At the end of his book The Essential Difference, which argues that autism and autistic-spectrum syndrome may result from an “extreme male brain,” Simon Baron-Cohen asks, “Is there an extreme female brain?” (Baron-Cohen Reference Baron-Cohen2003, pp. 170–85). Crespi & Badcock (C&B) have now answered this question. Contrary to Baron-Cohen, who speculates that, if there is an extreme female brain, it would not be pathological, C&B suggest, with a considerable amount of supportive evidence, that an extreme female brain, the polar opposite of autism, may amount to schizophrenia and psychotic-spectrum syndrome.

If C&B are correct, which all available evidence strongly suggests they may be, then their intragenomic conflict theory of the social brain immediately suggests a couple of novel hypotheses. First, recent evolutionary psychological theories (Atran Reference Atran2002; Boyer Reference Boyer2001; Guthrie Reference Guthrie1993; Kirkpatrick Reference Kirkpatrick2005; Miller & Kanazawa Reference Miller and Kanazawa2007, pp. 158–61) suggest that religion is not an adaptation in itself but a by-product of other psychological adaptations, variously named “agency-detection mechanism” or “animistic bias.” We may be evolutionarily designed to infer personal and animistic intentions behind natural phenomena, because the consequence of erroneously overinferring intentions – being paranoid about perfectly natural phenomena – is much less costly in evolutionary terms than the consequence of erroneously underinferring intentions – being killed by predators and enemies when we least expect them (Haselton & Nettle Reference Haselton and Nettle2006). We may therefore be designed to be paranoid because it can potentially save our lives, and we may be religious because we are paranoid and see the “hands of God” behind completely natural phenomena.

If this view is correct, then C&B's theory suggests that schizophrenics, who are hypermentalistic and are more “paranoid,” may be predisposed to be more religious, more prone to see the hands of God behind natural phenomena, just as some (McNamara Reference McNamara and Andresen2001; Miller & Kanazawa Reference Miller and Kanazawa2007, p. 206, n13) suggest that autistics should be less religious because of their hypomentalism (absence of theory of mind). C&B identify “overestimation of meaningfulness of naturally occurring coincidences” (sect. 6.1.3, para. 2) as one of the symptoms of schizophrenia.

In virtually all nations of the world, women are more religious than men. While Alan S. Miller and I (Miller & Kanazawa Reference Miller and Kanazawa2007, pp. 161–65; cf. Miller & Stark Reference Miller and Stark2002) explain this in terms of women's greater tendency toward risk aversion, C&B's theory suggests another explanation: If religion is mentalizing natural phenomena, and if the female brain tends toward mentalizing, then women should naturally be more religious.

Second, in an entirely different area, the generalized Trivers-Willard hypothesis (gTWH) (Kanazawa Reference Kanazawa2005) proposes that parents who possess any heritable trait which increases male reproductive success at a greater rate (or decreases male reproductive success at a smaller rate) than female reproductive success in a given environment, will have higher-than-expected offspring sex ratios (more sons). Conversely, parents who possess any heritable trait which increases female reproductive success at a greater rate (or decreases female reproductive success at a smaller rate) than male reproductive success in a given environment, will have lower-than-expected offspring sex ratios (more daughters). Because body size and tendency toward violence are distinct advantages in male intrasexual competition for mates and status, big and tall parents are more likely to have sons (Kanazawa Reference Kanazawa2005; Reference Kanazawa2007b), and violent men are more likely to have sons (Kanazawa Reference Kanazawa2006). Because physical attractiveness, while an advantage for both men and women, is even a greater advantage for women than for men, more beautiful parents are more likely to have daughters (Kanazawa Reference Kanazawa2007a). Because language and communication are more important to women, and thus language impairment is relatively more costly for them, mothers (though not fathers) with a developmental language impairment have more sons than daughters (Tallal et al. Reference Tallal, Ross and Curtiss1989).

I have applied the gTWH to brain types and shown that those with “strong male brains,” such as engineers, mathematicians, and scientists, are more likely to have sons, because the male brain is more adaptive for boys. In contrast, those with “strong female brains,” such as nurses, social workers, and kindergarten teachers, are more likely to have daughters, because the female brain is more adaptive for girls (Kanazawa & Vandermassen Reference Kanazawa and Vandermassen2005). If autism and autistic-spectrum syndrome are the result of “extreme male brains,” as Baron-Cohen (Reference Baron-Cohen2003) suggests, then autistics should have more sons. Conversely, if schizophrenia and psychotic-spectrum syndrome are the result of “extreme female brains,” or hyper-mentalism, as C&B propose, then they should have more daughters. Although C&B cite a large number of studies to demonstrate that first-degree relatives of schizophrenics have more children (sect. 6.4), they do not reveal (nor do I know) whether schizophrenics and their relatives have lower offspring sex ratios (more daughters). This prediction assumes that the genetic tendency toward schizophrenia is heritable, which is reasonable, given that the genetic tendency toward autism appears heritable (Baron-Cohen et al. Reference Baron-Cohen, Bolton, Wheelwright, Scahill, Short, Mead and Smith1998).

There is one potential confound, however. If lighter birth weight and slower postnatal growth of schizophrenics, and heavier birth weight and faster postnatal growth of autistics, which C&B document, continue throughout development, then autistics on average should have larger adult stature than schizophrenics. If this is the case, then autistics should be expected to have more sons, and schizophrenics more daughters, purely because of their body size (Kanazawa Reference Kanazawa2005; Reference Kanazawa2007b). It would therefore be necessary to control for body size in order to determine the partial effect of autism and schizophrenia on offspring sex ratios. While, given their low frequency in the population, it would be difficult to detect the effect of autism and schizophrenia on offspring sex ratios in representative samples, one should be able to determine, with clinical samples of autistics and schizophrenics with children, whether their offspring sex ratios significantly deviate from the normal secondary sex ratio of .5122 (105 boys for 100 girls).