Between week 8 and 24 of pregnancy, a male fetus is exposed to testosterone, whereas a female fetus is not. This prenatal testosterone peak in males is thought to lead to early ‘masculinization’ of the brain and behavior. That is to say, it enhances the development of typically male traits. For example, increased aggression has been attributed to relatively high prenatal testosterone exposure.

Because of ethical considerations direct measurements of prenatal testosterone in healthy human fetuses are carried out only rarely. Therefore, researchers have tried to find ways to estimate prenatal testosterone exposure using indirect measures. One of these ways is to study opposite-sex (boy/girl) twins: fetuses gestated together with a male co-twin are thought to be ‘masculinized’ in their development, arguably due to the influence of prenatal testosterone exposure. According to this so-called twin–testosterone-transfer (TTT) hypothesis, female fetuses with a twin brother are being exposed to higher levels of prenatal testosterone than fetuses with a twin sister. The hypothesis is based on the idea that within the uterus twins exchange hormones either via the maternal bloodstream or directly through the amniotic membrane. Even though hormonal transfer between fetuses has already been established in animals, it is still unclear whether a similar mechanism exists in humans.

Empirical evidence for the TTT hypothesis was reviewed by Tapp and colleagues (2011). The results of more than 25 studies using opposite-sex twins were examined for three domains: behavioral, cognitive and physiological/morphological traits. With respect to behavior, evidence for the TTT hypothesis is scarce. Aggression and toy preference, for instance, do not seem to be more ‘masculine’ in girls with a twin brother than girls with a twin sister. On the other hand, eating disorders (more prevalent in girls) seem to be less frequent in girls with a twin brother, whereas autistic traits (more prevalent in boys) occur less in boys with a twin sister.

In further support of the TTT hypothesis, physiological/morphological characteristics such as otoacoustic emissions (sounds produced by the inner ear either in response to a sound or in the absence of any stimulus (F>M)), tooth size (M>F) and brain size (M>F) also seem ‘masculinized’ through the intrauterine presence of a brother.

Interestingly, the TTT hypothesis holds not only for girls gestated with a brother: boys gestated together with another twin brother are also more masculinized than boys with a twin sister. However, the effects of gestation with a male co-twin are more pronounced in females than in males, possibly because females produce little testosterone themselves and therefore may be more susceptible to external testosterone.

The TTT hypothesis has been criticized because social factors such as growing up/playing together with a brother masculinize behavior as well and, in fact, have little to do with a possible prenatal hormone transfer. In an attempt to distinguish prenatal from postnatal (social) factors, some studies also include an additional singleton brother in the analyses. The results of these studies indicate that the masculinizing effects of a male co-twin can exceed the influence of the singleton brother, suggesting more pronounced prenatal than postnatal effects. Finally, the question whether the effects are only observable during early development and may be ‘overwritten’ by hormonal effects during puberty remains an important issue that needs to be addressed in future studies.

In sum, studying opposite-sex twins seems an interesting approach to studying the effects of prenatal testosterone on brain development, behavior, and psychopathology.