Human Pheromones

Pheromones most likely become airborne after being secreted in axillary (armpit) sweat glands. (

Pheromones have been demonstrated clearly in other species, but their existence and ability to affect humans is still under debate. Several studies have been conducted that have reached different conclusions, but some pheromone effects seen in other mammals are definitely present in humans. Many contradictory studies failed important aspects of pheromone experiments, such as common odor or musk control, so these studies are disregarded. Although exact conclusions differ, most studies have found that human sweat or putative pheromones increase physiological arousal, one way or another. This idea that something we can't even consciously smell is affecting our behavior is horrifying to some people, but in reality we are controlled by the combination of all our senses! Think of proprioception, the body sense. We are not always conscious of the fact that we are thinking about our position in space, but we are! Personally, I think pheromones are another sensory system (separate from olfactory) that we just didn't have the technology to recognize until recently.

It's important to realize that any odorant can affect behavior but pheromones are produced by conspecifics (members of the same species) and play a role in communication. Studies show they are processed differently than common odors.


Effects of hormones on the menstrual cycle. (

Neuroendocrine effects of pheromones were of the first to be recognized and are clearly demonstrated. (For effects on cortisol, see physiological effects.)

The first widely accepted evidence of human pheromones was research done by Cutler and Preti, 1986. They reported on the effects on male and female pheromones on the female menstrual cycle, continuing the research they had done earlier and the 1971 article on synchronization by McClintock. They concentrated on the idea that pheromones caused this phenomenon.

The menstrual cycle of women exposed to axillary secretions of other women is shortened or lengthened depending on the phase of the cycle the women who pheromones are detected are in. For example, during the late follicular phase, the cycle is shortened by a delay in releasing the luteinizing hormone, necessary for ovulation. During the ovulation phase the cycle is accelerated in the opposite fashion. The end result is that some pheromones of women are affecting each other so that eventually, females who spend enough time together have the same cycle. This synchronization is sometimes known as the McClintock Effect. (Tirindelli et al., 2009)

Male pheromones affect the human menstrual cycle as well, by accelerating and increases fertility. Cutler and Preti, 1986, claim that their research shows how important the presence of men's "essence," (now claimed to be pheromones), is to female's biology. They found that regular sex decreases fertility problems in women, regulates menstrual cycles and correlates with a milder menopause. The cycles of woman exposed to male sweat on their upper lip for a month, found their cycles shortened or lengthened so that they all were at the optimum length of time, about 29.5 days. Exposure to AND (a putative pheromone produced by men) accelerates the onset of the peak of LH necessary for ovulation, by affecting the hypothalamus to secrete gonadatropin releasing hormone (GnRH). The researchers also noted an increased relaxation. (Preti et al., 2003)

Some studies found no evidence of synchrony. Trevathan et al (1993) found no synchrony between lesbian couples. This may be caused by the difference in the way homosexual people perceive pheromones. McClintock continues to defend her research, bringing up numerous other issues that can affect the menstrual cycle, such as stress levels, presence of males, even circadian rhythms and the effect of light, and length of day (McClintock, 1998). Different people produce different levels of hormones at different times in their life, which suggests the amount of pheromones produced varies, too.

Compounds from lactating women "have the potential to function as pheromones," according to Jacob et al, 2004. Exposure to breastfeeding compounds affects hormonal levels, which affect the menstrual cycle. Aprocrine glands, which are believed to secrete pheromones, are also present on women's nipples, a possible pathway of mother-infant communication.


The chemical structure of androstadienone (AND), a putative pheromone produced by men. (

Pheromone like compounds are believed to be secreted via axillary (armpit) glands called apocrine glands, which create a nutritional environment for bacteria. The compounds can be further affected by the microbial flora, which may separate compounds to enable their transmission or make the steroid pheromones more odorous (Hays, 2003). There are differences between men and women in the types of glands, secretions and even microbial flora present in the axillary environment, suggesting a sex-specific role. Most of these glands do not become active until after puberty, suggesting a role in sexual communication. (Hays, 2003) Pheromone-like compounds are also found in salivary, seminal and urine secretions, but studies tend to focus on the most accessible: axillary sweat.

There are at least ten different steroid compounds secreted in men's sweat, but one compound has gotten the most attention. Progestorone derivative androstadienone (4, 16-androstadien-3-one, AND) found in male sweat, urine and saliva. This compound is the most frequently tested as a pheromone and has been demonstrated to affect endocrine levels, physiological arousal, context-dependent mood and affects depend on sexual orientation. The estrogen like compound, 1,3,5(10), 16-estratetraen-3-ol (EST), is a putative human pheromone found in female urine. There is less evidence for the natural secretion of EST. (Bensafi et al., 2003)

Some studies assume that since the putative pheromones are typically found in sweat, that showing the effects of sweat will show the effects of pheromones, without specifying which compounds act as pheromones, while other studies specifically used synthesized putative pheromones. Synthesized pheromones may lack the additional processing provided by microbial flora around natural apocrine glands.

To administer the potential pheromones, most studies choose passive inhalation of pheromones over topical application. They can either ask the subjects to sniff, place the sample above the upper lip or even apply it directly to the VNO. Grosser et al, 2001, conducted a study that applied directly to vomeronasal pits concluded that AND decreased sympathetic activity, which is contradictory to almost every other AND study. A possible explanation for the difference is because the VNO in humans is vestigial, so this method of administration was not the most efficient or even effective. The conclusions of several studies suggest an exquisitely sensitive sense for pheromones. Even though effects have been noted when sweat or synthesized putative pheromones are directly given to humans to sense, they may not exist in adequate levels in nature. It is possible that we still produce effective pheromones, just not in effective amounts. Cutler and Preti show effects on menstruation are most notable in nature when subjects are having regular sex. How close do we have to be to sense each different pheromone?


Most nonhuman animal species use a specialized olfactory system to detect pheromones called the vomeronasal organ (VNO). One of the major complications of demonstrating the use of pheromones in humans is that although embryos develop a vomeronasal organ, it degenerates in later development (Hays, 2003). The VNO is generally believed to be vestigial, nonfunctional in humans. The accessory olfactory bulb, where the VNO projects in most species, is absent in humans. Out of hundreds of possible VNO genes (V1R, V2R) all but five are pseudogenes (Trinidelli et al). No neural connections between human VNO and the brain have been found. If humans do indeed detect compounds from other members of the species, they are not detected with this organ. Studies have suggested that pheromones are detected by the same sensory organ as general odors (volatile molecules), the main olfactory epithelium (Savic, 2004). A study by Savic et al. (2009), found damage to the olfactory muscosa inhibited the effect of EST activation of the anterior hypothalamus in men. This supports the idea that the main olfactory system in used to detect and transmit pheromone signals.


Most frequently, an increase in physiological arousal in pheromone studies is successfully measured by an increase in skin conductance and a decrease in skin temperature. Bensafi et al (2003) also measured electrocardiogram, finger pulse, ear pulse, blood pressure, abdominal respiration and thoracic respiration, after subjects sniffed undiluted AND and EST. All increased but still, most significant changes occurred in skin temperature and skin conductance. EST did not significantly affect physiological arousal of men. AND increased arousal in women and decreased it in men, shown with skin conductance as well as hypothalamic activation shown in PET scans. They also used many methods of measuring arousal that involved self-reports by subjects, in mood and preference of compounds. When asked to rate the compounds in intensity, pleasantness and familiarity, there was no significant difference between any compound and each gender's perception. They used Eckman's scale, a well-accepted questionnaire to analyze mood and found no effects on mood. Bensafi's study was conducted over three days, with each day a different compound. Women's sensitivity to olfaction and pheromones (AND) can increase by a magnitude of eleven after repeated exposure (Diamond et al., 2005). Studies conducted over just one day may have different results than long-term studies.

Jacob et al., 2001, tested diluted AND and EST on men and woman. The putative pheromone chemicals were given in a strong odor carrier, so as to disguise the scent. Other researchers did not use this technique because almost always the vast majority of subjects do not report being able to smell the compounds. Pheromones are typically odorless. They concluded that AND positively affects mood and increase arousal. Jacob et al suggests that the effects of these compounds are context dependent, and this is why the results vary widely across different studies.

Grosser's 2000 study was mentioned under compounds. Grosser et al, 2001, conducted a study that applied diluted synthetic pheromones (AND, EST) directly to vomeronasal pits concluded that AND decreased sympathetic activity, almost opposite to the exact conclusions made by Bensafi and others. With AND, they noted decreased skin conductance and increased skin temperature, as well as decrease in negative mood.

Simply sniffing AND increases the level of the hormone cortisol in women, as detected in saliva (Wyart et al, 2007). Cortisol is normally released when a person in under stress, which may be a factor in the increased arousal noted in most AND studies. Cortisol also may affect serotonin levels in the brain, which would have an effect on mood. This study also found increased sexual arousal, physiological arousal and better mood.


Major regions in the hypothalamus, discussed here. (

Functional brain imaging studies can show how exactly the brain is activated when a person is processing putative pheromones. In contrast to combinational olfactory stimulation of common odors, pheromones are detected in narrowly tuned receptors (although no specific human pheromone receptors have been found) and activate sexually dimorphic neural circuitries (Touhara & Vosshall, 2009). Woman smell AND and have activated preoptic and ventromedial nuclei of hypothalamus. Men who smell estrogen-like substance have activation in paraventricular and dorsomedial nuclei hypothalamus. Hypothalamic activation is typical for ordinary odors as well, but the difference between men and women's activation is congruent with theories on sexually dimorphic area, involved with sex.

The previously mentioned brain regions are all located in the hypothalamus, a structure highly involved in physiological arousal controlled by the autonomic (mostly automatically controlled) nervous system. It also regulates most hornomes. The hypothalamus is often said to control the four f's: fighting, fleeing, feeding, and reproduction. The hypothalamus is full of sexually dimorphic regions, where they differ in size between men and women. Further research found that these areas are indeed related to sex. For example, the preoptic nucleus (activated in women) is twice as large in men, and is involved in the regulation of gonadatrophin releasing hormone, which causes the release of LH and follicle-stimulating hormone, involved in women's (see hormonal effects). The ventromedial nucleus is specifically involved in the feminine response to sex, lordosis (receptive to sex) in many mammals. More generally, the anterior hypothalamus is involved with thermoregulation and sweating, which we have seen are effects of pheromones.

Zhou and Chen (2008) used axillary sweat to explore differences in brain activation, to see if it would support their previous conclusions that chemosignals in sweat can affect mood, help distinguish unfamiliar from familiar women in men, as well as are involved with sexual communication. Using functional magnetic resonance imaging (fMRI), they found activation in the right orbitofrontal cortex (commonly associated with olfaction), right fusiform cortex (temporal lobe), and the right hypothalamus. The significance of right versus left hemisphere activation may suggest that this is mainly an emotional response, since the right hemisphere is more specialized to express and perceive emotion. The right hemisphere is also dominant in processing olfaction and social qualities. The OFC is involved with socioemotional regulation and the fusiform cortex is involved with human perception, face and body recognition. They exposed women to male sweat collected during sexual arousal versus neutral states. They found that the physiological arousal of woman was significantly higher exposed to "sexual sweat" compared to nonsexual sweat. The hypothalamus responded more to sexual sweat than the neutral sweat.

Savic et al. (2001) found that EST activated hypothalamic regions in men and regions typical of ordinary olfactory sense in woman (administered via sniffing). This was opposite for AND. This shows sex-specificity.


Photo shows Eckman's universal mood facial expressions (

The effects of pheromones on mood are even more inconclusive (see physiological arousal). No doubt this is partially because of the subjectivity of mood in contrast to measuring arousal, for example. Chen (2000) found that mood could be communicated purely via chemicals found in axillary sweat. Sweat collected from both men and women while viewing either a funny (happy) or frightening video was recognized later. Women successfully identified bottles filled associated with "happy" compared to frightened and unused pads when up to six samples were present. Men were able to make the distinction (happy) significantly only when the sweat was collected from women. The frightened pheromones were detected by both men and women only from sweat collected from men.

Jacob (2000, 2001) found increased physiological effects, only noticed a positive increase in mood and increased arousal in women when the test administrator present was male. Men were unaffected by the sex of the tester, but had slightly less effects than females overall. They concluded that pheromones are incredibly context dependent. Humans must integrate many different brain functions to affect arousal and mood.

A recent study by Marazziti et al (2010) found a direct correlation between axillary compounds and serotonin, which affects mood. Male sweat modulated the affinity of the platelet serotonin transporter measured by a decrease in the dissociation constant, which means serotonin stays longer in the receptor and has a longer effect (increased effectiveness of serotonin). They also noted an increase in impulsiveness in women. There is a positive correlation between Kd (dissociation constant) values and romantic attachments characteristics as measured by the "experiences in Close Relationships" questionnaire. We must be aware that subjective data given by the participants themselves may be less reliable than objective measurements.


Cutler (1998, 2002) found that topically applied synthesized pheromones (AND or EST) increases sexual attractiveness. In another study, (Cutler) men wearing the synthesized pheromones reported more sexual activity with romantic partners, but no increase in autoerotic behaviors.

Cornwell et al, 2003, found a correlation between preference for masculine facial features with preference for AND, and between a preference for feminine facial features with preference for EST. This leaves us with more questions about what senses predominate when it comes to sexual arousal.

Chemicals that communicate information of individuals' immune systems have been demonstrated but are not always considered pheromones, since they do not cause a specific reaction in conspecifics. Several studies have shown that animals and humans choose mates that have fewer matches in immunological genes, such as the major histocompatibility complex (MHC) (called MHC-based odor choices). (Hays, 2003, Wedekin et al. 1995) The MHC is highly polymorphic; there are many genes and therefore high diversity. HLA gene alleles are responsible for MHC. One study that compared the type of HLA allele with preference found that people prefer more matches (with alleles the woman had inherited from father, only), as long as the alleles are not identical to their own (Jacob et al, 2002). These conclusions are in contrast to others, such as Wedekind et al original t-shirt study (1995) found preference for MHC-dissimilar mates. Women were given t-shirts worn by heterosexual males for a few days. When asked which t-shirts they preferred, a significant correlation was found between preference and differences in immune genes. Evolutionarily, people want to find mates that can offer different immunities. Studies have shown birth defect risk goes up as genetic similarity increases, so it would make sense intuitively that humans would prefer fewer matches. It is typically agreed that people prefer a moderate amount of matches, not identical, in a potential mate. Some matches make sense. Since you are alive to reach the reproductive age, your genes must be doing something right! It is clear that humans can "smell" genetically compatible mates.


Shows brain activation after exposure to AND.
Note similarities between heterosexual women (HeW) and homosexual men (HoM) (

Human pheromones are sex-specific and sexual orientation-specific. In all the studies previously described, the subjects identified as heterosexual. Homosexual males respond to AND in the same way as heterosexual females. Savic et al noted increased brain activity in several sexually dimorphic nuclei, such as the medial preoptic area. AND elicits a response from lesbian women as would be expected from common odors, where EST partly activates the anterior hypothalamus, as in heterosexual men would. (Savic et al, 2001) Remember that the anterior hypothalamus is involved in physiological arousal. In homosexual men, AND elicits a response congruent with that of heterosexual women: activation of the preoptic area and the anterior hypothalamus. In contrast to homosexual women, homosexual males' sexual brain regions were activated similarly with AND and EST.

Martins et al (2005) tested preference between limited choices of sweat (rather than synthetic pheromones as Savic did) collected from heterosexual males, homosexual males, heterosexual females and homosexual females. The results were startlingly significant. Homosexual males and homosexual females prefer pheromones from each other to all other options. All but homosexual males prefer heterosexual male sweat to homosexual male. Heterosexual men didn't significantly prefer homosexual of heterosexual females, in contrast to females who prefer heterosexual men. All other groups prefer heterosexual females to homosexual. Gay males preferred heterosexual female sweat to heterosexual male sweat. This study also believes there to be a relationship between sexual orientation and MHC, which has already been shown to be able to be detected with pheromones (see mate compatibility). MHC might affect the sensitivity to testosterone. It is difficult to make specific conclusions from this so far, but it is clear that there is significant differences in the way humans react to pheromones based on sexual orientation.


The previously discussed studies have suggested that pheromones can communicate information about genetic compatibility, sexual orientation and gender by affecting humans unconsciously and physically. Reproduction is the most important aspect when it comes to being a successful species, comparable only, perhaps, to survival. Of course it makes sense, then, that there are mechanisms beyond our control that urge us not only to procreate, but with mates that will result in the most successful offspring. I urge those of you who are not yet die-hard determinists not to view the effect of pheromones on mate choice as a loss of agency, but as a gain of sensory ability, enabling us to make a more informed decision. Pheromones are just another way we can communicate.