I wish the study would be as simple as you hope. However...
1. Different headphones are said to break-in to different extents, especially phones with larger diaphragms, which are said to exhibit higher levels of break-in. It would, for example, be less than fruitful to break in, say, a Sennheiser PXC250. The larger, more sensitive headphones will likely provide the most satisfying results. Obviously, they're also going to be more expensive.
2. It would be useful to control for differences in batch. For example, some have noted differences between K701s separated by time, since they have variations in materials and, possibly, workmanship.
3. Once you've broken in the phones, the issue of judging sound is a potential problem. Even if you've deliberately switched diaphragms in some of your headphones, user ratings will make it difficult to get significant results; that is, to the <.05 level. The best solution would be take readings of the headphones themselves, and determine if changes are noted in the readings. (Studies on speakers showed differences pre- and post-break-in.) More than a few have claimed, however, that tests haven't yet devised to compare the full range of potential sound differences.
4. Given the controversial nature of the subject, it would of course be necessary to perform multiple studies on different headphones, as headphones are said to vary on their break-in. Although possibly expensive, this could really bolster the reliability of the results.
It's not my intention to trample on your idea, because the idea of validating break-in is quite laudable. But disagree on one point: testing headphone break-in isn't all that simple. To wit, a number of convincing studies have been performed to disprove cable break-in, but few - if any - on headphone break-in.
