Sean has pointed out that the change which occurs with initial use is a DETERIORATION, and he needs to readjust circuitry to make it work right. Tuned, adjustable circuits (which appear to be what he refers to) can be shifted away from optimum adjustment by the small changes of component electrical characteristics. This is not the same thing as saying that a power amplifier circuit improves with use.
The original posted question sought to find a scientific explanation for "breakin" improvement, but somehow the thread has degenerated into a debate as to whether such a thing happens. Frankly, I remain to be convinced, but my comments have been directed to pointing out that some of the theories put forward are, from an engineering perspective, obviously wrong, and detract, rather than add, to the pro-breakin point of view.
The original posted question sought to find a scientific explanation for "breakin" improvement, but somehow the thread has degenerated into a debate as to whether such a thing happens. Frankly, I remain to be convinced, but my comments have been directed to pointing out that some of the theories put forward are, from an engineering perspective, obviously wrong, and detract, rather than add, to the pro-breakin point of view.