As always, a reminder that correlation != causation. This shows an association between hormone levels and autism, which is suggestive. But keep in mind that the

As always, a reminder that correlation != causation. This shows an association between hormone levels and autism, which is suggestive. But keep in mind that there may be a confounding factor or factors (e.g. the same genes that control synapses that are dysfunctional in autism may be involved in hormonal regulation). Biology is decidedly unlike good code - no modularity, many pleiotropic effects. Something more definitive would be examining pregnancies where oestrogens were artificially elevated, maybe by looking at women who were unknowingly pregnant but continued to take oestrogenic contraceptives.