The single-neuronal basis and causal underpinnings of joint social behavior are still almost completely unknown, which directly affect our ability to understand and treat disorders with social impairments such as those seen in ASD. Keren Haroush plans to examine the neural and population basis of interactive social behavior and its modulation in a nonhuman primate model using neural recordings and deep brain stimulation.

John Huguenard has preliminary data to suggest that neuronal axons don't develop normally in two mouse models of ASD that result from maternal immune activation. In the current proposal, Huguenard will test if this is a common feature across ASD mouse models, by assessing axonal structure and function in four additional environmental and genetic models.

Each individual with ASD is unique, and understanding this variability is essential for both biomarker and treatment development. Joshua Hartshorne and Stefano Anzellotti plan to use Bayesian nonparametric techniques to analyze functional brain imaging data from individuals with ASD and related neurodevelopmental disorders. The goal of the study is to identify an optimal model that captures variability and clusters the data based on either subtypes of ASD or a set of dimensions; such a subclassification system will be useful for tailoring individual-specific interventions.