In accounting for the well-established sex differences on mental rotation tasks that involve cube stimuli of the Shepard and Metzler (Shepard & Metzler, 1971) kind, performance factors are frequently invoked. Three studies are presented that examine performance factors. In Study 1, analyses of the performance of a large number of subjects (n=1765) that performed the Vandenberg and Kuse (1976) mental rotation test (MRT) under standard conditions showed that the magnitude of the sex differences increases as subjects proceed through the sequence of problems, and that fewer females than males reach the last problems in a problem set. This supports the role of time constraints in sex differences on the MRT. Study 2 compared the magnitude of sex differences for subjects (n=212) who did the MRT under standard conditions (T1), and with double the time (T2) allowed for the test. No significant reduction in the magnitude of sex differences was observed-even though the overall scores under T2 increased markedly for both sexes. Study 3 examined the suggestion by Kerkman, Wise, and Harwood (2000) that mental rotation differences arise because females spend more time than males in assuring that stimuli that do not match do indeed not match, with no sex differences for matching stimuli. This hypothesis was not supported for a sample of 20 males and 26 females. We conclude that performance factors may play a role in sex difference on mental rotation tasks, but do not account for all of the differences.