In a paper published this month, MIT researchers suggest that many of the varied symptoms that characterize autism may be explained by a difficulty with making predictions. The ability to predict is fundamental to tasks as diverse as adjusting to sensory stimuli and inferring other’s mental states based on the context. When prediction is compromised, a person lives in a “seemingly ‘magical’ world wherein events occur unexpectedly and without cause,” write the authors, who include professors Pawan Sinha and Richard Held from the Department of Brain and Cognitive Sciences. Impaired predictive skills can make the world feel overwhelming and may lead to some of the behaviors linked to autism, such as repetitive behavior or difficulty gauging social situations.
In devising their hypothesis, the researchers reviewed more than 100 studies and accounts of autism over more than three decades, with the goal of finding a common and coherent basis for the disorder. A new theory of autism could help researchers design to more effective therapies to treat it.
“At the moment, the treatments that have been developed are driven by the end symptoms. We’re suggesting that the deeper issue is a predictive difficulty, which may, therefore, be a better target for interventions,” says Sinha.
Explore Professor Sinha’s research in the Open Access Articles collection in DSpace@MIT, where it is openly accessible to the world.
Since the MIT faculty established their Open Access Policy in March 2009 they have made thousands of research papers freely available to the world via DSpace@MIT. To highlight that research, we’re offering a series of blog posts that link news stories about scholars’ work to their open access papers in DSpace.