Tuesday May 25 2010
This simple lab experiment cannot fully predict behaviour in real life
“Testosterone reduces trust”, according to The Independent, which says a study has found that “small doses of the male sex hormone testosterone can make people less trusting of strangers”.
This study looked at the effects of testosterone compared to a placebo treatment in 24 healthy young women. It found that the women were more likely to rate unfamiliar faces as less trustworthy after receiving testosterone than after receiving a dummy treatment. This change in trust was only apparent in women who were more trusting at the start of the study.
This study suggests that testosterone can affect interpersonal trust, at least in women. However, it was relatively small and included only young healthy women, meaning that the effects may differ in males, different age groups or in less healthy individuals. Human behaviour is complex, and this type of research is a small step towards unravelling how different hormones affect it. However, simple laboratory experiments such as this one may not fully predict how people will behave in real-life situations.
Where did the story come from?
The study was carried out by researchers from Utrecht University and the University of Cape Town. Funding was provided by Utrecht University and the Hope for Depression Research Foundation. The study was published in the peer-reviewed journal Proceedings of the National Academy of Sciences of the USA.
The Independent and The Daily Telegraph both reported this study accurately.
What kind of research was this?
This was a randomised crossover trial looking at the effects of testosterone on interpersonal trust. The researchers say that the hormone oxytocin has been shown to increase trust, but that testosterone is associated with competition and dominance, and may have an opposite effect on trust.
This study used a good design for assessing the effects of testosterone. Importantly, the study allowed a three-day ‘washout period’ between testosterone and placebo treatments to ensure that the effects of the first treatment did not influence the results of the second test. The study also blinded participants and researchers to which treatment was received, so that any preconceptions they had about the effects of testosterone could not affect the results of the test.
What did the research involve?
The researchers enrolled 24 healthy female volunteers (average age 20.2 years), and gave them either 0.5mg testosterone or a placebo on two separate days. On each occasion they then assessed how trustworthy the women perceived a set of unfamiliar faces to be. They subsequently compared the trustworthiness ratings given after receiving testosterone and the placebo treatment.
Women were not eligible to participate if they had a history of psychiatric or other disorders, smoked or took medicines other than contraceptives. Participants received the placebo and testosterone in a random order (i.e. either the testosterone tablet first and the placebo second, or vice versa). The testosterone and placebo were given orally as a liquid placed under the tongue. After the testosterone or placebo was given, the researchers waited four hours before giving the participants the trust test. There was a three-day break between the first test and the second test to make sure that the first treatment was not still having an effect.
The trust test involved showing the women 75 faces and asking them to rate how trustworthy they believed the person to be, from “very untrustworthy” to “neutral” to “very trustworthy”. Two different sets of faces were used for the two tests, but these had been matched for their trustworthiness ratings based on the responses of 36 healthy adults.
What were the basic results?
The researchers found that women rated unfamiliar faces as less trustworthy after receiving testosterone than after they received the placebo. This was due to a reduction of trustworthiness ratings provided by ‘more trusting’ women that gave high trustworthiness ratings after a placebo but lower ratings after testosterone. Women who had lower trustworthiness ratings after placebo did not show a reduction in trust with testosterone. This effect was not related to the women’s natural testosterone levels at the start of the study.
Women were not able to correctly guess which treatment they had received more often than would be expected by chance.
How did the researchers interpret the results?
The researchers concluded that testosterone increases social wariness in trusting individuals, and that this may prepare them better for competition for status and resources. They say that their study provides “unique insights into the hormonal regulation of human sociality”.
The study used an appropriate design to assess the effect of testosterone. It randomly assigned the order in which women received testosterone and placebo; it blinded participants and researchers to the treatment received, and allowed a ‘washout period’ between treatments.
This study suggests that testosterone can affect interpersonal trust, at least in women. This study only assessed women because the dose of testosterone used is known to have a measureable effect on a woman’s body within four hours. The dose of testosterone needed for men and the time course of the effects in them are not clearly known. Further experiments will be needed to see whether the results also apply to men. Also, the study was relatively small and included only young healthy women, meaning that the effects may differ in different age groups or in less healthy individuals.
Human behaviour is complex, and these types of studies can incrementally help to unravel how different hormones influence perception and relationships. However, simple experiments such as this one may not fully predict how people will behave in real-life situations.