Fifty percent of children with autism spectrum disorder (ASD) do not respond to classic behavioral interventions. Why? We think it is because we are not considering the many differences in how these children learn. In previous studies, some individuals with ASD have had difficulty learning implicitly (without instructions) compared to typically developing individuals, while other studies have shown no such distinction. What explains these inconsistent findings? We predict that some, but not all children with ASD have difficulties learning without instructions, and we propose to use multiple types of analyses to test what factors contribute to this implicit learning. We can then use this information to predict who will respond to treatment and to customize behavioral interventions to each child's strengths and weaknesses. The first aim of the proposal is to build upon the broad learning literature and use a specifically designed non-verbal task on the iPAD with stimuli that children find fun and engaging. We will test (1) high and low cognitively able children with ASD (ages 3-7) and (2) children with Intellectual Developmental Disorder (IDD) without ASD (ages 3-7). We predict that children who are faster at learning patterns in the task will have higher social communication skills and language abilities. We will determine whether there are implicit learning differences that are unique to ASD by comparing the children. The second aim of the proposal is to collect data on a group of more cognitively able children with ASD (ages 5-7) who will perform the learning task during functional magnetic resonance imaging (fMRI). This will create a baseline for the fundamental neural learning circuitry in children with ASD. We hope these data, together with the behavioral study discussed above, will provide the foundation for future work that links implicit learning abilities with neural biomarkers. Identifying such biomarkers could lead to the potential for earlier intervention and also possible pharmaceutical targets. The third aim of the proposal is to collect data on a group of children who receive intensive behavioral intervention and determine whether faster implicit learning abilities predict a child's improvement in social communication skills. These data will provide an understanding of how to identify children who may respond better and more quickly to treatment. Thus, our findings will enable us to target children who may require a tailored intervention strategy and understand why some children respond better to treatment initially. This information would benefit therapists and clinicians working with children to optimize treatment planning. Our approach goes beyond the current methodological and analytical framework for studying ASD. First, testing on an iPAD deviates from conventional methods and will increase the likelihood of appealing to children who have lower cognitive abilities. Second, unlike prior studies that focused on more cognitively able, older individuals with ASD, we will test less cognitively able young children with ASD (ages 3-7) in order to target a population who have not been included in most studies and who may benefit most from an improved understanding of response to behavioral intervention. Determining individual differences in learning abilities in order to target children who require specific behavioral interventions in ASD will have significant ramifications for improving health and educational outcomes. The goal is to have a much larger percentage of children with ASD respond to behavioral intervention. This proposal brings together a team of clinical and pediatric neuroimaging experts to test whether individual learning differences is the key to increasing the success rates of treatment outcomes. The application addresses Fiscal Year 2013 autism Research Program Area "Understanding factors underlying the heterogeneity of response to treatment of ASD."
Interagency Autism Coordinating Committee (IACC)
Autism Research Database (AFD)
