When a teacher points out that something specific is wrong, say, your kua (hip socket region) isn't open, three things become immediately imperative. What is the test? What is the measure? and What is the game?
Unfortunately the more common follow up is, just do this movement 10,000 times and you'll understand how it fits with everything else. That is a hook with no bait in my book. Every student knows on the first day of class that there is a danger of conditioning the wrong thing.
A test is often a result that can be felt or seen on oneself or on another person. Often times you can easily be trained to say Yes that's it, or No that's wrong, long before you can pass the test yourself. For a teacher to say you are doing something wrong, they themselves must be performing some kind of test. If you don't have access to this test you are training in the dark, metaphorically speaking. There are certainly valid arguments for training a student in the dark, but they are rare. (see below)
Just how open does your kua need to be? A measure is a way of deducing the degree to which one has some particular attribute, either how much or how little, under increasing amounts of pressure/movement, or time/speed. A simple example would be, do you have the structural integration for a head attack. The test is very simple, can you move someone with your head. The measure would involve adding pressure and force gradually such that you have no feeling of pain or compression in your neck, spine or other joints. At the point when you have compression or pain you are out of your range. We can be taught to see this in others too. A measure is a little different from a test. If a test is qualitative, a measure is quantitative.
Thirdly, and most important is a game. Without a game conditioning is slow and of questionable value. A game automatically enters the part of your brain that makes learning fun, and drills it deep into the place where you can access it instantly and automatically.
The more complex or difficult the attribute is, the more important it is to use a game to condition it. Otherwise you are just conditioning frustration! And it's great to play the game before the test or the measure, if you can. In that sense it is fine to have students learning in the (metaphoric) darkness as long as they understand the test and the measure eventually. But just giving a correction is, like I said, a hook with out a worm.
Even with a simple form correction, the measure can be as simple as, it looks like this, not like that. The test answers the question why it's an important attribute and/or shows some sort of structural function. It becomes a game when is happens with music, timing, rhythm and variations of style. It can also be conditioned in a two person form or a limited push hands exchange, or a resistance drill that just works that position as a game.
Teasing, jostling, tricking, improvising, dancing, funky-grooviness--these are some of the most important ways of learning, and all fall under the games category. Think: Games, the sky is the limit. A good teacher alternates between too serious and too much fun. (In my humble, yet irreverent, opinion.)
The test, the measure and the game are important for the student to know for almost any correction or principle. This is what we should expect from a good teacher, and a good teacher will expect us to ask for it too.
Traditionally, getting a beating at the moment of transmission may have had a powerful conditioning effect. Few people want that experience these days, so we need games.
I feel strongly about everything I just said above, I don't mean to diminish it, but there is another case to consider. A teacher may present a puzzle for the student to solve. Like, Okay, now figure out how I just did that. But puzzles in martial arts classes sometimes last decades. That seems wrong to me. Puzzles are great, but if the students aren't solving it, it's time for a new puzzle or a different game.
Why do puzzles sometimes last so long? In Asia, it is often considered an attack on the status of the teacher to ask a question. It is a sad self-defeating custom. Also sometimes students want to stay in awe, because they get a kind of devotional high from it. That's not very productive, even if it does pay the bills. Puzzles cross over into the realm of secrets (and magic).
Kids learn at about age four that if you want to be more interesting, you need to get good at keeping secrets. Even just looking like you are hiding a secret can magnetize people to you. But oh heavens, trading secrets is even more funner than fun.