One thing I have noticed with groups of 4 or more in a listening session is the "AB" instant switching tends to annoy most of the people in the test.
The A/B switching, especially when done quickly, instantaneously, is definitely more of a solo activity. However, you could have a
short segment where you tell everyone that the switch will be toggled every 5 seconds for 1 minute (a total of 12 toggles), a fairly short test. If people focus on specific musicals tones (bass, treble, voices, etc.) after every 3 switches, they can get a more accurate observation of those specific things.
"Does X speaker immediately have more bass than Y?
Switch.
Listen again.
Switch.
Listen again.
Switch
Listen again.
Restart the same music piece.
Does X vocals sound brighter or muffled in comparison?"
Switch.
Listen again.
Switch.
Listen again.
Switch
Listen again.
The instantaneous method REALLY makes a huge difference in observing these variations between units and is much more similar to how Axiom has their setup as well, just that the Axiom switching unit isn't quite as quick a switch as the one i've used here.
This instantaneous switching is where level matching becomes very important because you can immediately notice if one speaker sounds louder than another. It also makes it easy to observe if the speakers are located on the right or the left which is why i suggest sitting a bit farther back (say 12' or more) and play the music in mono. This goes a long way to eliminating the left and right pull that can fool your ears into thinking it is a sound character difference vs. spatial location.
Again, this is why such a test is better done solo since having 12 guys located in different spots in the room would have that left vs. right issue the more off axis they were located, but it is worth trying in a group IMO, just for a very short period.
Reduces aggravation.
Hand out an extra beer before the test (or 'green chocolate' if you happen to live in Colorado).