Episode

Eric Trexler, welcome to Learning Bayesian Statistics.

Thanks.

Yeah.

Great to be here.

Yeah, thank you for taking the time.

I am really happy to have you on the show for a lot of reasons.

The main one is that you work on really interesting topics, at least to me.

I've been nerding out on your content for the last few months, and I really love it because it's basically scientific breakdown.

of the sports and nutrition sciences literature.

So I am pretty sure that my listeners will at least appreciate the very nerdy sciences side.

This is actually the very first podcast I'm doing about fitness topic, which I discovered as I dove into the literature.

It's actually a very nerdy topic, so I'm really happy about it.

And so I would encourage.

all of my listeners to actually pay attention to it a bit more if they are not, because it's extremely interesting.

And it's also good for your health and body.

So several birds with just one stone.

So let's start.

Lots of questions for you.

But as usual, let's start with your origin story, Eric.

So yeah, basically, how did you come to the world of?

sports and nutrition sciences and what was the egg and the chicken?

Did sports come first or did science come first in your interests?

Yeah.

I mean, so like a lot of people in my field, uh, the sport came first.

And then at some point you have to figure out what you're going to do with your life.

And then the science becomes a little bit more pertinent, uh, unless you're going to become a professional athlete.

Uh, which not, not particularly likely based on the, uh, based on the percentages there.

So I started out, um, really enjoying sports, uh, mostly football, baseball and wrestling.

So.

Mm-hmm.

are all in different seasons, so it'd keep me busy all year.

Football got me really interested in wanting to get bigger and stronger.

And then later I started wrestling and had to do weight manipulation.

I had to lose weight, stay lean, try to maximize not just total strength, but strength relative to body mass.

So I started different types of, you know, strategies to, to lose body fat while keeping muscle and then, you know, weight.

You know, water related strategies, kind of adjusting hydration based on when weigh-ins are and, you know, trying to dehydrate for weigh-ins safely and rehydrate when I needed to

really be at my best.

So yeah, I just kind of became fascinated with these topics and really tried to be very analytical about what I was doing.

Even I mean, I was obsessed with fitness by 12 and by 16 was starting to kind of dig into the nuances of nutrition.

And so my career trajectory really kind of took off from there.

Yeah, that's extremely early.

Were your parents already into fitness topics like that?

Or did you meet someone that was really into that and introduced you?

Or was it just you at 12 starting to look into all that?

Yeah, I mean, it was when I was 12, I just love sports.

It was not something that was really emphasized by my family.

My parents were not like fitness enthusiasts by any means.

So for me, it was just, I just wanted to be better at football, which I loved.

And I was the only person in my family that really cared that much about football.

Um, when I was wrestling, I will say there was a particular coach.

who was a really avid fitness enthusiast and we would train together and he would kind of give me, he was a subscriber to Muscle and Fitness Magazine and when he was done with an

issue, he'd say, here you go, you can have this one.

And that kind of, the way that we approached training, it was the first time, for four years, from 12 to 16, I was just really doing it alone.

Didn't even have a training partner really.

but probably around age 16 when we started training together.

I feel like then he kind of really took a passion for fitness that I had and really developed it into something more where it was not just passionate about fitness, but

passionate about digging beneath the surface and exploring different strategies.

It kind of planted that seed.

It wasn't rigorous science that he was giving me, but it was at least that fundamental concept.

of exploring, hypothesizing, testing, reassessing that kind of iterative process that I think planted the seed.

Mm-hmm.

Yeah.

Um, and I, I found that also fascinating when I started digging into that, um, so yeah, like one of the, one of the things I thought about when you said as, um, like, like a lot

of people in your field, sports, uh, came before science.

And I finally have the directed acyclic graph now, because I was always wondering, you know, like, of course, science doesn't have as a general.

You know, in the general population, when you say you're a scientist, people don't think you're extremely fit.

Like they mostly think you're very like kind of nerdy and stuff like that.

So, but then you look at the world of sports science is really, you know, on its own because people are extremely fit.

They do a lot of sports and they are scientists, which if you go to other fields, I can confirm it's not the same.

So it was always like.

But I'm wondering what the causality here, like the direction of the causality is.

You know, is that because these people applied those strategies, then they became very fit or were they fit already, and then they became interested in the science.

So.

Yeah, I will say though, it can go both directions in the sense that more conventional team sports got me into embracing a more scientific approach.

But then as I pushed further in that, it created an interest in bodybuilding for me.

So sports like football and wrestling brought me to science and then science and physiology brought me to bodybuilding because I was so fascinated by

how physiology changes under these extremely unusual constraints that we see with high levels of muscularity, low levels of body fat, severe caloric deprivation, things like

that.

Mm-hmm.

Yeah, yeah, yeah.

Definitely the application of the scientific principles then enhances also your training and makes it more efficient, in a way.

I mean, most of the time, that means you're not losing too much time at the gym, and you're not leaving gains on the table, which I guess everybody wants.

But not everybody thinks about it in a scientific manner, where it's like,

I just do the stuff, and then I grind at the gym.

But actually, I'm losing time and gains on the table.

But here, that's also the thing I really love, is applying these little experiments on myself, and then seeing the science replicating itself is always, for me, incredible.

I forgot what I wanted to ask next, but that's really cool.

I love your...

path and passion, basically.

Like you always, in all of your content, we can refill your passion for your topics.

And so I'm wondering basically from what you're saying, I guess what the answer is, but would you say your path was very random or was somewhat clear to where you are now?

And basically I'm asking you if...

Your career is replicable, Eric, which is what we all care about in this podcast.

I would say that it is broadly replicable, you know, in the sense of, you know, from starting point to outcome.

I don't think that there's anything particularly special or unique that I have accomplished in my career trajectory.

But I will say that zooming in a more granular way and looking not just at, you know, point A and point Z, but looking at the intermediate steps in between, that was all

random.

Uh, it would not be, I don't think you could replicate it.

I, there's just a bunch of anyone who looks back at, you know, how they, uh, went through life.

It's a bunch of random, random choices.

Most of the best mistake, the best decisions that I've made have been, uh, decisions that I made for the worst reasons possible.

Right.

And so like, I can look back and say, wow, what a great decision and give myself credit for the outcome.

But I know that the, the reasoning for it.

was completely dissociated from, you know, why it turned out to be a good decision.

So yeah, it's life's weird like that.

You know, you, you think, you know, when you're making an important decision, you rarely do.

You know, when you think something is a really important decision, it usually isn't, and then the mundane stuff turns out to be the most pivotable, most pivotal, uh, turning

points in your entire life.

So it's been weird.

It's been random, uh, wouldn't change it.

Like where I'm at.

Um,

And yeah, I mean, I think the biggest thing with the path, zooming out a lot and just saying, you know, you like fitness and you like science, of course this has been the path.

It's easy to do that in retrospect.

And so in the broadest sense, I'd say, yeah, totally predictable.

I'm gonna do something involving fitness and science, but I've been surprised every step of the way in terms of what that has looked like.

Yeah.

And I mean, perfect.

This is a show about uncertainty and randomness and how to estimate it, but also how to embrace it to better live with it.

So, you know, that's perfect.

We love random path here.

And mine is definitely very random.

So, you know, I definitely can relate to that.

So now let's turn a bit to what you're doing today.

And basically, first, a general perspective is that now you're part of Herman Ponzer's lab at Duke University.

Some listeners may have heard of Herman Ponzer's because he wrote a pretty well-known book called Burn, which I put in the show notes.

I haven't read it yet, but it's on my reading list.

I've been introduced to it.

Thanks to your writings mainly Eric, and especially the metabolic adaptation manual that is in the show notes also.

And of course, we're going to talk about metabolic adaptation because it's one of your babies.

But first, more generally, can you tell us what you're studying at Duke University?

And yeah, what's your job nowadays?

Yeah, it's really crazy to, like I said, didn't predict exactly where I'd end up professionally.

It's crazy because over the last several years, one of the things I do is the mass research review.

It's every month we review current research in exercise and nutrition.

We get pretty nerdy with some of the stats.

Depending on your experience level, you might say, wow, this is pretty thorough.

But if you're like a bona fide stats expert, you'd say, okay, this is...

adequate, you know, but, um, but we, we try to make the, the fitness and nutrition information very practical, very applicable.

And sometimes, you know, we'll see a meta analysis that everyone's talking about and say, no, we need to re-crunch the numbers and we'll actually, you know, we'll do it, uh,

begrudgingly, cause it's a lot of work sometimes, but, um, anyway, so in mass, I had been reviewing Herman's work.

Um, I had reviewed several of his papers in recent years and I, I just kept saying, man.

this guy in this lab is doing some of the most fascinating research that's happening in our fields.

And it happens to be right down the road from me.

Like I was living, yeah, I was living like 20 minutes from his lab and had never met him, had never talked to him.

And I was just like, you know, this is the most fascinating research that I'm seeing in the last five years or so.

And so a position opened up in his lab and I said, let's do it.

You know, let's go for it.

And so that's where I am now.

Um, it's great working with Herman and, um, just getting started there actually.

But we have some cool projects going on.

I mean, Herman, he's, he's really known for his work on the constrained energy expenditure model, which, uh, is kind of the focus of the book, burn.

Which basically indicates that there's kind of a surprising, uh, what appears to be a ceiling effect on total energy expenditure such that.

when people are doing a ton of exercise and thinking, man, I'm burning all these extra calories, they are increasing their total energy expenditure in the vast majority of

cases, but a lot of people underestimate the compensatory mechanisms by which, as you increase exercise energy expenditure, your body tends to do some very subtle things to

reduce your basal or resting energy expenditure to kind of offset some of that.

energetic cost.

It's kind of like if you increase expenditures in one area of your budget, it would be a nice conservative choice to reduce expenditures elsewhere to kind of make the budget work.

So that's what we're, that's kind of like the main primary focus of the lab is kind of testing out these different constraints on energy expenditure and looking at what's really

going on during various models of increased...

energy costs.

So looking at fairly intense exercise, looking at things like pregnancy, which are very energy intensive and looking at the combination of the two.

So that's kind of one consistent line of research is looking at those constraints on energy expenditure across many different species.

So we don't, in our lab, we don't only do human research.

You know, there's some animal research in the mix as well.

So that's one thing that's going on.

currently is mostly in some of the applied ramifications of energy expenditure regulations.

So I'm working on a project now where we're looking at folks in a fairly high stress occupation and we're doing a very comprehensive study to explore a variety of outcomes

that are remarkably wide ranging and comprehensive in nature.

So physical activity and energy expenditure are a component of the research.

but we're also looking at relationships with stress and diet.

And, um, you know, we're looking at a lot of psychological outcomes on top of our physiological outcomes to see, you know, when we look at a total picture of someone in a

stressful occupation, and we look at their sleep, their physical activity, nutrition, uh, their stress management, how do these affect the broader picture of overall health and

wellness?

Um, so it's fat.

It's really fantastic because I,

have always loved energy expenditure regulation, but one of the things that I think erroneously talked me out of diving straight into research after my PhD, I've been doing a

lot of research in my PhD, took four or five years, did more private business related things, and now I'm back in the research game.

One of the things that kind of dissuaded me from starting my own lab right away,

was I felt like I was perhaps too much of a generalist to really sink my teeth into a very narrow line of research.

I kind of looked at where are my passions?

What is my training background looking like?

How will I make this all work in a cohesive line of research that is narrow and focused enough to make incremental progress in the field?

And I was really coming up short to be totally candid.

And I said, well, if you can't...

really paint that picture, it's probably best to not go and start a lab and kind of try to figure it out at some point along the way.

After taking a few years and embracing the idea of being a generalist and having multifaceted interests, I feel totally reinvigorated as a scientist and as a researcher.

And this is a perfect project to allow me to integrate all those general

generalist things that I've been kind of exploring over the last several years.

I've been writing about not just physical activity and energy expenditure, but nutrition and sleep and stress.

To be able to wrap those all up into a big project is tremendously lucky.

Again, just totally random, but really fortunate.

Mm-hmm.

Yeah.

And so thanks for this introduction.

We're going to dive into these topics, but yeah, like something I learned also by reading your content is how broad basically is the whole field.

And it's not just looking at exercise basically in training, but it's also

about sleep, about nutrition.

Um, something I discovered myself is that nutrition is extremely important.

Uh, and, and not that much talked about in, in the end, like people really emphasize the training, but, um, if you forget nutrition, it can really, really hinder you or, uh, on,

on the contrary, really help you.

So that's definitely something I learned, like basically having that holistic approach to, um, to the.

to the training and not only the exercise part for sure.

And to relate basically what you said about the constraint energy expenditure model, which I found fascinating also.

Basically, it would be for my listeners, if you folks have worked with, for instance, a multinomial model, that could be like that.

Multinomial, you have to use a softmax link function and

the Toft Max makes it a zero-sum game in a way.

So if one category increases, then one other category has to decrease.

I'm not sure the constraint energy expenditure model is exactly a zero-sum game in the sense that would mean you cannot make some games after some point, which if I understood

correctly, it's not really the case.

So you could say, I don't know if it's a one-sum game.

So you still make gains, but very small.

because the body basically has all these checks and balances to basically prevent you from starvation, which in a way, it's pretty cool.

But I found really interesting because nowadays in our current environment, we don't really face that risk a lot, at least in our societies and what basically made us to

the...

One of the things that help us go to the top of the food chain is today something that really prevents us from making the most muscle gains that we could.

I find that kind of ironic.

I'm sure Darwin would appreciate it.

Yeah, and it's funny that you mentioned Darwin because the department that I'm in now is evolutionary anthropology.

So when we explore these topics of innate inherent physiological constraints, a lot of times we do come back to what is the evolutionary purpose?

You know, in what way would this make sense within an evolutionary framework?

And when it comes to regulation of body weight and, you know, by extension, energy balance, you know, body weight and body composition are basically a long-term expression

of energy balance over time.

Um, it's really fascinating because the two major constraints, we have constraints on both boundaries.

One is starvation and the other is, uh, thought to be predation, you know, uh, becoming prey of a hunting animal.

And it's really fascinating to see how those two, I think they're really elegantly described in the dual intervention point model by John Speakman, who does a lot of

research on energy expenditure regulation.

And it's really fascinating to look at that and to say, you know, we're now so far removed, fortunately, in a lot of developed areas and areas with, you know, a surplus of

resources.

It's not true everywhere in the world, but in many places, we are very much removed from these two important constraints that really dictate everything that we're looking at in

terms of energy expenditure and body composition.

Yeah, yeah, yeah.

I find that super interesting.

That also makes me wonder, if homo sapiens manage somehow to not destroy the planet and stick around for long enough, would evolution catch up at some point?

And we wouldn't see these constrained energy models.

But then,

who would live in a world where maybe that would be way easier to put on some muscle gains because muscle is not going to be such a drag on your probabilities of surviving.

Do you think about those kind of stuff?

Or am I the only nerdy guy here thinking about that?

I think largely in the field of evolutionary anthropology, we look at the timeline of evolutionary history and I think it kind of creates this perspective of perpetually

looking backward, because you say if we divide it up into the kind of modern applications and then all of the data before it, you look at it and you say, well, the vast majority of

what we can learn from this is all from the past.

It's kind of making sense of the present by looking through the prism of the past.

So I must admit as a clear blind spot, I don't spend a lot of time necessarily looking forward far enough to incorporate changes in evolutionary constraints.

Because you're talking many, many generations.

And like you said, I think before we get there, we have some...

very pressing applied questions to answer within the next two or three generations, before we start thinking about that stuff.

So it is fascinating though, on times when you're not quite as busy getting projects over deadlines to maybe ponder that and take some time to think through it.

Yeah, I mean, that would definitely make for a science fiction novel.

I would read, you know, like a world where these kinds of constraints have shifted.

I don't know.

I imagine me being a scientist at that time and, you know, having the constraint model in mind and I like, basically we have figured out more or less how to do those things and

then you find some specimen from Homo sapiens, which who do not have those constraints and then they can just...

build some so much muscle because they have adapted.

That would be a fantastic side fiction novel.

If anybody wants to write it, please contact me.

We'll get you on the show right now.

Yeah, but it even is like, it's really fascinating.

We don't have to go that far to entertain some of these really exciting and thought-provoking questions pertaining to evolution and the challenges we face as humans,

because you think about something like space travel, which theoretically could become much more relevant in the next several generations.

Maybe, I don't know.

I'm not an expert on that, but looking at physiological responses to

uh, to microgravity, uh, it becomes so abundantly clear that this human machine really assumes that you're going to have a particular amount of gravity and it affects a lot of

different systems in the body.

And once you remove gravity, I mean, when people come back from space, their bodies are tremendously different and there is a significant recovery period.

And some things I've seen some recent research indicating that some changes in the brain, uh,

take a very, very long time to restore back to kind of baseline characteristics.

So very, very fascinating stuff that, yeah, we can only really answer those questions by looking back and saying, well, how did we get here in the first place?

Why do our systems function the way that they do?

Which is, I think that's my favorite thing about transitioning from being in exercise science departments to being in evolutionary anthropology now is that

The questions are so much, they're the things that kind of inspire so much imagination and creativity.

I mean, you still have to then dig in and do the less imaginative, more robust, you know, actual science.

But just that kind of initial hypothesis generating conversation, the questions are so big.

And so, yeah, they just kind of make you smile when you think about the possibility and

and the overall scope of some of the questions that we have to answer.

Yeah, yeah, no, for sure.

I mean, I really love that field also for that, evolutionary anthropology.

And I think only physics also bring those kind of questions to mind for me also, where it's like, some might read something about evolutionary anthropology or physics, and like

at some point, I'm gonna end up asking myself some very existential philosophical questions, which is super interesting.

I don't know if you've been reading up on it, but physics is really heating up right now.

There's some muon research that people are getting really, really excited about and they're thinking some chapters of textbooks might need to be rewritten relatively soon,

which is not too often that you get to even entertain those kinds of ideas.

Yeah, yeah, for sure.

And so I think it's a perfect, thank you, Eric.

And I swear people, we didn't prepare that, but that's a perfect plug because I told listeners that there was a very special episode coming in and I didn't say what it was

about.

But this summer I was lucky to be invited to CERN to the Collider and I filmed over there.

It's gonna be the first video documentary episode we're gonna have.

here, and we were actually lucky to see the control room of the Atlas experiment at CERN, which is one of the experiments which is looking to muons, as you were talking about,

Eric.

So yeah, it's definitely fascinating.

And I'm working on editing that episode, folks.

It's going to take a lot of time because I'm not a video editor.

So it takes me a lot of time to do it, but it's really fun.

And I hope you'll enjoy it.

It's going to be a very long episode.

I think the longest of the whole podcast, but super fascinating.

Thanks a lot for.

Yeah.

getting back, that's awesome, but getting back to the randomness of careers, this, for years I would go on podcasts and I'd be one person in this kind of rotating cycle of like

fitness professionals.

This year, you know, I've been on two podcasts where one person says, hey, by the way, next week we've got Noam Chomsky coming on.

And I was like, what, what do you mean?

How did this happen?

Why am I on the same podcast as, you know, like, how is, how is there any overlap whatsoever?

And now, you know, hey, by the way, we're going to, you know, show you Inside CERN in a few weeks, like, or a few months.

I don't want to rush your video editing, but yeah, just weird.

Just weird.

Exciting, though.

Yeah, I mean, common denominator here, Bayesian stats.

Everybody needs them and use them.

But yeah, no, thanks for this amazing plug, Eric.

It was perfect.

And let's go back to you now and talk about metabolic adaptation.

So we've talked a bit about it now.

You've written a very extensive guide that I'll put in the show notes.

So maybe can you tell us a bit about what metabolic adaptation is and why that's important?

Absolutely.

Yeah.

So dieting is hard.

If you ask someone who's trying to lose weight and lose body fat, it is really difficult and there are many, many reasons for that.

But one area of research is very interested in kind of unraveling what those specific challenges are and why they arise.

So kind of uncovering some of these things that make dieting a little bit more challenging than we think it ought to be.

Because on paper, it's very simple, right?

You know, when there's too much energy, when there's a surplus of energy, you produce the intake or you increase the expenditure and there you go.

Now you're back in, you've got energy balance where you want it to be.

So it should be conceptually quite simple, but in application, it's tremendously challenging.

One of the many challenges relates to metabolic adaptation, which is that when people start to...

lose weight via caloric restriction, we tend to see that basically their energy expenditure goes down more than we would predict.

So what I mean by that is if a person, let's say a person is reducing their energy intake so that they can lose 30 pounds, we would expect that a person who weighs 30 fewer pounds

would burn fewer calories throughout the day.

broadly speaking, everything else staying equal, smaller bodies ought to burn fewer calories on a daily basis, but we should be able to generally predict what the magnitude

of that change should be.

What we see in practice is that the reduction in energy expenditure during dieting is considerably larger than we would mathematically predict based on changes in body mass

alone.

And what it essentially identifies is an adaptive mechanism.

where the body reduces to some extent, you know, resting or basal energy expenditure, but to a larger extent, non-exercise activity energy expenditure.

We're seeing that these are being reduced disproportionately in a way that basically applies a little bit of a break to the weight loss process.

It just adds some friction that makes it more challenging.

And so of course, this is important to understand.

I don't want to frame it as the most important constraint on weight loss because I don't believe that to be true, but it is an important element of understanding why weight loss

is so challenging, which is of critical public health importance at a time where rates of overweight and obesity are higher than they've ever been in a historical perspective.

As we see that, they are quite high in contributing to...

health challenges and even population level burdens with regards to keeping sustainable healthcare systems working effectively.

It's very important to understand these challenges that are associated with weight loss so that we can become more effective in managing obesity and the health-related implications

of obesity.

Yeah, exactly.

And to make that even more concrete, so that means that basically as you lose weight, you're going to decrease your energy expenditure.

I mean, your body is going to decrease its energy expenditure, as you were saying, this could be simple physics, lower volume, so lower energy expenditure.

But you're noticing with that model that basically people lose

the people's energy expenditure decreases more than what's predicted.

And so concretely, that means if you still want to lose weight after that inflection point, you have to reduce even more the caloric intake because otherwise you're not going

to lose weight anymore.

Is that what that means concretely for people and what makes it even more challenging to lose weight?

Yes, it means that the dieting process will become more restrictive than it otherwise would be.

And what's really to make matters worse, so we lose weight by creating an energy deficit, right?

So we're consuming fewer calories than we're burning on a daily basis.

So we introduce this deficit and being in a deficit alone causes a reduction in energy expenditure.

Our body can kind of tell with that short term shortfall of energy intake, oh, let's kind of slow down some expenditure, power down some processes that aren't totally essential.

So being in a deficit causes a reduction in energy expenditure.

And let's say that works and now you're losing weight.

Now you are a smaller person, which reduces energy expenditure.

And then you add on top of those, this adaptive reduction in response to decreases in fat mass, and now you have three separate factors that are contributing to this reduction in

energy expenditure, which means instead of dieting on, you know, perhaps what might feel like a comfortable diet of 1900 calories per day, you're pushing considerably lower than

that into caloric ranges that are much lower than you're comfortable with.

And now you're, you know,

ability to get through the day with a suitable energy level is threatened.

Your hunger is considerably increased.

It creates a lot of challenges in that regard.

Yeah, yeah, yeah.

And this is really fascinating to read about.

Also to live is quite interesting.

I've been myself on a cut lately, and that has been definitely at some point at the end of the cut was very challenging, these kinds of things where you're like, you have days where

you like literally think about food all the time and you're like, you're like trying to.

You know, optimize the macros and so on and like trying to find, I was literally food hunting in some of the shops here.

I was like trying to find the best foods, like the most protein, the less fat or the less carbs.

Which is kind of fun, but yeah.

If you do it in a controlled way, that's interesting.

But otherwise that definitely puts a challenge on, on top of already a challenging situation.

And what I found also really interesting is that.

metabolic adaptation.

I don't know if you still collect metabolic adaptation in that case, but that also kicks in these mechanisms that make sure your energy expenditure basically doesn't go too low

and doesn't go too high.

These they kick in also if you're in a caloric surplus.

So if you want to gain weight because you want to put on some muscles, for instance, so that would be the case for athletes, or for a lot of people in the general population,

These mechanisms also kick in, right?

Yeah, yeah.

And so when it comes to metabolic adaptation, I'm glad you frame the question that way because there is considerable debate or heterogeneity, whichever way you view it, in terms

of the terminology.

Some people talk about adaptive thermogenesis, which is specifically the reduction in energy expenditure that we've talked about up to this point.

But when I talk about metabolic adaptation, some people use that synonymously with adaptive thermogenesis.

It's a very narrow scope of focus that looks just at energy expenditure fluctuations in response to weight loss, or conversely, perhaps weight gain.

When I write about metabolic adaptation, I like to take a more holistic approach, and I talk about not just energy expenditure, but the entire milieu of endocrine and even

neuroendocrine changes.

that affect many body systems.

So of course, energy expenditure goes down.

That alone, honestly, is not that big of a deal.

It's not helpful, but it's not catastrophic by any means.

The larger changes that we see typically are just being in a deficit reduces expenditure and being smaller.

So if it were only an energy expenditure problem, I don't think it would be so important.

It matter, but to a lesser degree.

But with metabolic adaptation, the way I like to talk about it holistically, we see reductions in sex hormones as a response of this.

I mean, the hypothalamus is coordinating this widespread set of consequences across many different body systems.

So reductions in sex hormones that can impact potentially body composition, potentially the reproductive system.

These are things that impact quality of life.

And then when we look at the neurophysiological changes related to hunger and appetite regulation, I think those are the most important adaptations that we see during weight

loss in terms of just recalibrating our hunger and desire to eat and the types of food choices that we're inclined to make.

So you could set, you know...

cut those up into individual components and study them all separately, but I don't think it makes sense to do so, which is why when I write about metabolic adaptation, I think a

very fair criticism is some people might say, you've gone beyond the scope of metabolic adaptation.

And my response would be, you have too narrow a scope for metabolic adaptation because you really cannot uncouple these things from one another.

So yeah, I definitely want to acknowledge that,

metabolic adaptation the way I view it, not only is it very wide ranging across body systems, but it also does apply, like you were saying, in two directions.

So we see adaptations to underfeeding and weight loss, but we also tend to see adaptations to overfeeding and weight gain.

So there's fascinating experiments where we bring people in, not I didn't do it, but when scientists bring people in.

overfeed them intentionally to a considerable degree.

And there are tremendous adaptation, adaptive mechanisms that lead to dramatic increases in energy expenditure to maintain a relatively stable body weight.

And what we see is that appetite goes through, I mean, just to the floor, just no appetite whatsoever when people are dramatically overfed intentionally.

But

Hmm.

The really fascinating thing, the unfortunate thing, because I'm sure a lot of people are listening saying, no, your science is wrong.

The variation between individuals is immense.

And so they're even in these studies that are tightly controlled, we'll see some people defend a body weight quite effectively.

Other folks, you know, their body weight just changes quite readily, right?

So there are a lot of folks who say,

When I overfeed, I don't increase energy expenditure.

My appetite doesn't drop.

I just gain weight.

And so that observation is not at odds with the science.

It's very much compatible with it.

But what we do find is, you know, there is a lot of variation in people's adaptive mechanisms in both directions.

So there are some folks who, when they do a weight loss diet, they just lose the weight and it goes fine.

And other people will look at them and say, how is this possible?

Like, you feel like you're from a different planet.

when you look at the difference in responses.

And that I think is another reason that it's so important to understand these, um, you know, metabolic adaptation in general is.

I think first of all, there's the practical benefit of understanding what it is and what to do about it to facilitate, uh, folks who are trying to gain or lose weight, uh, to, to

find a healthier body weight range for them.

But also I think it, I think if you.

interact and engage with this literature seriously and you do so with a very critical eye You're almost forced to kind of decouple obesity and weight gain from Some of these

inaccurate and counterproductive assumptions that are really common in health care and really common in society you know a lot of people will look at

obesity or weight gain and inherently mentally connect them to essentially character flaws like gluttony or laziness.

And I think if you look at this literature and you do so critically with an open mind, what you tend to see is there are so many factors impacting body weight regulation that

appear to be innate.

These inter-individual differences are immense.

And what you start to see is as someone who has...

generally in my life had success gaining and losing body weight very readily.

When I was younger, I used to assume that there was, you know, oh, I'm so much tougher.

I can push through these things that other people can't push you.

That's not the case.

As I've looked at the literature more, I find that I am just a better responder to some of these weight regulation interventions.

You know, I lose and gain weight intentionally.

while encountering far less friction than other folks do.

And so that was unfortunate.

At first I thought I was really special and just had this ability to push through things.

But I think metabolic adaptation, as we understand these barriers to weight loss that are very inconsistent from one person to another, it helps us look at things like obesity and

weight gain and weight management.

in a much more empathetic way, in a much more accurate way, you know, so I think that's really important.

Yeah, yeah, for sure.

And you can still be special in the sense that you have a special genetic component.

You chose your parents, right?

Well done, Eric.

So many questions.

So what you were talking about, basically, that there is a huge inter-subject viability.

Yeah, I'm not.

surprised about that.

It's something I've noticed also in that literature.

Hopefully, we'll have time to talk a bit more about that later on in the show from a more statistical point of view.

Face dance, whispering.

Something I'm wondering is, do you notice a correlation between the people who can gain weight easily and then lose weight easily?

And then you have basically you...

that you would call responders in this case, and then people who respond way less.

And so if you have trouble gaining weight, then also probably you will have trouble losing weight.

That would make sense to me, but is that something that you also see in the data?

Yeah, so what we tend to see if we're looking specifically at physiological responses, so like changes in energy expenditure that we can kind of measure with physiological tools,

if we look at that, then we can broadly define people or kind of categorize them into two major phenotypes.

You know, thrifty versus spendthrift are the terms that are used.

So there are some folks who if they're overfed.

You know, they, they gain weight really easily, but then when they try to lose weight, they run into a lot of friction, right?

So they're kind of nudged toward these, um, you know, if overeating and then trying to diet, you know, they would readily gain weight and then kind of struggle to lose it.

There are other folks who are, who are the inverse where, um, if they try to overfeed, it will be, uh, you know, we'll see a lot of those physiological.

adaptive mechanisms that kind of keep them from or resist their ability to gain weight readily.

But they don't seem to run into as much friction when dietary intake is reduced.

And what's really fascinating is you can kind of look at these over very short time scales.

You can just do in a crossover study, you can look at overfeeding and underfeeding or even fasting responses over the course of just 24 hours each.

And you can kind of broadly categorize people and say, okay, if you, you know, had a huge increase in, um, energy expenditure when we did overfeeding, we can already start to make

some inferences about, you know, the fact that, you know, a huge increase during overfeeding probably means that we're not going to see a lot of friction during

underfeeding versus, you know, folks who, you know, when, when they overfeed, they just gain fat mass, you know, the, uh, energy expenditure doesn't really move much.

Um,

You know, when they go into a fasting or underfeeding stage, we expect a pretty considerable amount of friction in the sense that energy expenditure will drop.

So there there's kind of.

It only seems fair in the context of bodybuilding, you know?

Um, so what I mean by that is bodybuilders will go through phases where they really want to gain weight.

And then when they really want to lose weight, you know, they want to build all their muscle, gain a bunch of weight, lose fat.

And so bodybuilders usually, you can kind of count on having one or the other.

You know, you can say, oh, I can bulk up really well, but I struggle when I'm cutting.

Or some people will say, you know, it's really challenging for me to gain weight and put on muscle, but when I want to lose fat, it usually goes pretty smoothly.

So in bodybuilding, it seems fair.

Outside of that world, I think most folks in the general population feel like it's very unfair and they want to be the person who can lose weight easily and struggle to gain

weight when intentionally overfeeding.

But I did mention the caveat that I was speaking really about physiological responses there, because I think one other important factor that I've not seen really studied all

together comprehensively very frequently is we've got that kind of regulation system of these physiological parameters.

But I also think that there is work to be done in the kind of neurophysiological regulation of appetite and desire to eat.

And I think part of that is

reward responses to feeding and those kind of hedonic responses to feeding.

And I think for me, one of the reasons that I'm able to kind of do weight gain and weight loss relatively easily is that, you know, I think I probably do experience, I'm more of

the thrifty phenotype, meaning that I can gain weight intentionally quite easily.

I do have reductions in energy expenditure, pretty considerable ones when I diet.

But I think the way that I respond to food from a neurophysiological perspective, I'm just not as into it as most people.

So I don't run into the really tremendously challenging appetite and desire to eat challenges when I'm dieting.

So my energy expenditure goes down plenty.

It's just not that big of a deal.

Because if I need to cut calories more, that doesn't feel based on my neurotransmitters basically.

like it's too great a sacrifice to make.

So there are these competing systems going on that make things really quite fascinating.

Yeah, yeah, for sure.

And I really love also that you're talking about the neurological aspect because that's something I've seen also is, yeah, first, one of the things like when you go on a fat loss

diet and then go on a caloric surplus, you kind of have to have to change your mental thinking about food.

It's really weird.

Whereas when you're losing fat, you have to be very...

intuitive about the way you eat and being very aware of the society cues that you have and that actually often you can stop eating before you do.

And being very aware of those stuff, eating very slowly and all those things like that where meditation helps and so on.

And then when you want to gain weight, you have to kind of throw all that out the window.

And basically eat even though you're not hungry.

Continue eating even though you're not hungry.

It's really a mental shift.

That's really weird.

And also you have to change your habits.

And it's something I've really dove into a lot because basically how to build better habits and how to pair basically these incredible power that the brain has, which is

creature of habits because these are shortcuts and it makes just like...

life easier for the brain.

Well, that can be a very bad thing if you have really bad habits, but then you can change those habits.

How do you change them for the better?

An extremely good book that I found about that, I've read a lot of things about that, but the main one I would recommend for someone who's digging into that would be the one from

Kathy Milkman, How to Change the Science of Going from Where You Are to Where You Wanna Go.

This is a really good one.

really good breakdown of the current literature.

And it helps you also pair, basically what we're talking about with, okay, how do I try to develop that into routines that help me?

So developing cues, for instance, me, I'm really a coffee nerd and addict.

So when I started going to the gym, well, trying to find a coffee shop that's on the way to the gym so that, and not having coffee at home, so that I have to get to the coffee

shop to get the coffee.

And while the gym is just there, well, might as well get some training.

And then after the gym, well, you get another coffee because it's a reward.

And then at some point, the brain does the stuff automatically.

This is really a fascinating part also of that research.

But I think that I think also we're just starting to explore.

Yeah, yeah.

And I love the behavioral components because having worked with a lot of clients who have weight loss goals, it's so fascinating to see this fine line where an intuitive assumption

is that you should make the most minor, tiniest, most feasible change as possible if you want to start reducing energy intake and losing weight.

And in many cases, that is true.

But in some cases,

individuals are so locked into habits and routines that it's almost harder to do the same routines and same kind of things and just make these little changes within that.

Sometimes it is a lot easier to just completely restructure a day and just to say, let's create a completely different series of habits, a different schedule, a different routine.

Let's start from scratch and very intentionally bundle some of those things like you mentioned, where it's like, okay, yeah, we're going to...

create this association, this coupling between gym time and coffee.

And I know you want your coffee, so you will get to the gym.

Um, and I do those same kinds of things right now.

Like I, my favorite thing to kind of just, um, it's the least productive thing I do.

It's anti-productive, but, uh, when I really just want to turn my brain off and hang out.

I love football.

I'll play like a football video game.

And my.

Video game system where I play my football game is right in front of my treadmill.

If I'm playing it, I'm walking.

If I'm walking, I'm doing something positive for my health.

Just finding those ways to couple those things because I noticed my step count was too low.

It's like, well, how do we make that work in a way that doesn't feel like a chore?

We couple walking with one of our favorite things to do that when we get a chance, we're going to do it.

Yeah.

And also that I find that helpful in the way that it also gets us out of the, most of the time, unhelpful mindset of, you know, no pain, no gain, that you have to just grind

through.

And if you're not, then that means you're weak, you have weak character in all those things we were talking about, you were talking about, before, about weak gain and obesity.

And the weirdest thing though, like one of the, I almost, if I wasn't going to join, uh, Dr.

Ponser's lab, I gave serious consideration to doing another PhD in the social sciences and, uh, you know, health psychology, because one of the weirdest things I find working

with clients, you know, cause all the theory goes out the window when you're actually out in the field working with people.

Mm-hmm.

most people, I fully agree, and based on the textbooks, I fully agree, you know, the no pain, no gain thing, forget about that, let's make things approachable, let's make them

feasible.

There is a certain percentage of the population where if they cannot link what they're doing to an internal narrative, that they are grinding, and they are doing the hard stuff.

Like, there's this entire narrative that...

makes them excited to do it because it feels like it sucks.

And I want to understand better what that is and how we can sort the people with what I would consider a more typical average response, which is let's make it not suck.

I think most people like that.

But I'm very intrigued by, you know, early in my coaching career, I would try to fit people into that mold.

And I found enough of these people that I said, we have to have at least two molds, where some of these people do want to be grinding and they like the fact that it is hard.

That's what excites them about it.

And sometimes they wanna make it seem, to be totally honest, even harder than it is, which is very fascinating to me.

Yeah, this is super interesting.

I mean, like, and in a way, also, you could, like, I'm sure there are some tactics to make them think they are grinding, even though it's just a bit more friction, but it's actually

no harm.

Like, it makes me think about, do you know about the IKEA effect in this literature, which is where basically the idea is

For instance, in France, you have these all-made, pre-made crepes that you can buy in the supermarket.

And when they started selling those, it didn't sell as well as they thought.

And then they did all their studies and so on.

And they understood that it was because everything was made and then people couldn't basically claim that it was their own crepes that they were doing.

And so they just removed the eggs.

And now when you buy that, you have to put the flour on all made and so on.

And yet you just put the eggs and you bake it like you whisk them.

And then cells went through the roof because basically like people feel that they are part of the process and that, you know, they, they belong in the process.

And so that's called the IKEA effect because like you go to IKEA, it's not done.

You have to put your own furniture together.

And so basically, even though it's a really small thing, then that makes a derance to the program, which is of tremendous importance in sports science, much higher and in the end

that makes the program more effective.

Yeah, I never heard it framed as the Ikea effect, but I heard that same kind of story about cake mixes in the United States where they said, okay, you got to put the egg in.

And that does make sense.

And sometimes, you know, I'll even take it a step further and have clients, you know, design a little part of their program themselves that I set enough constraints that

whether they pick, you know, red or blue.

it's going to be equally efficacious, right?

So there's nothing at stake here, but it does increase that, um, that feeling of ownership and kind of having that autonomy and self-efficacy to really make an imprint on what we're

doing.

Yeah.

Yeah, yeah, yeah.

And yeah, I'm loving all that discussion and like I want to dive into more stats, but I really love it.

So just to continue a bit on that and on the grinding part that you talked about, also to me that resonates with a lot of stoicism.

I'm really interested in stoic philosophy.

And so like you see a lot of that where basically the idea

One of the ideas and principles of stoic philosophy is sometimes make you a bit more uncomfortable than you need to be, because that way it makes you more resilient in a way,

and that shows you that you can do it basically if you have to.

For instance, when it's chilly outside, well, get out with one layer less than you would be comfortable.

And actually, that makes you a more stronger and resilient person.

And I really love that.

that part of the stoic philosophy.

I put a link to the one of the best books about that I've read, which is called the stoic challenge by William B Irvine, really, really good book.

I'll put that in the show notes.

And about the, like the pleasure that comes from pain makes me think about another book I read recently, which is called the sweet spot, the pleasures of suffering and the search

for meaning by Paul Bloom.

And yeah, it's a bit about what you talked about where basically for some people and for some endeavors, the suffering is part of the experience.

And if you're not suffering at least a bit, then for some people that is not worth it.

Yeah.

Yeah, really, really fascinating.

So before we dive a bit more into the statistics side of things, I'm wondering about, to close up on metabolic adaptation, what are the current frontier regarding that field?

Basically, what are the questions that you and the people who are studying that, what are the questions you really want to know the answer to these days?

I mean, there are a few, some of them are boring.

They're about methods.

So, unless you're researching in the area, you don't care.

But one of the things that's really challenging with this area is that when we predict changes in energy expenditure, we typically have to use fat-free mass as one of the main

predictors.

But when you look at fat-free mass, you have tissues with very different energy expenditure.

amounts, you know, if you look at the rate of energy expenditure in the kidney versus the liver versus the heart versus muscle tissue, you can't just say it's all the same per

kilogram.

And so that one area of research is how much organ mass is lost during dieting and is it possible that is explaining at least some of this reduction in energy expenditure that

seems disproportionate.

So that's an active area of discussion.

Some of the more practical questions, you know.

Is there anything that we can do feasibly to attenuate or mitigate these changes in terms of metabolic adaptation?

Do we essentially reach an equilibrium after we've maintained our weight loss for some amount of time?

And if so, how much?

So far we've got longitudinal studies up to six years, I think, of follow-up where we say, yeah, it doesn't really look like there's much changing here.

It looks like it's a fairly persistent thing.

So that's another open question that is always top of mind.

And then the one that interests me the most is how do we...

How do we put together a unified model that includes the constrained energy expenditure model and like exercise energy compensation and metabolic adaptation at the same time?

What I mean by that is we know that as people are dieting, restricting energy intake to lose weight, metabolic adaptation will reduce their energy expenditure.

But we also know that if people do extremely large amounts of exercise...

we often see an attenuation of resting metabolic rate.

And so the question is, most interventions will include energy restriction and extra exercise.

So how do we kind of parse those two things that are almost certainly happening simultaneously?

And one hunch that I have that I'd like to explore in upcoming years is, I personally think that

Exercise energy compensation is greatest when we're in a caloric deficit.

So when we are under eating relative to our energy expenditure.

So I expect that in programs that involve weight loss and dietary restriction, or I mean exercise and dietary restriction for weight loss, we are seeing not just an additive

effect of the two things, but perhaps even kind of an amplification of that effect because

exercise energy compensation in line with the constrained expenditure model, it does seem to vary based on the energy status of the individual, whether they're in neutral,

positive, or negative energy balance.

So that's a frontier that I think is really important to explore.

Yeah.

And that, that's a really fascinating one.

I found, um, like from a statistician's point of view, basically, like, like everything we're talking about since the beginning makes me think like I have stuff in my head, like

interaction effects all over the place, basically.

That's what you just talked about, like interaction between dietary restriction and, um, and then, um, the exercise compensation, um, and, and also like, I just have

you know, logistic and logit curves in my head and logarithmic curves, basically where it's a lot of nonlinear effects combined to interactions, which make everything much more

complicated, at least for a homo sapiens, normal homo sapiens brain.

I'm guessing if you put that in a model in the computer, that will make much more sense, of course, but that's why we're doing that, right?

But...

Yeah, and actually, energy compensation is something you also worked on.

And I really find that super interesting.

So to try and make that more concrete for people, that would mean that, and that makes weight loss even more complicated if I understand correctly.

Because that means that basically you're trying to lose weight and mainly lose fat.

So you're going to have a deficit, calorie deficit.

And on top of that, you're going to add exercise, most of the time cardio.

And let's say you go for a run and you expect to lose 100 calories from that run.

Basically what energy compensation coupled to calorie deficit means is that, well, at the beginning of the weight loss, maybe you're going to lose 100 calories, but then the bigger

the deficit, the bigger the energy compensation.

And so that means that let's say that your body compensates.

So maybe I'm missing the sign here, but if the body compensates 80% of your exercise, that means you're only gonna lose 20 calories from that exercise bout instead of losing 100,

which makes weight loss even more challenging.

And in a way that makes sense, right?

Because if your body thinks you're beginning to starve, well then it's starting to put all those barriers so that you don't die.

Right.

So yeah, had I summed up that thing well?

And yeah, basically, can you talk a bit about that energy compensation and how that relates to what you're studying?

Yeah, absolutely.

So energy compensation really is the key factor underlying the constrained total energy expenditure hypothesis, meaning right now our best estimate is that if all we know about

someone is just the bare, the most simplistic information, which is that I'm a person, I'm a human being, and I'm going to do an extra 100 calories worth of exercise per day.

Mm-hmm.

best estimate based on that minimal information would be that their total daily energy expenditure will only go up by about 70 calories per day instead of 100, because 30% of

that will be compensated for by reducing resting factors of metabolism, trying to offset some of that energy cost.

But we really need to do a lot more work to unravel.

what makes a person compensate more or less.

And so for example, if we look at the energy expenditure of competitive athletes who are intentionally eating a lot to try to replace their calories, we do see that their energy

expenditure is considerably higher than a sedentary person who's weight stable.

So it's not to say that we cannot change our total daily energy expenditure at all.

But there is certainly some degree of compensation that occurs and my suspicion is that it is largely, the magnitude of compensation is largely dictated, not exclusively, but

largely by energy intake.

And I expect that, you know, when we're in a situation, I mean, we're seeing, you know, if we look in a vacuum, let's say you're not exercising, you're just reducing calories to

lose weight.

You're gonna get extra friction from metabolic adaptation, right?

Yeah.

to be true.

And let's say in a vacuum looking elsewhere, you're not doing any energy restriction necessarily, but you're exercising.

Well, you know, there is gonna be probably some degree of compensation where you're, if the goal was to lose weight just by doing exercise, you're not gonna lose as much as you

thought because some of that is gonna be compensated for.

Now, when we put those two together,

In light of observations that compensation seems to be greatest when energy intake is, you know, when there's a caloric deficit or negative energy balance, we don't just see, I

would expect, I don't expect that we would see just an additive combined effect.

I think we might see something that's more kind of synergistic in a way that the two kind of amplify each other in the context of a weight loss program involving diet and exercise.

So all of that is to say, yeah, when you're doing a holistic weight loss program with diet and exercise, a lot of folks say, why does this feel so hard?

And the answer is there's a lot of friction to be encountered along the way.

Yeah.

And also it makes more sense that these kicks in more in a deficit, a caloric deficit than a caloric surplus, right?

Because then if you're in a caloric surplus, why would the body try to compensate for the extra, extra bout of exercise that you're enjoying?

Yeah, yeah, the one exception though, the one exception that I would say is, I mean, of course, from an evolutionary perspective, when we're looking at starvation as the limiting

factor, that all makes sense.

And we'd say, yeah, in a caloric deficit, you see a lot of compensation, otherwise maybe not so much, but they've done some really fascinating work in athletes who are just

spending tremendous amounts of energy on exercise.

And I think that over the long term,

there are some upper boundary constraints that are limited just by the amount of energy that we can feasibly extract from our diet on a daily basis.

So I think they were working with like really competitive cyclists doing tremendously arduous races and finding that like, you know, there are some constraints on those

absolute upper boundaries but those are constraints that most of us mortals are never gonna encounter, right?

Yeah.

yeah, yeah.

Yeah, it feels a bit like you only feel the effects of relativity when you go really fast.

Well, it feels a bit like what you're talking about for these kind of like, yeah, really, really hard endurance athletes.

Yeah.

I know we're running a bit long here.

Are you okay to continue a bit more or do you have a hard stop?

Yeah.

Okay.

Awesome.

Yeah.

Thanks.

Yeah, because I'm having too much fun.

So I want to continue.

good if it runs long and you're right now saving me from getting back into writing, so I'll talk forever.

Well, I'm happy to do it.

So yeah, let's turn a bit more to the statistical side of things here, because of course, you're using statistical models for all this work to make sense of it.

So I'm wondering what are the main modeling challenges that your field is facing?

Yeah.

You know, when I first got into the field, I was coaching a special Olympics powerlifting team and the head coach was, he's a dear friend of mine and he had been doing research for

decades before I was born.

I mean, he was probably about 70 years old when we met and he was a statistician and so he would work with a lot of different fields and

I remember expressing frustration to him about the way statistics were approached in my field, and he kind of calmed me down about it.

He said, listen, different fields grow up at different rates, and the more mature a field gets, the more rigorous its science gets, the more it starts to embrace more nuanced

statistical approaches, and people get better at it.

It comes down, it starts with the demand for rigor in...

know, the publication process, then it trickles down into the training that's received for folks that are coming up through the field.

And so I think my field right now is at a really cool inflection point in its growth where there are a lot of people who are really lifting the tide for statistical analyses in

exercise and sports science.

So that's good.

I'm happy to see that.

And I don't want to be a hypocrite.

It's not like I was, you know, a first year master's student and I had this like innate expertise in statistics and said, Oh, I'll do it all perfectly.

And everyone else is dumb.

I knew that it should be done better, but of course I wasn't capable of doing it.

And part of my frustration was I want to get trained in the good stuff and no one seems to know the good stuff around here.

You know what I mean?

Um, and who knows if I would have even picked it up if I was exposed to that training, you know, I'm not, I'm not, uh,

not claiming that expertise before developing it.

So there are many challenges in our field and we're encountering them more as we kind of go through this growth process, to be honest.

I think aside from the fact that our field has not prioritized statistic previously, so the training reflects that in a lot of programs, I think that's a big challenge.

It's just that people are not being exposed to...

really nuanced in-depth statistical training.

And so a field, we got to start doing more of that training in the master's and PhD programs.

But more to the point of probably what you were getting at, I think we have two main challenges in exercise and sports science.

Number one, sample size.

I think a lot of folks in other fields, they think we work with small samples because we're dumb.

and we don't get it.

It's like, why don't you do more?

And it's like, well, if you're doing these really, really resource intensive study protocols, you'll get it in both directions.

If you do exercise protocols that can be feasibly scaled to large samples, people will say that your measurements are so imprecise as to be worthless.

And they'll say, you can't even make the inferences you're trying to make because you use the cheap measurement.

that you could actually use 400, 500 participants in the study.

So you get a lot of pushback if you go the route of saying, well, let's do the less intensive measurement protocols that are more affordable and more feasible and get a

bigger sample.

On the other end, if you want to do the really nuanced measurements, the time cost, the labor cost, the financial cost, it starts to get so...

remarkably large that the idea of bringing in more than about 22 people, it starts to become completely infeasible.

And so of course, money would help.

Money tends to help a lot of things in the research world.

So if people wanted to dump all their money into these studies, then sure, we could take six years and run 300 people through the protocol.

But right now, a big constraint on our statistical development and one that I think is fair, like...

I could imagine the chair of a department who's been studying, doing research since I was not even born yet, 30 years before I was born, they could have been doing research.

They would look at me and say, all right, hotshot, you think you know it all, but go learn all your stats.

Why doesn't our department teach all this nuanced stuff?

Because you can't even do it.

Some of this, the really...

nuanced statistical methodology, especially in the frequentist world, you need such large sample sizes to even really observe the benefit of doing a lot of those things.

So sample size is a limiting factor that I think unfortunately is also holding us back from embracing some of these statistical approaches because people are saying, well, I

could train you on that, but what are you going to do with it?

Your biggest sample is 30 people.

So...

Yeah, that's definitely, I would say, a big challenge.

And then another challenge is a lot of the questions we explore have a lot of inter-individual variation or at least we expect them to based on biological and

physiological theory.

And it's really challenging for us to try to really parse out what is a genuine difference between individuals versus...

measurement error and all the other sources of error and randomness that find their way into real world data, especially for metrics that change day to day.

And I mean, if I measure some of these parameters, it's going to be different at 8 a.m.

than it is at 11 a.m.

for the same person on the same day.

So there's so much noise and trying to get through that and say, well,

What is, how do we attribute this noise to all these various components that can create noisiness in this dataset?

It can be really challenging.

Yeah, I mean, this is very interesting because this is what I've also noticed reading some of the literature that we're talking about.

And that's also why I thought it was super interesting to have you on the show.

Because unfortunately for now, I cannot really help with the money, but I can help with the statistics and I'm pretty sure what you...

Just talked about a lot of listeners have started thinking, patient stats, dude, patient stats.

Because it helps for low sample size, it's perfect for that.

And also the inter-subject viability, and we're going to talk a bit about that.

Teaser for listeners, there is a really cool paper we're going to talk about, I mean mention, from someone you know very well.

Yeah, so basically that would be my question.

So you told me you know a bit about Bayesian stats.

I mean, you've heard of it, but that it's still not much used for now in your field, am I correct?

It is starting to get used more.

And I feel that it's been interesting because I did my graduate research between 2013 and 2018.

And I actually expressed an interest in diving deep into Bayesian statistics.

And due to various circumstances, just didn't really get the opportunity to sink my teeth into it.

And one of the things that makes it challenging is

As we know as people who's published science, you know, there's a great responsibility You you can't just willy-nilly say I'm gonna try this new thing and I don't really know how to

check if I did it Right, but whatever I'll try it and I'll just publish it.

It's you don't really want to do that, right?

I mean, it's kind of a you want to make sure you're doing it, right?

And you want to make sure that you are kind of learning with the right resources or under the right people to kind of

uh, help you transition into that new skill set.

And so I remember approaching my stats professor, uh, during my PhD and saying, Hey, I want to get into this class over in the biostats department, but I think I need, uh, like

a recommendation to say that, you know, that I should do it.

And he was like, dude, that's, that's like the hardest class in like a PhD level biostats program.

I don't think you really want to do that.

Like he was like, I like just don't.

And I was like, okay, I understand that.

Like, cause I was looking for an introductory kind of, you know, get your feet wet in Bayesian statistics.

And he was like, dog, that's you are going to hate your life and they're going to be like, who the hell told you, you could come here.

So, uh, cause you know, Hey, I know, I know a thing or two, but I'm not, I don't have a PhD in biostats.

Right.

Um, and it's really important to, to kind of.

understand and adhere to your own limitations as an applied scientist who uses statistics, but is not a statistician.

So I tried that and didn't really work out.

I also pitched at one idea, at one point, the idea of doing an independent study in a particular statistical approach that was popular in sports science.

This was when I was like a, I think a master student.

And my advisor said, yeah, don't do that.

And I was kind of bummed at the time, boy, was she right.

because the approach that all the smart folks were using that I wanted to sink my teeth into, it was called magnitude-based inferences.

And it was something that was being used in our field, and it was kind of framed on the surface as like a mixture between frequentist statistics and Bayesian statistics, this

kind of hybrid approach that allowed us to make better inferences about small sample size research while

still staying within the frequentist framework within which we're most comfortable.

Well, in the last year or two, the last couple of years, there's been some papers where a statistician heard that we were doing that and they dug around and they're like, wait,

what have you guys been doing?

And so they looked into it and they're like, yeah, it's better for small sample sizes because your type one error rate is like 20%.

You're just committing tremendous statistical errors.

and just not doing rigorous analyses with this approach.

So the statistical properties of this approach just completely failed and then people stopped doing it.

So all of that is to say, I tried to kind of poke and prod at it when I was a grad student, but our field just wasn't really embracing it yet.

But in the last couple of years, I'm seeing more and more papers in our field that use Bayesian statistics.

And to be totally honest, the thing that's really changed it,

in my view is a JASP software, having a point and click user interface that opens up Bayesian statistics in a way that feels a lot more accessible to folks who don't have a

coding background.

Cause a lot of folks in my field, they do not like any code based stat softwares.

They like SPSS.

And once they heard JASP was on the menu, they said, okay, fine, I'll try it.

Hmm.

But yeah, so all of that is to say I've been aware of it and quite interested in it for a while.

When I was doing more of my own stats on studies, I had an interest but not enough proficiency to feel comfortable just jumping into it.

And yeah, now that I'm back in the research game, I think perhaps I'll have better opportunities to finally do that.

Hmm.

Yeah, that's, that's super interesting.

Um, thanks for that.

Yeah.

Like lay of the, lay of the ground.

Um, yeah, I love that.

Um, and for sure, like I think, and that's something that's really, I see that as something really positive, uh, for a long time, the main barrier to using patient sense

was not even that.

How do we train people?

Uh, how do we teach them?

How do you make that easy to use for them?

So like, for instance, with software like, like JASP.

But how do you compute that?

Basically it was compute power, the main problem, because that integral on the denominator is just the devil.

So you have to use approximation methods and that was very hard.

Now we have extremely powerful computers, which can do that in just a few seconds.

So that...

opened a lot of doors basically for people who don't know about these stuff to then use the software that nerdy stat people like us develop to actually use patient stat in their

analysis and that's really cool because now I think the barrier to entry has shifted from a computer power issue to a basically manpower issue.

Okay.

Who do I find?

to be my mentor, which book should I read, which video should I look at, which package should I look at.

So that's really cool.

There is a variety of packages to do it in Python, mainly of course in my, well, PIMC a lot.

BAMBEE is extremely useful because it allows you to do basically PIMC models, but instead of writing completely the model, you use FORMULA.

Is that how you say that?

It's used in R a lot.

So formula syntax, yes, formula syntax.

In R, I know you use R a lot.

So I would recommend looking at BRMS, which is extremely good, very powerful, based on Stan, which is state of the art algorithms below that.

And we have had Paul Buechner, the founder, I mean, the main developer of B.

the creator of BRMS on the podcast.

I will link that in the show notes.

And also, since you mentioned JASP, we had EJ Wagenmarkers on the podcast, episode 61, for people who want to listen to that.

That was a really interesting one, because EJ is one of the main persons developing and coordinating the development of JASP.

And actually, that episode was really interesting because EJ is...

really versed into psychology research.

And basically, I remember I called that episode, why we still use non-Vision methods.

That's very EJ spirit.

So it's an interesting listen.

I'll link that in the show notes, too.

So yeah, to me, that's a really good sign that now this is basically the main thing.

And so yeah, in the R world, I would say

BRMS is one of the main things to look at.

Also, the book and video series on YouTube by Richard McAlrath called Statistical Rethinking.

Extremely good, very pedagogical.

And he's actually teaching in Dresden at the Max Planck Institute, so evolutionary anthropology.

And there is a lot of examples of what that, so that should be familiar to you.

I send that to you because it's a really extremely good resource.

Yeah, that'd be great.

Yeah.

And I mean, so to, and then to narrow down on what patient stats can do here.

Yeah, I do think that the low sample size thing and also the intersubject variability is tremendously important.

And the good thing is that it comes out of the box in the patient framework, because, well, you have the priors basically.

And on that note, I.

read recently one of the new papers by Andrew Gelman, which people are familiar with, of course, Jessica Holman and Lauren Kennedy.

I'll post that into the show notes and I sent it to you, Eric, because I found it really interesting because basically the idea...

So you know the correlation

So basically a lot of people are familiar with it.

Basically, the same correlation number can be explained by a lot of different data points patterns.

And here, what they do in the paper is that they develop causal quartets, where they basically show that the same average treatment effect can be explained by a variety of

different causal patterns.

where you would have completely different data set and data points.

But if you just look at the average treatment effect, you would think that these are the same data points in a way.

And it's the same experiment, even though it's absolutely not and it's explained by different causal factors.

So I'm aware I'm explaining that in a podcast.

But this is very, actually I can, oh, I'm forgetting, but I can.

share my screen here now.

And here, I'm going to do that for people who are watching on YouTube.

Going to share my screen here.

And here, I have the paper.

And at some point in the paper, you have these kind of quartets here, for instance, where these plots are basically explaining the same data.

Can you see my screen?

It's loading.

Ah, it's loading.

Okay.

So maybe that doesn't work, but.

I looked at the paper so I know what you're referring to.

Ah, there it is, I see it now.

Oh, yeah, there it is.

Yeah.

So basically, you've seen that, right?

Like basically, you have four graphs showing different patterns of causal effect, but the four graphs have the same average effect of 0.1.

And so I don't know, yeah, what did you think about that paper?

Is that something that you think is really different for your field?

Is that something that you've seen also in your field?

Well, yeah, I mean, you know, I think one of the big challenges that we have in our field being kind of attached to the frequentist approach to statistics is that we really don't

like to mess with the nuances of distributions.

You know, we like to condense a lot of things down to, you know, average observed effect with a

you know, with a symmetrical confidence interval around it.

And we like to cross our fingers and hope that all relationships are linear when we're looking at continuous data.

Um, even though we know deep down in our heart that they're, they're probably not.

Right.

So like, you know, we run into so many things, especially in sports science, where we talk about, um, you know, how much training volume should you do?

Of course we know that there's going to be at least a ceiling effect where

you reach just completely diminishing returns.

And if nothing, I mean, if not a ceiling effect, more likely we start to see that too much volume.

For a while it's more productive and then eventually becomes counterproductive as you exhaust your ability to actually effectively recover from that training stimulus.

And so, if you had a dollar for every paper that assumes a linear relationship between training volume and the resulting.

training adaptations, you'd be quite wealthy.

And so, yeah, I think there's tremendous benefit in saying, not just, I think there's benefit in branching out beyond this simplistic idea of assuming that all of these causal

effects we see in the literature can be boiled down to, like I said, an average effect with a symmetrical confidence interval or a perfectly linear relationship.

You know, plus of course the, you know, the error term in the model, but, um, yeah, I thought this was a really fantastic paper and a point that goes, uh, yeah, it doesn't get

discussed as much as it ought to, uh, you know, how these different distributions lead to, uh, tremendously different applications, you know, cause I mean, if you're in exercise and

sports science, um, you're an applied researcher almost by default.

You know, there's not, not a lot of, not a tremendous amount of basic science going on.

And so usually the question is, okay, you found this, you answered this question, you addressed this hypothesis.

Now what do we do with it?

How does it turn into an intervention for a healthy person or a clinical population or a person in a particular set of circumstances?

And the actual pattern of data becomes quite important.

Yeah.

And so thanks.

Thanks a lot for taking a look at that paper.

That's pretty cool.

And yeah, to echo what you were saying.

Also, I encourage people to read it.

The paper is really easy to read.

It's an easy read.

It's not a lot of math.

Don't worry.

It's a lot of graphs actually.

And they even developed an R package called causal quartets to dig a bit into that.

super interesting read and yes, something to really keep in mind each time you see that kind of paper only talking about causal average causal effects, treatment effects,

especially in the social sciences, as you were as you were saying, Eric, where variability is extremely big between subjects.

And I mean, there is a part in the paper that was it's just it's just extremely funny to me where

They call here, they talk about like basically, um, citing from the paper.

They cite a paper who performed two small survey and they found that women were three times as likely to wear red or pink during certain days of their monthly cycle and that

the result achieved conventional levels of statistical significance, but then like you dig into that and it's like mainly explained by.

by viability between people.

Because, well, some people wear a lot of red.

Some people don't wear red at all.

And that viability is going to be huge.

But then you get this kind of average treatment effect, and that doesn't work at all.

So that was a funny part of the paper.

And so this is a very pedagogical paper.

I encourage people to dig into it.

Because, yeah, it's something to keep in mind.

And

As you were saying, that's really something I notice also when I teach people patient stats, especially when they come from the classic ML side.

Once they get the posterior distributions out of the model, the question is always, how do I summarize that?

I'm like, no, that breaks my heart always, you know, because I'm like, you've worked so hard to get that posterior distribution.

It's not easy to get a whole posterior distribution sometimes.

And now that you have that, you want to throw away all that information just to get a point estimate.

That's a bit of a shame.

So try basically to throw away information as late as possible, basically.

And try to get used to the distributions.

I know that ant dimension.

I know our brain is not good for more than three dimensions.

But basically, the idea is not throwing information away, especially when it's hard to come by.

and try to throw it away as late as possible.

And if you really, really have...

Yeah, I was smiling.

You mentioned that the human brain doesn't like to work in more than three dimensions.

And I, I over, you know, I was scrolling through Twitter and I saw someone who had created a, uh, an R package, uh, for like power and sample size calculations.

And someone had asked them, you know, how come you didn't make it easier to use this, um, this R package to, to estimate.

the sample size requirements for a four-way interaction and he said well if you're able to accurately predict The effect size of a four-way interaction in a social science

experiment Then you're way smarter than me.

So you should be right in the program I Thought that was really very true.

Yeah, once you're getting into the fourth dimension Our brains just kind of implode or at least mine does

Yeah, yeah, no, I mean, and also like four-way interaction.

I would say that's like in the paper, they really argue for thinking more about nonlinear effect.

And so that means interactions.

But then, yeah, the other caveat you should have, then students get really excited about that.

And then you're like, oh, wait, like five-way interactions?

I'm not sure you can even interpret that if that exists.

So it's like, just calm down.

It also happens a lot with hierarchical models, because hierarchical models are really easy, I would say, in the Bayesian framework, because it's just how the framework works.

And so you just put parameters into parameters like Russian dolls, which makes it extremely powerful.

Hierarchical models are extremely powerful, especially when you have low sample sizes, because then you can pull information and have basically better models.

But then it's like, people are like, so can I do a five-nested, five-dimensional nested hierarchy?

You can.

On paper, you can.

Will you be able to identify the top-level hierarchy?

And then just also interpret it?

I'm not sure.

Start with a two-level nested hierarchy, then we'll see.

Yeah.

I put all of that in the show notes.

Lots of things in the show notes for this episode.

That's really cool.

I really encourage you to take a look at people.

And Eric, you are being extremely generous with your time.

I'm going to try to start winding down and close up the show.

So maybe.

Yeah, something I would like to ask you is more forward looking.

I had two questions about that, that I'm going to condense into one, and then, uh, and then we'll get more practical for, for people to, to end up the show.

So, um, looking ahead, basically, um, I'm curious about what you are mostly excited about in your, in your field of sports and nutrition science and basically like the,

the question if there is one big question that you'd like the answer to before you die.

Yeah, in terms of what I'm excited about, the field is growing up.

I mean, it's crazy because I have been kind of on the periphery of the field, I would say, for four or five years as I finished my PhD, went out, did some, you know, industry work,

started some companies and whatnot.

Now that I'm getting back into it, it's remarkable.

I'm excited because it's grown in the direction I had hoped.

And I most importantly didn't have to do any of the hard work.

which is great.

So I was one of those like hypocrites who just said, you guys ought to do this, that and the other thing.

And then I got the hell out of there.

And now that I'm coming back, I'm like, oh great, you did it all.

Awesome, thanks for that.

Now I'm ready to dive back in now that it's easier to kind of pile on rather than pave the path.

But yeah, I mean, I remember when I was leaving the, not leaving the, when I was taking a little detour, the things I was thinking were, and saying,

I was just like, man, we need to do low sample size is a problem, right?

Why are we not doing more multisite trials?

It's such an intuitive way.

We're all using the same equipment and the same methods.

We're all making the same manufacturing companies richer by using the same machines.

So why don't we pool our data in these multisite trials and then we can actually make much more robust inferences without having to...

land these grants that are five times more than we're actually feasibly going to get.

So I've been seeing more multi-site trials, which is really exciting.

I remember thinking, we need to be branching out statistically and specifically doing more Bayesian analysis, but also doing more hierarchical models, kind of adopting that linear

mixed model framework.

And I've been seeing so much more of that.

because, and it's perfect for our field, we had gotten locked into analysis of variance being the kind of, the default approach to what we do, frequentist analysis of variance.

And I'm seeing so many more linear mix models and hierarchical models and seeing so much more application of Bayesian statistics.

So those are things that excite me.

And one thing that I'd really like to see even more of, which we're getting more of,

is embracing some of those open science principles in two ways.

First of all, I'm seeing more preprints, which is very exciting because I just don't think there's any need for us to wait nine, 12 months for that thing to get into print.

I really like the preprint approach to publication.

And one area that I'm not seeing quite as much of, but it is happening more, is open data, you know, completely open data files, not...

not the sentence at the end that says, if you beg me, I will send it.

And then they never respond to emails.

But open data, I think is huge for our field because we're so interested in inter-individual variation.

A really fantastic way to get at that is by doing, ideally, if we could do more participant level meta-analyses, we'd be in really good shape to address some of those

things.

But you can't do participant level meta-analyses.

It's very infeasible to do.

until you get a little bit more buy-in with open data sets.

So that's what I'm excited about.

And then in terms of questions I wanna answer, I have to be honest, like I still am a bit of a generalist at heart.

And my overall focus is to make sure that we are empowering people to take control of their cardio metabolic health if they wish, right?

So I'm not into telling anyone what they should or ought to do.

But for someone who says, hey, I want to feel better every day or, you know, change my body weight in the following way, I want to empower people to be able to do that in an

informed way where they feel like they have a great deal of autonomy and self-efficacy.

And to make that work, we need effective interventions.

And as I look around, I really don't think that we need any paradigm shifting basic science to occur in order to make that happen.

I think.

you know, what we have now in terms of understanding energy balance is pretty robust.

I think the biggest things that would even come close to being like game changers would be just the fact that we now have a better understanding of metabolic adaptation, energy, you

know, exercise energy compensation, and now seeing this new wave of effective obesity drugs on the market.

I mean, now that we're seeing how these, you know, these different areas interrelate,

I'm not sitting around waiting for a huge paradigm shift like we're going to potentially see in physics, right?

Where we say, oh, now we need to rewrite all the textbooks.

I think really what I'd like to see is more of the just boring everyday science where we make incremental progress towards seeing how some of these pieces fit together.

I think seeing things related to...

Health psychology behavior change appetite regulation energy expenditure regulation exercise adherence I think those are the final pieces where we just need a little bit more

of that Regular old-fashioned boring science of incremental progress where we start to put some of these things together And what I'd love to see is a holistic approach where we're

focusing on Behavior change psychology sleep energy intake, you know diet and exercise factors

And frankly, right now I'm really optimistic about using those in conjunction with some of these new pharmacological interventions for weight reduction.

And what I'd love to see is if perhaps we can get to a point with these non-pharmacological interventions that they are all introduced at the beginning and we

have the aid of that pharmacological intervention to kind of do a lot of the heavy lifting at the beginning.

but then eventually offer an off-ramp where a person can say, yeah, I need to do maybe a six or 12 month stint of this pharmacological intervention, then get off of it and use

these behavioral approaches moving forward to maintain that progress.

Whether or not we can do that, I think we gotta get better.

I think we need to really bolster.

some of the interventions that we're currently doing, but I do see that as potentially something that isn't implausible in the near future.

Yeah.

Thanks for that optimistic note.

And I mean, and I do agree.

Like I was actually quite surprised when I started digging into the literature that we have, it's not, it's not, you know, cosmology and like Big Bang physics, like we have that

stuff not figured out, but I don't know if the Pareto effect is already here, but like really there is some...

already solid scientific evidence of things we know and reproducible.

Like you try it on people and it works.

So apparently we're there.

So that's really good.

And yeah, what that made me think is, eh, that's weird.

That will still seem in the general public, so many really crappy science and nutrition advice.

And basically it's now, it seems to me to be at a point where

It's more a question of, okay, how do we package the science that we actually have and the knowledge that we actually have and allow it to percolate into the general audience?

And it's more this than really paradigm shifting things that apparently we need right now.

Yeah, I agree.

I'm glad you brought that up because I honestly think the biggest failure of nutrition and exercise science collectively is, and I'm not even pointing the finger in terms of blame

or offering how to fix it.

Maybe I'll go away for four or five years and they'll fix it all and then I can come back.

The widespread failure to do some of that.

public outreach and engagement in an effective way.

The fact that right now there are so many knowledgeable folks in exercise and nutrition, but the people who hold the megaphone are the influencers on Instagram who are dictating

the public discourse of what fitness looks like and what health looks like.

It is such a catastrophic failure.

And I don't think our fields view it as their responsibility yet.

And I hope that changes within the next five or 10 years.

I hope that, for example, I think it's insane that a tenured full professor has all these expectations for what they do with their time professionally in terms of community support

or service to the university, their research, reviewing papers, mentorship, teaching.

They have all these different things that are...

codified into their role as a thought leader in this area in the academic world and it seems like the public outreach the public communication science communication is so

undervalued and Not incentivized in the slightest and then everyone looks around and says hey, how come no one listens to us And it's like because that communication you have never

made it part of the job.

You've never incentivized it You've never rewarded it in a meaningful way.

And now you're wondering why you don't

dominate the headlines and lead the discourse.

And it's because there's no reason for that to be the case.

And there are people who are spending all of their day just on the communication.

And they will out influence you every single day of the week if there's no meaningful attempt of these fields to actually lead that outreach.

So I'm glad that you brought that up because that is a huge failure.

Yeah, I mean, I completely agree.

That's what I saw.

I mean, it's hard work for sure.

This is a hard thing.

We've struggled with that in the vision stats world for three years.

I think we're getting way better now.

Surely, almost internally, due to this podcast, I think we can acknowledge it.

No, but joking aside, yeah, it's hard work, but it needs to be done because like,

Yeah, I've seen the same content you're seeing, of course.

And I'm like, it's so messy.

I'm like, this is shame because, well, apart from the fact that I still don't understand why people believe what celebrities say, just because they are celebrities.

You know, I don't know why you would believe it's celebrity because they are doing some workout or some diet, even though, you know, they are not, they like, they have no

knowledge in it.

But okay, that aside, it makes me sad always a bit, but I'm like, oh, but we know about that.

Like, we like...

science knows about that, this falls, and there is a better way to do it.

So that would be better for people who are listening to that person to actually do something that works for them and not spend their hard-earned money on something that we

know is not going to work.

And yeah, so there is definitely a lot of work to do on that.

Incentives, extremely important for sure.

And I mean, that's also related to something you mentioned before,

Also, it's very important on these podcasts, not only because myself, I am an open source developer, but like I know a lot of listeners are open source developers and are the

people building these open software.

And so of course, open data for us is extremely important.

And I think it's changing a bit.

I see more and more, you know, academics doing open data and even developing packages, open source packages.

Thanks to R and Python mainly, which are picked up and then it's so common in those languages to have an open source package somewhere that yeah, it starts to become a

custom, but yeah, like I think universities could make a better job incentivizing basically investing in open source and open source means also open data.

So yeah, for sure.

Yeah, well, my advice to you, if you want to spread the word of Bayesian stats, the thing that sold me on it, at least on a theoretical level, when you make someone, force them to

interact with the true definition and interpretation of a p-value and a confidence interval,

Yeah.

I feel like if they know what they're doing and they know what they want to do with their research, that alone they should say, wait a minute, there's got to be a better way.

And I do think Bayesian offers just a dramatically more intuitive interpretation when it comes to the final output.

Yeah, yeah, for sure.

For sure.

That's completely true.

Even though in my experience, the best way to introduce people to patient status when patient status solves one of the problems.

They have spent, often in like doctoral students, spent hours on a frequency software and it doesn't work.

And then they switch and use, I don't know, BRMS or BAMBi, and then just one line of formula.

And that works.

That's the best way.

That's the best trojan horse.

Yeah.

Absolutely.

Um, so to close up before I ask you the last two questions, um, we've talked a bit about that already, and I think it's something I wanted to cover also, because I know

personally, uh, a lot of people, um, these were not in the same circle.

So I'm guessing a lot of people, you know, are really already into, into exercise and things like that, but in my world, not necessarily.

And, uh, something I often encounter is.

Well, a lot of people who tell me that they would like to take better care of their health and their body, but basically they are a bit overwhelmed and they are intimidated by the

amount of discipline it requires, at least it seems to them, things like counting your calories, weighing yourself consistently, all those things.

So I wanted to close up the show by being practical and basically,

ask you what you would recommend to help get those people on their healthier path.

Yeah, I would say, you know, if you're just getting started, I'm going to give two pieces of advice that seem contradictory, but they're actually complimentary.

So first you want to choose the strategies or the interventions that kind of feel at the starting point to be most feasible or most accessible to you, or I guess most enjoyable

even, right?

So you might be someone who says,

I want to lose some weight, but I don't want to count my calories.

Then don't.

Let's just change the types of foods that you're eating and do some basic portion control.

When I won my professional status in bodybuilding, I did not count my calories during my contest preparation.

You don't have to.

If you know generally what you're eating, then you kind of have a sense of how many calories you're eating.

Hmph.

And even if you did calculate it down to the single calorie, guess what?

You're wrong.

I mean, the food labels are not that accurate.

There's variability in the energy content of food.

So anyway, bit of a tangent, but choose the strategies that feel like they're going to be the most accessible, most enjoyable, most feasible and build on that.

So same thing, if you don't want to change your diet at all, but you kind of enjoy doing some physical activities, go play basketball more.

Boom.

Now you've taken that first step.

And once you're doing more basketball, you're getting in better shape, maybe then that progress starts to fuel additional motivation to chip into some of those things that used

to seem like they were too much, but now you're like, you know, I've already come so far.

I'd really like to see what we can do here and kind of take this up a notch.

The other thing is, number two here, is that you want to start with goals that are hard enough to care about, but easy enough.

to build some degree of self-efficacy.

You know, you wanna get your confidence rolling as you go.

So, like I said, you wanna choose things that are feasible, enjoyable, realistic, but if they're too easy and they're too feasible, you're not actually gonna feel like you're

doing anything.

And that can actually be a bit of a dead end because the fact that we are making positive improvements and making changes to make ourselves better, that alone becomes a source of

motivation.

So what they find in the goal setting literature is that if a goal is too hard, you're gonna fail, your self-efficacy will plummet, you will start to believe, I actually cannot

do this.

But if you set a goal that's too easy, you do it, but you're like, am I, it's like playing basketball against a four-year-old and it's like, yeah, I won, but do I really feel good?

Of course I won, right?

It's not enough to get you excited and make you feel like you're actually making some improvements to feel really.

happy about and enthusiastic about.

So that would be my recommendation.

Mm-hmm.

Yeah.

Love it.

Thanks a lot.

Thanks a lot, Harry.

Again, the first one.

year olds.

I'm sure many of you are good at basketball.

Yeah, yeah.

If four-year-olds listen to this podcast, please message me, we'll get you on the show.

I want that on the show.

Oh yeah.

Oh my God.

And yeah, the first point made me think we have that saying in French, which is, l'appetit vient en mangeant, which means, I don't think it's true for hunger actually, but it means

hunger comes as you eat.

So yeah.

Like your first option, like it's basically, yeah, try something first.

And then you'll see that if you like it, you'll naturally get nerdier about it.

I would say.

Awesome, Eric, I think we're breaking a record of the longest episode here.

So I am not surprised it's with you because I know you have a history of very long, very long recordings behind you.

So, you know, is there any topic I didn't ask you about and that you'd like to mention?

No, it's a me problem.

I go on and on.

So I, almost every podcast I go on, they say, wow, this is our longest one ever.

So sorry about that folks, but yeah, we've covered everything that the audience can handle.

I'm sure.

Awesome.

Yeah.

So for people who...

Oh, no, wait.

I'm forgetting the last two questions.

Am I crazy?

So yeah, before closing up the show, I'm going to ask you the last two questions I ask every guest at the end of the show because as I always say, it's not the point estimate

that counts, it's the distribution of answers.

So first one, if you had unlimited time,

and resources, which problem would you try to solve?

Yeah, so I don't think that I'm personally equipped to solve really any of the pressing problems in the world, but You did give me unlimited time and unlimited resources So I

would just assemble a team to do it with all my time and resources I wouldn't tell them about how much time they had I'd put a fake sense of urgency But I would get together the

best and the brightest and try to figure out a really Powerful solution in terms of generating clean energy for the world

generating that energy and distributing that energy effectively.

I think when I pretend to know more about my specific area of expertise, which is always a bad idea, every time I look at the world's problems, for a large percentage of them, I see

them as energy problems.

And I think if we could find a tremendous solution to meeting the energy demands of the world in a way that is not...

detrimental to the environment and can be distributed effectively, I think we'd live in a much better world.

I would do my own version of the Manhattan Project, but it would be for clean energy, and I'd get the best and brightest together and pay them whatever they need.

Nice.

Yeah.

I love it.

And I think we're back to physics then.

When I understood like fusion, nuclear fusion, maybe one of our closest contenders in this.

And second question, if you could have dinner with any great scientific mind dead, alive or fictional, who would it be?

It would be Leonardo da Vinci.

I don't know if that's too cliche.

Have you gotten that one before?

Yes, but not that often.

Yeah, less often than you think.

I was going on a date with someone many years ago, and I wanted to seem smart and cultured.

So we went to an art museum.

It's always a good choice.

Good little conversation starters along the way.

It's great.

Good date idea.

The exhibit that they had was actually like, you know, they had the normal stuff that's always out, but they had a special exhibit where they had some of DaVinci's notes on

display.

Oh nice!

Mm-hmm.

And I was looking at these notes and reading the summaries of what they were, because I'm pretty sure they were in backwards cursive Italian, if memory serves, with all sorts of

shorthand scribbles, and it was chaotic.

And the notes were just absolutely, some of them were brilliant, some of them were just absolutely insane.

And if you look at some of his ideas, you look at some of his inventions, look at his multifaceted interests,

What more could you want from a dinner guest?

Just fascinating, a little bit insane, can talk about any topic.

Yeah, that would be a wild time.

Yeah, yeah, I know for sure.

And such a life also incredible each time.

Also a big fan.

I'm a big fan of Italian culture and the country also in Salso.

I often go there and yeah, I was in Rome again a few days ago and just, yeah, what a city incredible and yeah.

I was looking at looking at his notes and when you're on page like six of the ramblings about the lake and river systems on the moon, you're getting into the good stuff.

I'm just like, dude, what are you talking about?

hmm.

Yeah, my main question would be, did you write that clean?

Or were you on substance?

Because if you did that on a clean brain, that's even more impressive.

Yeah, kudos.

Yeah.

And yeah, so if anybody is happens to be in France, in actually my region where I was born, which is called the Loire Valley.

You have one castle, which is the castle of Amboise.

And really next to that castle, you have a not that small mansion, which is Leonardo da Vinci's last house.

He died there.

He spent his last few years at the court of Francois Premier.

And yeah, like this is called the Clos Lucé.

So if you had the opportunity to read it, that's really...

Incredible.

First, the domain is beautiful.

He had his own vines.

He was making his own wine.

You've got some of the replicas of some of his inventions, and you've got some notes also.

His bedroom and so on.

That's really a beautiful place to visit.

I encourage anybody to go there if they have the opportunity.

Well, Eric, I think now we can call it a show.

Thank you so much for your time.

It was absolutely fascinating.

Learned a lot, of course, and motivated to still learn a lot about all this.

As usual, I will put a link to all your websites, socials, and so on for people who want to get in touch with you and also a lot of show notes.

For those of you who want to dig deeper, thanks a lot, Eric, for taking the time and being on this show.

Thank you so much, I had a great time.