A Preliminary Structural Magnetic Resonance Imaging Study of Individuals with Schizotypal and Autistic Traits

Background: Although autism and schizophrenia are usually readily distinguished on the basis of severe impairments in communication skills in the former and the presence of marked positive symptoms in the latter, there exist spectrum forms of these conditions in which this differentiation is less obvious.  Individuals with either autism spectrum or schizophrenia spectrum conditions may display social difficulties, bizarre behaviour and idiosyncratic language, as well as a wide range of psychopathology including anxiety, obsessive-compulsive phenomena and mild positive symptoms.  As a result, approximately one third of people with an autism spectrum disorder (ASD) meet diagnostic criteria for schizotypal personality disorder and in such cases the primary diagnosis may be unclear.  Indeed it is possible that it is incorrect to even regard the spectrums as distinct entities. 

Objectives: To identify brain structural features which differ between individuals with schizotypal traits and those with autistic traits. 
To examine brain structure in individuals with comorbid schizotypal and autistic traits in order to determine whether it is possible to assign this group to a primary diagnosis. 

Methods: Individuals were recruited as part of larger study into the mental health of adolescents receiving educational assistance.  Only male participants are considered here as there were insufficient female participants with prominent autistic traits to examine the effects of gender.  All participants received the Structured Interview for Schizotypy (SIS) and a structural magnetic resonance imaging scan.  In addition, a parent of the child completed the Social Communication Questionnaire (SCQ) along with an investigator.  Participants were divided into four groups using cut-offs on the SCQ and the SIS – ‘schizotypal’, ‘ASD’, ‘comorbid’ and ‘neither’.  Volumes of the cerebral lobes and the cerebellum were derived using automated methods derived in our laboratory and validated against hand tracing (all ICCs > 0.9).  Between group differences in volumes were determined using analysis of covariance, with total brain volume as a covariate.

Results: 85 participants were recruited (age range 13-22; IQ range 40-131; SIS range 10-66; SCQ range 0-33).  There were no differences in age or IQ between the four groups.  The main effect of the ANCOVA revealed a trend towards a difference in left temporal lobe volume between groups (F=2.28, p=0.09).  Between group t-tests showed that the mean left temporal lobe volume in the ASD group was significantly smaller than in the schizotypal group (p=0.05) and the comorbid group (p=0.01).  There were no significant volume differences between the comorbid and the schizotypal groups. 

Conclusions: Although preliminary these data indicate that it is possible to differentiate individuals with schizotypal traits from those with autistic traits using brain anatomical measures, supporting the conceptualisation of the schizophrenia and autism spectrums as distinct entities.  In addition, individuals with comorbid schizotypal and autistic traits may be more correctly classified as schizotypal.  Ongoing work will further investigate these findings using more clearly defined groups.