To the Editor: The literature review by Harvey Sternbach, M.D. (1), on testosterone supplementation and andropause is an important contribution to our knowledge concerning the identification and management of hormone-related somatic and psychological disorders in men. We believe, however, that one of the author’s statements regarding the cognitive effects of testosterone supplementation in men may lead to an incorrect inference on the part of readers. Dr. Sternbach states that "low and high levels are associated with poorer performance" (p. 1314) on tests of spatial cognition. Indeed, this is a simple paraphrase of the conclusions of Moffat and Hampson (2), but it fails to mention that in their study, such a conclusion only applied to right-handers and reflected the combined data of the influence of testosterone level on each of the sexes, i.e., a negative correlation in men and a positive one in women (lower testosterone levels in men and higher levels in women were associated with better performance). Data from at least two studies in men (3, 4) have shown that performance on tests of spatial cognition is inversely correlated with testosterone levels—i.e., lower levels of endogenous or exogenous testosterone were associated with better performance on these tests. However, other studies with men (5–11) have either failed to show such an inverse correlation or have even described a positive correlation. While the data in the literature are far from consistent, it would be misleading to suggest that low and high testosterone levels are associated with poor spatial abilities in men.