Right now, diagnosing disorders like autism relies heavily on interviews and behavioral observations. But new research published in PLoS One shows that a much more objective measure—reading a person’s thoughts through an fMRI brain scan—might be able to diagnose autism with close to perfect accuracy.

Lead study author Marcel Just, PhD, professor of psychology and director of the Center for Cognitive Brain Imaging at Carnegie Mellon University, and his team performed fMRI scans on 17 young adults with high-functioning autism and 17 people without autism while they thought about a range of different social interactions, like “hug,” “humiliate,” “kick” and “adore.” The researchers used machine-learning techniques to measure the activation in 135 tiny pieces of the brain, each the size of a peppercorn, and analyzed how the activation levels formed a pattern.

READ FULL ARTICLE Curated publisher From Time