with the London Olympics just four months away, the best athletes in the world are heading for the hills – or, alternately, crawling into hermetically sealed sleeping tents and pressurized houses.

High-altitude training, aimed at boosting your blood's oxygen-carrying capacity, is seen as virtually compulsory for elite endurance athletes in sports like running, swimming and cycling, and has also caught on in other sports like boxing and rugby. But two new studies suggest that the benefits of training in thin air may be more mental than physical – if they exist at all.

The basic theory of altitude training is simple: Your body responds to the shortage of oxygen at high altitudes by enhancing its ability to transport oxygen from the lungs to muscles. Back at sea level, this adaptation boosts your endurance.

But there's a hitch: The lack of oxygen means you can't work out as intensely at altitude. Runners will run more slowly at any given effort level. For example, their lungs may get a better workout, but their legs will get weaker. The solution, first proposed in the 1990s, is called "live high, train low," or LHTL. It calls for spending as many of your waking and sleeping hours as possible at high altitude, but descending to a lower altitude for a few hours each day to do your workouts.

One option for LHTL is to find a location at the right altitude, within a short drive of lower-altitude training venues. Flagstaff, Ariz., is a popular option; Canada's London-bound marathoners, Reid Coolsaet and Eric Gillis, are spending a month there right now. Another option is to use an altitude tent (or in places like the Australian Institute of Sport and Nike's Oregon training centres, full-scale altitude houses), pressurized to allow athletes to sleep at the oxygen equivalent of thousands of metres above sea level, then step out the door for a sea-level workout.

Initial studies showed that LHTL worked, boosting aerobic performance in runners and other endurance athletes. But there was a problem: The studies weren't blinded or placebo-controlled, meaning everyone knew who was supposed to get better and who wasn't – knowledge that could affect how hard the subjects trained during the study period, and how motivated they were during the performance tests.

Still, few people had any doubt that LHTL worked. When Carsten Lundby, a researcher at the University of Zurich's Center for Integrative Human Physiology, began organizing a meticulous double-blinded, placebo-controlled study of LHTL, his main goal was to pinpoint exactly which physiological mechanisms were responsible for LHTL's performance boost.

The study, which was published in the January issue of the Journal of Applied Physiology, involved 16 well-trained cyclists, who spent a total of eight weeks at a training centre on the French-Swiss border. For four of those weeks, they spent 16 hours a day in "hypoxic" rooms in which the effective altitude could be adjusted. Ten rooms were kept at 3,000 metres above sea level, while the other six were kept at less than 1,200 metres. Neither the athletes nor the scientists responsible for testing knew which rooms were which, and the athletes were unable to guess at the end of the study whether they'd been living "high" or "low."

Shockingly, the researchers were unable to find any differences, either in blood measurements like hemoglobin mass or in cycling performance, between the two groups either during or after the training period. "It was a surprise to us," Dr. Lundby admits.

Critics were quick to dismiss the results – after all, the real-world experience of elite athletes and coaches is more relevant than artificially controlled lab studies, they argued. But that makes the results of another study, from researchers at the Australian Institute of Sport working with the country's world-famous swim team, harder to ignore.

The study, which will appear in a forthcoming issue of the European Journal of Applied Physiology, compared 37 elite swimmers divided into three groups. One group completed a three-week period of LHTL using the institute's altitude house; a second group did three weeks of "classic" (i.e. live high, train high) altitude training in the mountains of Spain or Arizona; and a third group did no altitude training.

Once again, there were no differences in race performance among the three groups, either immediately after altitude training or throughout the season. If anything, the altitude-trained groups swam slightly slower for up to a week following the training camp.

Again, there are several possible shortcomings in the study. For example, the swimmers were tested over 100 and 200 metres, which are short distances for which any gains in endurance would make only a minor difference.

But the question remains, if the gains from altitude training are either marginal or unreliable, why are so many coaches and athletes convinced it works? One possible explanation is the "training camp effect": If you leave your routine worries and stresses behind, and head to a remote mountain village where all you have to do is train, eat and rest, is it any wonder that you return three weeks later feeling fitter and stronger?

While the scientific debate about altitude training is far from settled, that insight underlies Dr. Lundby's advice to athletes trying to arrange LHTL training camps. "Spend the money on other, less complicated training camps," he says.

The message applies equally to recreational athletes seeking an edge in their training. Time and freedom from distraction aren't easy for anyone to secure, but they're more accessible for most people than altitude tents and mountain getaways – and they may be just as effective.