In principal then, without regard to other pluses and minuses of i-contrast, if one had the sense before a shot that the scene would benefit from it then he'd end up with a cleaner jpg if enabled in the Menu for shooting rather than applying it in Review after the fact.
Probably, since we now know for sure that the shooting mode version isn't just applying the same post-processing as in playback. When you use i-contrast in shooting mode, the camera has a full 12 bit raw buffer to work with. If you do it on the jpeg, some precision and/or dynamic range must already have been lost.
Would the same logic apply to My Colors processing?
Hard to know for sure, but doing it in the shooting process certainly has the potential for a cleaner result.
When I get a chance I'm going to experiment to see if a difference between the two ways of applying i-contrast can be seen, but understanding the camera's operation is interesting as well.
Some review sites may have already done this, I remember reading some dpreview reviews where they tested the i-contrast, but I don't recall if they compared the playback and shooting versions.
I would suggest using a tripod to take identical scenes, difference the resulting images, and then crank up the contrast.