Here's my half baked theory: We know that a window blockage is detected by the increased current draw caused by the motor being unable to spin. In such a condition, the controller lowers the window so the obstruction can be cleared. I don't know where the current draw is measured, but perhaps the current from both motors running, plus a slight extra resistance from a window dragging in it's channels fools the controller into thinking there's an obstruction?
I can think of a few questions that contraindicate this theory, but I have no good answers. As I said, it's only half baked. Maybe paranormal intervention is a better explanation after all. I've experienced this behaviour as well. I had thought it was totally random, but now that you mention it, I think it was always when I was raising both windows. Then when I reattempted to raise the one window, it always worked fine.