Maybe we are arguing from different sides of the same coin. Transfer of Learning seems to be an expansive topic that may have multiple, and possibly conflicting, definitions attributed to it. I first read of ToL in regards to the Dual n-Back test (http://www.gwern.net/DNB%20FAQ), which involves training in one very specific, fluid-intelligence reliant game with the hopes that improvements made within the context of the game will transfer out to more practical activities. And while I agree that this is rarely substantiated and that this particular game is probably useless, that doesn't mean that the definition is invalid, it just means that the Transfer of Learning, in this one case, is abortive.

The wikipedia article itself includes terms similar to yours: "meta-cognitive skills such as strategic knowledge, heuristics, self-monitoring skills, and self-regulation quickly became the road to learning and transfer." And Kalla's own example of 'reading/writing' would fall within the "vertical" transfer taxonomy, defined as "Knowledge of a previous topic is essential to acquire new knowledge." The ability to read is the sine qua non for . . . reading, obviously. I guess the broadest example is a bit tautological.

I am sorry for making such a fuss about it, it's just that LessWrong has the habit of coining its own terms even when there is no need, or when, such as in this case, there already exists a perfectly acceptable one.

But since I am just arguing off of wiki-knowledge, I may be completely wrong. Maybe ToL has a more univocal definition in the established psych community. Maybe you hail from there; and if so. . . please let me know how wrong I am.

Skill 'transference' is in fact the generally accepted term for kalla's topic. http://en.wikipedia.org/wiki/Transfer_of_learning

edit: Holy Kerning B-Man!