it seems that the white assignable encoders are keeping the parameter they have been assigned at the program level, unless you are on focus on one of them.
Unless i did not understood what you mean exactly, why do you need this , since you can use the dedicated volume encoders for each part in the mixer section?