research in brief Brain Scans May Predict ASD Researchers at Carnegie Mellon University have created brainreading techniques to identify neural representations of social thoughts, possibly establishing the first biologically based diagnostic tool to detect autism spectrum disorders. The study, led by psychologist Marcel Just and reported in PLoS One (bit.ly/neuromarkers-asd), predicted ASD diagnoses with 97 percent accuracy. It used magnetic resonance imaging and machinelearning techniques that use brain activation patterns to scan and decode the contents of a person's thoughts of objects or emotions. Previous work demonstrated that specific thoughts and emotions display a similar neural signature in people without brain disorders, suggesting that brain disorders may show detectable alterations in these usual activation patterns. The study detected changes in the way certain concepts are represented in the brains of people with ASD. These alterations-"thoughtmarkers"-indicate differences in brain representations when people with ASD think about certain social concepts that appear to be indicative of the disorder. The researchers scanned the brains of 17 adults with highfunctioning autism and 17 neurotypical control participants while the participants thought about 16 different social interactions, such as "persuade," "adore" and "hug." Control participants' thoughts of social interaction included activation in the brain's posterior midline regions, indicating a representation of the "self." However, the self-related activation was near absent in participants with ASD. Machine-learning algorithms classified individuals as "autistic" or "non-autistic" with 97-percent accuracy based on the fMRI thought-markers. A brain-based measure of the altered thoughts used in conjunction with clinical assessments could enable clinicians to make quicker and more certain diagnoses and more quickly implement targeted therapies. le ader .pubs. asha .org 15http://www.bit.ly/neuromarkers-asd http://leader.pubs.asha.org