There were sound rooms built to simulate a variety of real world environments, complete with realistic RF interference. They could test other equipment alongside their own, to be confident that they were producing real value for their customers. They’re pretty secretive so I won’t go into more details out of respect, but it was very cool and definitely made me a loyal customer.
All that aside, like you said the psychoacoustics thing is very real, and knowing them, they’re probably working on it.
https://community.bose.com/t5/Around-On-Ear-Headphones/Bose-...
There can be a thousand factors, and these 20 objective tests are easier to trust than subjective listeners.
If I was reading this thread without experiencing it for myself I would be saying the exact same thing. Unfortunately for me (and lots of others according to the Bose forums) my subjective experience is the complete opposite of the outcome of these tests.
Luckily it is looking like they tested the wrong upgrade path which is reassuring as I have no idea how I could refute the tests otherwise.
FWIW I can confirm that this post shows great expertise in acoustic engineering, from the focus on the customer interviews/fieldwork (it is still very difficult to predict through synthetic tests in the lab how actual human will rate sounds) to the paired acoustic tests (we suck at comparing more than two sounds, we need pairing like in the ophthalmologist), the use of a critical listener, etc.