I's like to add my 0.02 to this. Not because I know anything, but just because I have 2 cents
Please allow me to assert that speaker drivers are motors by virtue of the fact that they convert electricity into motion. Therefore, they generate heat from friction and will wear. No doubt about it. If you graphed the wear over time, it might not be much, it might not be linear, but it would be there throughout the product life-cycle. To some degree, however minor, everything with moving parts will exhibit "break-in" performance changes. So the argument is never really about speaker break-in. Physics demand that.
The argument is about hearing that break-in performance change.
And when that's the argument, it's no longer a single-body problem (measure the driver and know.) Now it's at least a three body problem: The speaker, the ear (or other measuring device), and the transmission medium (currently air). An answer for one set of variables isn't a universal answer either. Change any of the variables and you're right back where you started. The odds of resolving this in any sense other than statistical just went way downhill.
Fortunatly, I do happen to have the final answer. It's simply that-(insert another .02 to continue, please)
