That's really neat, although mouse memory is much simpler than human memory. Those furry little bastards don't have language, so they rely on implicit associative learning much more than we do. It's impressive as hell that these scientists can knock that out in mice, but a human memory is so much more complex, and so much more interlinked to other explicit and conscious experiences that we've had that I can't see how on earth you could possibly remove a single memory from a human mind. Or rather, I don't see how you could do it without also f**king up lots of other stuff. If you were in a traumatic car accident and your memory of it was removed, would you also have to remove the memory of spending two weeks in hospital? And if so, how would you account for the big and freaky two-week period of amnesia, or the serious injuries you were recovering from? And how would you explain how you met the cool other people in the rehab center, or why you no longer have a car? And as other people have said, if you happened to be a crap driver, it's surely in your best interest to remember that so you do something about it. Like not drive.
Really cool story though.