Alright, look, the reason I won't nominate bear bells for removal is FIRST, they don't pose an obvious and immediate threat to the user (unlike The Extractor). Second, we don't don't have any definitive data (proof) one or the other.
In '89, there was a study (experiment) that claimed bells do work. The pointed to the fact that a number of people who were charged were not wearing bells (sounds like faulty critical thinking, but I'll see if I can find it).
Then, in 2000, a guy in Alaska rigged a bell with fishing line and rang it on 15 occasions when one to several bears came near. NOT ONE bear reacted AT ALL! Not even when he rang it vigorously like an alarm!! (I'll look for that one, too). They apparently treated the sound as merely ambiant noise.
Now, if the bell's perceived (or imagined) value is in "warning" the bear you're coming, theoretically so they're not surprised and will move on, then the latter study casts some serious doubts!
THE MAN IN THE MONKEY SUIT
I have a personal theory why the bells didn't work: there was once a study where people were asked to watch a video of a basketball game and told to count how many times the ball changed hands. At one point, a MAN in a monkey suit appeared on one side, walked THROUGH the players, and proceeded out the other side.
When asked, many of the people had NO memory of the man in the monkey suit. Even when the video was replayed, they found it hard to believe.
Point being, when you're looking, or listening, for a particular thing, you may unconsciously ignore the obvious. You probably experience this ready; narrowly missing a motorcycle because you "didn't see him", etc.
I suspect bears are listening for snapping twigs, breaking branches, rustling leaves, and VOCALIZATIONS (voices, grunts, growls, etc.)
Therefore, while warning bears off in advance is prudent, bells may be merely a way to make the PERSON feel better.