Its less a case of them occupying the same space, and more one of each object running its movement code and stating where it would like to be in the next frame.

Its less a case of them occupying the same space, and more one of each object running its movement code and stating where it would like to be in the next frame. Collision detection then takes place, finds that two objects want to share the same position, and notifies the object of that fact so that it can compensate. That compensation might take the form of changing velocity, blowing into a hundred pieces, or just outright ignoring the fact and carrying on.