It seems to me if you were to define "audible difference," it would be what could be heard in a well-run DBT in an acoustically dead room. Of course, this oversimplifies the problem of directionality of speakers. You might have to take sampling points in different parts of the room.

I think that "if people had trouble telling apart even speakers, cartridges and preamps using blind testing, this would help prove the point of view that" there were no audible differences between the things being measured.