Assessing the relationship between eating disorder psychopathology and autistic traits in a non-clinical adult population

Previous research demonstrates a genetic and behavioural link between eating disorders and autism spectrum disorders, and a recent study (Coombs et al. in Br J Clin Psychol 50:326-338, 2011) extends this link to typical populations, showing a positive correlation between behaviours in typically developing children. The purpose of the present study was to examine whether this relationship continues beyond development, by studying the link between behaviours in a non-clinical adult population. We examined associations between performance on measures relating to autistic traits and disordered eating. Undergraduate students, equally balanced by gender and by subject studied (i.e. humanity or science), completed three tasks: to measure autistic traits, participants were administered the Embedded Figures Test (EFT) and the Autism-Spectrum Quotient (AQ). Eating disorder symptomatology was measured by the Eating Attitudes Test (Eat-26). Our data revealed a significant positive correlation between scores on the AQ and Eat-26. Multiple linear regressions showed that higher scores on the AQ were particularly associated with higher scores on the Bulimia & Food Preoccupation subscale of the Eat-26. EFT performance was positively related to behaviours associated with autism and eating disorders, although not reliably so. These data support the broader link between autistic traits and disordered eating in the non-clinical population, and demonstrate that it extends into adulthood (a time at which autistic behaviours can decrease). This work carries implications for the development of cognitive therapies for people with eating disorders.