Yesterday I began taking apart a splash of absurdity by Richard Gibbens, who claims that muscle power and something called the Central Governor are the chief determining factors in limiting long-distance running performance. I left off by explaining in physiological terms why Gibbens' claim that a runner's race times at various distances can be neatly predicted from his or her all-out pace is wrong, and now I'll turn to some data.
Gibbens, after making his claim about extrapolating sprint times to long-distance times, gives us numbers to play with:
"For example if we take your 100m sprint time to be your top speed (it's not exactly, but it is close enough for our purposes) then with proper training you can run...
"400 meters at 96% of your top speed
800 meters at 87% of your top speed
3 kilometers at 74% of your top speed
5 kilometers at 71% of your top speed
10 kilometers at 68% of your top speed
10 miles at 66% of your top speed
1/2 marathon at 65% of your top speed
Marathon at 61% of your top speed"
Gibbens is badly, empirically wrong. I'll start by using a certain non-professional runner as an example.
In terms of all-out speed, this guy -- decent in high school and a solid competitor on a regional level since college -- is neither especially fast nor especially slow compared to others in his range in his best events (10 miles and above). To be generous, his fastest all-out 100 meters in his peak distance-running condition of 2004 might have been 13.5 seconds, give or take a tenth or two. Using Gibbens' model, he should be able to run 5K at a pace of (13.5/0.71) = 19.0 seconds per 100 meters, or 15:50. This is slow by almost a minute. His predicted 10-mile, half-marathon, and marathon times of 54:52, 1:13:02, and 2:35:40 are off by 3:20, 4:33, and 11:23 respectively (and would be off by more were the longer events contested on a track, as are the sprints).
We can go the other way too. My 10-mile, half-marathon, and marathon bests predict 400-meter times in the 52-second range, which is absurd. Watching him sprint is like watching a duck with its ass set on fire, only less elegant and noisy.
Now let's look at some real runners. To cut to the chase, Gibbens' chart predicts that any male who can run the 100 in 10.5 seconds (a top-notch but not unusual high-school time) or faster could shatter all existing world worlds at the distance events. Perhaps Gibbens would pull out his "exceptions" card when looking at such athletes, but even if we move the upper bound to 11.0 seconds, we find that these guys should be able to break 27:00 for 10K, which no man on earth did until about 15 years ago.
On the other hand, Pete Pfitzinger, my own coach and a two-time Olympian with a personal marathon best of 2:10:43, never ran faster than 55.5 seconds for 400 meters, meaning that he's no faster all-out than the first runner in this series of examples. Alberto Salazar was reportedly no faster but once held American records in the 10K and the marathon. And given the number of sub-10.00 100-meter performances out there, someone should have brought the marathon world record (2:04:55, Paul Tergat) well under two hours by now.
I could go on all day, but the point is already clear -- the conclusion of Gibbens' entire second section, Other physiological systems influence how hard your muscles can work, is patently false. He states:
"The prediction table is telling us what our potential is in the marathon with proper training. Without the proper training our physiological systems and factors won't be fully conditioned and prepared to work at the level required to run a marathon in 3 hours, nor will our muscles be conditioned to perform optimally and, therefore, our performance will be less than predicted.
"The good news is that proper training can mostly remove these physiological limitations so that you are able to tap into the full potential of your muscles."
Inother words, he asserts that the best distance runners in the world are badly undertrained, given that their times don't conform to his chart.
Now, if you were to assess a sample of "average" runners, you probably would see a pretty close correlation between sprint times and long-distance times. But the fact of the matter is that most people who run, and run road races, just don't train very hard. It's a hobby for most, and understandably people aren't apt to start running 10 or 15 miles a day when they have full-time jobs, families, and little hope in any case of making it to even the top local level. But bearing in mind that Gibbens purports to maintain a site that's all about maximizing performance, we need to look strictly at physiological factors, not psychosocial ones.
There are training implications to be observed here. If I'm advising an aspiring marathoner known to be heavy in the fast-twitch area, he or she is unlikely to respond as well to an overabundance of "lactate-threshold" training (basically, 10K to half-marathon race-pace training) as would someone more naturally inclined to distance running. This is because the fast-twitch fibers are best prepared to participate in long-distance events by depleting them of glycogen at a modest pace and forcing them to adapt to oxidative metabolism as best they can. That way, late in a marathon, when their more modest complement of Type I fibers are more or less exhausted, their bread-and-butter Type I fibers can "kick in" and help out. They can handle a heavier burden of faster "interval"-style training than can true distance types; in a nutshell, it's a matter of training to personal strengths while accommodating genetic shortcomings in a way that suits the runner's chosen (longer) distance.
Before moving to the final part of Gibbens' wayward essay, understand that he's completely out to lunch in decreeing that performance limitations are intrinsic to muscle anyway. One can state that the muscles ultimately move the limbs and that their failure therefore spells performance failure, but this is akin to saying that it's the wheels and axle and not the engine that really determine the power and speed of an automobile. Unless the heart is trained to pump a maximal amount of blood (and hence oxygen) to the working muscles, it doesn't matter how powerful, by any measure, those muscles are. And this kind of adaptation is gained by doing exactly what Gibbens rails against at every opportunity: logging a lot of mileage, a surprising fraction of it at a very comfortable pace.
That hummingbird I mentioned earlier? Its maximum heart rate is around 1,260 beats a minute, over six times that of a normal human being.
Finally, Gibbens goes on to discuss a favorite concept of his, the "Central Governor" model of exercise limitation originally introduced by Tim Noakes. Noakes is a South African physician whose tome The Lore of Running is a fantastic resource. He is also regarded with more than a shade of skepticism by other learned runners, which doesn't categorically make his farther-reaching ideas wrong, but explains Gibbens' affinity for his yen for his more unworkable notions.
The central governor idea implies that when people get really tired but continue to push, their muscles or other organs don't max out, their subconscious mind puts a clamp on the whole circus in order to prevent tissue damage. In other words, he says, the brain doesn't merely survey the wreckage but attempts to forestall it.
Here we see why Gibbens introduced this outburst by dismissing the significance of "research study dissections" and "in-depth discussion of intricate physiological factors" and elected instead to "explain how it works in simple, straightforward language": He doesn't have any data for his central governor. Noakes has published a few papers dealing with the model (here's one), but not only has it not been applied to runners in the real world, it's hard to envision how this would even be possible in the first place or what the utility would be. This, for now, like the endurance world's less baroque, less dizzying version of string theory.
I shouldn't need to reproduce Gibbens' words here to convince anyone of this: The central governor, whatever its real or imagined role on race day, is a meaningless concept in the context of training to maximally improve individual performance. Even if it were true that a "central governor" is responsible for anticipating incipient fatigue and, at critical workloads, putting the brakes on the muscles, heart and lungs, it wouldn't matter, because with improvements gained in training, this governor would simply recalibrate itself to a higher level. It's an anti-Occam's razor construct to the extreme.
Not only that, but the governor premise is, in key and fundamental ways, not borne out by laboratory and real-life experience. The central governor would posit a fall in heart rate, decreased glucose release, decreased ventilation, and other physiological parameters in extremis -- otherwise there would be no way to detect such a governor. Even if these things happened, it would have to be determined that cortical rather than local mediating factors were responsible. Also, there are numerous instances of top athletes running themselves more or less into the ground. The central governor appears to be a rather slipshod job of protecting us, given the relatively small number of "false alarms" in the form of race DNFs ("did not finish") and the notable number of people who wind up on the side of the road in medical tents owing to dehydration, hyperthermia, glycogen depletion, and other derangements. When you don't have any fuel left to burn or sufficient blood to pump throughout your vascular tree, it doesn't matter what is allegedly going on upstairs; your performance goes friggo-bazoo.
If Gibbens were merely saying that the brain plays a role in recognizing "red-lining" and helping set a proper pace given the race length, weather conditions, course difficulty, and so on, it would be different (and not at all new). Of course the brain (including the conscious mind) determines a runner's behavior; we're not just creatures who run pell-mell as hard as we can until we drop, unless we're running from something deadly, like a tiger or a Power Running article. But this, of course, isn't what Gibbens is saying at all.
All a central governor posits, then, is that the human body cannot exceed its limits, which gloriously trivial. One might propose any number of other mechanisms responsible for shutting down in extremis: the length of one's fingernails on race morning, the number of Harry Potter books read in the past three months, the whims of a baleful three-headed god, or the central governor's subversive lieutenant governor and his peripheral mistress, shacking up in the hyoid bone. The fact remains that reaching that breaking point is strictly a function of the physical capacity of the runner in question, a capacity determined not by a governor but merely observed by it at all meaningful points.
Finally, if a central governor limits performance, what are the implications? Find a way to override it to make endurance events dangerous by playing Tim Allen and, in effect, putting a jet engine in a dishwasher?
What can be said about this model for certain is that whatever its representation, its polar opposite exists in the head of Richard Gibbens when it comes to writing about distance running. The man takes the outer limits of sensible speculation and discourse and takes care not to remain within in them, but to forever contemplate beyond their boundaries.





Comments
This, Mr. Beck, is what keeps me coming back for more.
Posted by: Bill from Dover | February 18, 2007 5:52 PM
Kevin,
Whilst I appreciate that the Central Governor theory has gaping inconsistencies, I believe that you are being somewhat (unintentionally?) harsh on Prof. Noakes. Were you not aware that he advised the Gert Thys training group? Certainly Gert, Tobias, and the late Ian are proof that Noakes has a bit of a clue. As regards the Central Governor theory itself, I would have thought that a scientist might at least be open to exploring the idea, even with extreme skepticism. You only have to look at Eddington's opposition to the "Chandrasekar Limit" proposal to realise the dangers inherent in out-of-hand dismissal of new ideas. I'm not sure that the drop-out rate at mass-market marathons is an adequate test of the theory, for the obvious reason that a bunch of undertrained people attempting to complete a serious physical task may have other physical defects which over-ride the governor? As far as I'm concerned the jury is out on the idea, but my strongest opinion is that the title of "Central Governor" may be a misnomer.
Posted by: hopper3011 | February 19, 2007 5:02 AM
hopper --
Yes, in my emphasis on trampling Gibbens' ideas I caught Noakes in the crossfire to an extent I did not really intend. I love his book (though it is pleasantly unwieldy) and in no way mean to say he has done more harm (or any harm, really) than good. It is Gibbens' own explicit belief that Noakes' findings, however they pan out, lend support to the wackiness on his site I find unsavory.
As you suggest, attempting to divine a whole lot of anything from the exploits of a seriously undertrained bunch of marathon particpants is usually hazardous. I am skeptical, though, that even if the governor notion turns out to be dead on that it will have any practical ramifications in terms of runners' training. Whatever the upper performance bound proves to be in any of us, there will always remain are optimal ways to get to the point of testing them against physiological limits, and Gibbens' ideas about emphasizing low mileage and raw muscular power simply do not fit in.
Posted by: Kevin Beck | February 19, 2007 5:15 AM
Kevin,
I tend to agree with the idea that the Central Governor theory may be more use as a description of what happens to the body during training and/or racing, rather than as a tool with which to plan schedules for said training and/or racing. Noakes needs to refine and correct the theory before it can become more than that.
At least we agree that Gibbens is a fucking idiot.
Posted by: hopper3011 | February 19, 2007 6:04 AM
The central governor model tries to explain some fairly obvious observations. Like why athletes begin races of different distances at different paces and why they can speed up at the finish - the so-called endspurt. I am not aware that any other model of how the body acts during running, can adequately explain these simple observations. Certainly these simple observations disprove the theory that "peripheral fatigue" determines exercise performance.
The model does not predict that the brain suddenly decides just before fatigue develops that it must shut everything down in order to prevent a catastrophic failure. Rather it predicts that this decision is taken either before or shortly after the beginning of exercise and that this "decision" is constantly updated as the exercise continues. Indeed this is a fundamental feature of how the body normally works - there is a feedforward component (based on previous experience - ie the speed that you might start a running race) and this is complimented and undated continuously as the exercise continues and this gives the subtle variations in pace that occur as the exercise continues.
Our paper in the Journal of Physiology (R Tucker et al. 2006) nicely shows how during exercise in the heat, the brain alters the pacing strategy in order to insure that the rate of heat production during exercise is regulated usually within the first few minutes of exercise in order that a catastrophic elevation in body temperature does not occur during the expected duration of the exercise bout. This study clearly shows that exercise performance is regulated "in anticipation" by a complex intelligent system, the goal of which is self-preservation and not reaching the finishing line as fast as is possible. Humans evolved to survive, not to run themselves to death and the central governor model nicely explains how this happens.
The human brain insures that everything we do is carefully planned even though we may not be aware of the complexity of that planning process and how it is organized. We were most certainly not designed purely to run as fast as possible until we collapse with oxygen-starved muscles. If such a regulation ever existed in any mammalian species, evolution would have selected it out, leaving as survivors only those able to regulate their exercise intensity within safe limits that would not damage the body.
Posted by: Timothy Noakes | November 12, 2007 10:30 AM
The question remains: if the "central governor" is, as you state, the description of a systemic response to stressors outside the control of the athlete, how would this description of involuntary physical response alter an athlete's training?
Posted by: hopper3011 | November 12, 2007 11:06 AM
The idea of a "central governor" is completely bogus. The fact is, that there are many ATP consuming pathways and those pathways are regulated extremely well. Not all of those ATP pathways need to be consuming ATP at every second. Some can be put off until "later". How much later? That depends on the pathway.
Lets consider something like mitochondria biogenesis. That takes ATP to synthesize proteins and requires the transition metals tied up in the enzymes of the old mitochondria, Fe, Cu, Mn, Zn. Mitochondria are reprocessed via autophagy, which dissociates them and recycles the metals. In the rat CNS, mitochondria have a half life of about a month. When would be a "good" time to recyle mitochondria? Not during a time of high ATP consumption. Old and "tired" mitochondria have to be disposed of to free up the metals and to make physical space for the new ones. That reduces potential ATP production between when the old ones are destroyed and the new ones are synthesized. In rats, mitochondria biogenesis occurs during their inactive time. In humans it likely occurs during sleep.
How long could you put off mitochondria biogenesis? Normally it is put off until the next evening. You could probably sustain putting it off for a few days, maybe a week, but certainly not a month or a year. That is, put it off without adverse effects.
We know there is a physiological effect known as ischemic preconditioning, where a brief episode of ischemia induces a physiological state that is protective against longer periods of ischemia. The low ATP of the ischemic state induces ATP conservation pathways which reduce ATP consumption. Some ATP consuming pathways are turned off. Obviously the pathways that are turned off are not completely dispensable, otherwise organisms would have evolved to do without them. They are necessaary, just not during the period that ischemic preconditioning is effective for.
I suggest that things like mitochondria biogenesis are turned off during ischemic preconditioning. It would make sense to turn off anything that couldn't contribute to more ATP during the ATP crisis the cell is experiencing.
I talk about this in my blogs on the placebo effect and on acute psychosis. In an extreme "fight or flight" state, an organism can run itself to death. This is a "feature". Any injury short of death is infinitely better than being caught and eaten by a bear. Pain is simply the signal from what ever organ system is being exerted that essential systems are being shut down to free up more ATP for voluntary pathways. The "optimum" control system shuts off the longest time constant pathways first, as ATP consumption continues, shorter and shorter time constant pathways are shut down. When the time horizon of the pathways being shut down reaches the present is when the organism drops dead from exhaustion.
Tim, organisms can run themselves to exhaustion. Normally it is quite painful, but that is what the stimulent drugs of abuse do, they mask the pain and turn off the non-voluntary pathways that consume ATP leaving more for consumption by voluntary pathways. That is why it is a lot easier to run yourself to death while high on stimulents. Stimulents of abuse invoke the same delusion as the "runner's high", the delusion that you are not tired. A very useful delusion when an organism is running from a predator and to be caught is to be killed and eaten.
It should be remembered that the first Marathon, the runner did drop dead just as he finished it.
Posted by: daedalus2u | November 12, 2007 3:49 PM