Hmmm... For now, I'm going to accept that CO2 produced by yeast makes different bubbles than CO2 that comes from a bottle (Since I can only assume that the bottled CO2 is impure, and the CO2 from yeast may be also).
But, once you hook the keg up to the bottle, the formula changes.
Does the keg system have a one-way valve? Or am I confusing guage pressure with absolute pressure or something.... even if it did, then the bottled CO2 would be valved off until the keg pressure dropped to 3 PSIG. And then your beer would gradually reach an equilibrium of 3 PSIG... and there would be a constant exchage of CO2 at the surface of the beer between the CO2 in the beer and the CO2 in the space above the beer.
If a pressure vessel is at 10PSIG and there are 2 outlets, 1 valved outlet at 0 PSIG and one at 3 PSIG, then depending on the nozzle effect of the valves... when you opened the 0 PSIG valve (spigot) the vessel would tend to drop in pressure to 0PSIG. when the valve was then closed the vessel would tend to increase to 3 PSIG...
The piece that I'm missing is, how does the "trapped carbonation" get effected.
I know that in my glass, the beer stays pretty consistent for a while, and this is a 10PSIG carbonated beer in an open air container... but eventually, it goes flat.
If you have 3PSIG over the beer in the keg, eventually it will lose carbonation until its at equalibrium with the 3 PSIG headspace... (of course this will happen gradually as you "use up" the 10PSIG pushing your beer into your glasses).
I did a little research on the web, and found only controversy over this subject. I found one guy who said he was going to do a double blind taste test natural vs. forced, but never found the results. I am keeping my mind open, but I tend to believe that the physics and chemistry around a CO2 bubble in beer are the same regardless of where the CO2 molecules came from.
I did find a nice article on kegging... of particular interest to me was the table showing the effect of temperature and pressure on carbonating.
http://www.leeners.com/kegginghow2.html
I'm going to do more research on this... If I find something of substance, I'll post it up. I know this issue has been debated, tested, probed, and hopefully resolved before....