We deal a lot with floor/ceiling assemblies where it may be a different dynamic. It has been suggested that with a 12" cavity half filled with pink fluffy there could be a subtle impedance mismatch that is created between the air and the insulation that causes some shear. It'll be tough to get any real data to back that up, though.
As you say, I'd like to see the data that shows such an effect!
Many people have the wrong idea about what insulation in the cavity does: some people think it just "stops sound" from getting through. That's not really true. If you took the same thickness of insulation and placed it directly in front of a speaker playing typical contemporary music, you'd hear the music fairly well. Muffled highs, yes, and muddy mids, it is true, but you'd still hear it with just a small reduction. Only slight TL drop like that.
Insulation by itself is pretty bad at "stopping the flow of sound" on its own. But what it DOES do very well, is to damp resonance.
I often use the analogy of a kitchen sponge: If you spilled some water in your kitchen, and need to mop it up, then a sponge is a good way of doing that: it picks up the water very well, then you can squeeze out the water in the sink (or wait for the sponge to dry), and mop up more spills. You can carry on mopping up and squeezing out all day, with no problem! Sponges are good at that. But if you then turn on the tap and hold the sponge over the end of it, the water comes pouring through, as though the sponge wasn't even there. It does nothing to "stop the flow".
That's sort of like trying to use porous insulation to "stop the flow" of sound: it does nothing much useful. But when you make it part of a SYSTEM, then it does what it is really good at: mopping up resonance, or "damping". Inside an empty wall cavity, there are multiple resonances going on at different frequencies. There's the MSM resonance, of course, between the two sides of the wall, plus probably some modal resonances at very different frequencies, plus several others. Insulation can damp all of those, very effectively.
The insulation does several things at once to help with that. Firstly, it slows down the speed of sound: sound waves travel quite a bit slower through insulation than they do through empty air, so from one point of view, the sound "sees" a longer distance between the leaves than is really there (it takes a longer time than it would have to cross the gap), secondly the presence of the insulation changes the way the air deals with heat from adiabatic to isothermal, and third, it provides impedance for the movement of sound, which isn't the same as "stopping the flow" of sound. All of those are "good things" for damping.
Now, think about your 12" cavity: if you fill it only 6" deep, then you only have half the cavity with all those "good things" happening in it. The other half is just empty air, where the resonances are all still happening... and robbing the wall of isolation. Filling the cavity entirely means that the above "good things" are happening in 100% of the cavity, not just 50% of it.
Filling the cavity complete with SUITABLE insulation, is actually one of the best things you can do to a wall, to increase TL.
If my words don't convince you, then maybe actual data will...
effect-of-various-wall-cavity-insulation-fill-percentages.png
I also know that I have a report somewhere titled "The Acoustic Role Of Glass Wool In Double-Leaf Walls" someplace, but I can't find it right now. I'll dig some more, and see if it turns up... All I have is this extract:
insulation-in-MSM-walls-cavity-fill-isolation-study-awg_paper_CH_1-extract.jpg
- Stuart -
) by data, right?