I'll try, but I remember this part from a long time ago. Please correct me if I got something wrong.
It was part of my graduate school systems neuroscience class, taught by Tom Thatch (
zl). One of our classes involved replicating the study and extending it by seeing if underhanded throws adapt (they don't in the same way, suggesting a different mechanism).
I also sort of overspoke. You can control most error-correction through top-down cortical mechanisms (including overhand throws), but such control may interfere with other mechanisms.
You can probably find it in a textbook, but it will take me some time to dig up the references, which is why I encouraged others to try the prism glasses.
Here's one, that shows how adaptation to overhand throws is limited and may relate to coordination between the eyes, trunk, and arm movements.
Here's one that shows how lesions to cerebellar areas containing climbing fibers impair error-correction for overhand throws.
Here's one that reviews the neuroscience literature on reaching movements and posterior parietal cortex.
There's an interesting discussion in the literature as to whether over-practiced overhand throwing (e.g. pitchers) leads to reduced sources of error for adaptation. If so, the timing of the release is the one most affected in over-practiced overhand throwing.
In any case, adaptation (i.e. compensation) to exogenous factors (e.g. wind) may be a different mental process for a knuckleball than a fastball.
EDITS for links, as I can find them.