This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5). Why do some children in susceptible families become autistic when others do not? How can understanding what goes wrong in autistic brain development help us to understand what goes right in normal brain development? And how might this knowledge lead to interventions that maximize developmental potential in every child? Brain cells, like telephone networks, communicate on more than one scale: though most communications are local within a neighborhood of cells, some cover long distances between widely separated brain regions. In order for the correct long-distance connections to be made, the local connections have to be working properly. Although a small problem in local connections might not cause much disruption in the laying down of long-distance lines, larger local abnormalities may cause a chain reaction of larger disruptions in long-distance connectivity. This project evaluates the hypothesis that the way brain cells function together locally within brain regions may be genetic, whereas a secondary abnormality in long-distance connections between brain regions may be what emerges during brain development to make a person autistic. If this is the case, then by using behavioral therapies, drugs, and/or other future treatments it might become possible to drive a wedge between genes and outcome, preventing susceptible children from developing severe autism. To test this hypothesis of abnormal neural connectivity, this project combines magnetic resonance imaging (MRI) to show brain structure and high-density electroencephalography (EEG) to show brain function while children perform experiments that are embedded in the engaging environment of a video game. In this game-based format, children with autism (as well as non-autistic siblings and unrelated, unaffected children) take the experiments home on a laptop computer and learn at their own pace and in their own time, free from time constraints and other sources of anxiety. Having learned the game (and the experiments embedded within it), children are given the opportunity to come to the laboratory for EEG and MRI. Behavioral data from the game and anatomy and physiology from the laboratory are then combined to differentiate behavioral and neurobiological traits that differ categorically in people with autism from those that differ only in degree between autistic and non-autistic family members. In addition to answering scientific questions directly, this project trains undergraduate computer science students in the growing area of "serious games" -- that is, therapeutic computer games designed as vehicles for skills development or treatment delivery, provides research opportunities to students in an advanced undergraduate course on the neurobiology of autism, and provides graduate and postdoctoral training opportunities in cognitive neuroscience as it relates to the study of autism.
Autism's severe disabilities, combined with a lack of understanding and appropriate accommodation, rob people with autism of the chance to be fully integrated and productive members of society, disrupt the family environment for parents and siblings, and incur an economic cost of up to $35 billion annually in the United States alone. Yet when one considers the entire autism spectrum, including Asperger syndrome and the many cognitive differences (in some cases superiorities) in relatives of people with autism, the line between autistic and normal development begins to seem a very subtle one. How can developing children be steered across that line? Answering this question depends on an integrative style of inquiry that will be essential to twenty-first century science. Autism research has shown that neural abnormalities are networked throughout the brain and not limited to any single piece of tissue, that autism is seldom produced by one gene alone but rather arises from networks of many interacting genes, and that autism involves abnormalities (deficits or superiorities) in a web of skills from low levels such as detail perception and muscle control to high levels such as speech and social cognition. Putting all the pieces together in these complicated biological networks depends on cooperation among networks of scientists whose combined backgrounds and insights amount to more than a sum of individuals. This project aims to further such a culture of scientific collaboration by producing open-source software, open-access publication of results, and free availability of all experimental data and methods. The hard problem of autism research pulls in many branches and disciplines of science and provides an integrative bridge between them. In this regard, the project drives scientific innovation, and helps to understand the diversity of human cognition in general.