I am a professional mapmaker and artistic conventions may also play a role here. In almost all shaded relief maps (easier to visualise than contours for the layman), the source of illumination is always in the northwest, where it never is in the real world. (Sorry, I almost forgot about our friends Down Under!). This is why aerial photographs are usually exposed in mid-morning, but often don't look right with north at the top to those who aren't used to working with them.
Still, for some reason, the upper left sun looks more realistic to the human eye. The crater/dome ambiguity is minimized and dendritic drainage patterns don't look like raised eskers or branching blood vessels on the landscape.
And of course, anyone who's looked at the lunar terminator with a telescope, or seen a familiar mountain range with the sun low behind him, will know the dramatic effect shadows can give to help bring out relief.
Space/Science » in reply to Agree with what you say about perception
"A Conspiracy of Cartographers"
