Quantifying behavior in children with Autism

Despite its unitary definition, Autism is an incredibly heterogeneous disorder such that different children with ASD exhibit a variety of core and secondary symptoms. The underlying biology of autism is also incredibly heterogeneous such that >300 distinct genetic abnormalities raise the risk of developing the disorder. This suggests that there are multiple underlying mechanisms that cause multiple types of Autism.

How can we identify different types of Autism and determine optimal interventions for each? Over the last 5 years, in a unique clinical-research collaboration, we have been building an extensive database containing clinical, behavioral, and biological information from over 1800 children with autism and their families. The database includes multiple behavioral measures (ADOS, IQ, language, adaptive behaviors, sensory sensitivities, etc…), whole exome sequencing of children with autism and both parents, anatomical MRI scans, eye tracking recordings, video and audio recordings, and EEG recordings during sleep.

In the talk I will focus on a series of projects focused on developing quantitative, objective measures of behaviors that are relevant to young children with ASD. Examples will include the use of eye tracking, motion tracking, speech analysis, and analysis of facial expressions as well as research regarding sleep disturbances. I will demonstrate how a combination of these measures may be used to quantify symptoms in different children in ways that will hopefully enable more informative research regarding their underlying neurophysiology and the development of personalized medicine approaches.