According to the 'critical period' hypothesis, which attempts to explain the observed discrepancies in the studies on estrogen and cognition, estrogen therapy effectively decreases cognitive decline in aging women when it is initiated around the time of menopause but not when it is started decades later. Here, I review studies in which the timing of the initiation of estrogen therapy was provided, to determine whether their findings support the 'critical period' hypothesis. The vast majority of the reviewed studies support the idea that early but not late initiation of estrogen therapy might prevent or delay cognitive decline in aging women. Nevertheless, numerous design issues, such as the specific drugs and doses that were used, the possible effects of progestins on cognition, and the failure to administer neuropsychological tests of specific cognitive domains that are sensitive to estrogen therapy confound the extant literature. In view of the reanalyzes of the Women's Health Initiative's data that show a beneficial effect of estrogen therapy on cardiac and breast diseases in women aged 50-59 years, more definitive evidence is needed to confirm that the early initiation of estrogen therapy that is continued for a few years provides enduring protection against cognitive aging 15-20 years later.