way to really understand what's how cr how important sort of crosstalk is compared to sort of individual regions their own role ~ in that. Yeah, if that makes sense.
00:53:45.166 --> 00:53:47.048
No, absolutely. Gosh, Awesome.
00:53:47.036 --> 00:53:55.362
that's so fascinating. ~ It's amazing that you did two of these big studies in a like a PhD. It feels like it could be two PhDs.
00:53:55.362 --> 00:53:57.664
It's insane. So you've all finished that now.
00:53:57.664 --> 00:54:03.103
~ Can we briefly talk a little bit about the hackathon life and Yeah.
00:54:02.847 --> 00:54:07.469
kind of what that, what you do, what it consists of and yeah, what made you get into it?
00:54:08.543 --> 00:54:10.713
Maybe what are hackathons Absolutely. ~ as well?
00:54:10.280 --> 00:54:11.280
Button.
00:54:10.349 --> 00:54:12.094
Yeah Yeah, what hackathons, yeah, of course.
00:54:12.094 --> 00:54:18.143
So ~ so ~ so so here was asking what actually are hackathons, and I think it's Yes.
00:54:17.615 --> 00:54:23.079
where is where so it's like you think like a marathon hackathon, like you come together and you hack.
00:54:23.079 --> 00:54:50.699
So people come together to an event, they have like a few hours or a weekend generally to come up at in teams to come up with an idea for a product that could be commercialized or that solves a scientific problem or a health problem and that can basically be eventually sold and you deployed in real world settings to act solve that problem. So ~ I went to one where I worked on a screening tool for osteoporosis, for example.
00:54:50.699 --> 00:55:09.143
~ we call it BoneBot. And generally you work on the you s you sort of~ you do something called vibe coding, where you code through AI tools like Claude, clawed code, you you and you use tools like ~ Devon by Cognition, which allows you to, for example, look at agents and develop MVPs for products.
00:55:09.143 --> 00:55:31.911
So you develop an MVP, which is a sort of working, minimal working prototype, or put an example of how the product would work in real life. So this is what the phone app would look like, for example. So in our case, it was a tool where an AI agent would talk to you for a phone app, get information about your demographics, blood tests, blah, blah, and make a prediction based on how likely you were to have osteoporosis and whether you need screening.
00:55:31.911 --> 00:55:33.532
That could be flagged by the clinician.
00:55:33.833 --> 00:55:36.014
So you'd be doing all the processes involved in that.
00:55:36.014 --> 00:55:50.157
You'd be working as a team to decide on, you know, what the app would look like, design in the user interface, the business model, how you would sell it, ~ you know, what tools you'd use to actually create it, you know, and you work on all those elements of basically, yeah, creating a business product.
00:55:50.157 --> 00:55:53.318
And each person in the team generally has a role.
00:55:53.318 --> 00:55:59.650
One person's a software engineer, one person has product design experience, one person has clinical experience or MBA business experience.
00:55:59.650 --> 00:56:06.447
And your sort of skills, the idea is your skills come together to To form a team where everyone can get their own new piece of knowledge.
00:56:06.447 --> 00:56:19.632
And then eventually you come up with a product at the end, which you have to submit along with the GitHub repository for that, the code, the the sort of readme of how things work, and a live demonstration where you can say press here to try using our app.
00:56:19.632 --> 00:57:01.181
So it's like a it's like a fast track, it's like a fast version of what you'd be doing in a in a in a business if you like, that would sort of like you can imagine For example, ~ APO Health, where they have lots of health data and they ~ you know, they ask for the tracks of data basically, and it'd be something like that, but over a weekend. So you do you be developing that infrastructure but in as fast a time as possible. So it's it's very much ~ a sort of fun challenge of how can we work together effectively as a team, if anything. How can you delegate tasks, how can you multitask, work really efficiently, and use the best of every person in the team to come up with a final successful product.
00:57:01.181 --> 00:57:07.155
So I love that kind of that sort of adrenaline pumping challenge, you know, and that's what I like most about it.
00:57:07.155 --> 00:57:35.041
And seeing how things can be deployed, how we can actually turn an idea into an actual product that people can use in everyday life. ~ it's what really drives me as well, the application of that. So I started doing hackathons and I sort of never look back since 'cause I've met so many friends through them and met so many cool people in tech as well who can help with work and answer sort of questions about careers and different avenues that come from that. So yeah, I I'd r highly recommend the experience.
00:57:36.744 --> 00:57:39.849
How did you get into it? How did you find it?
00:57:39.849 --> 00:57:41.755
somebody tell you about it or was you just looking yeah.
00:57:41.436 --> 00:57:42.436
up?
00:57:42.217 --> 00:57:50.157
Good question. So on LinkedIn, during the lockdown twenty twenty, for example, I'd first started searching for opportunities and I just saw it through a LinkedIn post.
00:57:50.157 --> 00:58:02.021
And if we're in the right algorithms and you, you know, if you I guess if you respond to people who organize them, like ~ you know, people in health tech and business and you kind of connect with them and you stay up to date with their network, you hear about opportunities like that.
00:58:02.021 --> 00:58:03.563
And often companies sponsor these things.
00:58:03.563 --> 00:58:27.972
So Anthropotic sponsored one of the hackathons I did, ChatGPT. So if you're kind of generally in the circles of understanding what's going on in the tech world, people will up would advertise these things through LinkedIn, through WhatsApp groups, through their LinkedIn pages, through so yes, you've just through knowing who just through connecting with the right people and following their activity and responding and engaging, you'll just hear about these opportunities through the algorithms as well.
00:58:27.972 --> 00:58:43.376
And it So yeah, so and yeah, and also through word of mouth to through Instagram, for you know, different different areas, but but I think LinkedIn initially was how I I thought that looked cool and that was yeah, the first one I did in lockdown and yeah, so Yeah. Yeah, my wife then.
00:58:42.713 --> 00:58:56.902
has my wife has many a hackathon shirt in her closet because she parti participated in a lot as a student, as a master's student at UCL. So I assume you do you do them through your university as well, or like people just come if you're a computer scientist.
00:58:57.632 --> 00:58:59.083
Mm-hmm. Yeah. Fantastic.
00:58:59.083 --> 00:59:08.737
Yeah, exactly. Yeah. You you you sort of do people people when they're a lot of computer science students seem to do them as a sort of more like a sort of regular expected thing.
00:59:08.737 --> 00:59:15.851
~ but like you stay up to date with the AI tools that are being deployed by doing them. You learn about this tool's being used or this company's doing this.
00:59:15.851 --> 00:59:23.835
And yeah, I think it's ~ it's a great learning exercise as well to learn what's the most popular tool being used now, what's the most popular.
00:59:24.242 --> 00:59:26.934
agent AI tool that's being used to develop user interfaces.
00:59:26.934 --> 00:59:33.980
So I've learned a lot about, you know, where in the landscape my sort of skills fit in, which I think is very important for job wise.
00:59:33.980 --> 00:59:39.835
So I'd highly recommend it for that sort of ~ purely sort of educational purpose.
00:59:39.981 --> 00:59:46.376
~ w I have a question. Well, it's a question which we always ask, but usually Beth asks it this time.
00:59:46.376 --> 00:59:53.440
I will ask it. ~ suppose na you've been going through all these hackathons and things.
00:59:53.440 --> 00:59:56.318
You've done a lot of you've you've Mm-hmm.
00:59:55.862 --> 00:59:59.253
done like a wide range of research and you've spoken about next steps a lot.
00:59:59.253 --> 01:00:03.206
So let's assume all ethics are on holiday.
01:00:03.206 --> 01:00:09.610
The Bet has two billion pounds for you to hand you to do whatever you want.
01:00:09.889 --> 01:00:13.400
You just need to convince her about what your research project is going to be.
01:00:13.400 --> 01:00:17.510
So if that's the case, what is your dream research project, Dr. Urez?
01:00:17.295 --> 01:00:20.250
no ethics. The ethics board have gone to sleep.
01:00:20.250 --> 01:00:22.199
They're sleeping on Yeah.
01:00:21.652 --> 01:00:23.994
holiday for a week, so anything goes.
01:00:23.715 --> 01:00:25.416
I mean, great question.
01:00:23.994 --> 01:00:25.987
But be mindful, respectful.
01:00:26.755 --> 01:00:28.586
Yeah. Something I always think about.
01:00:28.586 --> 01:00:41.552
yeah, because you know, you want to be as far forward thinking with your research as you can. I mean, I think I would take a thousand newborns,~ randomly assign half of them to grow up, never using GPS, and then follow them for forty years.
01:00:41.552 --> 01:00:52.916
~ so I think I think you wanna ~ so the qu so you want ~ everyone asks me at parties, or is GPS making us worse at navigating and is it doing something to our brains?
01:00:53.278 --> 01:00:55.920
Nobody can answer it because it's correlational always.
01:00:55.920 --> 01:00:59.483
People who navigate by phone are different to people to begin with.
01:00:59.483 --> 01:01:06.940
So you randomize at birth, control the technology exposure for decades, you measure navigation, hippocampal structure, confident decline across the lifespan.
01:01:06.940 --> 01:01:14.657
It's one of the cleanest experiments that you could do because it's it's completely unrunnable, which is why the question is open after 20 years of sort of arguments.
01:01:14.657 --> 01:01:21.771
And you could learn not just whether GPS hurts, but whether the effect is ~ the device or the interaction between the person and the device.
01:01:22.010 --> 01:01:30.875
and what it is that we we actually can do and engage in to reduce our risk of cognitive decline, as you say, in navigation and understand that from very early on.
01:01:30.875 --> 01:01:50.273
So ~ you want to be able to instrument an entire city, sort of every r resident could even wear continuous brain and body sensors everywhere they go for a decade, like a neural link sort of collaboration, no consent, you know, no sampling bias, no lab. You get a first real map of how humans navigate, coordinate, you just be able to watch it as it goes along through life.
01:01:50.498 --> 01:02:05.164
So you could get all the data you want and just see how peop why people have the trajectories they do. ~ coordinate how do they make decisions collectively and and you'd find health signals nobody has looked for because nobody ever had the data. So that's that's I think what I'd probably do.
01:02:06.101 --> 01:02:12.367
It would be very fascinating as well because y we know like before the GPS different cultures came up with different ways of navigating.
01:02:12.367 --> 01:02:14.132
So how people would Yeah.
01:02:14.047 --> 01:02:21.610
in like a completely randomized experiment, how people would develop their own sort of methods of navigation from Mm-hmm.
01:02:21.954 --> 01:02:29.530
from scratch versus the GPS group to see if like will they use the stars, will they use certain like east west coordinates?
01:02:29.530 --> 01:02:30.601
Well yeah.
01:02:30.800 --> 01:02:34.793
I i exactly. Like you could see how what what do they evolve to survive?
01:02:34.793 --> 01:02:40.967
You know, what are they how are people who are not using GPS, how many how much variation is there in the techniques that they use?
01:02:40.967 --> 01:02:45.789
And yeah, what predicts w w which mechanisms they develop.
01:02:45.789 --> 01:02:48.731
You know, why do some people develop this mechanism and other p other people?
01:02:48.731 --> 01:02:51.282
Yeah, you know, that'd be very interesting.
01:02:51.478 --> 01:02:55.257
Yes, sold. Have two billion. Enjoy, I believe.
01:02:56.434 --> 01:03:00.860
Yeah. Exactly.
01:02:56.987 --> 01:02:59.512
Might need more actually for that that kind of project.
01:03:00.376 --> 01:03:03.106
Well, we can start off with that and then I can be persuaded.
01:03:00.860 --> 01:03:03.090
Yeah.
01:03:03.746 --> 01:03:06.088
One trillion maybe. Yeah. Yeah.
01:03:07.346 --> 01:03:08.418
Hit hit We'll contact Elon.
01:03:09.019 --> 01:03:13.282
hit maskov, you might know. ~ I mean it is a Neuralink collaboration, Yeah. Mm-hmm. Exactly.
01:03:13.802 --> 01:03:18.766
so Yeah. Yeah, yeah, yeah. I mean the the the chips part for sure.
01:03:18.766 --> 01:03:20.016
Yeah, chips in the brain.
01:03:19.905 --> 01:03:22.766
Yeah, yeah, yeah, I mean, it's it's happening.
01:03:22.766 --> 01:03:33.396
And okay, so I guess, is there anything that we've not spoken about anything you wanted to speak about a bit more? Or have we covered everything that you wanted to discuss?
01:03:34.753 --> 01:03:39.505
~ sorry, you asking me? Sorry, is that? ~ yeah.
01:03:38.449 --> 01:03:41.422
Yeah, yeah, asking you, asking you anything, yeah.
01:03:41.246 --> 01:03:53.791
~ I mean that's all that ~ I mean that's been a very rich discussion, really enjoyable. ~ I feel we talked a lot about the field and ~ you know where things are going and I think we've covered lots of ground.
01:03:53.791 --> 01:03:57.012
So I'm ~ I'm I'm I'm happy with what we've discussed.
01:03:57.012 --> 01:03:59.373
I I don't have any specific burning questions.
01:03:59.373 --> 01:04:04.918
~ just really to thank you for ~ Having me on here, it's been a pleasure.
01:04:05.842 --> 01:04:08.650
No, it's been incredibly interesting.
01:04:08.650 --> 01:04:11.981
Thank you so much for giving us all this knowledge.
01:04:13.208 --> 01:04:14.250
No, thank you, thank you.
01:04:14.380 --> 01:04:16.972
I guess the one thing to ask is then do you have a hot take for us?
01:04:17.105 --> 01:04:19.016
Yeah, yeah, yeah. ~ for for sure.
01:04:19.016 --> 01:04:29.840
I mean it's something I think a lot of a lot of our field has~ not wouldn't necessarily think that a mobile game can be a better scientific instrument than an FMRI scanner.
01:04:29.840 --> 01:04:45.047
~ intuitively you might think that you know na navigation is one thing that you can't necessarily study whilst lying still and and the most precious tool in the field is requires you to lie still with your head clamped.
01:04:45.804 --> 01:04:47.675
Usually with about 30 participants.
01:04:47.675 --> 01:04:51.797
~ and obviously prestige finding a publication or track scanner work.
01:04:51.797 --> 01:04:57.340
Data collection in the wild gets dismissed as confounded, and the critics have a real point without experimental control.
01:04:57.340 --> 01:05:00.061
You have to work much harder to rule out other explanations.
01:05:00.061 --> 01:05:02.612
And control kind of cuts both ways.
01:05:02.612 --> 01:05:07.344
So a task that's so constrained that behavior can't vary naturally is measuring the lab, not life.
01:05:07.344 --> 01:05:13.487
So I think my study of mobile navigation, like we could detect, you know, game experience, but driving sleep.
01:05:14.259 --> 01:05:19.862
I mean these things got picked up by new scientists, and the same platform now fl flags early cognitive decline risk.
01:05:19.862 --> 01:05:32.641
So I think ~ it's whether, you know, you'll I I think we need to understand that there has to be a real of world route where this research is going.
01:05:32.641 --> 01:05:35.293
And we defin we need that to see both sides of the coin.
01:05:35.293 --> 01:05:42.829
We need to, regardless of it being messier, you know, data being more confound to other factors, you know.
01:05:42.833 --> 01:06:20.737
We need to understand that sort of coordinative approach, how we relate socially to other people, how you know, looking at brain coordination, brain synchrony, a lot of a lot of these all the most of the AI tools and things that exist out there r rely on one individual and don't necessarily track how we relate to other people, other individuals. And I think that's actually where the exciting sort of new avenues are in terms of looking at~ how we relate socially versus individually and could give us the most informative information about sort of cognitive So I think ~ yeah, I think it comes down to two things.
01:06:20.737 --> 01:06:33.047
A mobile app beats kind of FMRI scanner in many ways, and that ~ sort of two person social or two or more person social understood coordination ~ beats the individual.
01:06:33.047 --> 01:06:45.784
~ and I think those are things that many people would maybe disagree with and think ~ because all the reasons I've said and ~ But I think that that that's the way things are going.
01:06:45.784 --> 01:06:49.255
So I that that's my yeah, that's my hot take from my experience.
01:06:49.382 --> 01:06:50.382
Brilliant.
01:06:51.043 --> 01:06:53.438
And also I agree with you. And Yeah.
01:06:53.625 --> 01:07:04.614
I did a pilot study with some really, really good data, but I couldn't get very much because I wanted, yeah, it's interesting data, which is the two people I'd say is like the novelty and the nice.
01:07:04.614 --> 01:07:09.358
Yeah, it's either you want novelty or it's gonna be interesting or you want lots of numbers, don't you? And.
01:07:09.492 --> 01:07:11.061
Of course, yeah, exactly, exactly.
01:07:09.925 --> 01:07:10.925
Yay.
01:07:11.505 --> 01:07:12.505
Yeah.
01:07:11.893 --> 01:07:16.170
Awesome. I guess on that note, thanks so much, Ember, for coming on.
01:07:17.365 --> 01:07:19.447
Of course. No, thank you so much.
01:07:19.447 --> 01:07:25.063
It's been it's been great. Definitely.
01:07:22.275 --> 01:07:23.275
enjoyable.
01:07:24.184 --> 01:07:29.971
brilliant. ~ and thank you everybody for listening and see you all in the next episode.
01:07:29.971 --> 01:07:30.971
Take care.
01:07:31.608 --> 01:07:34.369
Perfect. Bye. Thank you. See ya.