Earlier quoted context omitted.
Depends on the thing. Children tend to be fine at self-directed learning, e.g. figuring out how to play a new video game. Some—most?—adults chafe at even the idea of engaging in an activity that requires self-directed learning. (To the point that we even have an English word for those adults that don't: "neophiles.")
Some children enjoy playing some new video games. (As do some adults.) Some children (and some adults) would refuse to play a video game that someone else told them to play. Their decisions and feelings are contingent on the context (including social context). If you slow down and have a conversation you can sometimes figure out what the specific emotional problem is, often involving shame or fear, and what triggers…
Well, yes, but coercion, extrinsic motivation, and general spite are all edge-cases, and outside the scope of what the core fact about "learning" that I'm trying to talk about here.
Let's look at a specific non-edge-case case: learning under self-directed intrinsically-motivated reward + time pressure — for example, learning to hunt/gather when stranded in the wilderness and hungry.
In situations like this, children do better than adults at deciding to begin experimenting in an attempt to learn how to survive. Adults will try to apply any skills they already have (however rusty), but will have a much harder time than children do in deciding to try to do things they haven't already ever been taught to do. And even when they do make that decision, they will repeatedly revoke it—i.e. give up much more quickly when a solution does not come easily to them. Even though, in education-theoretic terms, all the stars should be aligned in terms of motivating their learning process.
> The problem is seldom if ever “so-and-so is incapable of learning anything new”.
Of course not. That's the observed abstract pattern, not the reason. The reason for adults, AFAICT, is almost always pride/hubris — that is, the idea that they don't need to learn anything new, that there's no advantage in it to outweigh the learning cost, because they've got by just fine up to now without having learned it. (And also because learning is just plain harder for adults, since they have so many existing schemas that need to be broken+reshaped in the process of fully absorbing new ideas that don't fit the existing schema.)
This resistance comes almost always from adults who see themselves as having higher social status than whoever's trying to teach them the thing. They have no mental schema for learning from someone younger or more junior (or of a lower social class/caste) than themselves, since it's not something they've ever experienced earlier in their lives.
Instead, when someone in an inferior social positions knows a thing, and offers to teach it to this social-superior adult, the adult will instead try to manipulate their social inferior into doing the task for them any time they need it to be done — often under the pretext of demonstration for purpose of learning. But with no intention of actually learning, and a complete rejection on being tested for knowledge acquisition (because that would be a violation of the social dynamic!)
Note also that learning is separate from doing. I know many adults that are willing to do novel things (without really understanding what they're doing), and to memorize that specific series of actions by rote — but who have no desire to understand the more general skill behind the rote actions they're taking, such that they could perform related actions in any context other than the exactly memorized thing. In general, memorizing is easy, while learning a new skill is much, much harder. (And yet memorization without learning almost never "sticks", since there's no backing mental schema to encode the memorized steps in terms of.)