Increased testosterone may be the cause of the finding of Sundquist et al (Reference Sundquist, Frank and Sundquist2004) that ‘A high level of urbanisation is associated with increased risk of psychosis and depression for both women and men’. Two hypotheses are required to explain this.

It is my hypothesis (Reference HowardHoward, 2001a ) that human evolution is driven by testosterone. Based on this, I suggest the ‘secular trend’, the increase in size and early puberty of children, is actually an increase in the percentage of individuals of higher testosterone. The trend may actually be a change in percentage of individuals within our populations and their ‘characteristics’ may also be increasing. This phenomenon occurs when a ‘feed and breed’ environment occurs. In these situations, individuals of higher testosterone, both men and women, will increase more rapidly than those of lower testosterone over time. (Individuals of higher testosterone are more aggressive and sexual.) Urban areas are ‘feed and breed’ centres; I suggest urban centres are areas of higher testosterone.

I hypothesise that dehydroepiandrosterone (DHEA) is directly involved in growth and development, and subsequent maintenance, of all tissues, especially the brain. (The large brain of mammals may have resulted from an evolutionary increase in DHEA; Reference HowardHoward, 2001b .) Numerous reports of beneficial effects of DHEA on neurons and tissue-level structures of the brain exist in the literature. I have suggested in the past that depression and schizophrenia, among other mental disorders, result from low DHEA during growth and development, subsequently exposed by adverse circumstances during maintenance. In depression and schizophrenia DHEA is low. Two other hormones may adversely affect the function or availability of DHEA: cortisol and testosterone. Over the past few years a connection with low DHEA, along with increased cortisol, has been demonstrated regarding depression. It is known that schizophrenia is often characterised as resulting from a non-causal, but significant, stressful event (cortisol) usually beginning in the late teens or early twenties (testosterone of puberty, in men and women, along with the natural decline of DHEA which begins at around age 20). In individuals of low DHEA, increased cortisol and testosterone may expose underlying, silent pathology.

Therefore, I suggest that increased rates of psychoses and depression in urban areas may be the product of increased stress and testosterone in both men and women. As suggested above, the secular trend may be due to increasing numbers of individuals of higher testosterone. This increase in these individuals of higher testosterone, along with increasing stress of urbanisation, may account for the findings of Sundquist et al, as well as reports of recent increases in these mental disorders.