Can species-typical qualitative changes in learning (with ontogeny) be related to ANYTHING (distinct and directly observable in the environment and in the response) of the individual human?
If "NO", that leaves a very big open question, doesn't it? [(Please, don't think "no".)]
Is that "OK"? [(Please, don't think "yes".)]
This is a basic question behind MANY other major, key questions (and this is just one class of ramifications which makes the question important), so it's not OK not to have an answer.
Is it "OK" just to make-up (or work-to-"find" and indicate statistically) basically very tangential or just-supposed and UN-clearly associated/related "behavior/response", "'behavior'-and-'response'" not ever clearly of/in the functioning OF the individual human, nor even CLEARLY related to anything (directly observable) there? AND, _then_ call those 'explanations' for major behaviors, and indicate that it is just THOSE all those interested in psychology should attend to, deliberate on, and try to "study" further?
[(Please, don't think "yes", "yes".)]
IN ANY CASE: The only "therapy" I can recommend is the ENTIRE BOOK on this topic which I, in effect, wrote here in Qs and As (I gave -- about 400 essays), all here on researchgate in the last 11 months. Psychology theorists, recognizing you have a problem, read all that and see if the direct answer you end up finding there (and which is otherwise consistently and in many ways justified by the REST of the perspective) helps -- OR if it "triggers" in you another direct answer, with an empirical way to establish that answer: based on directly observable overt behaviors (responses) shown by an individual human AND (at the same time) aspects of the environment (some directly relevant, clearly observable parts/aspects of that organism's current environment) -- that the individual human is THERE directly responding to. (In short, in other words, provide an outline for YOUR answer of a way to answer it empirically in the science of behavior, aka "psychology" -- completely empirical, at the very core (completely, as just described), and involving and requiring, for each and any significant increment of behavior change observed or inferred, overt behavior patterns and aspects-of-the-environment ONLY.