This is a tale of two studies. The first, by Saul L. Miller and Jon K. Maner, was published in 2010 and you may remember it. Here's the abstract:
Adaptationist models of human mating provide a useful framework for identifying subtle, biologically based mechanisms influencing cross-gender social interaction. In line with this framework, the current studies examined the extent to which olfactory cues to female ovulation—scents of women at the peak of their reproductive fertility—influence endocrinological responses in men. Men in the current studies smelled T-shirts worn by women near ovulation or far from ovulation (Studies 1 and 2) or control T-shirts not worn by anyone (Study 2). Men exposed to the scent of an ovulating woman subsequently displayed higher levels of testosterone than did men exposed to the scent of a nonovulating woman or a control scent. Hence, olfactory cues signaling women’s levels of reproductive fertility were associated with specific endocrinological responses in men—responses that have been linked to sexual behavior and the initiation of romantic courtship.
Sniffing sweaty t-shirts worn by women near ovulation caused a higher testosterone response in men than sniffing either clean t-shirts worn by nobody or t-shirts worn by women not near ovulation, get it? Testosterone is associated with sexual desire in both women and men, so the results suggest that men can smell ovulation, on some deeply unconscious level.
Now fast-forward to the present. Another study, actually completed before the Miller-Maner study summarized above, didn't find any such relationship between men's testosterone and smelling the sweat of ovulating women. From this recently published study, by James R. Roney and Zachary L. Simmons:
Unlike Miller and Maner (2010), we did not find more positive testosterone
responses in men after exposure to ovulatory chemical stimuli than after exposure to
control stimuli. Importantly, our study was not an exact replication of the Miller and Maner
(2010) study, and so should not be construed as a direct replication failure. Nonetheless, the
discrepancy in findings raises important issues regarding which differences in methods may
have led to the distinct outcomes.
...
A second difference between the studies concerns the male participants’ knowledge
of the stimulus source. As explained in the Introduction, male participants in Miller and
Maner (2010) were explicitly told that they were smelling shirts worn by women, and this
knowledge may have triggered mental images that contributed to the observed hormonal
responses. In the present study, men appeared to be in a state of uncertainty regarding the
identity of the odor stimuli: men were not significantly more likely to report detection of
human odors in the experimental as opposed to the control condition, and participants
generally reported detection of nonhuman odors more often than detection of human odors.
Even in cases in which men reported detecting sweat or body odor, there was never any
reference to women’s scents in particular, such that there is little reason to think that the
male participants were visualizing women. Differential likelihood of having visualized
women thus stands as another possible explanation for the discrepant findings between the
two studies: perhaps chemoreception of ovulatory cues is by itself insufficient to trigger a
neuroendocrine response in human males, which may instead require additional stimuli,
such as visual images triggered by knowledge of the stimulus source.
The bolds are mine.
Get it? The Roney-Simmons study found no increase in the testosterone levels of men who smelled the sweat of a near-ovulation woman as opposed to just water.
As the authors state, this is not an actual replication of the original Miller-Maner study, but it is better in at least some ways, and the most important one of those is that the study subjects were not primed by telling that what it is they are smelling is the body odors of women. If the conclusions of the Miller-Maner study really are about some subtle biological cues women's sweat offers about how close those women are to ovulating, then the Roney-Simmons study basically finds those conclusions rejected.
Why am I writing about these two studies? Because the popular media prioritizes the first type of studies. I'm willing to bet you almost anything that this new study doesn't make Fox News change what it wrote about the earlier study:
Forget Perfume, Ovulation Is the Way to Attract Men, Study Says
Published February 15, 2010
Perfumes and scented lotions only go so far when it comes to attracting a man; the body does the rest, MyFox National reported Monday.
According to a study done by researchers at Florida State University, men are unconsciously attracted to the natural scent of a woman, even more so when the woman is ovulating, Discovery News reported.
This is what is wrong with the popularizations. They don't popularize studies which have null findings or which disprove earlier, gender-clickable ones. And of course all that about a perfume is utter rubbish. The initial study had nothing to do with perfumes and could not make any kind of statement about their value in attracting men.
Because of those popularization biases, the general public gets a distorted view of what the studies actually find, and are practically never told about a study which has been found quite wrong.