Differences in digit ratios between gay men who prefer receptive versus insertive sex roles indicate a role for prenatal androgen

Abstract

Among non-human mammals, exposure to androgens during critical periods of development leads to gynephilia (attraction to females), whereas the absence or low levels of prenatal androgens leads to androphilia (attraction to males). However, in humans, retrospective markers of prenatal androgens have only been associated with gynephilia among women, but not with androphilia among men. Here, we asked whether an indirect indication of prenatal androgen exposure, 2D:4D, differs between subsets of gay men delineated by anal sex role (ASR). ASR was used as a proxy for subgroups because ASR groups tend to differ in other measures affected by brain sexual differentiation, such as gender conformity. First, we replicated the finding that gay men with a receptive ASR preference (bottoms) report greater gender nonconformity (GNC) compared to gay men with an insertive ASR preference (tops). We then found that Tops have a lower (male-typical) average right-hand digit ratio than Bottoms, and that among all gay men the right-hand 2D:4D correlated with GNC, indicating that a higher (female-typical) 2D:4D is associated with increased GNC. Differences were found between non-exclusive and exclusive same-sex attraction and GNC, and ASR group differences on digit ratios do not reach significance when all non-heterosexual men are included in the analyses, suggesting greater heterogeneity in the development of non-exclusive same-sex sexual orientations. Overall, results support a role for prenatal androgens, as approximated by digit ratios, in influencing the sexual orientation and GNC of a subset of gay men.

Figure 1

ASR preference groups differ in right digit ratios. ASR Bottoms have more female-typical (higher) right digit ratios than ASR Tops (d = 0.62); Versatiles were intermediate, such that they did not significantly differ from either ASR Tops or Bottoms. *Significant difference, p < .05.

Figure 2

ASR preference groups differ in gender nonconformity. On both Recalled Childhood GNC and Adulthood GNC measures, ASR Tops rated themselves as more male-typical compared to Bottoms. On the Childhood GNC measure, Versatiles were intermediate, such that they did not significantly differ from Tops or Bottoms. On the measure of Adulthood GNC, Versatiles rated themselves as more GNC than Tops. For consistency across measures, score on the Recalled Childhood GNC scale were subtracted by 5 and multiplied by negative 1; Thus, a higher score on both the Childhood and Adulthood GNC indicates more gender nonconformity. *Significant difference, p < .05.

Figure 3

Right digit ratios correlate with gender nonconformity scores. A correlation between right digit ratios and Childhood GNC indicates that a more female-typical digit ratio is associated with higher gender nonconformity scores. Similarly, a positive correlation between right digit ratios and Adulthood GNC indicates that a more male-typical digit ratio is associated with higher gender conformity. For consistency across measures, score on the Recalled Childhood GNC scale were subtracted by 5 and multiplied by negative 1 for graphical depiction of results; Thus, a higher score on both the Childhood and Adulthood GNC graphs indicates more gender nonconformity.