Crespi & Badcock's (C&B's) novel and bold theory places autism spectrum disorders (ASD) and psychotic-spectrum disorders (PSD) on diametrically opposite ends of a spectrum, with underdeveloped social cognition in ASD and hyper-developed social cognition in PSD. The authors represent schizophrenia at the extreme end of PSD and discuss their genetic imprinting hypothesis of hyper-mentalizing in PSD mainly in the context of schizophrenia. However, although the authors acknowledge the multifaceted nature of the disorder, they fail to adequately consider it in the construction of their theory.
One underlying problem with placing schizophrenia and ASD at extreme poles is that the major symptoms of schizophrenia do not cohere with this method of distinction. Schizophrenia is characterized by positive symptoms such as delusions and hallucinations, and negative symptoms that reflect a diminution or loss of normal functions such as alogia, anergia, and anhedonia. Positive symptoms fluctuate over time and can be ameliorated by antipsychotic medication, but negative symptoms are treatment-resistant and enduring. C&B's theory focuses on schizophrenia to represent the psychotic spectrum, yet the more trait-like negative symptoms are largely ignored.
The process of hyper-mentalizing, which results in increased suspiciousness, is best related to one component of positive symptoms, namely, paranoia and persecutory delusions. Although it is a theoretically reasonable argument that hyper-mentalizing is related to paranoid delusions, it is difficult to make the same argument for the negative symptoms. Thus the borders of C&B's autistic-psychotic continuum must be redrawn.
It is possible to constrain C&B's theory to address the positive rather than negative symptoms, but empirical data pose further complications. Here we focus on one aspect of social cognition, theory of mind (ToM), which is the ability to attribute mental states to oneself and others, and we consider the nature of ToM deficits in schizophrenia and the relationship between these deficits and clinical symptoms. An abundance of empirical evidence points to mentalizing deficits in both PSD (i.e., schizophrenia, schizotypy) and ASD. C&B propose that the underlying deficit determining impaired social cognition in autism is an inability to theorize about the mental states of others (hypo-mentalizing). Indeed, evidence for ToM deficits in autism is robust (Baron-Cohen Reference Baron-Cohen1995). In the case of schizophrenia, there is also good evidence for the presence of ToM deficit (for review, see Brüne Reference Brüne2005), but very little evidence exists for the hypothesis that schizophrenia involves hyper-mentalizing about the mental states of others. Moreover, ToM deficits in schizophrenia may stem from executive or planning deficits (Hardy-Baylé Reference Hardy-Baylé1994) or an inability to integrate contextual information (Hardy-Baylé et al. Reference Hardy-Baylé, Sarfati and Passerieux2003). Thus, both ASD and PSD involve impaired social cognition, but the reasons for these deficits may not lie on the same continuum of mentalizing ability.
With respect to the positive symptoms, persecutory delusions may arise when individuals mislabel actions of others (Frith Reference Frith1992), and therefore one may expect a correlation between persecutory delusions and ToM deficits. However, evidence for this relationship is inconclusive at best, in contrast to the more consistent evidence of a relationship between negative symptoms and ToM deficits (Freeman Reference Freeman2007). C&B point out that psychotic symptoms are not just present in schizophrenia, but also present in other conditions, including depressive psychosis, bipolar disorder, posttraumatic stress disorder (PTSD), paranoid personality disorder, dementia, and epilepsy. Persecutory delusions are one of the most common features of psychosis, occurring in almost 50% of cases (Sartorius et al. 1986). If paranoid ideation and persecutory delusions are the components of psychosis most directly linked to hyper-mentalizing, then there should be an association between these experiences and ToM deficits. However, Walston et al. Reference Walston, Blennerhassett and Charlton(2000) found no evidence for ToM deficits in participants with pure persecutory delusions.
Furthermore, C&B argue that non-pathological points on the psychosis spectrum are most useful in evaluating their hypothesis. However, even in healthy schizotypal individuals, those who score high on psychometric measures of schizotypal personality, ToM deficits were related to the negative syndrome and not with the positive syndrome (Langdon & Coltheart Reference Langdon and Coltheart1999). With respect to the non-pathological individuals on the spectrum, there is an additional complication. Machiavellian intelligence, that is, social intelligence used as a means of social manipulation in order to attain goals, can improve social cognitive functioning when applied appropriately in context. Therefore, controlled and intentional hyper-mentalizing, even if extreme, should correlate with enhanced ToM and effective social behavior. In other words, is there room for Sherlock Holmes in C&B's spectrum?
Moreover, whether mentalizing impairments in PSD are state or trait features is still inconclusive. Although, generally, the data lend more support to a trait hypothesis, it is impossible to confirm without longitudinal studies (Harrington et al. Reference Harrington, Siegert and McClure2005b). This is perhaps further evidence for the importance of negative symptoms for ToM deficits in schizophrenia, given that negative symptoms are a more stable feature of the illness than positive symptoms, which vary over the course of the illness and with treatment.
To conclude, we believe there is not a strong enough empirical platform to assert impaired mentalizing, and especially hyper-mentalizing, as a phenotypic marker of PSD, which C&B's subsequent genetic imprinting theory requires. Evidence regarding mentalizing deficits being a state or trait marker of PSD is mixed, and it is equally unclear whether impaired task performance in PSD is caused by deficits in mentalizing or other cognitive problems. We argue that autism and schizophrenia cannot easily be conceptualized as polar opposites on a spectrum or continuum of mentalizing ability without changing the nature of this spectrum. However, with some modifications, C&B's theory lends itself readily to empirical testing and hence provides a valuable framework for future research.
