Clinically, the target article of Crespi & Badcock (C&B) does have a ring of truth. However, their hypothesis – that the development of autistic- and psychotic-spectrum conditions is mediated in part by alterations in “genomic imprinting” – does not explain or accommodate some potent molecular and genetic considerations. There are two main points to consider:
1. Both of the conditions (autism and psychosis) are highly heterogeneous (as C&B point out), involving many complex traits. It is unlikely that all these traits are the result of disruption to either maternal or paternal gene imprinting processes, as the number of the imprinted genes is very small and does not account for more than 1% of the genome at the best estimate. Further, if the Prader-Willi and Angelman syndromes form the central dogma of the hypothesis, then how does a disruption of the imprinting process in the same set of genes result in two conditions with divergent age of onsets – early in the case of autism and at adolescence or around the mid-twenties in the case of psychosis?
2. As yet there is no solid molecular evidence (linkage or association) to implicate imprinting or disruption of an imprinting process in either autism or psychosis. If, however, identified in the future, they may have a minor contribution to the whole process of developing the above conditions, as both are polygenic disorders with several genes (each of minor effect) contributing to their development.
C&B's key table (Table 1) is very persuasive in showing the differences between autism and psychosis, and this forms the central hypothesis of the article. However, it is frequently reported in the literature that features of neurodevelopmental disorders are non-specific, and that overlaps in the nosology of neurodevelopmental disorders exist. This is frequently reported in relation to autism and attention-deficit/hyperactivity disorder (ADHD) with attentional difficulties commonly occurring in autism. One of the authors of this commentary (Fitzgerald) has observed an overlap between elective mutism and autism-spectrum disorders as suggested in Figure 1 of the target article. Further, individuals with high functioning autism may show fixed and rigid beliefs reminiscent of delusions observed in psychosis. Moreover, creativity is frequently observed in the context of psychosis, particularly in bipolar illness. Although creativity in schizophrenia diminishes after the onset of the condition, family members with less severe traits of the condition can show great creativity, reflecting perhaps broader phenotype features. Asperger's syndrome is also associated with creativity; and although great creativity in old age is uncommon, the lack of decrease of gray matter volume with aging (McAlonan et al. Reference McAlonan, Daly, Kumari, Critchley, van Amelsvoort, Suckling, Simmons, Sigmundsson, Greenwood, Russell, Schmitz, Happé, Howlin and Murphy2002) in persons with Asperger's syndrome may explain the old age creativity sometimes seen with persons with Asperger's syndrome. There are small subgroups of persons with autism-spectrum disorders who are fearless. This group has been given insufficient attention in the target article. In addition, all novel stimuli are not avoided in autism, as there is an autistic type of novelty seeking which is partly responsible for their great creativity.
Finally, at times, it appears that the separation between autism and schizophrenia is too neat in C&B's article.
