The abstract is read at the publisher; the summary is JClub's.
TALQ enables dense, two-week scoring of 52 depressive and anxiety symptoms from text in 491 members of two online mental health communities, revealing stable symptom factors and longitudinal dynamics over a year.
- Why it matters: Depression and anxiety are often measured sparsely, leaving symptom changes between assessments unobserved and limiting insight into how expressed states evolve over time.
- What they did: A large language model rated 52 symptoms from text, and TALQ scored 491 community members every two weeks for a year, producing 12 replicable symptom factors and testing prediction against human and inventory annotations.
- The result: TALQ ratings were stable across samples, predicted interpersonal risk factors, low self-esteem, and depression severity, agreed with same-model inventories, and showed that sleep disturbance and self-harm content preceded later overwhelm and emotional pain.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.