Diagnosing anxiety and depression in children under 8 is notoriously difficult. The signs are subtle. And kids that young sometimes don't display noticeable symptoms. University of Vermont researchers have used artifical intelligence to detect these conditions — known as internalizing disorders — in young children by analyzing their speech patterns. Their study, published in the Journal of Biomedical and Health Informatics, shows that machine learning analysis of audio data can be used to identify children with internalizing disorders with 80 percent accuracy. "These results point toward the future use of this approach for screening children for internalizing disorders so that interventions can be deployed when they have the highest chance for long-term success," the abstract states. Using a test designed to induce moderate psychological stress in a laboratory setting, experimenters asked children ages 3 to 8 to tell them a three-minute story, and indicated that "they would be judged based on how interesting it was." Buzzers sounded at the 30- and 90-second marks. The data indicated that kids with depression and anxiety had lower-pitched voices, repeated speech patterns and content, and responded to the buzzer in a higher pitch. "A couple of decades ago, no one thought kids that young could be depressed," explained Ellen McGinnis, a clinical psychologist at UVM Medical Center and the lead author of the study. When these disorders are identified in children, she continued, they can be treated through cognitive behavioral therapy and lifestyle modifications, including reducing stressors and increasing positive family time.