Using Connectivity to Explain Neuropsychiatric Conditions: The Example of Autism

Abstract

The chapters in this Handbook reveal the breadth of brilliant imaging and analysis techniques designed to fulfill the mandate of cognitive neuroscience: to understand how anatomical structures and physiological processes in the brain cause typical and atypical behavior. Yet merely producing data from the latest imaging method is insufficient to truly achieve this goal. We also need a mental toolbox that contains methods of inference that allow us to derive true scientific explanation from these data. Causal inference is not easy as regards human brain function, where we are limited primarily to observational data, and our methods of experimental perturbation in the service of causal explanation are limited. As a case study, we reverse engineer one of the most influential accounts of a neuropsychiatric disorder that is derived from observational imaging data: the connectivity theories of autism spectrum disorder (ASD). We take readers through an approach of first considering all possible causal paths that are allowed by preliminary imaging-behavioral correlations. By progressively sharpening the specificity of the measures and brain/behavioral constructs, we iteratively chip away at this space of allowable causal paths, like the sculptor chipping away the excess marble to reveal the statue. To assist in this process, we consider how current imaging methods that are lumped together under the term “connectivity” may actually offer a differentiated set of connectivity constructs that can more specifically relate notions of information transmission in the mind to the physiology of the brain.