Adolescent and Childhood Autism Spectrum Quotient
Bonnie Auyeung, Carrie Allison, Sally Wheelwright, Rosa Hoekstra, Simon Baron-Cohen
Given the utility of the Adult-AQ we have adapted the AQ for use with two age-groups: adolescents (ages 12-16) and children (ages 4-11) in the Adol-AQ and Child-AQ respectively. We have found that these produce very similar distributions to the Adult-AQ and are exploring their use in a range of populations, as useful ways to test predictions about who might be expected to have fewer or greater number of autistic traits.
We are currently testing the relationship between the Child-AQ and the CAST (Childhood Autism Spectrum Test), since the latter has been used in a range of studies (including epidemiology). Current evidence suggests the Child-AQ has a more normal distribution than the CAST, which is skewed. As such, the Child-AQ may be more useful in quantitative studies of the broader autism phenotype.
Planned studies aim to test if the AQ (Adult-, Adol- or Child-versions) can be adapted for the low-functioning subgroup with autism, since most of the existing studies have focused on the high-functioning subgroup. Having a quantitative metric of autistic traits across the IQ range would be a useful way of comparing individuals in terms of degree of severity.
As with the Adult-AQ, the Adol- and Child-AQs are also useful in association studies with biomarkers. We have recently found for example that scores on the Child-AQ are positively correlated with levels of foetal testosterone.