Autism and schizophrenia (more precisely, the negative syndrome of schizophrenia as observed in early-onset schizophrenia), far from being diametrical disorders of the social brain as postulated by Crespi & Badcock (C&B), share common psychopathological dimensions in terms of social communication deficits. This does not contradict C&B's hypothesis that autism and schizophrenia (but the latter in its paranoid form) represent two extremes on a cognitive spectrum.

Historical diagnostic overlaps between autism and schizophrenia have existed since the appearance of autism in the nomenclature of psychiatric disorders. Originally autistic disorder and schizophrenia were intimately linked. Indeed, the term autism is derived from the Greek autos, which means self and was introduced for the first time by the Swiss psychiatrist Eugen Bleuler in 1911 to describe social withdrawal in adults with schizophrenia. In 1943, the American psychiatrist Leo Kanner borrowed from Bleuler the term autism to define a syndrome observed in 11 children – which, at that time, was a part of the diagnostic category “Childhood Schizophrenia.” Since the early 1970s, autism and schizophrenia have belonged to two different diagnostic categories. However, they still share some common features: They are both considered as developmental psychiatric disorders involving psychotic symptoms with impairments in the same main behavioral domains (especially in communication and social interactions). Communication impairments are reported in autistic disorder as well as in early-onset schizophrenia (Alaghband-Rad et al. Reference Alaghband-Rad, McKenna and Gordon1995; Asarnow et al. Reference Asarnow, Tompson and Goldstein1994; Baum et al. 1995; Cantor et al. Reference Cantor, Evans, Pearce and Pezzot-Pearce1982). Thus, impairments in verbal communication (delay in the development of spoken language, poor or disorganized speech) and nonverbal communication (reduced facial expression or body language; poor eye contact; and abnormal emotional expression such as flat, bizarre, or inappropriate affects) are found in individuals with early-onset schizophrenia or autistic disorder. In addition, social development is impaired in both schizophrenia and autistic disorder. The degradation of social skills is a hallmark of schizophrenia (Mueser et al. Reference Mueser, Bellack, Douglas and Morrison1991) and the distinguishing feature of subgroups within schizophrenia (Carpenter et al. Reference Carpenter, Heinrichs and Wagman1988). Deterioration of social skills is associated with the chronic phase of schizophrenia and its deficit form (Carpenter et al. Reference Carpenter, Heinrichs and Wagman1988) or negative syndrome. Social isolation observed in schizophrenia, in particular in the childhood of individuals with early-onset schizophrenia, is similar to the autistic withdrawal described by Kanner Reference Kanner(1943).

Tantam Reference Tantam(1988) even argued that individuals with autistic disorder display symptoms that can also be considered as schizoid personality traits. More recently, Konstantareas and Hewitt Reference Konstantareas and Hewitt(2001) reported that none of 14 men with paranoid schizophrenia met criteria for autism, whereas 7 out of 14 males with autism met criteria for schizophrenia on the Structured Clinical Interview (SCID). More precisely, symptom overlap concerned negative symptoms of schizophrenia, such as affective flattening, alogia, avolition, apathy, anhedonia, and poor communication. This symptom overlap can be observed in particular between high functioning autistic disorder or Asperger's syndrome and early-onset schizophrenia, which tends to be characterized by negative symptoms with social interaction impairments (Bailer et al. Reference Bailer, Brauer and Rey1996) and a chronic course (Krauss et al. Reference Krauss, Marwinski, Schulze, Mueller, Held, Rietschel, Maier and Freyberger2000). Furthermore, retrospective studies conducted on patients with schizophrenia or longitudinal studies of children with autistic disorder reported frequent associations between these two disorders (antecedents of autistic disorder in the childhood of the schizophrenic group and appearance of schizophrenia in the autistic group) (Alaghband-Rad et al. Reference Alaghband-Rad, McKenna and Gordon1995; Bender & Faetra Reference Bender, Faetra and Kaplan1972; Jansen et al. Reference Jansen, Gispen-De Wied, Van der Gaag, Ten Hove, Willemsenswinkels, Harteveld and van Engeland2000; Petty et al. Reference Petty, Ornitz, Michelman and Zimmerman1984; Van Engeland & Van Der Gaag Reference Van Engleand and Van Der Gaag1994).

From a biological point of view, abnormal stress responses (in particular responses of the hypothalomo-pituitary-adrenal axis to psychosocial stress) are reported in schizophrenia (Jansen Reference Jansen1998; Jansen et al. Reference Jansen, Gispen-De Wied, Van der Gaag, Ten Hove, Willemsenswinkels, Harteveld and van Engeland2000) as well as in autistic disorder (for a review, see Tordjman et al. Reference Tordjman, Anderson, Macbride, Hetzig, Snow, Hall, Ferrari and Cohen1997). This common biological feature is associated in both disorders with major anguish focused often on body image, difficulties to adapt to novel situations, and abnormal behavioral responses to environmental stimuli and stressful situations (Jacobson & Ackerman Reference Jacobson and Ackerman1990; Tordjman et al. Reference Tordjman, Anderson, Macbride, Hetzig, Snow, Hall, Ferrari and Cohen1997; Van Den Bosch et al. Reference Van Den Bosch, Van Asma, Rombouts and Louwerens1992; Wiedl Reference Wiedl1992). Interestingly, some studies suggest that the apparent pain insensitivity observed in both autism and schizophrenia is more related to a different mode of pain expression due to social communication impairments and body image problems, than to a real endogenous analgesia (Guieu et al. Reference Guieu, Samuelian and Coulouvrat1994; Tordjman et al. Reference Tordjman, Antoine, Cohen, Gauvain-Piquard, Carlier, Roubertoux and Ferrari1999; Watson et al. Reference Watson, Akil, Berger and Barchas1979).

Finally, genetic factors possibly involved in both early-onset schizophrenia and autistic disorder have been discussed by Yan et al. Reference Yan, Guan, Green, Nicolson, Yap, Zhang, Jacobsen, Krasnewich, Kumra, Lenane, Gochman, Damschroder-Williams, Esterling, Long, Martin, Sitransky, Rapoport and Ginns(2000). The serotonin transporter gene (HTT) is of particular interest given the reported association of a HTT promoter polymorphism with anxiety-related traits in the general population (Lesch et al. Reference Lesch, Bengel, Heils, Sabol, Greenberg, Petri, Benjamin, Müller, Hamer and Murphy1996). Furthermore, in a genetic study conducted on families of 71 children with autism, we showed that the HTT promoter polymorphism modifies the behavioral phenotypic expression of autism: the data consistently pointed to the short (s) allele being associated with the more severe impairments in the communication and social interaction domains (Tordjman et al. Reference Tordjman, Gutknecht, Carlier, Spitz, Antoine, Slama, Carsalade, Cohen, Ferrari, Roubertoux and Anderson2001). These data were confirmed by another study (Brune et al. Reference Brune, Kim, Salt, Leventhal, Lord and Cook2006). Of special relevance are reports that the HTT promoter polymorphism also influences the severity of hallucinations in schizophrenia: Intensity of hallucinations was positively associated with the long (l) allele (Malhotra et al. Reference Malhotra, Goldman, Mazzanti, Clifton, Breier and Pickard1998) and negatively associated with the s allele (Golimbet et al. Reference Golimbet, Alfimova, Shcherbatikh, Kaleda, Abramova and Rogaev2003). It suggests that the HTT promoter polymorphism influences the phenotypic expression of positive symptoms (hallucinations) and negative symptoms (social withdrawal and communication deficits). In this regard, our results fit within C&B's hypothesis: Autism and the deficit form of schizophrenia (observed in particular in early-onset schizophrenia) would be at one extreme (negative symptoms), and paranoid schizophrenia at the other extreme (positive symptoms) of this continuum. It opens the path to future common research on the genetics of autism and schizophrenia, and underlines the importance of using a methodology in which genetic transmission is studied in concert with a detailed examination of clinical phenotype.

In conclusion, I suggest that future research on the genetics of autism and schizophrenia should develop multi-trait models (Tordjman et al. Reference Tordjman, Drapier, Bonnot, Graignic, Fortes, Cohen, Millet, Laurent and Roubertoux2007) and a multidimensional approach, such as studying social and communication impairments, instead of focusing on diagnostic categories.