Crespi & Badcock's (C&B's) magisterial article sets up a host of important and novel hypotheses about the phenotypes of psychosis and autism, the genetic and developmental mechanisms underlying these pathologies, normal variation in milder psychosis-like and autistic-like traits, and sex and parent-of-origin differences in all of the above. Here, I wish to focus briefly on possible implications of their thesis for one related area: the role of artistic creativity in male sexual attractiveness.
A number of lines of evidence suggest that artistic creativity is found sexually attractive. Nettle and Clegg (Reference Nettle and Clegg2006) found that poets and artists had more sexual partners than community controls, a finding repeated for males in a different population of visual artists in Clegg's (Reference Clegg2006) doctoral study, which additionally showed that the increased number of partners stemmed from a more short-term mating strategy. Haselton and Miller (Reference Haselton and Miller2006) devised vignettes of poor but creative versus rich but uncreative men. Women in the fertile phase of their menstrual cycle shifted preference towards the poor but creative man for a potential short-term mate (long-term preferences were unaffected). Such menstrual cycle shifts are thought to maximise genetic quality at the expense of, say, propensity to provide parental investment. Finally, Griskevicius et al. (Reference Griskevicius, Cialdini and Kenrick2006) found that whenever men were given cues designed to activate mating goals, they displayed more creativity in a number of tasks. This suggests a male response to general female interest in male creativity as a mate-selection indicator.
A standard account of these findings might be the following: Creativity is closely linked to general intelligence, and general intelligence is a consequence of the overall functional efficiency of the nervous system. This in turn will be affected by numerous developmental factors, including genetic mutational load. The substantial correlation of measured general intelligence to physical symmetry, long used by biologists as a marker of developmental quality, supports the relationship of general intelligence to mutational load (Prokosch et al. Reference Prokosch, Yeo and Miller2005). Thus, creativity is signalling intelligence which is signalling good overall genetic quality (Haselton & Miller Reference Haselton and Miller2006; Miller Reference Miller, Bock, Goode and Webb2000).
This account appears plausible enough, but there is an unresolved issue. Why is it particularly artistic-type creativity – namely, remote verbal and visual associations – which is found to be attractive? Mathematical, technical, or engineering ability could equally well signal high intelligence, and indeed, these kinds of capacities would have been very useful for survival in the ancestral environment. It is here that the connection to the autism-psychosis spectrum comes in. Artistically creative individuals tend to score highly on measures of psychosis-proneness, while mathematicians have unusually low scores, and if anything are more autistic-like in their psychological profiles (Baron-Cohen et al. Reference Baron-Cohen, Bolton, Wheelwright, Scahill, Short, Mead and Smith1998; Nettle Reference Nettle2006). Both groups have high general intelligence, but it appears to be the artistic types who are attractive for short-term matings and have increased number of partners (Nettle & Clegg Reference Nettle and Clegg2006). Hence, it appears that it is not just general intelligence which is a mate-choice indicator, but also the position on the spectrum from hyper-mentalising, psychosis-prone, to hypo-mentalising, autism-prone, with the hyper-mentalising end of the spectrum the more attractive one.
Two possible explanations spring to mind. One is that mildly hyper-mentalising individuals are better partners to have, because, being people people, they take into account the needs and desires of others. Thus, creative behaviour could simply be signalling one's type in terms of post-reproductive cooperativeness (see Nettle & Clegg Reference Nettle, Clegg, Geher and Miller2007 for some discussion). However, if so, then creative individuals should be especially preferred in the long-term mating context, whereas all the evidence suggests that highly creative individuals tend to follow short-term mating strategy and are preferred in the short-term, genetic-contribution-only context, rather than for long-term investment.
C&B's article suggests an intriguing alternative: hyper-mentalising, psychosis-prone males are signalling that their genotype is such that their babies will not make excessive demands on their mothers. C&B show that psychosis-proneness is associated with smaller, slower growing, less demanding babies, whereas autism-proneness is associated with the opposite. They also argue that the remote associations found in schizotypy and artistic creativity are a direct (and therefore unfakable) consequence of having a brain whose growth pattern is of the psychosis-prone – and therefore less costly for the mother – type.
This hypothesis is not implausible. Human reproduction makes extremely high metabolic demands on females. Rates of maternal death in pregnancy, and during or soon after birth, are as high as 2% per pregnancy in the poorest countries even today (World Health Organisation 2007), and would have been higher under ancestral conditions. The Haigian model of paternal–maternal conflict under multiple paternity predicts that paternal interests will tolerate a greater risk of harm or death to the mother than would be optimal from the point of view of the maternal genome. This tussle is partly carried out at the level of genetic imprints and counter-imprints, but can also be carried out in the arena of mate choice. If the female can find cues that discriminate males on the basis of how strongly their genotype is going to favour the patrilineal interest, she should exploit them. Such cues should be of particular interest in the short-term mating case, where genetic material is all that is being provided. In the long-term case, the additional costs of offspring of autistic-prone individuals could be offset by their propensity to make paternal behavioural investment.
The hypothesis, speculative as it is, could lead to testable predictions. Further investigation is needed into whether the young offspring of highly schizotypal or creative males are indeed smaller or less demanding than the babies of more autistic-prone individuals, and also of women's intuitions about the desirability of such men as partners. One prediction might be that women whose own somatic resources are limited (health problems, small size) should be especially keen on schizotypal-creative versus autism-prone males as partners. This would indicate that these traits are being used as mate-choice indicators for the reasons invoked.
It is a great testament to the breadth and importance of C&B's target article that it allows novel predictions to be generated about something so far removed from their initial concern.
