Take a fairly standard D&D-style fantasy world. Vaguely medieval-esque technology, monsters, magic, etc.

Now add the following law of nature. For the purposes of this discussion, please treat it as an absolute law, with no loopholes or gaming possible. But one where the exact contours and mechanics aren't known more than anecdotally.

Growth requires risk. And growth scales with risk. Risk is the expectation value of danger: <chance of failure>*<cost of failure>.
Effectively, one makes personal progress in getting stronger by taking on risk. And it has to be genuine risk--doing something risky (high cost of failure) but with very low probability of failure, or taking safeguards to either reduce the cost or reduce the probability of failure diminishes your return.

A lock-pick practicing on his lock at the safehouse would see minimal change once he is over that first hump. But put him in a dungeon with a guard expected any second and give him the same lock would see more growth.

Of course the two (risk and reward) don't scale identically--there's a sweet spot, although a changing one. A beginning adventuring team could, in theory get huge gains by taking down an elder dragon (very very high risk). But most likely they'll just die, and you can't grow if you're dead. On the other hand, a wizard doing research in his lab, all carefully controlled, or a wizard who "plays it safe" by sending out simulacrums or ice assassins from his well-protected demiplane would see minimal, if any growth.

This is a thought I had to explain why adventurers seem to grow more powerful more quickly than most. Basically, by taking on higher risk than normal, they get higher growth. And those who "play it safe" do so in exchange for fast growth. On the other hand, I'd expect that you'd see a lot more people burn out quickly by taking on more risk than they could handle. And a higher culture of risk taking, at least in parts of the culture.