Does testosterone make men misread a woman's interest?
A dose of testosterone did not make men overestimate a woman's sexual interest overall, but it made their judgments track her friendliness more closely, especially in men who rated themselves as attractive.
Source
Under the influence: exogenous testosterone influences men's cross-sex perceptions of sexual interest
Study at a glance
- Design
- Human experiment — Double-blind single-dose intranasal testosterone vs placebo gel, followed later the same session by a 3-minute unscripted waiting-room encounter with a female confederate
- N
- N=190 · 190 heterosexual men analysed for perceived sexual interest; 175 for the sexual misperception bias score because the confederate's own ratings were missing for some sessions
- Population
- Healthy heterosexual men aged 18-40 recruited in Ontario, Canada, from a larger testosterone and decision-making study
- Outcome
- Sexual misperception bias (perceived minus the confederate's actual short-term interest) and total perceived sexual interest
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
Men taking part in a larger lab study received either testosterone gel or a placebo gel in the nose, with neither participants nor researchers knowing which. Later in the same session they waited for 3 minutes in a room with an attractive female confederate who was told to be friendly and warm. Afterwards the men rated her short- and long-term sexual interest in them, their own interest in her, and their own attractiveness, while blind judges coded her friendly (affiliative) behaviour from video.
What they found
Across the sample, men overestimated her interest compared with what she actually reported, and men with higher natural (basal) testosterone overestimated more; once basal testosterone was controlled, the bias disappeared. The testosterone dose did not change the size of the bias compared with placebo (a small, non-significant difference). However, only in the testosterone group did her friendliness predict how interested the men thought she was, and this link got stronger the more attractive the men rated themselves. Men's own interest in her and their self-rated attractiveness were the strongest predictors of perceived interest, consistent with men projecting their desire onto her.
The limits
What it doesn't show
The study was not designed to test sexual misperception; these analyses are exploratory and need confirmation. There was only one confederate, so the 'actual interest' benchmark reflects one woman's idiosyncratic judgments. The encounter happened well after dosing, after economic decision tasks, when testosterone levels were probably falling, so the context is artificial. Testosterone levels after dosing were not measured, and the key interaction effects come from many regression models, which raises the risk of chance findings.
Key terms
- Sexual misperception bias
- The tendency, found more in men, to believe a potential partner is more sexually interested than they really are.
- Error Management Theory
- The idea that when judgments are uncertain, evolution favours biases toward the less costly error, such as a man overestimating rather than missing sexual interest.
- Projection
- Attributing your own feelings or desires to another person, e.g. assuming she is interested because you are.
- Exogenous vs basal testosterone
- Exogenous testosterone is given from outside (here as a nasal gel); basal testosterone is a person's natural resting level, measured here from saliva.
- Affiliative behaviour
- Friendly, warm behaviour such as smiling and engaging in conversation, which can be mistaken for romantic interest.
- Zero-acquaintance paradigm
- A method where strangers meet briefly and then judge each other, capturing first impressions.
Flashcards
0 of 11 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
What was the design of this study?
Common questions
If testosterone didn't increase the bias, why do the authors say it matters?
Because it changed how men used cues: under testosterone, her friendliness predicted perceived interest, while under placebo it did not. Natural testosterone levels were also linked to more overperception.
How was 'actual interest' measured?
The female confederate rated her own short-term interest in each man, and the bias score was the difference between that and the man's estimate of her interest.
Does this prove testosterone causes men to misread women in real life?
No. It was one confederate in an artificial lab setting, the analyses were exploratory, and the main effect of the dose on the bias itself was not significant.
More on Social cognition