I firmly believe that people play wrong notes for the right reason: namely, that their ear leads them to a better proposition than the one chosen by the composer, for reasons known to him or her alone.
Today I had a perfect example of how this is true. My computer-animator student with the extraordinary ear kept playing a certain wrong note in the opening of the second movement of the "Moonlight." Anyone with a background in music theory would be incapable of playing that note wrong. I know that I could not do so. I look at the score and know in advance what it will sound like.
Not so this young man. By the time his lesson was over I had learned about the choices Beethoven made much more than I would ever dreamed possible, all because of this young man's unbiased and accurate ear. It is not accurate in any theoretical or conceptual sense; it is simply accurate. There is a huge difference between the two.
Odd, because just this morning I was asking myself why that seemingly "normal" movement sits between two such giant challenges, one slow, the other agitated. Now, thanks to him, I get it.