An example would be the slightly greater or the slightly lesser intelligibility of a solo vocalist. In one system you think you might be able to understand a certain word that is being sung, but in another system you're confident that you are hearing it more clearly.
Unfortunately things are not so simple. Intelligibility is strongly related to RT60 - in a dry sounding enthronement (RT close to .3 or .2s) you will understand vocal much easier, but envelopment will be low.
Another example would be the ability to hear and to make out the words of one or more individual voices in a choir, versus an aggregated mass of voices that cannot be teased apart.
Yes, the famous Cantate Domino is often referred as an example of this. I have to say that the best separation of individual voices I have ever experienced with this recording was listening to the FIM CD, re-mastered by Winston Ma. BTW, I never listened to the tape, but would not mind listening to it. But the curiosity is not worth the high cost of the current copies of the master tape.
The clearer the intelligibility of the speech, the higher the resolution.
In my experience, not true every time. Resolution is much more than intelligibility - but we are entering a semantics dependent subject.