Around the same time as this article, I remember participating in a double blind study that was trying to evaluate something similar for movies. They wanted to

Around the same time as this article, I remember participating in a double blind study that was trying to evaluate something similar for movies. They wanted to test how audible it was to "discerning" listeners, so they contacted local community orchestras, and I ended up in a little theater for a couple of hours with a handful of musicians watching and listening to movie clips over and over. The "control" was just plain old mp3 compression artifacts, which seems to match the effects observed in the case of the music watermark. In my case, there were definitely a few instances where things sounded a little off in some of the clips, but most of the time they sounded totally normal. It really made you wonder if you were just hearing things. I always wondered how accurate I was but they wouldn't share the results!