Abstract: Although it is widely assumed that men’s sexual desire and interest in casual sex (i.e., sociosexual orientation) are linked to steroid hormone levels, evidence for such associations is mixed. Consequently, we tested for both longitudinal and cross-sectional relationships between salivary testosterone, cortisol, reported sexual desire and sociosexuality in a sample of 61 young adult men, each of whom was tested weekly on up to five occasions. Longitudinal analyses showed no clear relationships between steroid hormones and self-reported sexual desire or sociosexual orientation. Cross-sectional analyses showed no significant associations between average hormone levels and self-reported sexual desire. However, some aspects of sociosexuality, most notably desire for casual sex, were related to men’s average hormone levels. Men with higher average testosterone reported greater desire for casual sex, but only if they also had relatively low average cortisol. These results support a Dual Hormone account of men’s sociosexuality, in which the combined effects of testosterone and cortisol predict the extent of men’s interest in casual sex.
Discussion
Our analyses showed no clear effects of within-subject changes in men’s
testosterone, cortisol, or their interaction on any aspects of sociosexuality or
sexual desire. There was a weak negative effect of current cortisol on
sociosexual desire, but this was not robust to correction for multiple tests
(uncorrected p-value = .044). Thus, we did not replicate Raisanen et al’s
(2018) recent finding that within-subject changes in men’s solitary, but not
dyadic, sexual desire track changes in their testosterone and cortisol.
However, our null results for sociosexuality and within-subject changes in
men’s testosterone are consistent with similar null results reported by Gettler
et al. (2019). While our results support the recent proposal that changes in
endogenous steroid hormones contribute little (if at all) to within-subject
changes in men’s sociosexuality (Gettler et al., 2019), they do not support the
claim that endogenous steroid hormones contribute to the regulation of men’s
sexual desire (Raisanen et al., 2018).
Our analyses of responses on the SDI-2 also showed no significant crosssectional
associations between aspects of men’s sexual desire and average
steroid hormones. Thus, we did not replicate previous findings in which men
with higher average testosterone reported greater solitary sexual desire (Das
& Sawin, 2016; van Anders & Dunn, 2009). However, the null results for
cross-sectional associations between aspects of men’s sexual desire and
steroid hormones are consistent with similar null results that have been
reported in other studies (van Anders et al., 2007; van Anders, 2012).
Collectively, these results suggest that associations between average steroid
hormone levels and sexual desire in men are not robust.
Some previous studies have reported that men with higher average
testosterone levels score higher on sociosexual desire (Edelstein et al., 2011;
Puts et al., 2015), but lower on sociosexual behavior (Puts et al., 2015).
These results have been interpreted as evidence for a feedback loop in which
rising testosterone levels increase sociosexual desire, but that engaging in
sexual behavior causes men’s testosterone levels to fall (Puts et al., 2015).
Neither our longitudinal nor cross-sectional analyses of men’s sociosexuality
support this proposal. However, our null results for sociosexuality and men’s
average testosterone levels are consistent with similar null results reported in
other studies (Kordsmeyer et al., 2018; van Anders et al., 2007).
Intriguingly, we found that average testosterone was positively related to
sociosexual attitudes, sociosexual desires, and global sociosexual orientation
(i.e., total scores on the SOI-R) among men with relatively low cortisol.
Although we did not predict this result, we note here that the interactions
between average testosterone and average cortisol for sociosexual attitudes
and global sociosexual orientation would be significant even if Bonferroni
corrected for multiple comparisons. Some previous research suggests that the
combination of high testosterone and low cortisol is associated with status
related behaviors (see Mehta & Prasad, 2015, for a review of this Dual Hormone Hypothesis).
If this is the case, our results present preliminary
evidence that attitudes to uncommitted sexual relationships might be similarly
related to high testosterone and low cortisol. Further research would be
necessary to shed further light on this possibility. Previous studies
investigating possible associations between steroid hormones and men’s
sociosexuality may not have detected these relationships because they did
not consider the interaction between average testosterone and average
cortisol (Kordsmeyer et al., 2018; van Anders et al., 2007).
Strengths of the current study include the longitudinal analyses and
consideration of testosterone, cortisol, and their interactions. However, there
are limitations that could be addressed in future work. For example,
replicating the cross-sectional findings for sociosexuality in a larger sample
may clarify whether our results are robust or false positives.
In conclusion, we did not replicate previous results linking aspects of men’s
sexual desire to their steroid hormones. However, we did find evidence that
aspects of men’s sociosexual orientation, most notably their attitude to casual
sex, was predicted by the interaction between average testosterone and
average cortisol. Men with higher average testosterone levels reported more
positive attitudes to casual sex, but only if they also had relatively low average
cortisol. While such a pattern of results is arguably consistent with the Dual
Hormone Hypothesis of men’s competitive behaviors, further work is needed
to establish whether this pattern of results is robust.
Our analyses showed no clear effects of within-subject changes in men’s
testosterone, cortisol, or their interaction on any aspects of sociosexuality or
sexual desire. There was a weak negative effect of current cortisol on
sociosexual desire, but this was not robust to correction for multiple tests
(uncorrected p-value = .044). Thus, we did not replicate Raisanen et al’s
(2018) recent finding that within-subject changes in men’s solitary, but not
dyadic, sexual desire track changes in their testosterone and cortisol.
However, our null results for sociosexuality and within-subject changes in
men’s testosterone are consistent with similar null results reported by Gettler
et al. (2019). While our results support the recent proposal that changes in
endogenous steroid hormones contribute little (if at all) to within-subject
changes in men’s sociosexuality (Gettler et al., 2019), they do not support the
claim that endogenous steroid hormones contribute to the regulation of men’s
sexual desire (Raisanen et al., 2018).
Our analyses of responses on the SDI-2 also showed no significant crosssectional
associations between aspects of men’s sexual desire and average
steroid hormones. Thus, we did not replicate previous findings in which men
with higher average testosterone reported greater solitary sexual desire (Das
& Sawin, 2016; van Anders & Dunn, 2009). However, the null results for
cross-sectional associations between aspects of men’s sexual desire and
steroid hormones are consistent with similar null results that have been
reported in other studies (van Anders et al., 2007; van Anders, 2012).
Collectively, these results suggest that associations between average steroid
hormone levels and sexual desire in men are not robust.
Some previous studies have reported that men with higher average
testosterone levels score higher on sociosexual desire (Edelstein et al., 2011;
Puts et al., 2015), but lower on sociosexual behavior (Puts et al., 2015).
These results have been interpreted as evidence for a feedback loop in which
rising testosterone levels increase sociosexual desire, but that engaging in
sexual behavior causes men’s testosterone levels to fall (Puts et al., 2015).
Neither our longitudinal nor cross-sectional analyses of men’s sociosexuality
support this proposal. However, our null results for sociosexuality and men’s
average testosterone levels are consistent with similar null results reported in
other studies (Kordsmeyer et al., 2018; van Anders et al., 2007).
Intriguingly, we found that average testosterone was positively related to
sociosexual attitudes, sociosexual desires, and global sociosexual orientation
(i.e., total scores on the SOI-R) among men with relatively low cortisol.
Although we did not predict this result, we note here that the interactions
between average testosterone and average cortisol for sociosexual attitudes
and global sociosexual orientation would be significant even if Bonferroni
corrected for multiple comparisons. Some previous research suggests that the
combination of high testosterone and low cortisol is associated with status
related behaviors (see Mehta & Prasad, 2015, for a review of this Dual Hormone Hypothesis).
If this is the case, our results present preliminary
evidence that attitudes to uncommitted sexual relationships might be similarly
related to high testosterone and low cortisol. Further research would be
necessary to shed further light on this possibility. Previous studies
investigating possible associations between steroid hormones and men’s
sociosexuality may not have detected these relationships because they did
not consider the interaction between average testosterone and average
cortisol (Kordsmeyer et al., 2018; van Anders et al., 2007).
Strengths of the current study include the longitudinal analyses and
consideration of testosterone, cortisol, and their interactions. However, there
are limitations that could be addressed in future work. For example,
replicating the cross-sectional findings for sociosexuality in a larger sample
may clarify whether our results are robust or false positives.
In conclusion, we did not replicate previous results linking aspects of men’s
sexual desire to their steroid hormones. However, we did find evidence that
aspects of men’s sociosexual orientation, most notably their attitude to casual
sex, was predicted by the interaction between average testosterone and
average cortisol. Men with higher average testosterone levels reported more
positive attitudes to casual sex, but only if they also had relatively low average
cortisol. While such a pattern of results is arguably consistent with the Dual
Hormone Hypothesis of men’s competitive behaviors, further work is needed
to establish whether this pattern of results is robust.