Click here to buy!

1,512,289 visitors since


Printer Friendly Version

Sexual Orientation and Smell - Reactions to Martins, et al (2005)

Warren Throckmorton, PhD

Research Review (a work in progress, write me at if you see something incorrect or have questions about what I mean)

Title: Preference for human body odors is influenced by gender and sexual orientation
Journal: Psychological Science, September, 2005 (Accepted; prepublication copy)
Authors: Martins, Y., Preti, G., Crabtree, C., & Wysocki, C.

Abstract (by the authors, from the journal article)

Human body odor may contribute to selecting a partner. If true, then sexual orientation may influence preference for and perhaps production of human body odors. To test these hypotheses, heterosexual and homosexual males and females made two-alternative, forced-choice preferences for body odors obtained from other heterosexual and homosexual males and females. In four sub-experiments, subjects chose between odors from (i) heterosexual males versus gay males, (ii) heterosexual males versus females, (iii) heterosexual females versus lesbians, and (iv) gay males versus lesbians. Results indicate that differences in body odor are detected and responded to based, in part, upon an individual’s gender and sexual orientation. Possible mechanisms underlying these findings are discussed.

What the study investigated:

The authors began the article with a review of literature designed to support the idea that “Pheromones are chemical signals that influence the behavior or physiology of others.” (p. 1). Further, they pose an if-then statement: “If humans use the odors of others, at least in part, to seek a mate, then sexual orientation may influence preferences for human odors and/or production of these odors.” (p. 1). They note that “odors of the opposite sex are highly attractive” for many mammals. This is the foundation given to investigate how sexual orientation might relate to odor preferences.

The authors secured odor samples of armpit sweat from four categories of people (gay men, lesbians, straight men and women). They then secured 80 participants; 20 in each category of sexual preference named above. These subjects then smelled the odors from each category through a series of forced choices and rated their favorites of the two options and the degree of intensity and pleasantness for each of the odor types.

What the study found:

I cannot evaluate the methods used since this is not my area of expertise. However, here are the results as I read them:

• HM (heterosexual males), HF (heterosexual females), and LF (lesbian females) preferred odors from HM over GM (gay males)
• GM preferred odors from GM over HM
• HM, HF, LF over age 25 preferred odors from LF over GM
• GM preferred GM over LF
• GM, LF, HF preferred HF over LF
• HM did not prefer HF over LF in comparison
• GM preferred odors from HF over those from HM

Participants also ranked the pleasantness of these odors and if I am reading the author’s chart correctly, readings could be below zero for unpleasantness and above zero for pleasantness.

Questions about this study:

Incredibly, the GM group rated the HF odors more positively than anyone else rated anything, much higher than odors from GMs were rated. What does that mean?

The GM preference for HF is great even when considered for pleasantness. According to the study, GM give higher pleasantness ratings for HF than GM! GM rated HF slightly higher on pleasantness than did LF. If these odors are a reflection of sexual interest or orientation, I fail to see how these results corroborate that hypothesis.

The authors write: “First, it is clear that gay males, as odor evaluators, are strikingly different from heterosexual males and females and lesbians – they (GM) prefer the odors of gay males and heterosexual females in each of the forced-choice preference tests in which the odors were present.” This statement contains results that support two seemingly contrary hypotheses. The explanation that seems the most obvious is that gay males prefer “feminine” smells. Given the fact that most hormonal theories of sexual orientation are inversion theories (gay male brains are feminized) then this could support that idea. However, when you look at the theory concerning odor preference that the authors advanced in the literature review, then these results are contradictory. One would expect bisexual males to be attracted to both gay males and straight women but not exclusively gay males.

Another puzzle: I am baffled about why the authors chose to segment the lesbian group into 18-25 years and over 25 year olds. There was one forced choice difference where age mattered for lesbians: LFs over 25 preferred LF over GM in a comparison.

Strangely the authors seem to be allowing for some impact of environment here or at least maturation. In other words, if there was some biologically hard-wired mechanism that determined a lesbian response, why would this not be true before age 25? This is an admission that experience as a lesbian may have an impact on the way these participants responded to the odors. I also wonder why age differences were not explored with the other groups?

The forced comparisons lead to some rather odd conclusions if one is looking for an innate sexual orientation as expressed through smell preference. For instance, why don’t HM prefer HF over other odors? Why do GM prefer HF over LF? In the discussion section, these matters are not discussed. The authors seem so excited that they have found differences that they spend their time giving possible neurobiological explanations for the differences. Some of these differences seem not to make sense or even be contradictory but this is not explored.

Contrasts with the Swedish study:

In contrast to the Swedish study, there were no controls for disease status of the donors or smellers. This could be impactful, especially with HIV/AIDS status uncontrolled or assessed.

The Swedish study (Savic et al) found PET scan differences in how gay men and straight men and women responded to pheromone compounds but there were no subjective differences in assessments of pleasantness due to sexual orientation. The Martins et al study (Philadelphia) did find differences, many of them, on assessments of pleasantness.

Additional problems:

- Race/ethnicity/kinship/friendship status was not controlled. Any of these could impact the preferences for body odors.

In fact, the study methodology indicates that participants could recruit their friends into the study, apparently either as odor donors and/or smellers. Given the research on kinship and odor distinction that friends could have preferred the odors of friends. According to the study methodology, gay participants were recruited via friendship networks. It seems highly plausible that this uncontrolled factor could have contaminated the study.


Ask Dr. Throckmorton

Submit a question for Dr. Throckmorton.

Questions & Answers: Archives

Email Updates

Enter your email address to subscribe to Dr. Throckmorton's Email Updates.