Polys have highest testosterone, monos lowest

Discovery Channel News

It's easy to make junk science out of real science, so I hesitated to post this yesterday, but it's all over the poly lists anyway. Keep in mind that the hormone (endocrine) system is very complex, not always what it appears, and is only a small part of what we are and do. But. . . .

Study: Sex Makes People Feel Sexier

Jennifer Viegas, Discovery News

March 1, 2007 — Sexual activity for men and women, straight or gay, raises testosterone levels, which, at least in women, fuels the desire for intercourse, increases the likelihood of experiencing an orgasm and heightens the individual’s belief in her own sexiness, recent studies have found.

The findings are among the first to suggest that men and women can alter their own hormone levels based on how often they cuddle or copulate, both of which can lead to testosterone rises. . . .

"Our [first] paper does suggest that there might be sexual benefits to higher levels of testosterone," lead author Sari van Anders told Discovery News.

Van Anders, a researcher in the Department of Psychology at Canada’s Simon Fraser University, and her team conducted two separate studies. Both have been accepted for publication in the journal Hormones and Behavior. . . .

"We don’t know how testosterone increases after sex and close physical intimacy might benefit women, but some possibilities to examine in future studies include increased sexual desire, more positive moods, or more energy," said van Anders.

The second study also looked at testosterone increases related to sexual activity, but this time the test subjects were single, monogamous or polyamorous. A person is said to have a polyamorous lifestyle when he or she is involved in multiple, committed relationships.

The researchers determined partnered men and women had the lowest overall testosterone levels, while polyamorous men and women both had higher amounts of testosterone than single or monogamously partnered individuals.

Read the whole news article.

If this holds up, and if there is a cause-and-effect relationship here, one has to wonder which way it goes? Does polyamory increase testosterone? Or do people who have high testosterone to begin with find their way into poly? Or is there back-and-forth reinforcement between the two?

The two papers in question are:

Associations between testosterone secretion and sexual activity in women. van Anders, Sari M.; Hamilton, Lisa Dawn; Schmidt, Nicole; & Watson, Neil V.; in press. Hormones and Behavior. Abstract:

Some studies show an increase in testosterone (T) after sexual activity; this literature has inconsistent findings, focuses mostly on men, and does not employ control activities. The present study examined within-subject effects of intercourse versus control activities (cuddling; exercise) on salivary T. The initial sample included 49 women (mostly heterosexual), though not all participants returned all samples or engaged in all activities, leaving a smaller sample for endocrine analyses (n = 16). Participants attended an initial session in the laboratory where they completed questionnaires, and then engaged in the activities on their own. On three separate nights, they provided pre-activity, post-activity, and next-morning saliva samples and completed brief questionnaires at the last two timepoints. Women's T was higher pre-intercourse than pre-control activity. Women's T was also higher post-intercourse than post-control activity, though the percent change in T from pre- to post-activity was highest for cuddling, then intercourse, then exercise. Next-morning T did not differ by activity. Data pointed to an association between T and orgasming, sexual desire, and relationship commitment. Analyses on post-activity appraisals suggest that the close intimate physicality of a sexual and non-sexual nature can affect T and be beneficial in short-term and perhaps longer-lasting ways for women's sexuality and relationships.

Preprint of whole paper (uncorrected proof).

Multiple partners are associated with higher testosterone in North American men and women. van Anders, Sari M.; Hamilton, Lisa Dawn; & Watson, Neil V.; in press. Hormones and Behavior. Abstract:

Previous research has shown that being partnered is associated with lower testosterone (T) in men and women. To address how multiple partners may be associated with T, we examined 47 men and 48 women who were single, monoamorously partnered (partnered), polyamorous (having multiple committed relationships), or in a polyamorous lifestyle but not currently multipartnered. Men who were partnered had lower T than all other men, and polyamorous men had higher T than single men. Polyamorous women had higher T than all other women. Measures of sociosexual orientation (SOI) and sexual desire differed in women by relationship type, but not in men. Findings are interpreted in light of ‘competitive’ and ‘bond–maintenance’ relationship orientations and statuses.

Preprint of whole paper (uncorrected proof).

I asked van Anders about the statistical significance of these studies in light of the small numbers of subjects. She replied:

"There is a general misconception about sample size. Finding statistically significant differences with smaller samples sizes means that the relationship between the two variables of interest is actually quite strong, i.e. that you only need a small sample to see evidence of the association. A small sample size can, however, be problematic in terms of generalizability, i.e. does the finding apply to all the different people outside the sample? Statistical analyses take sample size into account, so findings of statistical significance cannot be undermined by sample size. But, as I note, generalizability can be limited with smaller sample sizes.

"People are also generally used to hearing about large public health studies in the media. Studies, for example, of cancer will routinely use thousands of participants because the effect size (roughly meaning the incidence of cancer and how it is related to variables of interest) is so small. So, the more participants you need to see a relationship between two variables (e.g. cancer-happiness, or partnering-hormones), the weaker the relationship is. I hope this clears up a very common misunderstanding."