Menstrual synchrony pheromones: cause for doubt.

Abstract

Stern and McClintock’s (1998) recent finding of ‘definitive evidence of human pheromones’ and ‘confirmation’ of the mechanism underlying ‘menstrual synchrony’ is indeed spectacular. However, they disregarded the methodological critiques (Wilson, 1992; Strassmann, 1997) and negative evidence (Jarett, 1984; Wilson et al., 1991; Trevathan et al., 1993; Strassmann, 1997) that undermine the original report of McClintock (1971), as well as subsequent studies (Graham and McGrew, 1980; Quadagno et al., 1981; Preti et al., 1986). When flawed statistical methods are taken into account, there is scant empirical evidence that the phenomenon of menstrual synchrony exists in the first place. Popular belief in menstrual synchrony stems from a misperception about how far apart menstrual onsets should be for two women whose onsets are independent. Given a cycle length of 28 days (not the rule—but an example), the maximum that two women can be out of phase is 14 days. On average, the onsets will be 7 days apart. Fully half the time they should be even closer (Wilson, 1992; Strassmann, 1997). Given that menstruation often lasts 5 days, it is not surprising that friends commonly experience overlapping menses, which is taken as personal confirmation of menstrual synchrony. McClintock’s original study (McClintock, 1971) reported that onsets for dormitory friends became 2 days closer together over a 4–6 month period. Calling this result ‘menstrual synchrony’ she encouraged the impression of menstrual concordance. In later critiques, Wilson (1991, 1992) noted three statistical errors in the evidence for menstrual synchrony: (i) failure to control adequately for the convergence of onsets by chance; (ii) inflation of the initial difference in onsets resulting in the spurious conclusion of synchronization over time; and (iii) sampling biases. Studies that correct for these statistical errors have found no evidence for synchronization, even as a weak effect (Jarett, 1984; Wilson et al., 1991; Trevathan et al., 1993). Although McClintock’s (1981) view is that menstrual synchrony is functionless, others postulate that it is an adaptive feature of human reproductive biology (Burley, 1979; Turke, 1984). These hypotheses assume that menstrual synchrony implies ovulatory synchrony or at least overlapping fertile periods, but neither has been reported. The suggestion that menstrual synchrony is adaptive is also diminished by the absence of evidence for synchrony in natural fertility popula-