I've conducted several A/B tests back when I had nothing better to do and it was a very education experience. I used a recording of a harpsichord, using lame V5 mp3s, and came away with a statistically significant result (I remember taking 14 trials, though I don't remember what my actual true positive rate was)
After all of that, though I had double blind evidence I could hear a difference, I knew I didn't really care about that difference. I couldn't even describe the difference - it was almost subliminal. And that was with a harpsichord, which is a known difficult sound to encode. I've been a lot more comfortable knowing lossy encoding isn't destroying my music enjoyment since those tests.
After all of that, though I had double blind evidence I could hear a difference, I knew I didn't really care about that difference. I couldn't even describe the difference - it was almost subliminal. And that was with a harpsichord, which is a known difficult sound to encode. I've been a lot more comfortable knowing lossy encoding isn't destroying my music enjoyment since those tests.