I think you're right, actually. I've always treated the room's corner as a conical section, but this changes at the ceiling. When looking at it like this, the expansion into the room from a corner apex appears to act like a conical section that changes to a parabolic section at the ceiling junction. See the post called "Room corner characteristics". Then there's also the opposing wall and the standing waves setup by it as well.It would be a good excercise to take a loudspeaker's anechoic response as a baseline and then measure it in various rooms of different sizes to compare their effects. I'd be very interested to see that kind of data.