Time limits and gender differences on paper-and-pencil tests of mental rotation: a meta-analysis