Here's an interesting graph to consider:
It comes from a paper by Nigel Stepto and his colleagues, then at the University of Cape Town, back in 1999. This particular version of it is redrawn in a review on physiological adaptations to high-intensity interval training (HIT) by Martin Gibala in the Journal of Physiology last year—Gibala mentioned this figure in a recent talk, and I thought it was interesting enough that it merited a blog post even though it's more than a decade old.
The study involved taking 20 cyclists, assigning each of them to one of five different interval training routines, twice a week for three weeks. The graph above shows how much each cyclist (represented by dots) improved in a 40-km time trial. The interval routines were:
More From Runner's World
(1) 12 x 30s @175% of peak sustained power (calculated in an incremental test to exhaustion), with 4:30 minutes rest
(2) 12 x 1:00 @100% with 4:00
(3) 12 x 2:00 @ 90% with 3:00
(4) 8 x 4:00 @85% with 1:30
(5) 4 x 8:00 @80% with 1:00
The result that jumps out is that the 30-second and the 4-minute intervals produced significant improvements, and the other ones didn't do that well. The success of the 4:00 intervals isn't surprising: that pace is very close to the 40-km time-trial pace, so it has the advantage of specificity. It's also consistent with findings that 3-5 minutes is a bit of a sweet spot for interval length in other studies.
The fact that the super-short and intense intervals with long rest were also effective (while the superficially similar 1:00 intervals weren't) is a bit more surprising. Stepto et al. speculate that this may indicate that a different mechanism for improved endurance performance is at work here, for example improved muscle buffering capacity instead of mitochondrial or central cardiovascular adaptations. This is an interesting idea (and certainly consistent with what many coaches and athletes do)—that you should spend time doing what are often referred to as VO2max intervals, but also spend some time doing very short and intense work, because it offers a separate means of improvement.
There are some obvious caveats here. Maybe the 1:00 and 2:00 intervals weren't effective because the intensity chosen was too low, or the rest chosen was too long. I have a hard time believing that's not at least partly what explains that funny-looking curve. But at the same time, I'm open to the possibility that this graph is telling us something real. When I do short intervals, I often have the tendency to keep the rest very short too, so that the workout ends up being just another “flavor” of the stimulus offered by my usual longer intervals. This is a reminder that there may be benefits taking to a deliberately different approach to short intervals, where the long rests allow you to access a completely different range of intensity.
* * *
Read the Sweat Science book, and follow the latest posts via Twitter, Facebook, or RSS.
