Original article By Alex Hutchinson, New Yorker Magazine, December 12, 2014

Roger Bannister at the end of his record-breaking mile run at Oxford, in 1954.
Photograph by Norman Potter

When, on a blustery day in Oxford in 1954, Roger Bannister ran the first sub-four-minute mile, measuring out the full capacity of his lungs and legs and collapsing across the finish line, he felt, as he later wrote, “like an exploded flashlight.” That was the feeling researchers were trying to evoke, recently, when they paid thirteen volunteers at Bangor University, in Wales, to pedal a stationary bike at a predetermined pace for as long as they could. Such “time to exhaustion” trials are a well-established method of measuring the limits of physical endurance, but in this case the experiment also had a hidden psychological component. As the cyclists pedalled, a screen in front of them periodically flashed images of happy or sad faces in imperceptible sixteen-millisecond bursts, ten to twenty times shorter than a typical blink. The cyclists who were shown sad faces rode, on average, twenty-two minutes and twenty-two seconds. Those who were shown happy faces rode for three minutes longer and reported less of a sense of exertion. In a second experiment, the researchers demonstrated that subliminal action words (go, lively) could boost a subject’s cycling performance by seventeen per cent over inaction words (toil, sleep).

The study, which was published last month (2013) in the journal Frontiers in Human Neuroscience, by Samuele Marcora, who heads the University of Kent’s Endurance Research Group, and two of his colleagues at Bangor, Anthony Blanchfield and James Hardy, is the latest salvo in an ongoing debate about the very nature of fatigue. According to one study, fatigue is “the inability of the contracting muscles to maintain the desired force.” But what causes it? Physiologists in the early twentieth century studied exhaustion by cutting off the hind legs of frogs and electrically stimulating the muscles over and over until they couldn’t contract anymore. In 1907, the Nobel Laureate Frederick Hopkins and one of his colleagues showed that the depleted frog muscles were bathed in lactic acid. Their experiment gave rise to an enduring—and incorrect—explanation for muscle failure; scientists now know that lactate, the form in which lactic acid occurs in the body, actually fuels muscular contraction rather than inhibiting it. Nevertheless, the view of fatigue as a mechanical breakdown has persisted. You max out your ability to pump oxygen, the acidity of your blood creeps up, and the neuromuscular signalling between your brain and your muscles gets weaker: one way or another, you hit a limit.

Marcora believes that this limit is probably never truly reached—that fatigue is simply a balance between effort and motivation, and that the decision to stop is a conscious choice rather than a mechanical failure. This, he says, is why factors that alter a person’s perception or motivation (monetary rewards, for example) can affect performance, even without any change in muscle capacity. In the subliminal experiments, the cyclists’ heart rates and lactate levels rose at the same rate no matter which faces they saw, indicating that nothing had changed from the neck down. Considerations like heat, hydration, and muscle conditioning, Marcora says, “are not unreal things, but their effect is mediated by perception of effort.” In other words, they don’t force you to slow down, as happens with the failing frog muscles in the petri dish; they cause you to want to slow down—a semantic difference, perhaps, but a significant one when it comes to testing the outer margins of human capability.

Marcora calls his theory the “psychobiological model.” It’s one of several attempts, in the past decade, to incorporate the brain into the understanding of endurance. This isn’t to say that previous generations of scientists discounted the mind’s influence on physical performance; as Michael Joyner, a physiologist at the Mayo Clinic, in Minnesota, told me, “There were people talking about this stuff in the eighteen-eighties, and coming up with very good thought experiments.” The Italian scientist Angelo Mosso, for instance, showed that the muscular endurance of two of his fellow physiology professors was diminished after they had given a series of lectures and oral exams to students. In the more than a century since, researchers have tried everything from hypnosis to curare toxin to alter the correspondence between mental effort and muscular output. But only recently have brain-imaging tools such as functional magnetic resonance imaging and electroencephalography become advanced enough to allow observation of the brain during intense exercise.

Consider, for example, the sub-two-hour marathon, which is starting to look like the contemporary equivalent of the four-minute mile. Back in 1991, Joyner published an influential paper in which he combined the upper observed limits of several aspects of running performance into a calculation of the fastest possible marathon time. He settled on 1:57:58, almost nine minutes faster than the world record at the time; a discrepancy that suggested, Joyner wrote, that “our level of knowledge about the determinants of human performance is inadequate.” At the Berlin Marathon, this September, a thirty-year-old Kenyan man named Dennis Kimetto, a former subsistence farmer who started competing internationally just three years ago, set a new world record, completing the race in a time of 2:02:57—still almost five minutes short of Joyner’s prediction.

Why does fatigue prevent athletic wonders like Kimetto from closing the five-minute gap? One possibility, proposed by Tim Noakes, a professor at the University of Cape Town, is that the brain has a subconscious safety mechanism that kicks in to prevent the body from getting too close to dangerous limits. Noakes calls this mechanism the “central governor.” In his view, fatigue is a protective emotion rather than a reflection of the body’s physiological state; its action is preëmptive and involuntary. That’s why, if you go for a run on a hot day, your pace is slower right from the start—not because you’re already overheating but because you might do so later. It’s possible, Noakes might argue, that what holds Kimetto back from a 1:57:48 marathon is hardwired self-preservation.

Some clues about how this protective circuitry may work are beginning to emerge. A study published in November by researchers at the University of Utah showed that leg exercise makes the arms tired, a brain-mediated phenomenon known as nonlocal fatigue—unless you inject the painkiller fentanyl into the spine to block nerve signals travelling upward from the legs, in which case the arms are unaffected. Other studies have shown that acetaminophen, the pain inhibitor that is used in Tylenol, can boost cycling performance by some two per cent. Last year, a group of Brazilian and international scientists used a weak electric current, directed at a region of their subjects’ brains that monitors effort and pain, to produce an improvement of around four per cent in cycling endurance. In each case, altering the brain’s ability to monitor distress signals from the body seemed to increase the level of fatigue that the central governor was willing to tolerate.

Marcora views the idea of a subconscious governor as unnecessarily complicated. He cites his subliminal-messaging study as a counterargument. Seeing a smiling face for a fraction of a second doesn’t change the fact that your pulse is, say, a hundred and eighty beats per minute and your blood-lactate concentration is seven millimoles per litre. It simply alters your conscious perception of those physiological extremes. In previous studies, Marcora has used caffeine gum, motivational self-talk, and what he calls “brain endurance training”—daily doses of cognitively challenging computer tasks—to tinker similarly with the feeling of exertion. (His initial interest in fatigue research was sparked by his mother’s struggles with unexplained fatigue after a kidney transplant, a common clinical occurrence in which perception is out of synch with physiology.)

As laboratory demonstrations of the brain’s role in endurance have accumulated, the sports world has begun its own experiments. In May, Red Bull brought four élite cyclists and triathletes and two dozen researchers, led by a team of neuroscientists from Weill Cornell Medical College and Burke Medical Research Institute, in New York, to its Santa Monica headquarters. There they explored the endurance-boosting potential of transcranial direct-current stimulation, the technique used in the Brazilian study. Members of the U.S. national BMX team are testing a program that was developed by neuroscientists at the University of California, San Diego, to encourage mindfulness. Marcora, meanwhile, is in discussion with Recon Instruments, which bills its Recon Jet as “the first heads-up display for sports”—a Google Glass-like contraption that is ideal for flashing subliminal encouragement.

Of course, coaches and athletes have long known to focus their efforts on the brain. I contacted Steve Magness, a cross-country coach at the University of Houston and the author of “The Science of Running: How to Find Your Limit and Train to Maximize Your Performance,” to ask him about Marcora’s study. It was the eve of the N.C.A.A. championships, and he was at a hotel in Indiana. “It’s intriguing that a seemingly subliminal cue could impact performance,” he told me in an e-mail. But he wasn’t surprised. “That’s what coaching is all about.” For months, Magness had been preparing his runners for the critical point in a race, the moment at which fatigue threatens to eclipse motivation. He planned to look his star runner in the eye the next morning and tell him that he was ready for the challenge. “That reinforcement from a coach, if it is genuine, I’m sure has a bigger psychological effect both consciously and subconsciously than presenting smiley faces,” he said.