Wow, what a thread. I didn't hear a difference in my system (humble as it is: Micromega Stage 6, Creek passive volume control, Bryston 4B, Apogee Centaur Majors, Nu-Vista interconnects), but I wouldn't say that this phenomenon can't occur with someone else's CD and transport.
If I did hear this phenomena, however, I would have presumed that there is something wrong with my CD player and would be looking to repair or replace it. In other words, it doesn't sound enhanced after you restart the track, just less busticated. You might attribute this problem to error correction or buffer issues or rotation speed induced jitter, but it's a mechanincal/logical problem that should just be fixed, not a quaint little tweak that has you evoking arcane digital incantations with your remote keypad. It's not cool - like Fonzy hitting the Jukebox to make it work. It's not cool at all.
I don't believe that "bits are bits", but I do believe that that is a worthwhile goal. A digital signal exists in an analog physical reality, but we should expect the analog characteristics of the signal to be tempered in the engineering phase of a high-end CD player, obviating the need for resamplers, buffers and other such jitter boxen. A player ought to sound its best whether it's just started a track or starting it over.
Anyway, if this problem is something that you can live with and you think your player is a fine value despite it, then bully for you. Heck, when I press the open/close button on the remote for my Micromega, it does just that: the drawer opens and then it closes. I have to press the button really quick to get it to just open the drawer and gimme the disk, but I'm going to try to fix this problem.
If anyone really cares, they could absolutely convince me and others here of this transport phenomenon if they were to record the SPDIF output from the player to a computer under both conditions and then generate a graphical delta of the two output files. While you're at it, try it with a scratched-up disc and a pristine disk under both conditions - one might conclude whether it's the error correction system. I don't really expect anyone to do this, I'm just conjecturing, but I wonder what free software exists to do this test. You'd need an SPDIF input on your computer.