Elsevier

Physiology & Behavior

Volume 74, Issues 1–2, 1–15 September 2001, Pages 15-27
Physiology & Behavior

Context-dependent effects of steroid chemosignals on human physiology and mood

https://doi.org/10.1016/S0031-9384(01)00537-6Get rights and content

Abstract

We examined the physiological and psychological effects of nanomolar amounts of steroids applied directly under the nose (Δ4,16-androstadien-3-one and 1,3,5,(10),16-estratetraen-3-ol). These potential human chemosignals were not consciously discernible in a strong-odor carrier (clove oil and propylene glycol). In a double-blind, within-subject, repeated-measures experiment with 65 subjects, we demonstrated that both steroids produced sustained changes in digit skin temperature and palmar skin conductance (an indicator of sympathetic nervous system tone) while the subjects were completing psychological questionnaires or reading. These effects, however, did not follow the sex-stereotyped pattern predicted by a sex attractant function. Both androstadienone and estratetraenol raised the skin temperature of men's hands and lowered it in women. Likewise, each steroid increased skin conductance, with a significantly greater effect on women than men. Women's responses were observed only in the sessions run by the male tester, an effect that may or may not be solely attributable to tester sex. Men's responses, in contrast, were not affected by this difference in socioexperimental condition. Similarly, women experienced an immediate increase in positive mood only in the presence of the male tester, while men's responses were unaffected by this socioexperimental context. One source of this sex difference may be the fact that the majority of women were in the late follicular phase of their menstrual cycle. Although it is premature to classify these steroids as pheromones, our data suggest that they function as chemosignals that modulate autonomic nervous system tone as well as psychological state.

Introduction

Chemosignals play a significant role in social interactions and communication in many animal species, although their role in human behavior has previously been considered minimal. Since human behavior is rarely determined by any single signal and strongly depends on the social context, we have previously postulated that chemosignals are more likely to modulate psychological responses to a given situation, such as mood or attention, rather than release stereotyped behavior or specific thoughts [24], [31], [32]. In many animals, pheromones also modulate behavior rather than simply releasing it [21], [44]. Moreover, since individual responses to the environment can be fine-tuned and enhanced by social information, we have further hypothesized that the social context may determine whether or not a chemical signal will have any detectable effect.

Recently, much attention has been given to two steroids that have been claimed to be human behavioral pheromones. These were reportedly isolated from human skin [1], [38] and considered pheromones because of their reported sex-specific effects on the surface potential of the putative human vomeronasal organ (VNO; [34] but see [23]). One compound, Δ4,16-androstadien-3-one (in US Patent No. 5,278,141) [1], only affected women, increasing surface potential of the VNO almost six-fold compared to odor control (clove oil in propylene glycol). Another compound (1,3,5,(10),16-estratetraen-3-ol, (same patent)) affected just men by producing a nine-fold increase in VNO surface potential. Neither affected the surface potential of the olfactory epithelium.

Both chemicals are naturally occurring steroids in humans. Androstadienone has been measured in peripheral plasma of men at 0.01–0.06 μg/100 ml unconjugated, 0.05–0.1 μg/100 ml sulfate-conjugated [6], 98 ng/100 ml [5], and 2.05 pmol/ml [14]. It is also found in men's sweat [29], semen [28], and axillary hair [36], [40]. In women, androstadienone is measurable in plasma at 36 ng/100 ml [5] and estratetraenol has been isolated from the urine of pregnant women in the third trimester [46].

We fully recognize that a change in epithelial surface potential is neither a receptor potential nor sufficient evidence that humans have a functional vomeronasal system. Nonetheless, these preliminary reports gave us a specific starting point for determining whether these two steroids have sex-specific effects on behavior or psychological state. In our first set of studies, we established that 9 nmol of androstadienone or estratetraenol can modulate the mood responses of women and men to participating in our experiment, as measured by widely used and established questionnaires [22]. When compared with two carrier controls, subjects' mood states were modulated within the first 10 min after initial steroid exposure and continued up to 2 h later, orders of magnitudes longer than the typical adaptation period for odorants. This effect was observed in comparison to a weak-odor carrier of propylene glycol. Moreover, these effects were still evident even when the steroids were heavily masked by a strong-odor carrier solution (clove oil in propylene glycol) and not verbally recognized as an odor [22], [24], [32]. These psychological findings led us to propose corresponding physiological responses to these two steroids.

Androstadienone and estratetraenol, however, do not have sex-exclusive or sexually stereotyped effects [22] on psychological state as has been claimed, based on electrophysiological data [1], [34]. Moreover, no specific social cognitions or emotions were released [24], [31], [32]. Instead, the psychological data supported our hypothesis that chemosignals modulate positive and negative mood states [24], [31], [32]. Because our data documented long-lasting changes in mood, we were interested in the possibility that these human chemosignals would have comparably long-lasting tonic effects upon the tone of the autonomic nervous system.

Androstadienone and estratetraenol may affect the autonomic nervous system. Preliminary data suggested an increase in sympathetic tone [34]. However, androstadienone was later reported to have the opposite effect in women, namely sympatholytic and parasympathomimetic effects [16]. This interpretation of data, however, was supported by cardiorespiratory measures but was not consistent with the reported skin conductance results [16]. We sought to resolve this controversy over effects of these compounds on the sympathetic nervous system by measuring skin conductance. Skin conductance reflects sympathetic outflow exclusively because the eccrine sweat glands are innervated only by the sympathetic and not the parasympathetic system or hormonal variables [20].

In this experiment, we again tested the hypothesis that these steroids have sex-specific effects (androstadienone on females and estratetraenol on males exclusively [34], [35]). We exposed women and men to both steroids in the carrier, as well as the carrier alone, with the expectation that the general positive shift in affect observed previously during the first 20 min [22] would be accompanied by changes in the autonomic nervous system. We chose two sympathetic nervous system biomarkers used in prior studies [34], [35]: skin conductance (positively correlated with SNS tone [20]) and skin temperature (which negatively correlates with SNS tone, although in some circumstances skin temperature can be overridden by other factors such as variation in core temperature and renal hormonal output).

We also tested our hypothesis that socioexperimental context (specifically sex of tester) might shape responses to human chemosignals. The reproductive function of chemosignals in other species suggested that differences in a reproductively relevant social context might lead to variations in observed responses. Therefore, we compared effects of chemosignals when a man presented them and was in the testing room, as compared to when a woman performed this role. We emphasize that any contextual effects reported for this study could be mediated by specific attributes of the testers and not necessarily be attributable to their sex. Nonetheless, for describing the results of this study, sex of tester was the most relevant and measurable social and physical difference among the testers.

Section snippets

Participants

Recruited participants were university students or staff with the mean age of 23.7±5.3 years of age (range 18–48 years), in good health, and blind to our specific hypotheses. Forty-four women and 21 men met our further screening criteria: nonsmokers with no history of nasal or sinus problems, and, for women, a history of regular and spontaneous ovulation and not utilizing hormone-based birth control at the time of the study (more women than men were recruited to accommodate potential menstrual

Testing sessions with carrier odor

The effect of each steroid was assessed in terms of a difference from ANS responses to smelling the carrier, clove oil, and propylene glycol, while participating in the experimental protocol in our human subjects testing room (see Fig. 1). The following patterns of ANS reactivity are typical sex differences. At the beginning of the carrier-control sessions, women had cooler hands than did men (35.33±0.21°C vs. 36.00±0.12°C, t=2.15, P≤.04) as well as lower skin conductance (0.05±0.01 μS vs.

Discussion

This assessment of skin conductance and temperature responses to androstadienone and estratetraenol confirms the conclusion of our psychological study [22]. The two steroids do not have sex-exclusive effects; androstadienone does not affect only women, nor does estratetraenol affect only men. Both steroids modulated the participants' ANS responses to being in the study in comparable ways, with androstadienone having stronger effects than estratetraenol in both men and women.

Women's responses to

Acknowledgements

This work was supported by the Mind–Body Network of the John D. and Catherine T. MacArthur Foundation, the NIH MERIT award R37 MH41788 to Martha K. McClintock, as well as the NIH MD/PhD Training Grant HD-07009 and Olfactory Research Fund's Tova Fellowship to Suma Jacob. We also thank Sheila Garcia for her commitment and significant help with many aspects of data collection and processing. Bethanne Zelano's assistance with graphing is greatly appreciated. Thanks to Harriet de Wit for input on

References (49)

  • L Monti-Bloch et al.

    Effect of putative pheromones on the electrical activity of the human vomeronasal organ and the olfactory epithelium

    J Steroid Biochem Mol Biol

    (1991)
  • L Monti-Bloch et al.

    The human vomeronasal system

    Psychoneuroendocrinology

    (1994)
  • A Nixon et al.

    Simultaneous quantification of five odorous steroids (16-androstenes) in the axillary hair of men

    J Steroid Biochem Mol Biol

    (1988)
  • G Preti et al.

    Human axillary secretions influence women's menstrual cycles: the role of donor extract of females

    Horm Behav

    (1986)
  • MJ Russell et al.

    Olfactory influences on the human menstrual cycle

    Pharmacol Biochem Behav

    (1980)
  • JC Schank et al.

    A coupled-oscillator model of ovarian-cycle synchrony among female rats

    J Theor Biol

    (1992)
  • SS Schiffman et al.

    The effect of pleasant odors and hormone status on mood of women at midlife

    Brain Res Bull

    (1995)
  • JP Signoret

    Chemical communication and reproduction in domestic mammals

  • B Thysen et al.

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

    Steroids

    (1968)
  • JP Zacny et al.

    Effects of setting on the subjective and behavioral effects of d-amphetamines in humans

    Addict Behav

    (1992)
  • Berliner DL. Fragrance compositions containing human pheromones. US Patent,...
  • GE Bigelow

    Human drug abuse liability assessment: opioids and analgesics

    Br J Addict

    (1991)
  • MM Bradley

    Emotion and motivation

  • LH Brauer et al.

    High dose pimozide does not block amphetamine-induced euphoria in normal volunteers

    Pharmacol Biochem Behav

    (1997)
  • Cited by (127)

    • Oxytocin and its links through scientific lineage

      2024, Comprehensive Psychoneuroendocrinology
    View all citing articles on Scopus
    View full text