Sometimes two distinct problems can be analyzed as having essentially the same search space, and yet one of the problems might be much easier for a human subject than the other. How might such differences in difficulty of equivalent problems be explained?