Matt Perryman Matt Perryman

A Systems View of Exercise

This article began to take shape after reading another well-intended internet complaint about how mock-quote “science” has no relevance to practical get-in-the-gym exercise.

As pro-science as I am, I have to admit there’s a lot of truth to that point of view. You don’t have to look much further than the papers passed around the strength and fitness blogs and Facebook updates to see why. While there’s occasionally interesting stuff turning up, there’s also a lot of crap. By crap I mean papers looking at how Molecular Signal X jiggled in hungover college students when exposed to a lab trial resembling no workout you will ever do.

While I personally find a lot of the biochem research interesting, there’s no shame in admitting that it’s exactly that: a personal interest. I don’t think that material has any relevance at all to doing things at the gym, at least not in the way most folks seem to expect.

Still, there’s something not quite right about the blanket anti-science, anti-intellectual perspective that characterizes some corners of the strength and fitness field. The stereotypical Bro, the musclehead who believes the pseudo-science in supplement ads but turns hostile toward any attempt at debunking it, isn’t our ideal role model. There’s rejecting the irrelevant, on the one hand, and then there’s needless hostility towards intellectual curiosity.

The former I can get behind. The latter, that’s just typical internet posturing — or, at best, an over-reaction to bad science — and in either case an attitude best ignored. The problem is, it’s not always clear which is which, or why there’s a difference at all.

Starting From The Wrong Place

The more I’ve thought about it, the more it’s occurred to me that confusion about what science is, and what it implies for exercise and nutrition, comes down to perspective. There’s a tremendous gulf between the public perception of science, created by bad science reporting and unrealistic expectations and plain old stupid, and the reality of science.

Many of the apparent disagreements we encounter between mock-quote “science” and the equally inane “common sense” aren’t actually disagreements at all.

Most of what you see paraded out on the internet, in blogs and science-journalism built on press releases, is mock-quote “science”, the Justin Bieber to the Mozart of actual research done by real scientists. That research, which is what scientists in labs, clinics, and research programs do and which is published in journals after a process of peer review, forms the foundation of our knowledge about the natural world.

The public perception of science and the real thing are often two different species, and both of those are removed again from the philosophy of science. That way of thinking about the natural world, influenced by luminaries like Francis Bacon and Karl Popper, doesn’t always come into the picture in most scientific back-and-forth, though, and there’s our first glimpse of the real issue.

If you don’t think about the world in the same way as I do, or the biologist writing the paper you’re citing, then you miss crucial parts of the message. It’s like two people standing on the edge of the Grand Canyon but ten miles apart. They’re looking at the same “thing”, but they’ll have very different glimpses of it and, once back at the hotel, two different stories to tell about it.

Everything I say and think comes from my perspective on the world. I’m standing on top of a mountain made out of all the life-experiences and books and research papers that have shaped my thinking, and it’s from that vantage point that I look at the world. You stand on your own, and a third person stands on still another.

We share much of that background thanks to common culture (and common biology, to go a level deeper). Still, it’s not hard to find spots where the overlap isn’t seamless and we find it harder to agree. It’s easy to agree that gravity points down or that we need food to live. More abstract ideas aren’t always so clear.

You look at the world with one viewpoint, and I look at it with another. Throwing more facts around within our personal sandbox gets us nowhere when we each play by our own set of rules. If you don’t look at the problem in the same way as I’m looking at it, then we may never agree.

It isn’t what you think, but why you think it.

Right and Wrong

One of my favorite essays from Isaac Asimov is The Relativity of Wrong. In it, Asimov, by way of addressing the letter of an angry student, dismantles a common complaint about the scientific process. Just look at how scientists used to think the world was flat, the student writes. Clearly scientists don’t know anything. Asimov goes on to politely eviscerate that suggestion.

To ancient peoples with limited powers of observation — you can’t see far if you’re a hunter-gatherer with no tools or transportation besides your own feet — the idea of a Flat Earth was very nearly right. The Earth’s curvature per mile is so low that a Flat Earth is “right enough” to Paleo Man.

It was only with the advent of new tools that a different picture began to emerge. The Greeks learned to measure the curvature of the Earth’s surface. Ships circumnavigated the globe, mapping it along the way. Finally we launched airplanes and satellites to take pictures and measure the Earth’s shape with unprecedented accuracy. Each step added a little more to our model of Earth, refining the observations and etching in ever-greater levels of detail.

“It might be that to describe the universe, we have to employ different theories in different situations. Each theory may have its own version of reality, but according to model-dependent realism, that is acceptable so long as the theories agree in their predictions whenever they overlap, that is, whenever they can both be applied.”
Stephen Hawking & Leonard Mlodinow

In their book The Grand Design, Stephen Hawking and Leonard Mlodinow introduced an idea they call model-dependent realism. Phenomena, they say, can be described equally well by many different models and, like the hunter-gatherer on the Flat Earth, we don’t have enough data (or the tools to gather the data) to say which is right or wrong. Thanks to that uncertainty different models, even models that seem to contradict themselves, can explain observations equally well.

A model, in this meaning, is meant to describe some aspect of the world around us. Aristotle’s Earth-centric model of the solar system, and the Copernican model that replaced it by sticking the Sun at the center, are both well-known examples. Observations are taken, guesses are made, and we create a picture that explains what we’ve seen.

A model is a representation of some part of the world. Models exist for everything in science, and exercise physiology is no exception. I’m using model instead of “theory” as theory comes with its own baggage, but they’re more or less interchangeable for these purposes.

Instead of throwing out competing models as nonsense, as they can’t all be right, Hawking and Mlodonow argue that we should look for meaning according to the usefulness of each one, rather than how well it reflects absolute correctness. As long as any particular model of reality works where it’s intended, then it works.

I want to clarify that I’m not (and Hawking and Mlodinow aren’t) arguing against the existence of objective facts. In the realms of fitness and strength & conditioning, the existing body of knowledge is good-enough for us to say some things with certainty. There are wealths of knowledge about muscle and nerve and connective tissue and how those tissues respond to exercise, and that does put some hard limits on the possibilities.

Not every last thing is true just because someone believes it or writes a blog about it. When an alternative theory doesn’t match reality, it’s no longer useful. This is genuine Bro-science, pseudo-science stemming from a worldview that simply doesn’t match reality.

At the same time, a good number of things treated as True Fact aren’t so much, and it’s easy to sacrifice the useful in chasing down the correct. That’s what we need to avoid.

Level Crossings

Rightness and wrongness — and usefulness — depend on scale. Does it matter to a hunter-gatherer with no airplanes, no satellites, not even a telescope, that the world isn’t flat? Besides intellectual curiosity, not really. The wrongness is irrelevant to any conceivable situation Paleo Man would ever encounter. It just doesn’t matter.

But there are other situations where the correctness does matter. The lady avoiding any kind of conditioning exercise because she heard that cardio raises cortisol and makes her fatter has crossed into the territory where the model’s correctness is important. Her behaviors — the exercise she does or doesn’t do — are directly affected by that model of exercise.

There are “right wrongs”, in which you may be wrong but it doesn’t matter, and then there are “wrong wrongs” where you’re wrong in a way that causes you grief. Anyone discouraged from exercising “because cortisol” has encountered one of the many wrong-wrongs in the fitness industry.

Leaps from one scale to another happen all the time in the on-going argument between ScienceTM and Get It Done lifting experience. One camp expects direct and immediate application from research. They expect to read an abstract or two off Pubmed and have an awesome “scientific” program. The other side notices that these nerds are fooling themselves and aren’t saying anything interesting about training or eating. The real info, they believe, comes from hard work and getting it done.

I don’t see why that argument needs to exist at all. We should, ideally, draw useful knowledge from any and every possible resource. Not all knowledge learned through experience can be written off as “Bro-science”, just as not all science is useless ivory-tower academia. By keeping scale in mind, we can avoid nonsense statements like “aerobic exercise makes you fatter”.

It’s all down to perspective. If you stand on your mountain and look down on a world where Science and Experience are two competing camps with irreconcilable differences, of course you’re going to see conflict.

Imagine there are three levels of possible knowledge, one right above the other and ranked according to how zoomed-in you are relative to the real life world we live in. The lowest level contains living tissues, individual cells, and the molecular biochemistry that makes up the inside of living cells. This is the raw biology, and it explains how muscle fibers and fat cells respond to exercise or protein intake or fasting. It explains the various signaling networks and genes that switch on or off when we lift weights. You know all this stuff is there but unless you’re a biologist you probably don’t think about it much (or at all).

A step above that, we’ve got a roughed-in sketch of the whole body but, like an anatomical drawing in a textbook, we’re still not quite at real life. Experiments at this level include all those papers in the Journal of Applied Physiology or the Journal of Strength & Conditioning Research, the ones that involve real people doing exercise (or exercise-like things) but always with those strange protocols that never look like a real workout. You’ll see the machine biceps curl or isokinetic leg extension where one side is tested with 10 sets of the eccentric 1RM and the other limb is left untrained as a control.

Since we’re seeing real live human beings doing exercise-ish things, this gives us a lot of data about how human bodies respond to physical activity. Still, as critics rightly note, this kind of research is a step or two removed from the training most of us would do at the gym.

Relating personal opinions and experiences is easy to write off as Bro-science, but that’s not always the case. Bro-science happens when bad science is “proven” by 20″ biceps and 500 lb benches and being hardcore. Sharing a workout that added 20kg to your bench isn’t Bro-science. Believing that your kooky ideas on insulin or cortisol — topics at a lower level on the ladder — are right because you won a pro card, that’s Bro-science.

Experience is just fine as long as it stays on the experience level. It’s when the levels cross, when big arms are used as support for bad 8th grade ideas on biology, then you’ve invoked the power of Bro.

We need real In The Gym knowledge. Even hard sciences can’t always stage controlled experiments. If that were the case, we’d have no field of astrophysics. Fortunately, we’ve got a whole universe full of stars and energetic objects zipping around and, with good instruments, astronomers only have to watch and take notes.

At the Real World level of stars, we have endless data to draw on. But astrophysicists aren’t working in a vacuum, either (so to speak). They rely on other domains — other levels of knowledge — to create the framework for their theories and models. They need to understand Newtonian gravity and relativity, chemistry, and even particle physics. The information of the lower levels provides support for the high-level real-world observations. There’s no conflict at all; the bottom-up data and top-down theory bounce off each other and reinforce the astrophysicist’s models.

Exercise is no different. Research gives us plenty of information on how biological systems, from whole bodies down to individual muscle fibers and their genes, respond to physical exercise. But that’s all at the bottom. To make it useful at the top, we need to treat that information as the groundwork it is.

I briefly discussed all this with Mike Tuchscherer a few weeks ago. As meticulous as Mike is about collecting data on his training and the lifters working with him, you’d have to be out of your mind to ignore all that just because it’s not a controlled double-blind trial with statistical analysis.

Russian coaches and scientists generated tremendous amounts of data on Olympic weightlifters with “natural pedagogical (coaching) experiments”. Prilepin’s Table, that chart we all like to use to figure out how to train, comes from Alexey Prilepin’s coaching of the national weightlifting team from the late 70s and early 80s. Prilepin had some ideas on strength and power development, which he implemented with real live weightlifters, and he eventually wrote that up into the famous table that tells us how many reps to do at a percentage of our 1RM.

The natural experiment, in which we see things happen in the wild and record what we saw — Getting It Done, in other words — is currently the only way to really understand what happens in a workout and what happens after months and years of training. No published research can quite compare to the playground of real life, provided you approach it with the scientific mindset.

Real live training is invaluable for generating ideas on how to train in real life. It doesn’t say anything about what insulin is doing, just as the phosphorylation of mTOR doesn’t give you any directly useful ideas for your next deadlift workout. And that’s alright. You shouldn’t expect that anyway.

It’s when the levels cross in ways they shouldn’t — like expecting useful deadlifting ideas from abstract biological data — that “science is useless!” gets a chance to show itself.

Keep everything on the level and you’re fine.

Hitpoints At Critical

Supercompensation. That word, even if you don’t know what it means, defines how you think about recovery from exercise.

You’ve seen the graph, the one in every exercise science textbook and website. You do a workout, depleting all the biochemicals in your body and leaving you in a state of fatigue. On the graph, the line showing your state of recovery takes a dip. While you recover over the next few days, the recovery line tracks back up to the original point, and then reaches up to a new peak.

That peak is supercompensation, an over-adaptation in which your body stores more of the magical biochemical stuff that makes you go. The result is a bigger muscle, a stronger deadlift, a faster sprint. Deplete, recover, supercompensate. Whether you know it or not, that’s the model of recovery that drives very nearly every program and training philosophy you’ve ever heard of. Train hard, get tired, recover.

There’s a history behind this model, having to do with Hans Selye’s original discovery of the universal stress-response back in the 1930s and how, for many years, it was believed that living bodies exhaust themselves when placed under chronically-stressful conditions. For now, it’s enough to know that this model of depletion and exhaustion has been challenged by contemporary stress research. The organs and nerves responsible for the adrenaline rush aren’t burned out or fatigued or anything like that.

What we’re seeing is actually an overactive stress-response. Living tissues aren’t depleted at all, but actually revved up into “stress mode” to cope with threats in the environment.

Threats, in this context, can be anything, physical or psychological. You can feel the adrenaline rush when you narrowly miss a snake-bite, or when you sprint away from the twig you mistook for a rattlesnake.

In the short term, to handle an immediate threat, the stress-response is good. Over longer spans of time, this is bad. When the stress systems stay on, they eat up resources that would otherwise go to essential life-processes. No depletion. No exhaustion. Only a gradual process of wear-and-tear, what stress researcher Bruce McEwen calls allostatic load.

The supercompensation model treats recovery like hitpoints in a video game. Sitting in the morning traffic jam takes some. Angry boss and kids and bills coming due all take their cut. Training, if you have any left to give, takes its share. Rest and relaxation are power-ups that bring you back to full health.

Under the model of allostasis, there are no hitpoints. Your body sits somewhere on a continuum ranging from mostly-normal, at one end, to entirely stressed-out. The closer you are to that stressed-out state, the harder it is for your body to function properly. Your whole system gets knocked out of equilibrium and otherwise normal bodily functions just don’t work right.

To the untrained eye, the distinction between “fatigued tissues” and “disrupted state of entire body” may not mean much. As far as you can tell, you feel beat up and worn down after a hard workout. Does it really matter what ScienceTM says about that?

As with Paleo Man’s Flat Earth, the question isn’t which is right, but which is useful. Does the supercompensation model stand up to the allostatic model on that front?

In some ways it does. I don’t think there’s any harm in using supercompensation as a model to design workouts for bodybuilders, as one example. I think bodybuilders, whether bulking or cutting, can benefit by thinking of how individual muscle groups recover between workouts. The biological processes relevant to muscle-building and fat-metabolizing are compatible with that thinking. When left alone there, I have no gripe.

Supercompensation has a way of creeping outside its useful box, however. Viewing your body as subject to exhaustion leads to ideas like “adrenal fatigue”, a phenomenon of chronic fatigue (fatigue in the mental sense of feeling tired, which need not imply any fatigue in living tissues) that in all likelihood results from an overactive stress-response intersecting with vulnerable personalities. The adrenals aren’t fatigued at all — how could they be when dumping out corticoids at record levels? — and it makes no sense to address the problem as if they are.

Much the same can be said for common ideas on overtraining and the monstrosity that “CNS fatigue” has become. Both of these conditions sit out towards the stressed-out end of the continuum. You rack up wear-and-tear that makes you feel bad and takes resources away from growth processes.

Exhaustion? Burn-out? Recovery doesn’t work that way.

Life processes continue on. Muscles and connective tissues and nerves repair damage and rebuild themselves with fresh proteins. Meanwhile, the stress-response kicks your body into adrenaline-rush high-gear, jacking up immune signals and catecholamines and glucocorticoids (including cortisol). Stress-mode makes you feel bad and, if left unchecked for too long, worn out. Not because you “aren’t recovered” but because you spent too much time coping with tremendous stress levels.

That’s allostasis: stability through change. You can be “burned out” and recovering simultaneously. It’s the net condition — how much the wear-and-tear of stress-mode takes away relative to your current fitness level — that matters.

A Systems View

Supercompensation appeals because it’s linear. There’s One Single Cause. You’re either recovered or you aren’t. Easy.

It isn’t the idea, so much as the thought process behind the idea, that interests me. Supercompensation as a concept is fine for what it is, but it reveals the reductionist views that define the fitness world. Our field is sitting three to four decades behind modern biology and it shows in our ideas.

I’m trying to get to the heart of the philosophical divide between what coaches and trainers are taught, on the one hand, and what contemporary biology is doing. There’s a whole new field out there going by the unwieldy name psychoneuroimmunology (PNI), a cross-disciplinary study of the interaction between the mind and the nervous and immune systems. We’re discovering that there’s little use to describing these systems as independent actors. Everything works so much more elegantly if we view them as integrated parts of the same cloth.

The PNI field couldn’t exist if biologists were still locked into reductionism like most personal trainers and coaches. If we only zoom in a little more, take all these little nuggets of wisdom about muscle fiber types and biomechanics and hormones and molecular signaling, we’ll have a complete Theory of Training that we can use to create Ultimate Workouts.

Modern thinking invokes the arcane powers of complexity science and chaos theory. You can’t take all the little pieces and add them up to get the complete picture. Information organizes into levels of meaning, and those levels aren’t easily crossed. To study the thing, you must study the thing itself.

Think of a fisherman’s net. The thick ropes knit together into a pattern, and it’s that pattern that matters. Remove any of one of the strands and you still have a net. Pull on one, and the whole thing moves in response. It makes no sense to talk about the net by talking about any one rope or thread within it. The “net” object is distinct from the ropes and threads that it’s built from.

The fitness field still thinks in reductionist terms. We see questions like “What’s more important: diet or training?” We hear things like “don’t spike your cortisol, cortisol is catabolic and bad”. What’s the best exercise, squats or deadlifts or bench press? Do deadlifts go on back day or leg day?

What’s more important to stay alive: your brain, your heart, or your kidneys? The answer isn’t important because the question itself makes no sense. You won’t live long missing any one of them, so why try to rank them — linearly — by importance? In a living body, the connections between the pieces are as important as the pieces.

A living body is nonlinear.

I want the fitness field to understand this perspective, what’s called the systems view of biology, and adopt it. The systems perspective challenges all the lovingly-held reductionist views, in which we can zoom in to One Single Cause and identify it as the source of all our woes. Insulin. Cortisol. Adrenal gland fatigue. Too many Type I muscle fibers. There’s one thing we can point to and say “that’s it”.

Looking back on my old articles, many of which were complaints or outright flame-posts, the one common factor I can see between all of them is that the “stupidity” I address always emerges as a consequence of reductionist thinking. Reductionism covers all kinds of fitness and nutritional kookery, from low-carb zealots to Paleo dieters to supplement hucksters and the dumbest of training ideas, even the new-age quackery of adrenal fatigue. These are all level-crossing offenses, with details from one level brought to a place they don’t belong.

In keeping with my desire to explain rather than condemn, I think this makes for an elegant explanation. It’s not stupidity as much as a conceptual or philosophical gulf. People simply aren’t looking at the problem from the right stance.

What are we missing out as coaches and trainers by not viewing life on its terms? Even the common ideas on recovery don’t quite work right when viewed reductively. Supercompensation is fine as a description over a small window of time, but how much does it hold you back if that’s the only way you can think about recovery? You’d certainly never be able to squat to a max every day, for months on end, and see constant improvements if recovery really worked like that.

Life is nonlinear. To understand it, we have to think nonlinearly.

The systems perspective extends to your workout programs and diets and all the detail-questions that you want to ask. Should I do this or that or that over there? Don’t ask if it’s right. Ask if it’s useful. Think in terms of real life cause and effect. If you do this at the gym, are you really going to die of cortisol poisoning? Are you really going to get fatter by eating fruit?

Systems theory tells us that we don’t know enough about those details to make level-crossing guesses, and anyone who claims they can is probably making it up. System theory, ironically, tells us to ignore all that mess and just go lift some weights and eat some food.

The only ideas that have any importance are the useful ideas. Train and eat and see what happens. Keep science in mind when testing it and, if it works, keep it. If it doesn’t, ditch it.