Please note the neutrality of this post is lost. It goes into the ideas of the believers, the non-believers and their doubt, then an attack at the doubt (their main argument against burn in). This is not a neutral post, it wasn't meant to be.
I do believe that all IEMs burn in (all headphones too for that matter). However, I'm a firm believer that it isn't always possible for everyone to hear the change in sound over time (EG the change in the change of sound). This is my main argument against people who say "BAs don't burn in" or "I just don't hear it". The fact of the matter is that everyone's ear has different sensitivity, and that fact needs to be accepted in order for this argument to work at all. With BAs, the change over time (change in change in sound) is so small that people will normally not notice it. However, even listening to my HF2s, the change in sound was evident as the bass smoothened out and the treble settled down. And again with my Apple IEMs, the bass punch went away over time, but the body and texture came in; those were the main changes I heard. With my PFE 232s, I could definitely hear the bass settle down (bass body mainly) while the treble thinned out and the midrange gained more energy.
Believers usually give the evidence through their observations which match up with everyone elses; even people who have never heard the term burn in before. That second part is crucial since it becomes a control group. They essentially are able to predict (not out of luck) what will happen with certain drivers. The second piece of evidence they normally give are small tests that are done measuring frequency curves before and after.
Non-believers will go out and deny this evidence by just saying it's a placebo effect, and a psychological effect that plays into why people hear the same things. They almost defeat the entire first argument using this up until they can't explain why total noobs are able to hear these changes; not only hear the changes, but the descriptions match those of people with audio experience that believe in burn in. Their second argument attacks the frequency graphs stating that they are inaccurate. The main argument against this one is that change has been spotted time and time again with each frequency graph. However, all of these are doubts. What ends up happening is that the believers give the non-believers information, they doubt that information. They give more information. They doubt that information too. Once more, they give even more observations, once again it's doubted. Not only are these observations each done scientifically, they are done under different circumstances (different headphones, different drivers, with/without crossovers, different number of drivers, different brands, etc). They keep giving more and more observations... They keep getting doubted, you get the cycle. Who do we end it? The philosopher David Hume wrote a something that sums it up pretty nicely. A doubter must relinquish his stance on a subject once it's been verified objectively multiple times over. Since at this point, his stance becomes an absurdity since the other stance has been proven objectively. The skeptic Nietzsche also says something about doubting. Something along the lines of, a doubt without an end is not a doubt at all. These non-believers are basically doing just that. They keep doubting, and doubting, and doubting. It doesn't end. So at this point, they aren't doubting, thusly, their argument returns void.
Non-believers do bring in one valid argument back into play though, why don't they hear the difference? And I explained this one above, it is a universally, objectified statement as well. Not everyone can hear the change in the change of sound due to the differences in people's hearing sensitivity. As always, they will doubt this as well, even though they prove it themselves when they make the statements, "I can't hear the change" while others can hear this change. I know someone's going to try to attacking this and somehow say it's "invalid" or "not accurate enough" or somehow some way incorrect. This argument along with the test that shows than total noobs of audio can hear these differences, despite never knowing about this phenomenon called burn in.
Now, does this mean I'm right about burn in? Doubters, who aren't doubters any longer, would say no. I do see this as enough evidence to disprove the doubter's stance (of doubt). By breaking that one, it leaves no arguments against any of the objective evidence that the believers have shown.