Freely Generated Word Responses Analyzed With Artificial Intelligence Predict Self-Reported Symptoms of Depression, Anxiety, and Worry

Abstract

This study explores the use of question-based computational language assessments (QCLA) to measure mental health symptoms of depression and anxiety through freely generated word responses. Results show that QCLA correlates significantly with DSM-5 criteria for MDD and GAD, suggesting its utility as a complement to rating scales.