It makes no sense that the guitar would be the only machine ever made that did not wear with age. I often think that 'playing in' is really the first stage of 'playing out'. It's sort of like 'stone washed' jeans; more comfortable than the stiff, new ones, but destined for a shorter life.
Generally speaking, the more heavily thhe instrument is built the longer it takes for it to play 'in', or 'out'. The Hills, in their book on the violins of the Amati family, reported that, based on dealer records, Amati and Stainer instruments tended to 'play in' in about 15-20 years, while Strads tended to require longer, on the order of 30-40, and some of the more robust Garnari 'Del Gesu' instruments took more like 60-75. With violins it's more then just how thick the wood is; arching shape makes a big difference, and possibly other things.
Even on a guitar wood thickness per se probably doesn't matter as much as you'd think. Stiffness along the grain at a given thickness for softwoods varies with the density of the wood, and that can range widely even within a given species. Segovia's '37 Hauser is famous for having a thick top, but without knowing how dense the wood is it's hard to say whether it's 'too thick'. One of the things a lutier does is to adjust the top thickness to get the 'proper' stiffness: Hause thought 3mm was about right for that particular piece, and who am I to argue with him?
At any rate, acoustic measurements I've made of guitars as they are played over long period suggest that the top tends to 'loosen up' a bit. Some of the lower order resonances seem to drop a bit in pitch, and the guitar moves more air in the low 'bass reflex' range. Guitars that start out sounding a bit 'tight' tend to improve, while those that are 'bassier' move toward 'tubby' sound. These measurements are hard to make reliably, since you need to use the same setup and equipment every time, and control things like humidity carefully.