> No, one is a form of natural experiment (changing the hormone profile of a single person who can compare the causal effects). But this is subject to placebo

> No, one is a form of natural experiment (changing the hormone profile of a single person who can compare the causal effects). But this is subject to placebo effect, as the OP said, so you can't directly infer a casual influence on behaviour. > The other is an assumption that someone's observed characteristics are caused by one particular hormone level, with no evidence to support that. The OP did not say the lack of maternal instincts was caused by lower than normal testosterone, they simply noted that she had both characteristics.