Effects of androstadienone on dominance perception in males with low and high social anxiety


Increasing evidence suggests that humans can communicate both trait-dominance and state-dominance via body odor. Androstadienone (androsta-4,16,-dien-3-one), a chemosignal found in human sweat, seems to be a likely candidate for signaling dominance in humans. The aim of the current study was to investigate the effects of androstadienone on the perception of social dominance. Moreover, we examined whether high levels of social anxiety, a psychopathology involving concerns that specifically pertain to social dominance, are associated with increased sensitivity to androstadienone as a chemical cue of dominance. In a double-blind, placebo-controlled, within-subject design, 64 heterosexual male participants (32 with high social anxiety and 32 with low social anxiety) viewed facial images of males depicting dominant, neutral and submissive postures, and were asked to recognize and rate the dominance expressed in those images. Participants completed the task twice, once under exposure to androstadienone and once under exposure to a control solution. The results indicate that androstadienone increased the perceived dominance of men’s faces, specifically among participants with high social anxiety. These findings suggest a direct influence of androstadienone on dominance perception and further highlight the preferential processing of dominance and social threat signals evident in social anxiety.


Social chemosignaling is an important means of communication for all mammals (Dulac and Torello, 2003; Brennan and Zufall, 2006). Such social chemosignaling is especially prominent in behaviors related to social hierarchy, territory and reproduction (Dulac and Torello, 2003; Wyatt, 2003; Brennan and Zufall, 2006). Social status, namely social dominance and social submissiveness, is known to be conveyed through chemosignals in rodents (Jones and Nowell, 1973; Novotny et al., 1990). Moreover, over the last decade increasing evidence suggests that humans also can communicate both trait-dominance (Havlicek et al., 2005; Sorokowska et al., 2012) and state-dominance (Adolph et al., 2010) via chemical compounds. For example, Sorokowska et al. (2012) showed that participants’ level of trait-dominance was accurately rated based on their body odor, and Havlicek et al. (2005) showed that the axillary odor of men high in trait dominance was rated as more attractive for women during the fertile phase of their menstrual cycle. Furthermore, sweat samples taken from male donors in a state of dominance (induced after winning a sports competition) were shown to elevate the skin conductance levels of perceivers (Adolph et al., 2010), suggesting that a state of dominance (e.g., victory) can also be conveyed via chemosignals.

Since victory and dominance are among the various social states associated with increased levels of testosterone (Booth et al., 1989; Mazur and Booth, 1998), it is possible that testosterone metabolites are emitted in such states through apocrine glands, which are known to be active in various emotive situations (Wilke et al., 2007). Androstadienone (androsta-4,16,-dien-3-one, AND) is a testosterone metabolite (Stylianou et al., 1961; Rennie et al., 1989) found in human axillary sweat (Labows, 1988), semen (Kwan et al., 1992) and plasma (Brooksbank et al., 1972) and has been suggested as a human chemosignal (Grosser et al., 2000), specifically as a human chemosignal of dominance (Huoviala and Rantala, 2013).

AND has been shown to induce some behavioral changes compatible with its potential role as a chemosignal of dominance, aggression and mate quality (Saxton et al., 2008; Zhou et al., 2014; Ferdenzi et al., 2016; Hornung et al., 2017; Banner et al., 2018). For example, Ferdenzi et al. (2016) showed that compared to control odor exposure, AND increased the level of men’s attractiveness, both among women who viewed images of men and among women who met several “speed-date” partners (Saxton et al., 2008). Moreover, Zhou et al. (2014) reported that exposure to AND biased heterosexual females and homosexual males to perceive digital point-light displays of walkers as more masculine, an attribute closely related to dominance. Furthermore, Hornung et al. (2017) showed that under the influence of AND men showed reduced interference in processing of angry target faces by non-relevant emotional words and suggested that AND prepares individuals for a potential conflict by highlighting threatening facial expressions. Banner et al. (2018) recently used the social orientation paradigm (SOP) to examine the effects of AND on avoidance. SOP is a competitive game that involves aggressive provocations from a fictitious partner and allows three types of responses: an aggressive response, an individualistic withdrawal response, and a cooperative response. The results indicated that exposure to AND increased individualistic withdrawal responses, suggesting that it may operate as a threatening signal of dominance that elicits behavioral avoidance and social withdrawal tendencies, possibly as a submissive response.

Thus, our first aim was to investigate whether AND has a direct impact on visually perceived dominance of others, highlighting its potential role in dominance signaling. Our second aim was to examine the possibility that social anxiety, a psychopathology involving hypersensitivity towards dominance signals, is also associated with increased sensitivity to AND, a putative chemical signal of dominance.

Social Anxiety Disorder (SAD) is defined as a persistent fear of one or more social or performance situations in which the individual is exposed to unfamiliar people or to possible scrutiny by others (Diagnostic and Statistical Manual of Mental Disorders, American Psychiatric Association, 2013). An evolutionary model explaining the generation and maintenance of SAD suggests that individuals with SAD tend to over-utilize the “social rank (defense) system,” a psychological system that has evolved in order to maintain social order through constant monitoring of one’s place in the social hierarchy (Trower and Gilbert, 1989). Simply put, individuals who are socially anxious tend to view the world from a hierarchical perspective. They see themselves as being low on the hierarchical ranking and see others as dominant competitors for social status. In line with this, research has shown that individuals with social anxiety exhibit increased sensitivity to dominance cues as compared to individuals who are not socially anxious (Maner et al., 2008; Aderka et al., 2013). For example, Aderka et al (2013) showed that after they read a narrative description, individuals with SAD rated dominant protagonists as higher in social rank than did controls who read the same narrative. Furthermore, Maner et al. (2008) provided evidence showing that people with high social anxiety respond to dominance threat by displaying signs of enhanced social submission expressed by decreased levels of testosterone. Another type of sensitivity towards cues of men’s dominance—visual dominance perception—was shown to be greatest among men with low trait-dominance (Watkins et al., 2010a,b). Since socially anxious individuals perceive themselves as having low social rank, being inferior, and behaving submissively (Weisman et al., 2011), it is reasonable to postulate that they too will show this type of sensitivity. Indeed, Adolph et al., (2010) reported that high levels of social anxiety were associated with high levels of skin conductance when exposed to chemosignals of victory, suggesting that socially anxious individuals are hypersensitive to another type of dominance cue—chemical signals. Moreover, socially anxious individuals were shown to have an increased sensitivity to anxiety chemosignals (Pause et al., 2009, 2010; Adolph et al., 2013), suggesting that some of the social difficulties evident in social anxiety may reflect enhanced reaction to social chemosignaling.

Therefore, the current study sought to investigate the involvement of AND in modulating perceived dominance in males with low and high social anxiety. In a computerized task, participants were asked to recognize dominance levels of visually presented protagonists depicting dominant, neutral, and submissive facial postures and to rate these for their dominance, once under exposure to AND and once under exposure to a control solution. We hypothesized that when exposed to AND, participants would rate other males as more dominant than when they were exposed to a control solution. Furthermore, we expected the aforementioned effect to be more prominent among participants with high social anxiety (HSA) compared to those with low social anxiety (LSA).


Sixty-four males were recruited to participate in the study. They were recruited via the internet and advertisements at the University of Haifa and surrounding areas. Participants ranged in age from 18 to 39 years (mean: 25.74, S.D: 4.3); all had 12–16 years of education and all were fluent Hebrew speakers. The participants reported that they were in general good health, heterosexual, non-smokers, not under chronic or acute medication and not suffering from any mental or somatic disease or from 

Subjective ratings of solutions

Androstadienone and the control solutions did not differ in their perceived intensity [Intensity-control M = 5.27, SD=2.07; Intensity-AND M = 5.13, SD = 1.84; t(63)= 0.51, p=0.61], familiarity [Familiarity-control M = 5.39, SD=2.39; Familiarity-AND M=5.20, SD = 2.23; t(63) = 0.68, p=0.49], or pleasantness [Pleasantness-control M = 5.56, SD 1.57; Pleasantness-AND M = 5.56, SD = 1.53; t(63)= 0.0, p = 1.0], indicating that behavioral differences did not stem from perceived odor qualities.


The aim of this study was to investigate the effects of androstadienone, a chemosignal found in men’s sweat, on dominance perception in individuals with high and low social anxiety. We used a within-subjects, double blind, placebo controlled design to determine whether AND influences participants’ dominance assessment and recognition of male targets depicting submissive, neutral and dominant postures. We hypothesized that if AND indeed acts as a chemosignal of dominance, male participants

Conflict of interest

The authors declare that there is no conflict of interest regarding the publication of this article.


We wish to thank Dr. Nicholas Rule and his lab members for generously providing the visual stimuli for this study.

Pheromones For Men

Leave a Reply

Your email address will not be published. Required fields are marked *