The Presence of an Attractive Woman
Elevates Testosterone and Physical Risk
Taking in Young Men
and William von Hippel
The authors report a field experiment with skateboarders that demonstrates that physical risk taking by young men increases in
the presence of an attractive female. This increased risk taking leads to more successes but also more crash landings in front of a
female observer. Mediational analyses suggest that this increase in risk taking is caused in part by elevated testosterone levels of
men who performed in front of the attractive female. In addition, skateboarders’ risk taking was predicted by their performance
on a reversal-learning task, reversal-learning performance was disrupted by the presence of the attractive female, and the female’s
presence moderated the observed relationship between risk taking and reversal learning. These results suggest that men use
physical risk taking as a sexual display strategy, and they provide suggestive evidence regarding possible hormonal and neural
risk taking, evolutionary psychology, decision making, social neuroscience, neuroscience
The archetype of the femme fatale appears in the religious
texts, art, and literature of a range of cultures. She appears as
men from their reason, in essence embodying the relationship
between female sexuality and loss of self-control among men.
Although such a relationship might be nothing more than a
dubious defense, concocted by men to mitigate their own
behavior, recent research supports the possibility of such an
association: Attractive women have the power to shift men’s
time perspective away from the long-term consequences of
their choices and focus their attention on the here and now
(Wilson & Daly, 2004).
A theoretical account of why attractive women inspire such
a myopic time perspective in the male mind can be found in
Trivers’s (1972) theory of parental investment. Because of
unequal gamete size, females invest more than males in repro-
duction in most species, with the result that males typically
compete with each other for access to females (Trivers,
1972). Sexual selection consequently favors males who engage
in competitive behavior and costly displays to attract females
(Andersson, 1994). Although such displays can enhance repro-
ductive success, they can also be detrimental in terms of
survival (Brooks, 2000; Kokko, Brooks, McNamara, &
Houston, 2002). Such effects of sexual selection have been
documented in a variety of nonhuman animals, but they should
also be apparent in humans, as differential parental investment
is exacerbated in humans by lengthy gestation and lactation and
by the extended period of development of dependent young.
Consistent with this logic, human evolution shows signs of
recurrent male–male competition for access to females, such
as sexual size dimorphism (Alexander, Hoogland, Howard,
Noonan, & Sherman, 1979) and the size and shape of male gen-
italia (Gallup et al., 2003). Presumably as a consequence of
these evolutionary pressures, men are more predisposed than
women toward risk taking, same-sex competition, and aggres-
sion (Byrnes, Miller, & Schaeffer, 1999; Eagly & Steffen,
1986; Wilson, Daly, & Pound, 2002).
Although physical risk taking can bestow important repro-
ductive and reputational benefits on young men (Chagnon,
1988), there are obvious costs as well, particularly in a contem-
porary context. For example, men are 2.5 times more likely
than women to be killed in road accidents (World Health Orga-
nization, 2002), and this sex difference exceeds a factor of 3
among 15- to 29-year-olds (Roads and Traffic Authority of
New South Wales, 2001). Same-sex homicides are also predom-
inantly committed by young single men (Daly & Wilson, 1990),
with disputes over respect and saving face being the typical
University of Queensland, St Lucia, Australia
Richard Ronay, School of Psychology, University of Queensland, St Lucia, QLD
Social Psychological and
ªThe Author(s) 2010
Reprints and permission: http://www.
catalyst (Daly & Wilson, 1988). As a result of this greater ten-
dency toward risk taking, young men occupy the highest demo-
graphic risk category for early mortality in industrialized
nations (Kruger, 2004).
If men take risks in pursuit of mating opportunities, then
such risk taking should be attuned to cues that signal the poten-
tial for successful mating (Baker & Maner, 2008; Wilson &
Daly, 2004). As female attractiveness conveys statistically
reliable and observable cues to fertility (Rhodes, 2006; Singh,
1993), we hypothesized that young men would engage in
greater physical risk taking when in the presence of an attrac-
tive woman. To test this possibility, we recruited young adult
male skateboarders and recorded whether the presence of an
attractive female experimenter affected their risk taking.
Performing tricks on a skateboard gives young men the
opportunity to display mastery, physical prowess, and bravado,
but skateboarding tricks also involve the potential for physical
harm and embarrassment. When skateboarders attempt their
tricks, there is a decisive moment at which they must choose
to abort the trick or try to land it. If there is any doubt of suc-
cess, the safest option is to abort the trick and land on one’s
feet. This decision cannot be made in advance but rather must
be made in midair, based on a split-second evaluation of the
likelihood of success and on the physical costs that failure
might bring. It was this split-second decision making in the
face of risk that we sought to examine, in part because it resem-
bles the type of risky decisions that young men make when
behind the steering wheel of a car or when in physical confron-
tations with each other.
To examine a possible proximal mechanism for the
hypothesized increase in this sort of risk taking in the presence
of an attractive female, we focused on the role of testosterone.
Testosterone fuels competition; high-testosterone males strive
for positions of dominance, and short-term increases in testos-
terone help high-testosterone males achieve dominance by
reducing fear while increasing assertiveness, violence, and
competitiveness (Dabbs & Dabbs, 2000). Increases in testoster-
one have been shown to focus attention on rewards and reduce
sensitivity to losses (van Honk et al., 2004), a volatile combi-
nation likely to enhance risky decision making. Testosterone
also fuels sexual interest, arousal, and activity (Tuiten et al.,
2006). High-testosterone males pay more attention to sexual
stimuli (Rupp & Wallen, 2007), have more sexual partners (van
Anders, Hamilton, & Watson, 2007), and are more likely to
seek nonmonogamous sexual relationships (McIntyre et al.,
2006) than low-testosterone males. Importantly, testosterone
levels also increase in response to brief interactions with attrac-
tive women (Roney, 2003; Roney, Lukaszewski, & Simmons,
2007). Because male testosterone levels rise in the presence
of attractive women, and because testosterone is associated
with increased competition and risk taking (Coates & Herbert,
2008; Dabbs & Dabbs, 2000), we hypothesized that increased
risk taking in the presence of an attractive woman might be
induced by elevated testosterone.
Finally, we sought to test the possibility that this increase in
risk taking might also be associated with decreased executive
control, as facilitated by the prefrontal cortex (Barkley,
2001). Specifically, we wished to target the ventral medial pre-
frontal cortex (VMPFC) because of its role in the processing of
rewards and punishments that are essential to decision making
under risk (Bechara, Damasio, Damasio, & Anderson, 1994;
Hare, Camerer, & Rangel, 2009; Xue et al., 2009). The
VMPFC has been shown to play a role in the provision of
somatosensory feedback, and these visceral cues appear to be
critical in learning to avoid punishment (Bechara, 2004). In
addition, the VMPFC is activated in response to a range of
reward cues including food (Hare et al., 2009), attractive mem-
bers of the opposite sex (O’Doherty et al., 2003), and money
(O’Doherty, Kringelbach, Rolls, Hornak, & Andrews, 2001).
Taken together, these findings suggest that the VMPFC is inte-
gral to the appraisal of a range of potential rewards and punish-
ments that underlie effective decision making.
One possible consequence of the VMPFC’s generalized
response to rewards is misattribution. Note that we do not refer
here to a misattribution of physiological arousal, as in Dutton
and Aron’s (1974) classic study but rather to a misattribution
of the source of neural activity. Specifically, it is possible that
during decision making, the presence of reward cues unrelated
to the decision under consideration might contribute to activa-
tion of the VMPFC and thereby exert an influence on the deci-
sion at hand. This potential for misattribution may be greater
during decisions providing less time for deliberation and reflec-
tion, such as those made in midair by our skateboarding partic-
ipants. We therefore suspected that the presence of an attractive
female experimenter might lead to increased activation of our
skateboarders’ VMPFCs and thereby interfere with the manner
in which the region guides the rapid decisions required for suc-
Although functional magnetic resonance imaging (fMRI)
would enable the most direct test of such a hypothesis, our field
experiment ruled against this possibility, and so we sought a
proxy measure of VMPFC function. One task that has been
shown in fMRI studies to target the VMPFC is a ‘‘reversal-
learning’’ procedure (Fellows & Farah, 2005). Reversal learn-
ing requires participants to choose between two options that
differ in terms of the rewards and punishments provided by
their selection. One option results in larger rewards, smaller
punishments, and more of the former than the latter, whereas
the other results in an even distribution of smaller rewards and
larger punishments. Given these contingencies, most partic-
ipants quickly learn to choose the more profitable option. How-
ever, once they begin to repeatedly choose the profitable
option, the contingencies are switched without notice, and par-
ticipants’ ability to adjust their choices in accordance with this
shift is the measure of interest.
Such reversal-learning tasks measure participants’ capacity
to process dynamically changing reward–punishment contin-
gencies and use this information to guide decision making.
Thus, although the surface features of reversal-learning tasks
are quite different from those involved in performing tricks
on a skateboard, there may be overlap in the neural regions
involved in both tasks, with the result that poorer reversal-
58 Social Psychological and Personality Science 1(1)
learning performance should predict riskier skateboarding.
Furthermore, because the female experimenter’s presence
should lead to increased activation of the VMPFC (O’Doherty
et al., 2003), we expected that her presence would interfere
with both the reversal-learning task and the skateboarding task.
That is, because her presence should cause task-irrelevant acti-
vation of the VMPFC, this region should no longer be able to
guide reversal learning or skateboarding as effectively as if she
were not there, leading to poorer performance in the reversal-
learning task and greater risk taking on the skateboard. If the
degree of disruption caused by the female experimenter in
reversal learning is commensurate with the degree of disruption
on the skateboarding task, the relationship between the two
should remain intact. Alternatively, if the degree of disruption
on the two tasks is incommensurate, we would expect attenua-
tion of the relationship between reversal learning and skate-
boarding. Thus, we also examined the possibility that the
female experimenter’s presence might moderate the predicted
relationship between risky skateboarding and reversal-
A total of 96 young adult male skateboarders (age M¼21.58,
SD ¼3.99, range ¼18–35) completed the experiment in skate-
boarding parks in Brisbane, Australia. Participants were
recruited at skateboard parks and offered Aus$20 (*US$16)
as compensation for their time. Of the participants, 43 were
assigned to the male-experimenter condition (age M¼21.26,
SD ¼3.37) and 53 were assigned to the female-experimenter
condition (age M¼21.85, SD ¼4.44). Testing was conducted
between 2:00 p.m. and 6:00 p.m. to control for diurnal variation
in testosterone concentrations.
Skateboarders were asked to choose one easy trick (i.e., one
they could successfully complete on most attempts) and one
difficult trick (i.e., one they were currently learning and that
they could successfully complete approximately 50%of the
time), each of which they then attempted 10 times while being
video recorded by a male experimenter (Block 1). Following a
short break, they were asked to make 10 attempts of both of
these same tricks again (Block 2), either for the same male
experimenter or for an attractive 18-year-old female experi-
menter who was blind to hypotheses. Attractiveness of the
female experimenter was established by having 20 independent
male raters (age M¼21.05, SD ¼3.58) view a photograph of
the female experimenter and rate how attractive they found her
on a scale from 1 (very unattractive)to7(very attractive). The
mean attractiveness rating was 5.58 (SD ¼0.84), which was
significantly higher than the scale midpoint of 4, t(19) ¼
8.31, p< .01, d¼3.81 (these attractiveness ratings were
corroborated by many informal comments and phone number
requests from the skateboarders).
Skateboarders’ attempts were subsequently coded by two
raters (k¼.81) for one of three possible outcomes: success,
crash landing, or an aborted attempt, with the latter an inverse
indicator of risk taking.
Saliva samples were collected by passive drool at the conclu-
sion of the experiment—thereby providing enough time for tes-
tosterone changes induced by the second experimenter to
appear in participants’ saliva (Schultheiss et al., 2005). Sam-
ples were frozen and stored until the experiment was com-
pleted, at which point they were sent to the lab for analysis.
As a consequence, samples were stored from 1 to 5 months
at –20"C until assay with RadioImmunoAssay by Pathlab Inte-
grative Medicine in Burwood, Australia.
As noted by Granger,
Shirtcliff, Booth, Kivlighan, and Schwartz (2004), storage at
–20"C for this duration does not lead to significant degradation
A modified version of the reversal-learning task was developed
for this study, as the standard version was found to be insuffi-
ciently interesting to maintain the attention of the skateboard-
ers. In this modified version, inspired by the Balloon
Analogue Risk Task of Lejuez et al. (2002), participants
pumped up a series of 40 cyber-balloons on a laptop, earning
money for each pump but losing the money earned for each bal-
loon if it popped before participants decided to move on to the
next balloon. In the first half of the trials, pink balloons popped
at a smaller size (M¼14 pumps, SD ¼2.56) than did blue bal-
loons (M¼45 pumps, SD ¼11.20). In the second half of the
trials these contingencies were reversed. Successful learning of
this reversal is indicated by fewer pumps on the blue balloon
and more on the pink balloon after the reversal. The learning
component involved in the reversal-learning task precluded
administering it more than once, and thus its placement in the
experimental order was counterbalanced; 49 of the participants
completed the task before the first block of skateboarding trials
and 47 of the participants completed it before the second block
To test the possibility that arousal might be responsible for
changes in skateboarding performance in front of the female
experimenter, participants’ pulses were recorded with a Nordic
sports watch that measured heart rate at the tip of the index fin-
ger through an electronic sensor.
Measurements were taken
immediately prior to the first block of skateboarding and then
again immediately prior to the second block of skateboarding.
Ronay and von Hippel 59
Consistent with predictions, participants took greater risks on
the difficult tricks in the presence of the female experimenter,
as indicated by fewer aborted tricks (see Figure 1). This reduc-
tion in aborted tricks led to an increase in both crash landings
and successful tricks (see Figure 1). Significant interactions
emerged between gender of the second experimenter and trial
block on all three measures of skateboarding performance on
the difficult tricks: aborted tricks, F(1, 94) ¼13.18, p< .001,
d¼0.75, successful tricks, F(1, 94) ¼4.14, p< .05, d¼
0.42, and crash landings, F(1, 94) ¼5.91, p< .02, d¼0.50.
In all three of these interactions, no differences in performance
emerged in front of the male experimenter across the two
blocks of trials (all ps > .20), but the presence of the female
experimenter led to fewer aborted tricks, t(52) ¼–5.39, p<
.001, d¼1.49, more crash landings, t(52) ¼3.01, p< .01,
d¼0.84, and more successes, t(52) ¼4.23, p< .001, d¼
1.17. Controlling for age had no effect on any of these results.
As predicted, testosterone levels were significantly higher
among men who skateboarded in front of the female experi-
menter (M¼295.95 pmol/L, SD ¼143.69) than among men
who skateboarded only in front of the male experimenter
(M¼212.88 pmol/L, SD ¼101.62), F(1, 69) ¼5.99, p< .02,
d¼0.67. Importantly, this increase in testosterone partially
accounted for the decreased likelihood of aborting tricks in
front of the female experimenter (see Figure 2). A bootstrap-
ping procedure (Preacher & Hayes, 2004) with 10,000 resam-
ples indicated that testosterone significantly mediated the
relationship between experimenter gender and the number of
aborted tricks (indirect effect ¼–.14, SE ¼.07, 95%confi-
dence interval ¼–.33, –.03). These analyses suggest that
increased risk taking in front of the female experimenter was
partially mediated by increased testosterone. All analyses were
then rerun while controlling for age, and all effects remained
Reversal-learning data were log transformed prior to analysis
to overcome positive skew, although for ease of interpretation
raw means are reported below (all results remain significant
when the raw data were analyzed without log transformation).
Because of counterbalancing of the reversal-learning task,
order of administration was controlled in all reported analyses.
Consistent with expectations, reversal learning was better when
the task was performed in front of the male experimenter (M¼
2.41, SD ¼0.44, n¼70) than when it was performed in front
of the female experimenter (M¼2.21, SD ¼0.28, n¼26),
t(94) ¼2.23, p< .05, d¼0.44. Reversal-learning performance
did not differ significantly across the three conditions in which
the task was performed in front of the male experimenter (ts#
1.5, ps$.14). Better reversal learning was also associated with
increased frequency of aborting the tricks in front of the male
experimenter, b¼.31, t(40) ¼2.07, p< .05, but not in front
of the female experimenter, b¼–.11, t(50) ¼–0.73, p¼
.47, and this difference between conditions was itself signifi-
cant, b¼–.23, t(91) ¼–2.38, p¼.02.
An alternative proximal mechanism for the current results is
that the attractive experimenter increased men’s arousal and
Figure 1. Performance on difficult tricks in Block 1 and Block 2
by experimenter gender. For examples of aborted tricks, failed tricks,
and successful tricks see supplementary files at http://spp.sagepub.
Note: Error bars represent 1 standard error.
Figure 2. Mediation of the effect of experimenter gender on number
of aborted tricks via testosterone
Note: Path coefficients represent standardized regression weights. The coef-
ficient below the path from experimenter gender to aborted tricks represents
the direct effect with no mediator in the model; the coefficient above the path
represents the effect when testosterone is included as a mediator.
60 Social Psychological and Personality Science 1(1)
that this increased arousal led to greater risk taking. However, if
arousal were the proximal mechanism, then participants would
have performed better on the easy tricks and worse on the diffi-
cult tricks in the presence of the attractive female (Zajonc,
1965). In contrast to this possibility, participants showed a mix
of greater failure and greater success on the difficult tricks in the
presence of the attractive female. Furthermore, experimenter
gender had no main or interactive effects on performance
outcomes for the easy tricks, aborted tricks, interaction
F(1, 94) ¼0.46, p¼.50, successful tricks, interaction F(1,
94) ¼1.34, p¼.25, and crash landings, interaction F(1, 94)
¼1.95, p¼.17 (see Figure 3), although ceiling and floor effects
with the easy tricks make this absence of an effect somewhat
ambiguous. In addition, an increase in arousal should be marked
by an elevated heart rate in the presence of the female experi-
menter. Tentative evidence emerged for an increase in heart rate
from Block 1 (M¼80.04, SD ¼18.14) to Block 2 (M¼84.63,
SD ¼21.04) of the experiment, F(1, 88) ¼3.55, p< .07, but no
evidence emerged for this change in heart rate being affected by
the gender of the experimenter, F(1, 88) ¼1.32, p> .05. Corre-
lations between heart rate measurements at Block 1 and Block 2
were r¼.59, p< .01 and r¼.34, p< .05 for the control and
experimental conditions, respectively. A comparison of the two
correlations following Fisher’s rto ztransformations (Preacher,
2002) revealed no significant difference between the two condi-
tions (z¼1.47, p¼.14). Finally, none of the skateboarding vari-
ables were correlated with the change in heart rate from Block 1
to Block 2, nor with either of the independent measures of heart
rate (all rs<.06,allps>.57).
Another alternative to the mediational model proposed in
Figure 2 is that attractive women might lead men to take greater
risks, and this enhanced risk taking might itself lead to elevated
testosterone. Because testosterone increases with success com-
pared to failure (Archer, 2006), this alternative model suggests
that increased testosterone in the presence of the female
experimenter should be mediated by the increased number of
successful landings, change in the ratio of successful to crash
landings, or the decreased number of aborted landings. None
of these alternative models showed evidence for mediation.
The results of this field experiment provide evidence that
young men take greater physical risks when in the presence
of an attractive woman and that increases in circulating testos-
terone partially explain this effect. Such displays of physical
risk taking might best be understood as hormonally fueled
advertisements of health and vigor aimed at potential mates and
signals of strength, fitness, and daring intended to intimidate
potential rivals. The finding that increased risk taking led to
both more successes and more crashes suggests that although
sexual displays in human males might be adaptive in terms
of reproductive success, they might also be costly in terms of
survival, as has been found in other species (Hunt et al.,
2004). Other instances of physical risk taking that contribute
to men’s early mortality, such as dangerous driving and phys-
ical aggression, might also be influenced by increases in testos-
terone brought about by the presence of attractive women. The
possibility that male risk taking emerges in part because of an
adaptive legacy that wages survival against reproductive suc-
cess offers a Darwinian perspective on the causes of such costly
expressions of risk taking.
Our finding that reversal-learning performance predicted
risk taking among the skateboarders in this study provides evi-
dence that reversal learning is linked to a real-world instance of
risk taking. It is notable that the split-second decisions underly-
ing our measure of risk taking afforded participants little
opportunity for choosing a course of action in advance. The
dynamic evaluation of potential rewards and losses required
by the skateboarding task is likely to be at least partially facili-
tated by the functioning of the VMPFC (Bechara et al., 1994;
Glimcher & Rustichini, 2004; Hare et al., 2009; Xue et al.,
2009), and, as noted, reversal-learning tasks such as the one
completed by the skateboarders in this study have also been
linked to activation of the VMPFC (Fellows & Farah, 2005).
It is therefore possible that the VMPFC provides a source of the
relationship observed between the two tasks in the current
study. Nevertheless, replication and extension are clearly
necessary to confirm this interpretation of these findings.
The results of the current study also suggest possible under-
lying mechanisms for Wilson and Daly’s (2004) finding that
attractive women increase delay discounting (also see Baker
& Maner, 2008; McAlvanah, 2008). First, these data suggest
that increases in young men’s testosterone levels in response
to an attractive woman provide a partial explanation for shifts
Figure 3. Performance on easy tricks in Block 1 and Block 2 by
Note: Error bars represent 1 standard error.
Ronay and von Hippel 61
toward greater risk taking. Future research might seek to estab-
lish whether similar hormonal changes following exposure to
attractive women mediate shifts in delay discounting. Second,
if delay discounting and increased risk taking in the presence of
attractive females have an evolutionary origin (Baker & Maner,
2008; Wilson & Daly, 2004), then it is possible that the sort of
misattribution that might take place in the VMPFC may have
evolved in part to facilitate young men’s risky decisions that
can be necessary to get into the mating game. That is, the
VMPFC might have evolved to facilitate risk taking during
those circumstances when it was most adaptive to risk one’s
own survival in the pursuit of reproduction.
Caveats and Limitations
There are important limitations to this study that should be
noted. First, because of the collection of only a single posttest
sample of testosterone, it is possible that the presence of the
male experimenter caused participants’ testosterone levels to
decrease rather than the female experimenter causing them to
increase. Such a possibility seems unlikely, given that no evi-
dence supports such a possibility, whereas there is evidence
that women elevate male testosterone levels (Roney, 2003;
Roney et al., 2007); nevertheless, future research should repli-
cate these results using changes in pretest to posttest measures
of testosterone. Second, and related, the experimental group
experienced a change in experimenter (from male to female)
but the control group did not, and thus it is possible that the
higher levels of testosterone and risk taking found among the
experimental group were because of the novelty of having a
new experimenter present for the second half of the study.
Although we have no theoretical reason to expect novelty alone
to have such an impact, it remains a possibility that would be
best addressed in future research by the assessment of pretest
and posttest testosterone levels.
Third, as only one female experimenter was used throughout
the study, we do not know whether the presence of any female
would have led to the same results and thus whether her attrac-
tiveness was a relevant detail (Wells & Windschitl, 1999). We
suspect that her attractiveness is important, given Baker and
Maner’s (2008) research demonstrating that male risk taking
increases only after exposure to attractive females. Nonethe-
less, although we anticipate that the current effects are likely
to be stronger in the presence of attractive women, it is possible
that most women of reproductive age could serve as a sufficient
mating cue to lead to the increase in risk taking seen in the cur-
Fourth, we have no evidence of whether all of our partici-
pants were indeed heterosexual and not in committed relation-
ships, and thus we do not know if all of them were potentially
interested in the attractive female experimenter. We chose not
to ask our participants about these issues, in part because of the
questionable veracity of their responses in this context. Never-
theless, the unintended inclusion of homosexual participants or
participants in committed relationships should only weaken our
pattern of results.
Finally, it is important to keep in mind that the reversal-
learning task used in this study is a distal measure of VMPFC
function. Although previous fMRI studies suggest that reversal
learning involves activation of the VMPFC (Fellows & Farah,
2005), no task is process pure, and thus it is quite possible that
other aspects of the task beyond its relationship with VMPFC
functioning underlie its correlation with risk taking. Future
research with other measures and procedures will be necessary
to corroborate the effects reported here.
The current experiment provides evidence for an effect that has
existed in art, mythology, and literature for thousands of years:
Beautiful women lead men to throw caution to the wind. Our
data extend this ancient literature in two directions, by suggest-
ing that increased male risk taking in the presence of an attrac-
tive woman is mediated by increases in circulating testosterone
and by suggesting that the VMPFC might play an intermediary
role in these processes. These findings suggest that, for men,
the adaptive benefits gained by enticing mates and intimidating
rivals may have resulted in evolved hormonal and neurological
mechanisms that facilitated greater risk taking in the presence
of attractive women.
1. Consistent with the challenges of field experiments, not all partici-
pants were willing to provide saliva samples, and not all collected
samples were free of contaminants, as some participants appeared
not to rinse thoroughly and some chose not to wait the requested 5
minutes after rinsing their mouth with water prior to providing a
saliva sample. Thus, of the 89 assayed samples, 13 were below and
5 were above the normal range of 100 to 720 pmol/L for young
adult males established by Pathlab with its assaying procedures.
Because of concerns about the causes of these outliers, they were
excluded from subsequent analyses. When the missing values (out-
liers, those who refused to provide a sample, or both) were replaced
with the sample mean for testosterone (270.21), all reported anal-
yses replicated significantly. When the outliers were retained in the
analyses, the effect of experimenter gender on testosterone
remained significant, but the mediation analysis did not.
2. The watch was validated against heart rate measured 10 times
directly at the radial artery with a stopwatch (r¼.72, p¼.02). Sen-
sitivity of the device was established by comparing 10 measure-
ments taken at rest (M¼83.4, SD ¼2.50) to 10 measurements
taken after climbing six flights of stairs (M¼89.5, SD ¼6.85),
t(9) ¼3.05, p¼.01.
We thank Rob Brooks, David Buss, Adam Galinsky, Don Moore,
Robert Trivers, and Frank von Hippel for comments on this article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to
the authorship and/or publication of this article.
62 Social Psychological and Personality Science 1(1)
The author(s) disclosed receipt of the following financial support for
the research and/or authorship of this article: Preparation of the article
was supported by a fellowship from the Institute for Advanced Study,
Berlin, and the research was funded by a grant from the Australian
Alexander, R. D., Hoogland, J. L., Howard, R. D., Noonan, K. M., &
Sherman, P. W. (1979). Sexual dimorphism and breeding systems
in pinnipeds, ungulates, primates and humans. In N. A. Chagnon &
W. Irons (Eds.), Evolutionary biology and human social behavior:
An anthropological perspective (pp. 402-435). North Scituate, RI:
Andersson, M. (1994). Sexual selection.Princeton,NJ:PrincetonUni-
Archer, J. (2006). Testosterone and human aggression: And evaluation
of the challenge hypothesis. Neuroscience & Biobehavioral
Baker, M. D., & Maner, J. K. (2008). Risk-taking as a situationally
sensitive male mating strategy. Evolution and Human Behavior,
Barkley, R. (2001). Executive functions and self-regulation: An evo-
lutionary neuropsychological perspective. Neuropsychology
Bechara, A. (2004). The role of emotion in decision making: Evidence
from neurological patients with orbitofrontal damage. Brain and
Bechara, A., Damasio, A. R., Damasio, H., & Anderson, S. W. (1994).
Insensitivity to future consequences following damage to the
human prefrontal cortex. Cognition,50, 7-15.
Brooks, R. (2000). Negative genetic correlation between male sexual
attractiveness and survival. Nature,406, 67-70.
Byrnes, J. P., Miller, D. C., & Schaeffer, W. D. (1999). Gender differ-
ences in risk taking: A meta-analysis. Psychological Bulletin,125,
Chagnon, N. A. (1988). Life histories, blood revenge, and warfare in a
tribal population. Science,239, 985-992.
Coates, J. M., & Herbert, J. (2008). Endogenous steroids and financial
risk taking on a London trading floor. Proceedings of the National
Academy of Sciences, USA,105, 6167-6172.
Dabbs, J. M., & Dabbs, M. G. (2000). Heroes, rogues and lovers:
Testosterone and behavior. New York: McGraw-Hill.
Daly, M., & Wilson, M. (1988). Homicide. New York: Aldine.
Daly, M., & Wilson, M. (1990). Killing the competition. Human
Dutton, D. G., & Aron, A. P. (1974). Some evidence for heightened
sexual attraction under conditions of high anxiety. Journal of Per-
sonality and Social Psychology,30, 510-517.
Eagly, A. H., & Steffen, V. J. (1986). Gender and aggressive behavior:
A meta-analytic review of the social psychological literature. Psy-
chological Bulletin,100, 309-330.
Fellows, L. K., & Farah, M. J. (2005). Different underlying impair-
ments in decision making following ventromedial and dorsolateral
frontal damage in humans. Cerebral Cortex,15, 58-63.
Gallup, G. G., Jr., Burch, R. L., Zappieri, M. L., Parvez, R., Stockwell, M.,
device. Evolution and Human Behavior,24,277-289.
Glimcher, P. W., & Rustichini, A. (2004). Neuroeconomics: The con-
silience of brain and decision. Science,306, 447-452.
Granger, D. A. Shirtcliff, E. A., Booth, A., Kivlighan, K. T., &
Schwartz, E. B. (2004). The ‘‘trouble’’ with salivary testosterone.
Hare, T. A., Camerer, C. F., & Rangel, A. (2009). Self-control in deci-
sion making involves modulation of the vmPFC valuation system.
Hunt, J., Brooks, R., Jennions, M. D., Smith, M. J., Bentsen, C. L., &
Bussiere, L. F. (2004). High quality male field crickets invest heav-
ily in sexual display but die young. Nature,432,1024-1027.
Kokko, H., Brooks, R., McNamara, J. M., & Houston, A. I. (2002).
The sexual selection continuum. Proceedings of the Royal Society
of London Series B,269, 1331-1340.
Kruger, D. J. (2004). Sexual selection and the male: Female mortality
ratio. Evolutionary Psychology,2, 66-85.
Lejuez, C. W., Read, J. P., Kahler, C. W., Richards, J. B.,
Ramsey, S. E., Stuart, G. L., et al. (2002). Evaluation of a beha-
vioral measure of risk taking: The balloon analogue risk task
(BART). Journal of Experimental Psychology: Applied,8, 75-84.
McAlvanah, P. (2008). Are people more risk-taking in the presence of
the opposite sex? Journal of Economic Psychology,30, 136-146.
McIntyre, M., Gangestad, S. W., Gray, P. B., Chapman, J. F.,
Burnham, T. C., & O’Rourke, M. T. (2006). Romantic involve-
ment often reduces men’s testosterone levels—But not always:
The moderating role of extrapair sexual interest. Journal of
Personality and Social Psychology,91, 642-651.
O’Doherty, J., Kringelbach, M. L., Rolls, E. T., Hornak, J., &
Andrews, C. (2001). Abstract reward and punishment representations
in the human orbitofrontal cortex. Nature Neuroscience,4,95-102.
O’Doherty, J., Winston, J., Critchley, H., Perrett, D., Burt, D. M., &
Dolan, R. J. (2003). Beautyin a smile: The roleof medial orbitofron-
tal cortex in facial attractiveness. Neuropsychologia,41,147-155.
Preacher, K. J. (2002, May). Calculation for the test of the difference
between two independent correlation coefficients [Computer soft-
ware]. Retrieved July 15, 2009, from http://www.quantpsy.org
Preacher, K. J., & Hayes, A. (2004). SPSS and SAS procedures for
estimating indirect effects in simple mediational models. Behavior
Research Methods, Instruments, & Computers,36, 717-731.
Rhodes, G. (2006). The evolutionary psychology of facial beauty.
Annual Review of Psychology,57, 199-226.
Roads and Traffic Authority of New South Wales. (2001). Road traffic
accidents in New South Wales. Haymarket, Australia: Roads and
Traffic Authority, Road Safety Strategy Branch.
Roney, J. R. (2003). Behavioral and hormonal responses of men to
brief interactions with women. Evolution and Human Behavior,
Roney, J. R., Lukaszewski, A. W., & Simmons, Z. L. (2007). Rapid
endocrine responses of young men to social interactions with
young women. Hormones and Behavior,52, 326-333.
Rupp, H. A., & Wallen, K. (2007). Relationship between testosterone
and interest in sexual stimuli: The effect of experience. Hormones
and Behavior,52, 581-589.
Ronay and von Hippel 63
Schultheiss, O. C., Wirth, M. M., Torges, C. M., Pang, J. S.,
Villacorta, M. A., & Welsh, K. M. (2005). Effects of implicit
power motivation on men’s and women’s implicit learning and tes-
tosterone changes after social victory or defeat. Journal of Person-
ality and Social Psychology,88, 174-188.
Singh, D. (1993). Adaptive significance of female physical attractive-
ness: The role of waist-to-hip ratio. Journal of Personality and
Social Psychology,65, 293-307.
Trivers, R. L. (1972). Parental investment and sexual selection. In B.
Campbell (Ed.), Sexual selection and the descent of man, 1871-
1971 (pp. 136-179). Chicago: Aldine-Atherton.
Tuiten, A., Van Honk, J., Koppeschaar, H., Bernaards, C., Thisjen, J.,
istration on sexual arousal in women. Archives of General Psychia-
van Anders, S. M., Hamilton, L. D., & Watson, N. V. (2007). Multiple
partners are associated with higher testosterone in North American
men and women. Hormones and Behavior,51, 454-459.
van Honk, J., Schutter, D. J. L. G., Hermans, E. J., Putman, P.,
Tuiten, A., & Koppeschaar, H. (2004). Testosterone shifts the bal-
ance between sensitivity for punishment and reward in healthy
young women. Psychoneuroendocrinology,29, 937-943.
Wells, G. L., & Windschitl, P. D. (1999). Stimulus sampling and
social psychological experimentation. Personality and Social
Psychology Bulletin,25, 1115-1125.
Wilson, M., & Daly, M. (2004). Do pretty women inspire men to dis-
count the future? Biology Letters,271, 177-179.
Wilson, M., Daly, M., & Pound, N. (2002). An evolutionary
psychological perspective on the modulation of competitive
confrontation and risk-taking. Hormones, Brain and Behavior,
World Health Organization. (2002). Gender and road traffic injuries.
Geneva, Switzerland: Author, Department of Gender and Women’s
Xue, G., Lu, Z., Levin, I. P., Weller, J. A., Li, Z., & Bechara, A.
(2009). Functional dissociations of risk and reward processing in
the medial prefrontal cortex. Cerebral Cortex,19, 1019-1027.
Zajonc, R. B. (1965). Social facilitation. Science,149, 269-274.
Richard Ronay received his BA and BPSYCH from Macquarie Uni-
versity in Sydney, Australia and is currently a doctoral candidate in
Social Psychology at the University of Queensland.
William von Hippel received his BA from Yale University and his
PhD in Social Psychology from the University of Michigan and is
Professor of Psychology at the University of Queensland.
64 Social Psychological and Personality Science 1(1)