Sniffing for Human Sex Pheromones

One of the very earliest known works of art is an 11cm tall limestone figurine called the Venus of Willendorf, made some 25,000 to 30,000 years ago, and it is erotic art. Humans have had sex on the brain for a long time. Not only did we carve figurines and make lewd cave paintings, we've turned to our leading minds — whether they were the tribal shaman ten thousand years ago, or the pheromone scientist today — to find some way to get potential partners in the mood. It may have evolved from superstition into science, but its fundamental goal has never changed. Today the way we ask the question is whether human sex pheromones are real, and whether they can be employed to achieve this goal.

As usual, we'll begin with a basic science overview of the topic at hand. Pheromones are scented chemicals that trigger some social behavior in others of the same species. One of the most familiar examples is the smell-based food trail that ants lay down. Another would be the alarm pheromone that a bee might secrete when attacked, telling the hive to swarm. A dog peeing in its yard to mark its territory is also a pheromone at work; other dogs will sense that and not go there. It's basically any biological odor that communicates and that triggers some behavior in others. Pheromones are found in plants, animals, and even in microbes. But not all of them; many species do not appear to use any known pheromones. Today's question is whether humans have sex pheromones: which, if they exist, would literally be aphrodisiac scents that put a potential partner into the mood, exclusive of other influences. Biologically, there is nothing carved in stone that says humans (or any other given species) must have these.

The established science on this question is pretty easy to find, all you need to do is do a search for a term like "do humans have pheromones" and you'll quickly learn that there's no evidence that we do. In animals, pheromones are detected by the vomeronasal organ (VNO), which is a second olfactory sensor in the nasal cavity. This organ is present in most reptiles, in lots of mammals, and in a few primates; but in humans, it is vestigial only and not functional. Many humans don't even have that vestige, and even in those who do, no anatomy connecting it to the brain has ever been found. It's a tiny strip of nonfunctional flesh. So, in a sense, that could be the end of this episode: human's can't use pheromones, sexual or otherwise, because we lack the needed hardware to receive the chemical signals.

But that wouldn't be very satisfying, and wouldn't answer the many lingering questions. Magazines are full of ads selling pheromones that will supposedly make you irresistible to members of the opposite sex (more on these products later). Many people report anecdotal experiences with a pheromone-like smell when they become sexually stimulated. What might that be, if not a pheromone? What about the ability of mothers to recognize their own infants by smell alone — and the infants' corresponding ability to recognize their mothers? Wouldn't that be analogous to a pheromone?

So this is where we need to go a little deeper into the difference between a pheromone and an ordinary smell. Yes, lots of people can do lots of things with scents and odors. Food smells can trigger our appetites. Lots of studies have exposed people to various body odors and measured their biological responses, which do exist; wouldn't that have to be considered a pheromone? And there's the old idea that the menstrual cycles of women living together synchronize, presumably via pheromones.

By definition, pheromones function autonomically, both on the sending end and the receiving end. They cause a behavior, like the bees swarming or the ants following the trail. But humans, with our advanced brains and conscious thought and willpower, would be able to overrule those commanded behaviors. This is why we don't have military-grade pheromones that would, say, make all the enemy soldiers run away or go to sleep. And we know from literally centuries of trying that love potions don't exist.

So researchers, in hunting for human pheromones, have relied on study designs that look for more subtle, involuntary responses. A study published in 2017 in Royal Society Open Science took two compounds that have, for a long time, been considered the leading contenders for human pheromones: AND, found in male sweat, and EST, found in women's urine. Researchers wondered whether exposure to these chemicals would alter the way subjects looked at members of the opposite sex. The test subjects were all adult men and women, all heterosexual, and all white, to control for racial preferences. In the first of two tests, they took a bunch of photos of people made androgynous by blending male and female faces with a computer. On consecutive days, some subjects were given a control scent and some were given either AND or EST (AND to the women, EST to the men). Researchers figured that if the compounds were indeed pheromones, people would be more likely to interpret the androgynous photos as members of the opposite sex, i.e., as potential sex partners. But unfortunately, the data showed no such trend.

For the second experiment, they took a whole bunch of unaltered photos of men and women. They showed male subjects pictures of women, again giving them either the control scent or the EST on consecutive days; and showed female subjects pictures of men, giving them either a control scent or AND. Researchers wondered if the subjects, when exposed to the potential pheromone, would be more likely to see members of the opposite sex as attractive or as potential sex partners; and so asked them to rate the attractiveness of each person in the photo and also estimate that person's likelihood to be unfaithful, i.e., to have a hookup. But once again, exposure to the potential pheromone had no impact on the results. It was definitely a blow to the advocates of human sex pheromones. The study's lead author, who believes human pheromones probably do exist, said "I've convinced myself that AND and EST are not worth pursuing."

Some of the most interesting evidence in favor of the existence of human pheromones — though not sex pheromones — comes from the tangled world of synchronized menstrual cycles. I say tangled because numerous large studies have found no evidence for this, while others have. Many anecdotal cases of perceived synchronization can be put down to simple confirmation bias. There was an interesting study done in 1997 of women of the Dogon population in Mali, West Africa. In their culture, menstruating women are sequestered into a menstrual hut, which made it easy for researchers to track menstruation in the community over a long period without having to interview anyone. Over 736 consecutive days, no evidence of synchrony was found; all the women menstruated independently of one another.

But then we contrast this finding with those of Kathleen Stern together with Martha McClintock, who has a pretty good share of the literature written on pheromone candidates and menstruation. In 1998 they published in Nature the results of their study of the armpit sweat of menstruating women and its effect on other menstruating women. Odorless compounds from the sweat were collected — note, odorless — from earlier in their menstrual cycle, and again from later in their cycle. When these compounds were exposed to other menstruating women, the compound from earlier tended to shorten a cycle, while the compound from later tended to extend the cycle. Interestingly, this influence — if real — would tend to nudge cycles apart from one another, not synchronize them. The authors concluded "By showing in a fully controlled experiment that the timing of ovulation can be manipulated, this study provides definitive evidence of human pheromones."

And of course, it's necessary to point out that every one of these studies — and the countless others making similar findings, both positive and negative, has been criticized for flawed methodology, biased data analysis, or bad study design.

An interesting thing to note about the Stern & McClintock study is that they believed they'd proven the existence of a human pheromone, even though we humans have no functional VNO capable of detecting pheromones. The obvious question to ask is what if our normal olfactory organ has picked up the slack, and is somehow able to detect compounds we didn't think it could?

We know that people are pretty good about identifying each other by smell. A 1989 study found that adults can identify a T-shirt that they wore themselves by smell alone, picking it out of 100 T-shirts worn by other people; and a 1981 study found that family members can do the same thing, correctly identifying shirts worn by partners, siblings, and children, by smell alone. But simple smell has nothing to do with pheromones; no behavior is triggered.

Many report anecdotally that there's that stimulating, pheromone-like smell when people are getting hot and heavy with one another. But sorry to burst your bubble, that is simply the smell of the other person's sweat and saliva. We've come to associate it with arousal because that's the condition we're already in when we first detect it. This introduces a complexity to the study of human pheromones, which, by definition, function without the conscious knowledge of the recipient. But humans have complex brains and thought processes, and are aware of many things going on. When we do become sexually aroused, we are very well aware of it, and very well aware of the other person. There is so much conscious knowledge of what's happening that any pheromone effect, if it existed, would be lost like a drop in the ocean. This ability for the human intellect to override any possible pheromone effect has long been a complication for researchers. As the authors of one paper concluded:

As the pheromone field matures, so will study methodology. The ideal study does not disclose the purpose of the study at all and does not involve skin application of compounds with masking odors, but introduces the experimental pheromone subliminally and with no additive to the participants (in the room air) while strongly controlling the psychosocial setting of the experiment.

The fact is that the history of human sex pheromone research is only the newest chapter in the much longer history of love potions and aphrodisiacs. From humanity's very earliest recorded literature, we've sought potions and hexes and spells to make ourselves irresistible to members of the opposite sex. It was probably the original worthless snake oil ever sold, and it still is today: countless perfumes for women, and colognes for men, are sold as pheromones promising to boost your attractiveness to your date. Some of these contain AND; still more contain known pheromones — known because they're animal pheromones proven to exist, usually synthetic versions of secretions from civet cats, beavers, pigs, or musk deer — none of which have any effect on humans. But at least you're actually getting what's claimed on the product label.

The darkest end of this spectrum of products goes all the way to date rape drugs, which don't do anything other than knock the victim out, cause them to be too confused to understand what's happening, and/or cause temporary amnesia. Kind of the opposite of what a love potion is supposed to do: make the victim love you, or at least feel sexually attracted to you.

That leaves us with exactly zero functional love potions. Considering the incalculable amount of money and research that has been poured into this for millennia, I'd argue that the sheer quantity of bad evidence constitutes evidence that no good evidence exists. We certainly don't have proof from the world of science, let alone from the world of snake oil product marketing. Where we find ourselves is sitting squarely in the center of the null hypothesis. We may have been nudged one way or the other by some promising research over the years, but still nothing has ever survived scrutiny and proven to be that magic potion so many have longed for. And that null hypothesis — that there is insufficient evidence for the existence of human sex pheromones — is going to be our official Skeptoid conclusion for today.

By Brian Dunning
Follow @BrianDunning