Speech Markers as Longitudinal Predictors of Youth Mental Health: A Systematic Review
Bibliographic record
Abstract
Abstract Introduction: Severe mental illnesses in young people (0–25 years) are often preceded by subtle changes in communication and thinking, detectable in speech and language. Speech and language markers are promising for early detection; however, no systematic review has evaluated their prospective utility in predicting mental health disorders in youth. We systematically reviewed longitudinal studies assessing speech and language markers as predictors of major mental health disorder onset or symptom progression in youth. Methods: We searched for longitudinal studies using recorded speech samples to predict diagnostic changes or symptom severity in major depressive disorder (MDD), psychosis, ADHD, substance use disorder, bipolar disorder, OCD, and eating disorders. Risk of bias was assessed using the Newcastle Ottawa Scale. Our protocol was pre-registered [CRD42024579798]. Results: Of 2,259 articles, 10 studies met inclusion criteria, covering MDD (n=2), psychosis (n=5), and ADHD (n=3). No eligible studies were found for OCD, substance use, bipolar, or eating disorders. Both manual and computational speech analyses were used, with speech samples from parents and youth. Predictive speech markers included parental expressed emotion (MDD, ADHD), formal thought disorder (psychosis), and acoustic/linguistic features (psychosis, ADHD). Study quality was moderate to good (mean score: 5.5/8). Conclusions: Externally validated longitudinal studies on the predictive value of speech and language markers of youth-onset mental disorders are scarce and restricted to a few target disorders. Nonetheless, existing studies highlight the potential of applying Natural Language Processing methods to speech samples from both youth and parents for early identification. Keywords: prediction; speech; youth mental health
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.009 | 0.045 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.008 | 0.009 |
| Bibliometrics | 0.007 | 0.008 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".