correction. i read the manual and i'm fairly sure this is it, actually:
this pair of caps still engages via the rym7, relay for mode 7, iic+ mode. according to the manual,
"There was actually a difference between the EQ model and those non-EQ models. It all came down to a coupling capacitor at the end of the EQ circuit that feeds the driver. In the EQ model, it was a great big cap that let a lot of sub-low pass, slowing down the sound and making it fatter. In my amp - a non-EQ version - this cap was smaller and didn’t let as much sub-low through - which speeds up the sound and makes everything tighter and more urgent. There it was, a simple part…but it made all the difference in the time domain."
labels for the mode next to the switches or relays on the schematic indicate that's when the mode is selected, so the "m7" is by the side that would engage the .22u cap, while not in m7 mode (mark iv/extreme) selects the 10uF.
it's probably polarized only because of how large a value it is--10u is not a what is usually considered a common coupling cap value since it's probably about 2 decades larger than the "normal" range (i.e. 100x). in marshalls for instance the coupling caps are on the order of .022 or .1 depending on the models and circuits. usually this can distinguish between the "vintage" 60's/70's plexi and the modern ("harsher") circuits of the 80's or later.
however, note that the caps in the previous post right next to v6a also engage with the same m7 relay, so there's more than one cap swap when the switch is thrown. keep this in mind it's not "just one" part.
anyone else out there concur?