Hormones in speed-dating: The role of testosterone and cortisol in attraction

Hormones in speed-dating: The role of testosterone and cortisol in attraction. Leander van der Meij et al. Hormones and Behavior, Volume 116, November 2019, 104555. https://doi.org/10.1016/j.yhbeh.2019.07.003

Highlights
•    Testosterone and cortisol levels are probably related to romantic attraction.
•    We conducted a study with a romantic speed-dating condition and control condition.
•    We found strong anticipatory hormonal responses.
•    In women, but not in men, testosterone levels increased during speed-dating.
•    Cortisol was related more to the attraction of a romantic partner than testosterone.

Abstract: There is evidence that testosterone and cortisol levels are related to the attraction of a romantic partner; testosterone levels relate to a wide range of sexual behaviors and cortisol is a crucial component in the response to stress. To investigate this, we conducted a speed-dating study among heterosexual singles. We measured salivary testosterone and cortisol changes in men and women (n = 79) when they participated in a romantic condition (meeting opposite-sex others, i.e., potential romantic partners), as well as a control condition (meeting same-sex others, i.e., potential friends). Over the course of the romantic speed-dating event, results showed that women's but not men's testosterone levels increased and cortisol levels decreased for both men and women. These findings indicate that men's testosterone and cortisol levels were elevated in anticipation of the event, whereas for women, this appears to only be the case for cortisol. Concerning the relationship between attraction and hormonal change, four important findings can be distinguished. First, men were more popular when they arrived at the romantic speed-dating event with elevated cortisol levels. Second, in both men and women, a larger change in cortisol levels during romantic speed-dating was related to more selectivity. Third, testosterone alone was unrelated to any romantic speed-dating outcome (selectivity or popularity). However, fourth, women who arrived at the romantic speed-dating event with higher testosterone levels were more selective when their anticipatory cortisol response was low. Overall, our findings suggest that changes in the hormone cortisol may be stronger associated with the attraction of a romantic partner than testosterone.

4. Discussion

4.1. Testosterone change

Our findings showed that testosterone levels increased in women
during romantic speed-dating and decreased in women during the
control condition. Although these changes were small-medium effect
sizes, they are in line with theoretical models predicting that high
testosterone levels relate to more mate acquisition (Archer, 2006;
Roney, 2016; Zilioli and Bird, 2017) and more competitive behavior
(van Anders et al., 2011). However, surprisingly, in men, testosterone
levels did not change during romantic speed-dating and remained high
throughout the event. This is not in line with some previous research, as
numerous studies have shown that men experience an increase in testosterone
levels when talking to a potential mate in a waiting room
situation (Roney et al., 2010, 2007, 2003; van der Meij et al., 2008),
although one other study also showed that testosterone levels did not
change during romantic speed-dating (Lefevre et al., 2013). A speculative
explanation for these divergent findings is that romantic speeddating
is a much more arousing social context than a waiting room
situation. Unlike a waiting room situation, romantic speed-dating is an
unambiguous dating context where individuals scan each other as potential
mates. While the waiting room situation is unlikely to trigger
prior expectations because participants do not know that they will be
waiting together, participants of a romantic speed-dating do know that
they will be evaluated as a potential romantic partner.
Thus, it could be that, in contrast to women's testosterone levels,
men's testosterone levels did not increase further due to negative
feedback from already high testosterone levels on the hypothalamuspituitary-
gonadal (HPG) axis. This may also have held true for the
control condition, as testosterone levels were similar in this condition.
In both conditions, men may have experienced greater amounts of social
evaluative stress than women, as they were being evaluated on
either suitability as a romantic partner or were checking the competition
in the control condition. This finding is in line with other recent
studies showing that testosterone levels increase in men during stress
tasks with a social evaluative component (Bedgood et al., 2014;
Lennartsson et al., 2012; Phan et al., 2017; Turan et al., 2015), although
some older studies found no change in testosterone levels after psychosocial
stress (Gerra et al., 2000; Heinz et al., 2003; Schoofs and
Wolf, 2011) and one other study showed a decrease (Schulz et al.,
1996). This increase in testosterone levels may be part of an adaptive
response that assists an individual to cope with social challenges
(Salvador, 2005; Salvador and Costa, 2009). Indeed, previous research
has shown that the more men experienced a testosterone increase the
more they affiliated with women (van der Meij et al., 2012).

4.2. Testosterone and attraction

An important finding is that testosterone levels were unrelated to
popularity and selectivity in both men and women. This null finding for
men may be related to the previously discussed elevated hormonal levels.
Male testosterone levels may have been too elevated for most
participants even before the romantic speed-dating began, which reduced
variance in testosterone levels such that we were unable to detect
a relationship with their behavior (either in selectivity or popularity).
However, it is important to note that we may have lacked the power to
detect smaller effect sizes, since men and women have different testosterone
levels, and thus we had to analyze their testosterone data
separately. In men, we did find an indication that a larger anticipatory
testosterone response was related to less selectivity, although this effect
was statistically not significant. Future studies with larger sample sizes
may untangle if heightened testosterone levels during romantic speeddating
makes the relationship between attraction and testosterone undetectable
in men.
For women, there was also no relationship between attraction and
testosterone levels. This null finding is more difficult to explain, as
testosterone levels in women did increase during romantic speeddating.
Additionally, previous research shows that, in a lab setting, an
increase in testosterone levels was related to more sexual arousal in
women (Tuiten et al., 2000), which suggests that increased testosterone
levels could decrease selectivity. A speculative explanation for this null
finding in women is that temporal changes in their testosterone levels
had less of an effect on their behavior in an ecologically valid environment
such as romantic speed-dating. Perhaps women more rationally
deliberated the pros and cons of a potential romantic partner
and were not so much affected by their own bodily and psychological
state. Also interesting was that female popularity was unaffected by
their testosterone levels. A possible explanation here could be that that
men's selectivity is not so much influenced by female behavior during
these speed dates. Men may largely determine beforehand if they will
say yes to a date based on physical appearance. For example, in one
particular study, BMI predicted 25% of female popularity alone
(Kurzban and Weeden, 2005). Another explanation could be that variance
in female popularity was limited and this reduced the power to
detect an effect of anticipatory testosterone. Indeed, men said yes to
72% of their dates whereas women said yes to 48% of their dates.

4.3. Cortisol change

Results showed that both men and women arrived at both the romantic
speed-dating and control condition with elevated cortisol levels
and that during the course of both conditions their cortisol levels decreased.
Furthermore, this decrease was a very large effect size in the
control condition and less so in the romantic speed-dating condition
(small-medium effect size). Also, cortisol levels were higher at the end
of romantic speed-dating than at the end of the control condition.
Together these findings indicate that participants perceived the romantic
speed-dating as more challenging and stressful than friendship
dating. This implies that being judged as a potential romantic partner is
more stressful, and requires more impression management than when
being judged as a potential friend. Furthermore, results showed that
cortisol levels decreased during the course of romantic speed-dating
and control condition. These results contrast with other studies showing
that a brief social contact with a potential romantic partner produces an
increase in cortisol levels in heterosexual men (Roney et al., 2010,
2007), although another study showed that cortisol levels only increased
when in such encounters men perceived their potential partner
as attractive (van der Meij et al., 2010).
There are two speculative explanations for these different results.
First, our speed-dating study took over an hour to complete, thus cortisol
levels may have started decreasing towards more normal values
due to negative feedback from high cortisol levels on the hypothalamuspituitary-
adrenal (HPA) axis. Second, unlike these other studies, our
participants probably arrived with relatively high cortisol levels in
anticipation of the event. Thus, after having experienced several speed
dates they may have habituated. Adding to this, social affiliation may
have reduced anxiety through the release of oxytocin (for a review see
Heinrichs et al., 2009). Indeed, previous research has shown that oxytocin
administration reduces cortisol secretion during social evaluative
stress (Heinrichs et al., 2003).

4.4. Cortisol and attraction

There are two interesting findings concerning cortisol and attraction.
First, only in men, cortisol release in anticipation of romantic
speed-dating was related to more popularity. This effect was substantial
as women said “yes” to 34% of their dates when men experienced a
small anticipatory cortisol response, whereas they said “yes” to 65% of
their dates when men experienced a high anticipatory cortisol response.
A possible explanation is that men who arrived with these high levels
were more interested in dating women. Consequently, they may have
put more energy into making positive impressions during the speeddates.
Additionally, they may have had more energy at their disposal
since cortisol secretion increases local cerebral glucose utilization and
cardiovascular tone (Sapolsky et al., 2000). However, it is important to
note that a causal effect of cortisol on mate attraction could not be
established in the current study. Other third variables, such as a high
speed-dating motivation, may have produced more mate attraction
behaviors as well as a rise in cortisol levels in anticipation of the event.
Why women with elevated cortisol levels were not more popular may
have to do with the small variance in female popularity. Similar to the
function of testosterone, the function of these elevated cortisol levels in
men may help them cope with social challenges (Salvador, 2005;
Salvador and Costa, 2009). Furthermore, it could also reflect an effort
to affiliate, as it has been shown that, in men, increased cortisol secretions
during social evaluative stress predicted their feelings of closeness
to a stranger in a subsequent interaction (Berger et al., 2016).
Thus, our finding lends support for a “tend and befriend response” in
men during stressful times (Geary and Flinn, 2002). Finally, this finding
is in line with the Physiology of Romantic Pair Bond Initiation and
Maintenance Model, as this model posits that HPA-axis activation in
mating contexts is necessary to improve evaluations by potential mates
(Mercado and Hibel, 2017).
Second, contrary to our hypothesis, in both men and women, a
larger cortisol change during romantic speed-dating was related to
more selectivity (controlling for baseline and cortisol change in the
control condition), although this effect was small to medium. A speculative
explanation is that romantic speed-dating was not a positive
experience for all participants. Those men and women that experienced
an increase in cortisol levels may have been worried that they would
end up with no or very few matches. This would be in line with the
stress literature as cortisol release is more prominent in social situations
that are uncontrollable and pose a social-evaluative threat (for a review
see Dickerson and Kemeny, 2004). Romantic speed-dating has both
these elements: participants can only guess whether their interaction
partner likes them (low control) and they are being evaluated as a
potential partner at each date (high social-evaluative threat). In such a
scenario, two different effects can be argued. The most rational strategy
would be to say “yes” to many dates (low selectivity), to increase the
chances of a match. However, our data shows the opposite: a larger
cortisol change was related to more selectivity. This shows that a different
process may have been occuring. Perhaps those participants who
experienced a larger increase in cortisol levels during speed-dating
were more preoccupied with impression management and found it,
therefore, more difficult to connect with their dates. As a result, they
could have subjectively experienced fewer matches and said “yes” to
fewer dates.

4.5. Testosterone×Cortisol

Our results showed overall weak support for the dual-hormone
hypothesis (Mehta and Josephs, 2010) in a mating context. The most
direct prediction from this hypothesis would be that popularity in romantic
speed-dating was related to the interaction between testosterone
and cortisol levels, yet we did not find evidence for this. These null
findings could mean that the dual-hormone hypothesis is limited to
social contexts in which social status can be gained more openly, for
example in competition with others (Zilioli and Watson, 2012) or in
leadership positions (Sherman et al., 2016). A potential alternative
explanation for these null findings is that saying yes or no to other dates
may depend on unique conversation dynamics for which we could not
control. Perhaps this reduced our power to detect the interaction between
both hormones. Indeed, many of the studies showing support for
the dual-hormone hypothesis use laboratory tasks (Mehta et al., 2015)
in which it is far easier to control for confounding variables.
Nonetheless, we did find support for one of our mutually exclusive
predictions based on the dual-hormone hypothesis. Only in women, a
higher anticipatory testosterone response was related to more selectivity
when their anticipatory cortisol response was low. Women
with this hormone profile may not have been motivated to gain social
status by going for more matches (thus by being less selective). Instead,
these women may have been motivated to gain social status by appearing
exclusive. This finding would also be in line with the sexual
double standard (Sagebin Bordini and Sperb, 2013). Women feel they
are being valued more highly as a partner when they are restrictive in
their sexual contacts, whereas for men this is less of a concern.