What I mean is, has anyone taken the time to formally think over and analyze what kinds of in-dream behavior are conceivably possible, and categorize different types of behavior?

Clearly, the human mind is capable of a lot; hypnosis shows that you can cause yourself to hallucinate, or at least believe you're hallucinating (hmm, any hypnosis experts know if cognitive science has made the distinction?). The ability to access a part of your mind/consciousness in a unique way may open the way for a lot of potentially very useful or even (subjectively speaking) wonderful experiences, and you don't have to follow any kind of mysticism to agree the idea of that isn't completely far-fetched.

So, my question is, has anyone thought about this developed a framework for what kinds of in-dream exercises could be considered? Involving the extent to which a person can affect the dream world and themselves and be creative, the ways in which a person's state of mind may differ and how they might take advantage of how the mind differs, and what kinds of sensations are accessible in dreams, for instance.

I'll note that in the Research forum I'm seeing a lot of research about how to approach lucid dreams, some more research about testing for supernatural aspects of dreaming, and just a few interesting topics about in-dream activities, such as derlacher's post about improving motor skills (not sure if he ever really completed that, though?).

(Just to be clear, again, when I say ability I /don't/ mean OBEs or psychic powers, I mean practical applications of affecting your mind within the context of itelf, if that makes any sense.)