The term ‘pheromone’ is derived from ancient Greek pherein ‘to transfer’ and hormōn ‘to excite’. Pheromones are known as chemical signals that convey information between members of the same species (Karlson and Luscher, 1959). Less appreciated are their intricate interplays with the central neural and neuroendocrine systems. In insects and rodents, specific neuropeptides and hormones are critically involved in the production, release, as well as perception of pheromones (Cusson and McNeil, 1989; Lin et al., 2016; Mugford and Nowell, 1971; Raina et al., 1989; Tang et al., 1989). Little is known as to whether this link holds for non-rodent mammals or how it is manifested there. In humans, particularly, the very existence of pheromones has long been an issue of debate (Wyatt, 2015). Since the first report of human menstrual synchrony in 1971 (McClintock, 1971), nearly five decades of research has shown that human body secretions, particularly axillary sweat, subconsciously convey emotion and reproductive state (de Groot et al., 2015; Stern and McClintock, 1998; Zhou and Chen, 2009), which in turn are driven by complex neurochemical changes (Blouin et al., 2013; Jones and Lopez, 2013). Of the many components of human body secretions, two endogenous steroids, androsta-4,16,-dien-3-one, a non-androgenic derivative of gonadal progesterone (Gower and Ruparelia, 1993), and estra-1,3,5 (10),16-tetraen-3-ol, related to the estrogen sex hormones but with no known estrogenic effects (Thysen et al., 1968), emerge as candidates of human sex pheromones. Aside from affecting autonomic responses and mood states in women and men, respectively (Bensafi et al., 2004; Jacob and McClintock, 2000; Lundström et al., 2003a; Olsson et al., 2006), androstadienone and estratetraenol are found to effectively convey sex-specific information and activate the hypothalamus, a structure critically involved in sexual reproduction (Simerly, 2002), in distinct patterns. Specifically, androstadienone signals masculinity and estratetraenol signals femininity to their targeted recipients (Zhou et al., 2014). They also prime the identification of emotionally receptive states for the potential mates with whom they are associated (Ye et al., 2019). In parallel, androstadienone activates the hypothalamus in heterosexual women and homosexual men, but not in heterosexual men or homosexual women, whereas estratetraenol activates the hypothalamus in heterosexual men and homosexual women, but not in heterosexual women or homosexual men (Berglund et al., 2006; Savic et al., 2001; Savic et al., 2005).

The hypothalamus contains several sexually dimorphic nuclei and is well documented to coordinate neuroendocrine responses with sensory cues that regulate motivated behavior (Simerly, 2002). It produces, among various hormones, two structurally similar nonapeptides called oxytocin and vasopressin that function both as neuropeptides and hormones and have ancient roles in reproductive behaviors. Oxytocin and vasopressin are heavily implicated in sexual arousal and subsequent copulatory behavior in rats, rabbits, rams, bulls, and humans, with their homologs mediating comparable activities in nematodes, snails, and bony fish (Carter, 1992; Donaldson and Young, 2008). Moreover, the two peptides, especially oxytocin, are believed to underlie the evolution and expression of human sociality including trust, love, and cooperation (Carter, 2014; Donaldson and Young, 2008). Based on the animal literature, receptors for oxytocin and vasopressin are expressed in the olfactory system and other downstream sexually dimorphic nuclei under the modulatory influence of gonadal steroids, and are involved in the processing of social odors (Stoop, 2012). These findings have led us to suspect that the effects of androstadienone and estratetraenol, considered by some as putative human sex pheromones, could be regulated by oxytocin and/or vasopressin.

To examine the above hypothesis, we employed a gender identification task of point-light walkers (PLWs) (Figure 1A)—dynamic point-light displays portraying the gaits of walkers—that had been shown to be sensitive to the interactive effects of androstadienone, estratetraenol, and the recipients’ sex and sexual orientation (Zhou et al., 2014), and combined it with pharmacological manipulations of central oxytocin and vasopressin. We also measured each participant’s social proficiency (the opposite of autistic-like tendency) with the Autism Spectrum Quotient (AQ) (Baron-Cohen et al., 2001), in view of the positive association between social proficiency and endogenous oxytocin level (Koven and Max, 2014; Lancaster et al., 2015; Parker et al., 2014) and research showing that individual differences in social proficiency or AQ score can moderate the effects of exogenously administered oxytocin (Bartz et al., 2019; Bartz et al., 2010; Bartz et al., 2011). From the data we obtained psychometric curves that depicted the probability of making male judgments as a function of the physical gender of the PLWs (Figure 1B) under different combinations of olfactory stimuli and drug treatments. Systematic comparisons of these psychometric curves (Figure 1C) across five experiments totaling 1056 testing sessions enabled us to assess the roles of oxytocin and vasopressin in chemosensory communications of masculinity and femininity through androstadienone and estratetraenol in both heterosexual (Experiment 1, sample size n = 72) and homosexual (Experiment 2, n = 72) men (Figure 1D) as well as in socially less proficient (Experiments 3 and 4, n = 24 in each experiment) and socially proficient (Experiment 5, n = 24) individuals (Figure 1E).

Figure 1

Download asset Open asset

General procedures of the gender identification task have been described elsewhere (Zhou et al., 2014). In brief, genders of the PLWs were quantified as Z scores (Troje, 2002) and ranged in seven equal steps from feminine (−0.45 SD) to masculine (0.45 SD) with 0 marking the approximate gender-neutral point individually adjusted for each participant prior to the actual experiment in the absence of olfactory stimulus and drug treatment. In each trial of the task, participants viewed a PLW for 500 ms (0.5 walking cycle) and made a forced choice judgment on whether it was a male or a female walker (Figure 1A). To characterize how gender perception criteria were influenced by androstadienone or estratetraenol under different manipulations of oxytocin and vasopressin, we fitted gender judgments for all PLWs separately for each olfactory condition and each participant under each drug treatment condition with a Boltzmann sigmoid function that contained two parameters: point of subjective equality (PSE) and difference limen (Figure 1B). The PSE, an index of response criterion, is the Z score (which denoted PLWs’ physical gender) corresponding to a probability of 50% on the fitted psychometric function, where the participant perceived a PLW as equally masculine and feminine. The difference limen is half the interquartile range of the fitted function and indicates response sensitivity. For each drug treatment condition, participants’ PSEs when smelling androstadienone or estratetraenol—each dissolved in a clove oil carrier solution (1% v/v clove oil in propylene glycol)—were compared with those when smelling the clove oil carrier solution alone (control). A negative PSE shift relative to the carrier control condition corresponds to an overall leftward shift of the psychometric curve and reflects an increased tendency to judge the PLWs as male, hence a masculine bias in gender perception (Figure 1C). Conversely, a positive PSE shift indicates a feminine bias in gender perception.

Oxytocin, vasopressin, and subconscious chemosensory decoding of sex in heterosexual and homosexual men

We first recruited 72 heterosexual males (Kinsey scores = 0) in Experiment 1 and randomly assigned them to three groups of 24 each to receive different intranasal drug treatments: 24 IU oxytocin, 24 IU vasopressin, or 60 µg atosiban—a desamino-oxytocin analogue and a competitive antagonist of both oxytocin and vasopressin receptors (Manning et al., 2012). Nasally delivered oxytocin and vasopressin can directly access the cerebrospinal fluid (Born et al., 2002; Freeman et al., 2016); there has been some limited indication that intranasal atosiban is also centrally available (Liu et al., 2018; Lundin et al., 1986). In a double-blind procedure, each participant was tested in 3 sessions held at around the same time of the day on 3 consecutive days. During each session they performed the gender identification task (Figure 1A) while being continuously exposed to either androstadienone (500 μM, 5 ml), estratetraenol (500 μM, 5 ml) or their carrier solution alone (control condition, 1% v/v clove oil in propylene glycol, 5 ml), one per session in a counterbalanced manner, 35 min following drug administration (Figure 1D). The three olfactory stimuli were perceptually indiscriminable as assessed in a separate panel of 48 male participants (mean triangular discrimination accuracy = 0.33 vs. chance = 0.33, t₄₇ = 0.02, p>0.9), which was also in line with earlier reports (Ye et al., 2019; Zhou et al., 2014). Our previous study had shown that heterosexual males were affected by estratetraenol, but not androstadienone, in making gender judgments. Specifically, estratetraenol subconsciously biased them toward perceiving the PLWs as more feminine (Figure 2A; Zhou et al., 2014). We reasoned that the effect of estratetraenol, if regulated by the oxytocin/vasopressin system, would be blocked by intranasal atosiban and modulated by intranasal oxytocin and/or vasopressin. It was difficult to predict the directions of their effects (if any), as both positive and negative effects of oxytocin and vasopressin have been reported in the literature, depending on dose, context, and personal characteristics (Bartz et al., 2011; Carter, 2014; Donaldson and Young, 2008). It was also possible that only one of oxytocin and vasopressin plays a role, in which case the chemosensory effect would not be altered by the administration of the other.

Figure 2

Download asset Open asset

Figure 2—source data 1

Experiments 1 and 2.

Androstadienone- and estratetraenol- induced visual gender judgment biases after the nasal administrations of 60 µg atosiban, 24 IU oxytocin, and 24 IU vasopressin in heterosexual and homosexual male participants.

https://cdn.elifesciences.org/articles/59376/elife-59376-fig2-data1-v1.data.xlsx

Download elife-59376-fig2-data1-v1.data.xlsx

Indeed, in the atosiban-treated heterosexual men, we found that smelling estratetraenol relative to the carrier solution alone failed to influence their gender perception criteria (indexed by the PSEs, t₂₃ = 1.37, p=0.18; Figure 2B,E). Interestingly, this was also the case for those treated with 24 IU oxytocin (t₂₃ = 0.17, p=0.87; Figure 2C,E). Only in the participants treated with 24 IU vasopressin did estratetraenol induce a systematic bias towards perceiving the PLWs as more feminine (t₂₃ = 4.25, p<0.001, Cohen’s d = 0.87; Figure 2D,E). The size of the estratetraenol-induced gender perception bias under vasopressin was similar to that found earlier without drug treatment (Zhou et al., 2014), and a direct comparison of the two sets of data showed no significant difference (p=0.35). In other words, vasopressin did not significantly enhance or diminish the effect of estratetraenol on heterosexual males. On the other hand, smelling androstadienone relative to the carrier solution alone did not influence gender perception criteria regardless of drug treatments (i.e. 60 µg atosiban, 24 IU oxytocin, or 24 IU vasopressin, ps > 0.34; Figure 2B–E), consistent with earlier findings (Zhou et al., 2014).

These results presented a complex picture. Antagonizing oxytocin and vasopressin receptors with atosiban abolished the known effect of estratetraenol on heterosexual males (i.e. biasing them toward perceiving the PLWs as more feminine), yet 24 IU oxytocin appeared to produce the same effect. On the flip side, the administration of 24 IU vasopressin did not significantly alter the processing of estratetraenol in heterosexual males—they remained biased toward perceiving the PLWs as more feminine under the exposure of estratetraenol, to the same extent as when no drug was administered. We wondered if this pattern of drug influences would hold for the chemosensory decoding of masculine information carried by androstadienone. To this end, we turned to homosexual males in Experiment 2, who had been shown to be subconsciously biased by androstadienone, but not estratetraenol, in making gender judgments (Figure 2F; Zhou et al., 2014).

Except for the participants’ sexual orientation (homosexual males, mean Kinsey score ± SD = 5.26 ± 0.63), Experiment 2 was identical to Experiment 1, and revealed a similar pattern of drug effects. Gender perception criteria were unaffected by the exposure to androstadienone, relative to the carrier control, in the homosexual men treated with 60 µg atosiban (t₂₃ = 0.88, p=0.39; Figure 2G,J) as well as in those treated with 24 IU oxytocin (t₂₃ = −1.29, p=0.21; Figure 2H,J). By contrast, in the 24 IU vasopressin group, smelling androstadienone induced a systematic bias toward perceiving the PLWs as more masculine (t₂₃ = −2.99, p=0.007, Cohen’s d = 0.61; Figure 2I,J). The strength of the androstadienone-induced masculine bias under vasopressin was again not different from that found earlier (Zhou et al., 2014) in homosexual males receiving no drug treatment (p=0.18). Meanwhile, in line with earlier findings (Zhou et al., 2014), smelling estratetraenol did not influence gender judgements in these homosexual men irrespective of drug manipulations (ps > 0.58; Figure 2G–J).

An omnibus ANOVA of the PSEs from both Experiments 1 and 2 (olfactory condition × drug treatment × sexual orientation) identified a significant interaction between olfactory condition and drug treatment (F_{4, 276} = 3.58, p=0.007, partial η² = 0.05), which reinforced that the effects of androstadienone and estratetraenol on gender judgments were modulated by the drug manipulations, and that this modulation was similar in both heterosexual and homosexual males (olfactory condition × drug treatment × sexual orientation: F_{4, 276} = 0.80, p=0.53). To facilitate comparison, we highlighted the central tendencies of the androstadienone- and estratetraenol-induced PSE shifts (x- and y- axes, respectively) under different drug treatments in Figure 3 (see also Figure 3—figure supplement 1), generated by using a standard bootstrapping procedure (Davison and Hinkley, 1997; see Materials and methods). Under 60 µg atosiban and 24 IU oxytocin, neither androstadienone nor estratetraenol induced any significant change in gender perception criterion in heterosexual (cyan dots) or homosexual males (lime dots), and the bootstrapped sample means of the two groups of men overlapped around the origin. Conversely, under 24 IU vasopressin, they formed two discrete clusters: estratetraenol biased heterosexual, but not homosexual, males toward perceiving the PLWs as more feminine (cyan dots fell around the vertical axis on the positive side), whereas androstadienone biased homosexual, but not heterosexual, males toward perceiving the PLWs as more masculine (lime dots fell around the horizontal axis on the negative side). In addition, the difference limens of the heterosexual and homosexual males did not differ (F_{1, 138} = 0.47, p=0.50) and were comparable across olfactory conditions and drug treatments (olfactory condition × drug treatment: F_{4, 276} = 1.19, p=0.31; olfactory condition: F_{2, 276} = 0.48, p=0.62; drug treatment: F_{2, 138} = 0.24, p=0.79). So were their self-reported mood states on the Profile of Mood States (McNair et al., 1971) (POMS; total mood disturbance: ps = 0.84, 0.51 and 0.90, respectively; all subscales: ps > 0.05, corrected). Thus, it was the criterion (reflected in the PSEs) rather than the sensitivity of gender judgment (reflected in the difference limens) or transient mood state that was swayed by the interplays between the chemosignals and the drug manipulations.

Figure 3 with 2 supplements see all

Download asset Open asset

Figure 3—source data 1

Experiments 1 and 2.

Degrees of gender perception biases induced by chemosensory sexual cues in high AQ and low AQ individuals.

https://cdn.elifesciences.org/articles/59376/elife-59376-fig3-data1-v1.data.xlsx

Download elife-59376-fig3-data1-v1.data.xlsx

The combined results of Experiments 1 and 2 thus suggested that the decoding of feminine information carried by estratetraenol and that of masculine information carried by androstadienone, while contingent upon the recipients’ sexual orientation, were subserved by similar neuroendocrine mechanisms that were disrupted by intranasal atosiban—the competitive antagonist of both oxytocin and vasopressin receptors, as well as by 24 IU oxytocin, and were unaffected by 24 IU vasopressin. Since vasopressin seemed to exert no effect (participants’ response patterns to the chemosignals under 24 IU vasopressin were comparable to those previously obtained without drug treatment), by deduction, such mechanisms involved oxytocin. The question remained as to why the administration of 24 IU oxytocin, like atosiban, exempted the participants from the influences of the chemosignals, and we explored it in more detail.

If, regardless of one’s sexual orientation, oxytocin plays a role in the processing of chemosensory sexual cues associated with the preferred sex, given the relationship between social proficiency and endogenous oxytocin level (Koven and Max, 2014; Lancaster et al., 2015; Parker et al., 2014) and the heterogenous effects of intranasal oxytocin on individuals with different levels of social proficiency (Bartz et al., 2011), specifically as assessed by the AQ (Bartz et al., 2019; Bartz et al., 2010), it follows that individuals with different AQ scores could differ in their susceptibility to such chemosignals as well as to the effect of exogenous oxytocin. Put differently, social proficiency could be related to the subconscious extraction of chemosensory sexual information, a link hitherto unsuspected. As an initial attempt to test this inference, we reexamined the data from Experiments 1 and 2 to see if high AQ and low AQ participants differed in their susceptibility to the chemosignals. The participants’ AQ scores ranged from 9 to 35. We adopted a relatively strict criterion—AQ scores ≥ 25, that is, 1 SD or more above the reported mean for males (Baron-Cohen et al., 2001)—for high AQ individuals in hopes to better capture the effects of social proficiency (Bartz et al., 2019). Whereas an AQ score above 32 is used as the cutoff for distinguishing individuals with clinically significant levels of autistic traits (Baron-Cohen et al., 2001), we did not opt for this stringent cutoff value here due to pragmatic considerations: Such individuals are rare in the general population and only 1 out of the 144 participants scored above 32. Those with AQ scores below 25 were classified as low AQ individuals. Overall, AQ scores were comparable among the drug groups and also between the heterosexual and the homosexual participants in Experiments 1 and 2 (drug treatment: F_{2, 138} = 0.48, p=0.62; sexual orientation: F_{1, 138} = 1.39, p=0.24; interaction: F_{2, 138} = 0.097, p=0.91). Since no significant effect of androstadienone or estratetraenol was observed under the 60 µg atosiban condition or 24 IU oxytocin condition, we focused our examination on the 24 IU vasopressin condition where the effects of the two chemosignals were comparable to those obtained earlier without drug treatment (Zhou et al., 2014). The majority (85.4%) of the participants were low AQ individuals (AQ scores < 25). Our supplementary analysis revealed that, relative to the high AQ participants (n = 7, 14.6%), estratetraenol and androstadienone induced larger gender perception biases in the low AQ participants, driving the overall effects of the chemosignals on gender perception (Figure 3—figure supplement 2). The results thus lent preliminary support to the postulated link between social proficiency and the subconscious processing of chemosignals, and led us to assess the effect of exogenous oxytocin separately in high AQ and low AQ individuals.

Dose-dependent modulation of chemosensory decoding of sex by oxytocin but not vasopressin in high and low AQ individuals

It has been reported that the effect of intranasal oxytocin is more pronounced in socially less proficient individuals as measured by the AQ (Bartz et al., 2019; Bartz et al., 2010; Bartz et al., 2011) and that a lower dose of oxytocin could exert a more positive effect than a higher dose (Cardoso et al., 2013; Quintana et al., 2017; Spengler et al., 2017). To better probe the role of oxytocin in the chemosensory processing of sexual information and not miss any possible effect of vasopressin, we recruited 24 heterosexual males with high AQ scores (range: 25–36) in Experiment 3 and assessed the extent to which their gender perception was biased by estratetraenol under different doses of oxytocin and vasopressin in a within-subject design (Figure 1E). Specifically, each participant was tested in 10 sessions held around the same time of the day on 10 days that comprised of five drug conditions (2 sessions each): 12 IU oxytocin, 24 IU oxytocin, 12 IU vasopressin, 24 IU vasopressin, and no drug treatment. Under each drug condition, they performed the gender identification task while being continuously exposed to estratetraenol in one session, and to the carrier solution alone in the other session. Androstadienone was not included as it consistently showed no effect on heterosexual men’s gender perception (Experiment 1) (Zhou et al., 2014).

In the absence of drug treatment, the high AQ heterosexual men failed to utilize the feminine information carried by estratetraenol in making gender judgments (t₂₃ = 0.70, p=0.49; Figure 4A,F), which echoed with the preliminary result from the high AQ individuals in Experiments 1 and 2 (Figure 3—figure supplement 2). Remarkably, this chemosensory ability was restored by the administration of 12 IU (t₂₃ = 2.72, p=0.012, Cohen’s d = 0.55; Figure 4B,F), but not 24 IU (t₂₃ = −0.36, p=0.72; Figure 4C,F), intranasal oxytocin. Meanwhile, the participants remained insusceptible to estratetraenol regardless of whether they were treated with 12 IU (t₂₃ = 0.76, p=0.46; Figure 4D,F) or 24 IU vasopressin (t₂₃ = −0.24, p=0.81; Figure 4E,F). The results thus more directly pointed to the involvement of oxytocin, but not vasopressin, in the processing of chemosensory sexual information. Critically, they suggested that the effect of oxytocin was not monotonic. At 12 IU, intranasal oxytocin facilitated the decoding of the feminine information carried by estratetraenol in high AQ heterosexual men; but at 24 IU, the effect disappeared.

Figure 4 with 1 supplement see all

Download asset Open asset

Figure 4—source data 1

Experiments 3 and 5.

Estratetraenol-induced visual gender judgment biases in the absence of drug treatment and after the nasal administrations of 12 IU oxytocin, 24 IU oxytocin, 12 IU vasopressin, and 24 IU vasopressin in high AQ and low AQ heterosexual male participants.

https://cdn.elifesciences.org/articles/59376/elife-59376-fig4-data1-v1.xlsx

Download elife-59376-fig4-data1-v1.xlsx

Figure 4—source data 2

Experiment 4.

Estratetraenol-induced visual gender judgment biases after the nasal administrations of saline, 12 IU oxytocin, and 24 IU oxytocin in high AQ heterosexual male participants.

https://cdn.elifesciences.org/articles/59376/elife-59376-fig4-data2-v1.xlsx

Download elife-59376-fig4-data2-v1.xlsx

The nonmonotonic effect of oxytocin on chemosensory decoding of sex was striking, and we sought to replicate it before continuation. In Experiment 4, we adopted a fully double-blind placebo-controlled within-subject design and recruited another 24 high AQ (range: 25–39) heterosexual men, who underwent three drug conditions: saline, 12 IU oxytocin, and 24 IU oxytocin. Their results mirrored those of Experiment 3 (Figure 4—figure supplement 1). The combined data from Experiments 3 and 4 affirmed that high AQ heterosexual men were unsusceptible to estratetraenol at baseline (no drug treatment/saline; t₄₇ = 0.95, p=0.35; with no difference between no drug treatment and saline, p=0.95), yet showed a robust estratetraenol-induced gender perception bias when treated with 12 IU oxytocin (t₄₇ = 4.10, p<0.001, Cohen’s d = 0.59) and not 24 IU oxytocin (t₄₇ = −0.025, p=0.98). There was overall a significant quadratic effect of intranasal oxytocin dose on estratetraenol-induced shift of gender judgment criterion in these participants (F_{1, 47} = 6.58, p=0.014, partial η² = 0.12). We hence deduced that oxytocin level and the chemosensory processing of sexual cues follow an inverted-U-shaped relationship. It follows that low AQ individuals, who presumably have higher levels of endogenous oxytocin (Koven and Max, 2014; Lancaster et al., 2015; Parker et al., 2014), would benefit less, if any, from 12 IU oxytocin, and not from 24 IU oxytocin. Furthermore, we predicted that their decoding of chemosensory sexual information would not be influenced by vasopressin regardless of the dose.

Our hypotheses were confirmed in Experiment 5, which was identical to Experiment 3 except that the participants were 24 heterosexual males with low AQ scores (range: 8–24). In sharp contrast to the high AQ men in Experiments 3 and 4, the low AQ individuals decoded the feminine information carried by estratetraenol in the absence of drug treatment (t₂₃ = 3.15, p=0.005, Cohen’s d = 0.64; Figure 4G,L), again in line with the preliminary result from the low AQ individuals in Experiments 1 and 2 (Figure 3—figure supplement 2). The chemosensory effect was nonetheless diminished following the administration of 12 IU oxytocin (t₂₃ = 1.30, p=0.21; Figure 4H,L) and was abolished by the administration of 24 IU oxytocin (t₂₃ = 0.47, p=0.64; Figure 4I,L); that is, exogenous oxytocin at these doses hampered rather than facilitated the utilization of chemosensory feminine information in these low AQ individuals. Vasopressin, as expected, exerted no influence. The participants remained capable of detecting the feminine information carried by estratetraenol under the treatments of 12 IU and 24 IU vasopressin (t₂₃s = 4.02 and 3.10, ps = 0.001 and 0.005, Cohen’s ds = 0.82 and 0.63, respectively; Figure 4J–L).

To further characterize the role of oxytocin in the chemosensory processing of sexual information, we analyzed the pooled PSEs across different combinations of oxytocin doses (no drug treatment/saline, 12 IU oxytocin, 24 IU oxytocin) and olfactory stimuli (estratetraenol vs. carrier control) from both the high AQ and low AQ heterosexual men in Experiments 3–5. We found that relative to baseline (no drug treatment/saline), the high AQ and low AQ individuals showed opposite patterns when treated with 12 IU oxytocin, resulting in a significant three-way interaction among social proficiency, oxytocin treatment (no drug treatment/saline vs. 12 IU oxytocin) and olfactory condition (F_{1, 70} = 5.65, p=0.020, partial η² = 0.075), on top of a strong main effect of olfactory condition (F_{1, 70} = 20.63, p<0.0001, partial η² = 0.23) and a marginally significant interaction between social proficiency and oxytocin treatment (F_{1, 70} = 3.13, p=0.081). To illustrate these effects, we plotted in Figure 5A the central tendencies of the estratetraenol-induced PSE shifts (relative to the carrier control) at baseline (x axis) and under 12 IU oxytocin (y axis) from both the high AQ and low AQ heterosexual men in Experiments 3–5. The bootstrapped sample means of the two groups of men formed two discrete clusters: those of high AQ individuals (dark brown dots) fell around the vertical axis on the positive side, indicating that they utilized the feminine information carried by estratetraenol only after 12 IU oxytocin treatment, and those of low AQ individuals (light brown dots) fell around the horizontal axis on the positive side, indicating a significant effect of estratetraenol in them at baseline but not after 12 IU oxytocin treatment. A closer inspection of the data revealed that AQ score, while negatively correlated with estratetraenol-induced PSE shift at baseline (r₇₂ = −0.26, p=0.028), was positively correlated with the increase of estratetraenol-induced PSE shift post 12 IU oxytocin treatment (i.e. estratetraenol-induced PSE shift under 12 IU oxytocin minus that at baseline; r₇₂ = 0.29, p=0.014) (Figure 5B). In other words, one’s social proficiency was predictive of his response to 12 IU oxytocin treatment. Under 24 IU oxytocin, both groups of men became insusceptible to estratetraenol (olfactory condition × social proficiency: F_{1, 70} = 0.13, p=0.72; olfactory condition: F_{1, 70} = 0.11, p=0.74; social proficiency: F_{1, 70} = 0.072, p=0.79), consistent with the results of Experiment 1. Overall, as summarized in Figure 5C, the effect of exogenous oxytocin on the chemosensory processing of sexual cues was dose-dependent, nonmonotonic and contingent upon the recipient’s social proficiency. By contrast, exogenous vasopressin did not exert any significant impact.

Figure 5

Download asset Open asset

At the same time, the difference limens of the high AQ and low AQ heterosexual men in Experiments 3 and 5 did not differ (F_{1, 46} = 0.078, p=0.78) and were comparable across all olfactory conditions and drug treatments (i.e. 12 IU oxytocin, 24 IU oxytocin, 12 IU vasopressin, 24 IU vasopressin, and no drug treatment) (olfactory condition × drug treatment: F_{4, 184} = 1.48, p=0.21; olfactory condition: F_{1, 46} = 1.00, p=0.32; drug treatment: F_{4, 184} = 1.66, p=0.16). Their mood states, as reflected by self-reported ratings on the POMS, were also stable across olfactory conditions and drug treatments (total mood disturbance: ps = 0.76, 0.86, and 0.45, respectively; all subscales: ps > 0.05, corrected). We also specifically examined the difference limens and POMS ratings across different doses of oxytocin (no drug treatment/saline, 12 IU oxytocin and 24 IU oxytocin) and olfactory conditions for all the participants in Experiments 3–5, and obtained the same results (ps > 0.5). Moreover, these participants could not tell apart estratetraenol and the carrier solution alone by smell (mean accuracy = 0.36 vs. chance = 0.33, t₇₁ = 0.93, p=0.36). We therefore concluded that the chemosensory processing of sexual information took place below olfactory awareness. Through interactions with the oxytocin system, it shifted one’s criterion, but not sensitivity, of gender perception, without significantly altering his transient mood state.

The current study represents an initial effort to unravel the neuroendocrine mechanisms underlying human chemo-signaling of sex. Data collected through formal psychophysical testing of 216 individuals over a total of 1056 testing sessions jointly demonstrate that the decoding of chemosensory sexual cues, including that of estratetraenol in heterosexual men and of androstadienone in homosexual men, is modulated by oxytocin instead of vasopressin in a dose-dependent manner, and is blocked by atosiban, a competitive antagonist of both oxytocin and vasopressin receptors (Manning et al., 2012; Table 1).

We are aware of a recent criticism of studies using androstadienone and/or estratetraenol that show a positive effect, which states that these studies were underpowered and problematic due to small sample sizes, lack of a priori evidence of effects, and lack of full replication, and the results are likely false positives (Wyatt, 2015). We, along with other researchers in the field (Endevelt-Shapira et al., 2018), do not agree with this criticism and believe that the current study presents a strong counter-argument. We had clear a priori evidence of the effects of androstadienone and estratetraenol from an earlier study (Zhou et al., 2014). Using the same task as in that study, we consistently replicated the original findings (Table 1), namely, androstadienone signals masculinity to homosexual men (Experiment 2) and estratetraenol signals femininity to heterosexual men (Experiments 1, 3, 4, and 5). The overall effect size was comparable to that of gender adaptation using visually presented faces or bodies (Ghuman et al., 2010). Moreover, whereas the natural occurring concentration of estratetraenol in sweat has not been measured, the concentration of androstadienone (500 µM) presented to the participants was similar to that in freshly produced apocrine sweat (mean = 0.44 nmol/μl = 0.44 × 10⁻³ mol/l = 440 μM) (Gower et al., 1994) and thus arguably ecologically relevant (although likely significantly higher than those encountered in non-intimate social interactions). It was indeed the reliability of these chemosensory effects that allowed us to probe into the underlying neuroendocrine mechanisms and uncover their modulation by oxytocin rather than vasopressin.

Oxytocin and vasopressin differ from each other at only two amino acid positions. Both are strongly implicated in a range of reproductive and social behaviors in animals as well as humans (Carter, 1992; Carter, 2014; Donaldson and Young, 2008). Many of the roles of oxytocin have been associated with female-typical behaviors in animals, and many of the behaviors associated with vasopressin have been demonstrated in males (McCall and Singer, 2012). Nonetheless, the specific behaviors they influence show extensive variation among different species, and to which extent the sex-related differences hold for humans is scantly known (Donaldson and Young, 2008). Here, we found in heterosexual and homosexual men that the utilization of the feminine information carried by estratetraenol, while ‘male-typical’, was not modulated by vasopressin regardless of its dose or the recipient’s social proficiency, but was instead modulated by oxytocin, like the ‘female-typical’ utilization of the masculine information carried by androstadienone. These results hence argue against a clear-cut sex- or sexual orientation-based division of labor between the oxytocin and vasopressin systems in humans.

Importantly, we observed that the oxytocinergic modulation of human chemosensory decoding of sex was dose-dependent, non-monotonic, and roughly followed an inverted-U-shaped function, and that the dose effect of exogenous oxytocin depended on the recipient’s social proficiency. At 12 IU, nasally administered oxytocin restored the ability to utilize the feminine information carried by estratetraenol in socially less proficient heterosexual men, yet hampered the very ability in socially proficient ones. Overall, the less socially proficient a man was, the less he utilized chemosensory sexual information at baseline, the more he became to do so after 12 IU oxytocin treatment, and vice versa. Increasing the dose of oxytocin to 24 IU eliminated the processing of the chemosensory sexual cues irrespective of the recipient’s sexual orientation or social proficiency. It has been shown in rats that the dose-response curve for oxytocin is bell-shaped. Moderate doses of oxytocin facilitate, whereas high doses of oxytocin inhibit penile erections (Argiolas and Gessa, 1991). Similarly, moderate doses of oxytocin promote, whereas high doses of oxytocin disrupt social memory (Benelli et al., 1995). Moreover, a marked increase of oxytocin has been suspected to mediate or signal satiety and contribute to a refractory state in animals (Carter, 1992). In humans, there has also been hints that the effect of intranasal oxytocin is dose-dependent (Cardoso et al., 2013; Quintana et al., 2017). A recent study indicates that pretreatment blood oxytocin concentrations predict treatment response to a set dosage of intranasal oxytocin in children with autism (Parker et al., 2017). Several others note that the influence of intranasal oxytocin on socially proficient individuals is complicated, and can be null or even antisocial (Bartz et al., 2011). Our findings dovetail with these reports. Moreover, they provide strong behavioral evidence for a non-monotonic effect of intranasal oxytocin that interacts with the recipient’s social proficiency. While we tested healthy male volunteers, the non-monotonic property of oxytocin’s effects has significant implications in the design of oxytocin-based treatment protocols for conditions like autism spectrum disorder.

Following the work of Kosfeld et al., 2005, the majority of studies on the effect of oxytocin in humans have used a single intranasal dose of 24 IU and shown a prosocial effect. Most have not simultaneously evaluated the effect of vasopressin or an antagonist of oxytocin or vasopressin receptors. In the limited studies noting that a higher dose of oxytocin produces a smaller effect, the phenomenon is speculated to be a result of cross-binding to vasopressin receptors that ‘cancels out’ the oxytocinergic effect (Cardoso et al., 2013). In our study, only at 12 IU did intranasal oxytocin significantly facilitate the decoding of sexual information in socially less proficient individuals. There was no significant effect of vasopressin irrespective of the dose. Thus, the lack of a positive effect of 24 IU oxytocin could not be explained by cross-binding of oxytocin to vasopressin receptors. Rather, we argue that it reflects the dose-response characteristics of oxytocin. Moreover, in view of the well-documented prosocial effect of 24 IU intranasal oxytocin, our results suggest that the optimal oxytocin level for the decoding of chemosensory sexual information differs from that for the promotion of prosocial behavior. After all, social behavior evolves as a result of effects upon the reproductive competition among group members (Alexander, 1974).

In rats, oxytocin has been found to enhance social recognition by modulating cortical control of early olfactory processing (Oettl et al., 2016). What neural pathways subserve the observed oxytocinergic modulation of human chemosensory communication? How does oxytocin act on sexually dimorphic sensory processing? These interesting questions await further research to clarify. Given the evolutionary and anatomical intimacy between the olfactory system and the neuroendocrine system, particularly through the hypothalamus (Gorbman, 1995), analyses of their interplays will open up new avenues for the regulations of both sensory perception of the external world and physiological processes within the human body.

All data analyzed during this study are included in the manuscript and supporting files. Source data files have been provided for Figures 2, 3 and 4.

1. 1. Alexander RD

   (1974) The evolution of social behavior

   Annual Review of Ecology and Systematics 5:325–383.

   https://doi.org/10.1146/annurev.es.05.110174.001545

   • Google Scholar
2. 1. Argiolas A
   2. Gessa GL

   (1991) Central functions of oxytocin

   Neuroscience & Biobehavioral Reviews 15:217–231.

   https://doi.org/10.1016/S0149-7634(05)80002-8

   • PubMed
   • Google Scholar
3. 1. Baron-Cohen S
   2. Wheelwright S
   3. Skinner R
   4. Martin J
   5. Clubley E

   (2001) The autism-spectrum quotient (AQ): evidence from asperger syndrome/high-functioning autism, males and females, scientists and mathematicians

   Journal of Autism and Developmental Disorders 31:5–17.

   https://doi.org/10.1023/a:1005653411471

   • PubMed
   • Google Scholar
4. 1. Bartz JA
   2. Zaki J
   3. Bolger N
   4. Hollander E
   5. Ludwig NN
   6. Kolevzon A
   7. Ochsner KN

   (2010) Oxytocin selectively improves empathic accuracy

   Psychological Science 21:1426–1428.

   https://doi.org/10.1177/0956797610383439

   • PubMed
   • Google Scholar
5. 1. Bartz JA
   2. Zaki J
   3. Bolger N
   4. Ochsner KN

   (2011) Social effects of oxytocin in humans: context and person matter

   Trends in Cognitive Sciences 15:301–309.

   https://doi.org/10.1016/j.tics.2011.05.002

   • PubMed
   • Google Scholar
6. 1. Bartz JA
   2. Nitschke JP
   3. Krol SA
   4. Tellier P-P

   (2019) Oxytocin selectively improves empathic accuracy: a replication in men and novel insights in women

   Biological Psychiatry: Cognitive Neuroscience and Neuroimaging 4:1042–1048.

   https://doi.org/10.1016/j.bpsc.2019.01.014

   • Google Scholar
7. 1. Benelli A
   2. Bertolini A
   3. Poggioli R
   4. Menozzi B
   5. Basaglia R
   6. Arletti R

   (1995) Polymodal dose-response curve for oxytocin in the social recognition test

   Neuropeptides 28:251–255.

   https://doi.org/10.1016/0143-4179(95)90029-2

   • PubMed
   • Google Scholar
8. 1. Bensafi M
   2. Brown WM
   3. Khan R
   4. Levenson B
   5. Sobel N

   (2004) Sniffing human sex-steroid derived compounds modulates mood, memory and autonomic nervous system function in specific behavioral contexts

   Behavioural Brain Research 152:11–22.

   https://doi.org/10.1016/j.bbr.2003.09.009

   • PubMed
   • Google Scholar
9. 1. Berglund H
   2. Lindström P
   3. Savic I

   (2006) Brain response to putative pheromones in lesbian women

   PNAS 103:8269–8274.

   https://doi.org/10.1073/pnas.0600331103

   • PubMed
   • Google Scholar
10. 1. Blouin AM
    2. Fried I
    3. Wilson CL
    4. Staba RJ
    5. Behnke EJ
    6. Lam HA
    7. Maidment NT
    8. Karlsson KÆ
    9. Lapierre JL
    10. Siegel JM

    (2013) Human hypocretin and melanin-concentrating hormone levels are linked to emotion and social interaction

    Nature Communications 4:1547.

    https://doi.org/10.1038/ncomms2461

    • PubMed
    • Google Scholar
11. 1. Born J
    2. Lange T
    3. Kern W
    4. McGregor GP
    5. Bickel U
    6. Fehm HL

    (2002) Sniffing neuropeptides: a transnasal approach to the human brain

    Nature Neuroscience 5:514–516.

    https://doi.org/10.1038/nn0602-849

    • PubMed
    • Google Scholar
12. 1. Cardoso C
    2. Ellenbogen MA
    3. Orlando MA
    4. Bacon SL
    5. Joober R

    (2013) Intranasal oxytocin attenuates the cortisol response to physical stress: a dose-response study

    Psychoneuroendocrinology 38:399–407.

    https://doi.org/10.1016/j.psyneuen.2012.07.013

    • PubMed
    • Google Scholar
13. 1. Carter CS

    (1992) Oxytocin and sexual behavior

    Neuroscience & Biobehavioral Reviews 16:131–144.

    https://doi.org/10.1016/S0149-7634(05)80176-9

    • PubMed
    • Google Scholar
14. 1. Carter CS

    (2014) Oxytocin pathways and the evolution of human behavior

    Annual Review of Psychology 65:17–39.

    https://doi.org/10.1146/annurev-psych-010213-115110

    • PubMed
    • Google Scholar
15. 1. Cusson M
    2. McNeil JN

    (1989) Involvement of juvenile hormone in the regulation of pheromone release activities in a moth

    Science 243:210–212.

    https://doi.org/10.1126/science.243.4888.210

    • PubMed
    • Google Scholar
16. Book

    1. Davison AC
    2. Hinkley DV

    (1997) Bootstrap Methods and Their Application

    Cambridge: Cambridge University Press.

    https://doi.org/10.1017/CBO9780511802843

    • Google Scholar
17. 1. de Groot JH
    2. Smeets MA
    3. Rowson MJ
    4. Bulsing PJ
    5. Blonk CG
    6. Wilkinson JE
    7. Semin GR

    (2015) A sniff of happiness

    Psychological Science 26:684–700.

    https://doi.org/10.1177/0956797614566318

    • PubMed
    • Google Scholar
18. 1. Donaldson ZR
    2. Young LJ

    (2008) Oxytocin, vasopressin, and the neurogenetics of sociality

    Science 322:900–904.

    https://doi.org/10.1126/science.1158668

    • PubMed
    • Google Scholar
19. 1. Endevelt-Shapira Y
    2. Perl O
    3. Ravia A
    4. Amir D
    5. Eisen A
    6. Bezalel V
    7. Rozenkrantz L
    8. Mishor E
    9. Pinchover L
    10. Soroka T
    11. Honigstein D
    12. Sobel N

    (2018) Altered responses to social chemosignals in autism spectrum disorder

    Nature Neuroscience 21:111–119.

    https://doi.org/10.1038/s41593-017-0024-x

    • PubMed
    • Google Scholar
20. 1. Freeman SM
    2. Samineni S
    3. Allen PC
    4. Stockinger D
    5. Bales KL
    6. Hwa GG
    7. Roberts JA

    (2016) Plasma and CSF oxytocin levels after intranasal and intravenous oxytocin in awake macaques

    Psychoneuroendocrinology 66:185–194.

    https://doi.org/10.1016/j.psyneuen.2016.01.014

    • PubMed
    • Google Scholar
21. 1. Ghuman AS
    2. McDaniel JR
    3. Martin A

    (2010) Face adaptation without a face

    Current Biology 20:32–36.

    https://doi.org/10.1016/j.cub.2009.10.077

    • PubMed
    • Google Scholar
22. 1. Gorbman A

    (1995) Olfactory origins and evolution of the brain-pituitary endocrine system: facts and speculation

    General and Comparative Endocrinology 97:171–178.

    https://doi.org/10.1006/gcen.1995.1016

    • PubMed
    • Google Scholar
23. 1. Gower DB
    2. Holland KT
    3. Mallet AI
    4. Rennie PJ
    5. Watkins WJ

    (1994) Comparison of 16-androstene steroid concentrations in sterile apocrine sweat and axillary secretions: interconversions of 16-androstenes by the axillary microflora--a mechanism for axillary odour production in man?

    The Journal of Steroid Biochemistry and Molecular Biology 48:409–418.

    https://doi.org/10.1016/0960-0760(94)90082-5

    • PubMed
    • Google Scholar
24. 1. Gower DB
    2. Ruparelia BA

    (1993) Olfaction in humans with special reference to odorous 16-androstenes: their occurrence, perception and possible social, psychological and sexual impact

    Journal of Endocrinology 137:167–187.

    https://doi.org/10.1677/joe.0.1370167

    • Google Scholar
25. 1. Insel TR
    2. Young L
    3. Witt DM
    4. Crews D

    (1993) Gonadal steroids have paradoxical effects on brain oxytocin receptors

    Journal of Neuroendocrinology 5:619–628.

    https://doi.org/10.1111/j.1365-2826.1993.tb00531.x

    • PubMed
    • Google Scholar
26. 1. Jacob S
    2. McClintock MK

    (2000) Psychological state and mood effects of steroidal chemosignals in women and men

    Hormones and Behavior 37:57–78.

    https://doi.org/10.1006/hbeh.1999.1559

    • PubMed
    • Google Scholar
27. Book

    1. Jones RE
    2. Lopez KH

    (2013)

    Human Reproductive Biology (Fourth Edition)

    Academic Press.

    • Google Scholar
28. 1. Karlson P
    2. Luscher M

    (1959) Pheromones': a new term for a class of biologically active substances

    Nature 183:55–56.

    https://doi.org/10.1038/183055a0

    • PubMed
    • Google Scholar
29. 1. Kosfeld M
    2. Heinrichs M
    3. Zak PJ
    4. Fischbacher U
    5. Fehr E

    (2005) Oxytocin increases trust in humans

    Nature 435:673–676.

    https://doi.org/10.1038/nature03701

    • PubMed
    • Google Scholar
30. 1. Koven NS
    2. Max LK

    (2014) Basal salivary oxytocin level predicts extra- but not intra-personal dimensions of emotional intelligence

    Psychoneuroendocrinology 44:20–29.

    https://doi.org/10.1016/j.psyneuen.2014.02.018

    • PubMed
    • Google Scholar
31. 1. Lancaster K
    2. Carter CS
    3. Pournajafi-Nazarloo H
    4. Karaoli T
    5. Lillard TS
    6. Jack A
    7. Davis JM
    8. Morris JP
    9. Connelly JJ

    (2015) Plasma oxytocin explains individual differences in neural substrates of social perception

    Frontiers in Human Neuroscience 9:132.

    https://doi.org/10.3389/fnhum.2015.00132

    • PubMed
    • Google Scholar
32. 1. Lin HH
    2. Cao DS
    3. Sethi S
    4. Zeng Z
    5. Chin JSR
    6. Chakraborty TS
    7. Shepherd AK
    8. Nguyen CA
    9. Yew JY
    10. Su CY
    11. Wang JW

    (2016) Hormonal modulation of pheromone detection enhances male courtship success

    Neuron 90:1272–1285.

    https://doi.org/10.1016/j.neuron.2016.05.004

    • PubMed
    • Google Scholar
33. 1. Liu R
    2. Yuan X
    3. Chen K
    4. Jiang Y
    5. Zhou W

    (2018) Perception of social interaction compresses subjective duration in an oxytocin-dependent manner

    eLife 7:e32100.

    https://doi.org/10.7554/eLife.32100

    • PubMed
    • Google Scholar
34. 1. Lundin S
    2. Akerlund M
    3. Fagerström PO
    4. Hauksson A
    5. Melin P

    (1986) Pharmacokinetics in the human of a new synthetic vasopressin and oxytocin uterine antagonist

    Acta Endocrinologica 112:465–472.

    https://doi.org/10.1530/acta.0.1120465

    • PubMed
    • Google Scholar
35. 1. Lundström JN
    2. Gonçalves M
    3. Esteves F
    4. Olsson MJ

    (2003a) Psychological effects of subthreshold exposure to the putative human pheromone 4,16-androstadien-3-one

    Hormones and Behavior 44:395–401.

    https://doi.org/10.1016/j.yhbeh.2003.06.004

    • PubMed
    • Google Scholar
36. 1. Lundström JN
    2. Hummel T
    3. Olsson MJ

    (2003b) Individual differences in sensitivity to the odor of 4,16-androstadien-3-one

    Chemical Senses 28:643–650.

    https://doi.org/10.1093/chemse/bjg057

    • PubMed
    • Google Scholar
37. 1. Manning M
    2. Misicka A
    3. Olma A
    4. Bankowski K
    5. Stoev S
    6. Chini B
    7. Durroux T
    8. Mouillac B
    9. Corbani M
    10. Guillon G

    (2012) Oxytocin and vasopressin agonists and antagonists as research tools and potential therapeutics

    Journal of Neuroendocrinology 24:609–628.

    https://doi.org/10.1111/j.1365-2826.2012.02303.x

    • PubMed
    • Google Scholar
38. 1. McCall C
    2. Singer T

    (2012) The animal and human neuroendocrinology of social cognition, motivation and behavior

    Nature Neuroscience 15:681–688.

    https://doi.org/10.1038/nn.3084

    • PubMed
    • Google Scholar
39. 1. McClintock MK

    (1971) Menstrual synchorony and suppression

    Nature 229:244–245.

    https://doi.org/10.1038/229244a0

    • PubMed
    • Google Scholar
40. Book

    1. McNair DM
    2. Lorr M
    3. Droppleman LF

    (1971)

    Profile of Mood States

    San Francisco: Educational and Industrial Testing Service.

    • Google Scholar
41. 1. Mugford RA
    2. Nowell NW

    (1971) Endocrine control over production and activity of the anti-aggression pheromone from female mice

    Journal of Endocrinology 49:225–232.

    https://doi.org/10.1677/joe.0.0490225

    • Google Scholar
42. 1. Oettl LL
    2. Ravi N
    3. Schneider M
    4. Scheller MF
    5. Schneider P
    6. Mitre M
    7. da Silva Gouveia M
    8. Froemke RC
    9. Chao MV
    10. Young WS
    11. Meyer-Lindenberg A
    12. Grinevich V
    13. Shusterman R
    14. Kelsch W

    (2016) Oxytocin enhances social recognition by modulating cortical control of early olfactory processing

    Neuron 90:609–621.

    https://doi.org/10.1016/j.neuron.2016.03.033

    • PubMed
    • Google Scholar
43. 1. Olsson MJ
    2. Lundström JN
    3. Diamantopoulou S
    4. Esteves F

    (2006) A putative female pheromone affects mood in men differently depending on social context

    European Review of Applied Psychology 56:279–284.

    https://doi.org/10.1016/j.erap.2005.09.010

    • Google Scholar
44. 1. Parker KJ
    2. Garner JP
    3. Libove RA
    4. Hyde SA
    5. Hornbeak KB
    6. Carson DS
    7. Liao CP
    8. Phillips JM
    9. Hallmayer JF
    10. Hardan AY

    (2014) Plasma oxytocin concentrations and OXTR polymorphisms predict social impairments in children with and without autism spectrum disorder

    PNAS 111:12258–12263.

    https://doi.org/10.1073/pnas.1402236111

    • PubMed
    • Google Scholar
45. 1. Parker KJ
    2. Oztan O
    3. Libove RA
    4. Sumiyoshi RD
    5. Jackson LP
    6. Karhson DS
    7. Summers JE
    8. Hinman KE
    9. Motonaga KS
    10. Phillips JM
    11. Carson DS
    12. Garner JP
    13. Hardan AY

    (2017) Intranasal oxytocin treatment for social deficits and biomarkers of response in children with autism

    PNAS 114:8119–8124.

    https://doi.org/10.1073/pnas.1705521114

    • PubMed
    • Google Scholar
46. 1. Quintana DS
    2. Westlye LT
    3. Hope S
    4. Nærland T
    5. Elvsåshagen T
    6. Dørum E
    7. Rustan Ø
    8. Valstad M
    9. Rezvaya L
    10. Lishaugen H
    11. Stensønes E
    12. Yaqub S
    13. Smerud KT
    14. Mahmoud RA
    15. Djupesland PG
    16. Andreassen OA

    (2017) Dose-dependent social-cognitive effects of intranasal oxytocin delivered with novel breath powered device in adults with autism spectrum disorder: a randomized placebo-controlled double-blind crossover trial

    Translational Psychiatry 7:e1136.

    https://doi.org/10.1038/tp.2017.103

    • PubMed
    • Google Scholar
47. 1. Raina AK
    2. Jaffe H
    3. Kempe TG
    4. Keim P
    5. Blacher RW
    6. Fales HM
    7. Riley CT
    8. Klun JA
    9. Ridgway RL
    10. Hayes DK

    (1989) Identification of a neuropeptide hormone that regulates sex pheromone production in female moths

    Science 244:796–798.

    https://doi.org/10.1126/science.244.4906.796

    • PubMed
    • Google Scholar
48. 1. Savic I
    2. Berglund H
    3. Gulyas B
    4. Roland P

    (2001) Smelling of odorous sex hormone-like compounds causes sex-differentiated hypothalamic activations in humans

    Neuron 31:661–668.

    https://doi.org/10.1016/S0896-6273(01)00390-7

    • PubMed
    • Google Scholar
49. 1. Savic I
    2. Berglund H
    3. Lindström P

    (2005) Brain response to putative pheromones in homosexual men

    PNAS 102:7356–7361.

    https://doi.org/10.1073/pnas.0407998102

    • PubMed
    • Google Scholar
50. 1. Scheele D
    2. Plota J
    3. Stoffel-Wagner B
    4. Maier W
    5. Hurlemann R

    (2016) Hormonal contraceptives suppress oxytocin-induced brain reward responses to the partner's face

    Social Cognitive and Affective Neuroscience 11:767–774.

    https://doi.org/10.1093/scan/nsv157

    • PubMed
    • Google Scholar
51. 1. Simerly RB

    (2002) Wired for reproduction: organization and development of sexually dimorphic circuits in the mammalian forebrain

    Annual Review of Neuroscience 25:507–536.

    https://doi.org/10.1146/annurev.neuro.25.112701.142745

    • PubMed
    • Google Scholar
52. 1. Spengler FB
    2. Schultz J
    3. Scheele D
    4. Essel M
    5. Maier W
    6. Heinrichs M
    7. Hurlemann R

    (2017) Kinetics and dose dependency of intranasal oxytocin effects on amygdala reactivity

    Biological Psychiatry 82:885–894.

    https://doi.org/10.1016/j.biopsych.2017.04.015

    • PubMed
    • Google Scholar
53. 1. Stern K
    2. McClintock MK

    (1998) Regulation of ovulation by human pheromones

    Nature 392:177–179.

    https://doi.org/10.1038/32408

    • PubMed
    • Google Scholar
54. 1. Stoop R

    (2012) Neuromodulation by oxytocin and vasopressin

    Neuron 76:142–159.

    https://doi.org/10.1016/j.neuron.2012.09.025

    • PubMed
    • Google Scholar
55. 1. Tang JD
    2. Charlton RE
    3. Jurenka RA
    4. Wolf WA
    5. Phelan PL
    6. Sreng L
    7. Roelofs WL

    (1989) Regulation of pheromone biosynthesis by a brain hormone in two moth species

    PNAS 86:1806–1810.

    https://doi.org/10.1073/pnas.86.6.1806

    • PubMed
    • Google Scholar
56. 1. Thysen B
    2. Elliott WH
    3. Katzman PA

    (1968) Identification of estra-1,3,5(10),16-tetraen-3-ol (estratetraenol) from the urine of pregnant women (1)

    Steroids 11:73–87.

    https://doi.org/10.1016/S0039-128X(68)80052-2

    • PubMed
    • Google Scholar
57. 1. Troje NF

    (2002) Decomposing biological motion: a framework for analysis and synthesis of human gait patterns

    Journal of Vision 2:2–387.

    https://doi.org/10.1167/2.5.2

    • Google Scholar
58. 1. Wyatt TD

    (2015) The search for human pheromones: the lost decades and the necessity of returning to first principles

    Proceedings of the Royal Society B: Biological Sciences 282:20142994.

    https://doi.org/10.1098/rspb.2014.2994

    • Google Scholar
59. 1. Ye Y
    2. Zhuang Y
    3. Smeets MAM
    4. Zhou W

    (2019) Human chemosignals modulate emotional perception of biological motion in a sex-specific manner

    Psychoneuroendocrinology 100:246–253.

    https://doi.org/10.1016/j.psyneuen.2018.10.014

    • PubMed
    • Google Scholar
60. 1. Zhou W
    2. Yang X
    3. Chen K
    4. Cai P
    5. He S
    6. Jiang Y

    (2014) Chemosensory communication of gender through two human steroids in a sexually dimorphic manner

    Current Biology 24:1091–1095.

    https://doi.org/10.1016/j.cub.2014.03.035

    • PubMed
    • Google Scholar
61. 1. Zhou W
    2. Chen D

    (2009) Fear-related chemosignals modulate recognition of fear in ambiguous facial expressions

    Psychological Science 20:177–183.

    https://doi.org/10.1111/j.1467-9280.2009.02263.x

    • PubMed
    • Google Scholar

Author details

1. Kepu Chen

   State Key Laboratory of Brain and Cognitive Science, CAS Center for Excellence in Brain Science and Intelligence Technology, Institute of Psychology, Chinese Academy of Sciences, Beijing, China

   Contribution

   Data curation, Formal analysis, Validation, Investigation, Writing - original draft

   Contributed equally with

   Yuting Ye

   Competing interests

   No competing interests declared

2. Yuting Ye

   Department of Psychology, University of Chinese Academy of Sciences, Beijing, China

   Contribution

   Data curation, Formal analysis, Validation, Investigation, Writing - review and editing

   Contributed equally with

   Kepu Chen

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-9068-8045

3. Nikolaus F Troje

   Centre for Vision Research, York University, Toronto, Canada

   Contribution

   Software, Writing - review and editing

   Competing interests

   No competing interests declared

4. Wen Zhou

   1. State Key Laboratory of Brain and Cognitive Science, CAS Center for Excellence in Brain Science and Intelligence Technology, Institute of Psychology, Chinese Academy of Sciences, Beijing, China
   2. Department of Psychology, University of Chinese Academy of Sciences, Beijing, China
   3. Chinese Institute for Brain Research, Beijing, China

   Contribution

   Conceptualization, Supervision, Funding acquisition, Methodology, Writing - original draft, Writing - review and editing

   For correspondence

   zhouw@psych.ac.cn

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-6730-2116

• 2,500

  views

• 201

  downloads

• 4

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.