Journal of psychopathology and clinical science最新文献

{"title":"Rebuilding HiTOP from the ground up: Symptom-level analyses and a revised mapping to the DSM.","authors":"Miriam K Forbes, Tam Pham, Matthew Roberts, Carly J Johnco","doi":"10.1037/abn0001125","DOIUrl":"10.1037/abn0001125","url":null,"abstract":"<p><p>The evidence for the structure of the Hierarchical Taxonomy of Psychopathology (HiTOP) model relies heavily on analyses of correlations among <i>Diagnostic and Statistical Manual of Mental Disorders</i> diagnoses, which entrenches some of their limitations. In this study, we deconstructed the mental disorders covered in HiTOP into individual symptoms-with personality disorders represented via item-level assessment of the Alternative Model for Personality Disorders facets-and rebuilt the HiTOP model from the ground up. Participants self-selected into an online mental health study and were randomly split into primary (<i>n</i> = 11,762) and holdout (<i>n</i> = 3,000) samples to identify a robust hierarchy of empirically derived constructs, ranging from symptoms and syndromes to very broad \"superspectra\" of psychopathology. In each sample, both hierarchical clustering and hierarchical principal components analyses were used; the final hierarchical structure was derived based on points of cross-sample and cross-method agreement. The resulting model was remarkably similar to the HiTOP model, including nearly all of the same subfactors, spectra, and superspectra-albeit with some reorganization in the structure. The inclusion of broad disinhibition/hypomania and pathological introversion dimensions represented the most prominent differences compared with the current higher order structure of HiTOP. The detailed lower order constructs found here also mirrored many of the existing components and traits in HiTOP, as well as indicating some potential areas for revision. Finally, we remapped the disorders included in the Structured Clinical Interview for the <i>Diagnostic and Statistical Manual of Mental Disorders-5th Edition</i> onto the new framework model. Overall, the results suggest that the structural foundations of the HiTOP model are largely sound. (PsycInfo Database Record (c) 2026 APA, all rights reserved).</p>","PeriodicalId":73914,"journal":{"name":"Journal of psychopathology and clinical science","volume":" ","pages":"908-921"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147824416","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A systematic review and meta-analysis of the late positive potential and internalizing psychopathology.","authors":"Sarah B Barkley, Jadyn Trayvick, Mohan Li, Daniel N Klein, Roman Kotov, Brady D Nelson","doi":"10.1037/abn0001169","DOIUrl":"10.1037/abn0001169","url":null,"abstract":"<p><p>The present study leveraged the Hierarchical Taxonomy of Psychopathology (HiTOP) framework to conduct a systematic meta-analysis to determine the association between the late positive potential (LPP) index of emotional reactivity and internalizing psychopathology. PRISMA guidelines were followed. Articles were identified through PubMed, APA PsycInfo, and Web of Science online platforms in May 2025. Included articles examined associations between the LPP to positive and/or negative stimuli and internalizing psychopathology. Risk of bias and publication bias were assessed. Results were examined for individual disorders, distress and fear subfactors, and the internalizing spectrum using two approaches: standard analyses that examined aggregate effects and hierarchical analyses that examined direct and indirect relationships. We conducted moderator analyses for sample, task design, LPP quantification, and psychopathology measurement. We included 63 studies across 5,360 participants (<i>M</i>_age = 19.65, <i>SD</i> = 11.1; 58.7% female). In standard meta-analyses, depression was associated with a smaller LPP to positive stimuli (<i>r</i> = -.06, 95% confidence interval [CI; -.12, -.003]). Specific phobia was associated with a larger LPP to negative stimuli (<i>r</i> = .21, 95% CI [.02, .37]). Distress was associated with a smaller LPP to both positive (<i>r</i> = -.12) and negative (<i>r</i> = -.11) stimuli when measured via clinical interview, and fear was associated with a larger LPP to negative stimuli (<i>r</i> = .10, 95% CI [.03, .16]). Hierarchical analyses indicated that the depression results were specific to the disorder, whereas the fear disorder-level results were due to the higher order fear subfactor. The LPP demonstrates discriminant relationships with distress and fear disorders and subfactors. Results were largely robust against methodological factors. (PsycInfo Database Record (c) 2026 APA, all rights reserved).</p>","PeriodicalId":73914,"journal":{"name":"Journal of psychopathology and clinical science","volume":" ","pages":"933-952"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148762722","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using smartphone surveys to predict next-week suicide attempts.","authors":"Matthew K Nock, Evan M Kleiman, Kate H Bentley, Rebecca G Fortgang, Alexander J Millner, Kelly L Zuromski, Adam Bear, Alexis Christie, Merryn Daniel, Dylan DeMarco, Lia Follet, Flynn Kelly, Hope Neveux, Onyinye Obi-Obasi, Jordyn R Ricard, Narise Ramlal, Tida Tambedou, Yaniv Yacoby, Suzanne A Bird, Ralph Buonapane, Abigail Donovan, Patrick Mair, Jukka-Pekka Onnela, Rosalind Picard, Jordan W Smoller","doi":"10.1037/abn0001117","DOIUrl":"10.1037/abn0001117","url":null,"abstract":"<p><p>Clinicians are tasked with predicting and preventing suicidal behavior among their patients; however, there is currently no method for accurately predicting whether a person will make a suicide attempt (SA) in the near future. We tested whether brief, smartphone-based surveys, combined with passively collected survey metadata, could predict the occurrence of suicidal behavior over the next 7 days among those at elevated risk. Participants were 619 patients presenting to the hospital with suicidal thoughts/behavior. They were sent brief (20-item) smartphone-based surveys 6 times/day for 3 months. Survey responses (<i>N</i> = 79,448) and metadata (e.g., time since last survey submission) were used as predictors of next-week SA and suicide-related event (SRE; which also included hospitalization to prevent an SA) in a series of machine learning models. The most accurate prediction was achieved using bidirectional long short-term memory and simple lasso-penalized logistic regression models, with the best performing model using bidirectional long short-term memory to predict SRE, which with specificity at .90, had area under the curve = .94, sensitivity = .87, and positive predictive value = .30, and SAs with area under the curve = .90, sensitivity = .74, positive predictive value = .16. Prediction accuracy was higher than has been achieved in prior studies and was strongest for models that predicted SREs (vs. SAs), included more sources of data, focused on adults (vs. adolescents), and when participants' own data were included in the model training process (vs. being held out). The strongest and most consistent predictors of next-week SA included within-study history of SREs (from adult lasso regression: <i>OR</i> = 1.47) and self-reported agitation (<i>OR</i> = 1.11), whereas odds of next-week SA were decreased for surveys submitted on weekends (<i>OR</i> = 0.87) and in the context of feelings that one could resist suicidal urges (<i>OR</i> = 0.88-0.96). Brief smartphone-based surveys can predict next-week SAs/SREs with a fairly high degree of accuracy. Future work is needed to further improve accuracy and test just-in-time interventions targeting high-risk periods. (PsycInfo Database Record (c) 2026 APA, all rights reserved).</p>","PeriodicalId":73914,"journal":{"name":"Journal of psychopathology and clinical science","volume":" ","pages":"922-932"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13308188/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147965819","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Urges matter: Building an ideation-to-action approach to nonsuicidal self-injury.","authors":"Sarah E Victor, Elizabeth C Hoelscher, Amanda N Bianco","doi":"10.1037/abn0001162","DOIUrl":"10.1037/abn0001162","url":null,"abstract":"<p><p>This viewpoint article clarifies the relevance of urges as an independent focus of study and proposes a novel ideation to-action approach to the study of non-suicidal self-injury (NSSI) urges and behaviors. NSSI is a highly prevalent, trans-diagnostic clinical problem that contributes to poor mental health, physical injury, and perhaps most concerning, increased risk of suicidal thoughts and behaviors. Relatively less work has conceptualized the underlying cognitive states that precede or prompt NSSI behaviors, variously described as NSSI thoughts, urges, or ideation. Although no formal nomenclature exists for these constructs, we view NSSI thoughts as inclusive of multiple cognitive states related to NSSI (e.g., thinking about past NSSI, desire for NSSI in the present, and NSSI expectancies); in contrast, we define NSSI urges as thinking about NSSI with at least some desire to engage in the behavior. Clinical science and practice can be improved by integrating models from other fields to better operationalize, examine, and, ultimately, ameliorate NSSI urges among people struggling with NSSI. The dramatic growth in daily life research on NSSI over the last 10 years has elucidated underlying mechanisms relevant to theory and clinical practice. Now, we can take even more exciting steps forward, by recognizing NSSI urges as a meaningful independent outcome and by more carefully clarifying when and how these urges are most likely to lead to NSSI itself. In turn, we can better understand the lived and living experiences of people who struggle with NSSI and identify personalized interventions that can interrupt NSSI urges and behaviors in daily life. (PsycInfo Database Record (c) 2026 APA, all rights reserved).</p>","PeriodicalId":73914,"journal":{"name":"Journal of psychopathology and clinical science","volume":" ","pages":"901-902"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148521568","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Momentary persistence of negative interpretations as a dynamic marker of emotional distress.","authors":"Lisa M W Vos, Peter Kuppens, Tom Smeets, Jonas Everaert","doi":"10.1037/abn0001166","DOIUrl":"10.1037/abn0001166","url":null,"abstract":"<p><p>Inflexible revision of negative interpretations has been implicated in depression and social anxiety. Yet evidence primarily stems from studies employing controlled cognitive tasks that rely on standardized fictional scenarios. To track how negative interpretations persist over time in an everyday life context, this study developed the Daily Life Interpretation Inflexibility Task (DL-IIT) and embedded this cognitive task in an ecological momentary assessment (EMA) protocol. The DL-IIT was designed to capture the emergence and persistence of momentary negative interpretations and to test their dynamic links with momentary emotions and symptom severity. Community adults (<i>N</i> = 179) reporting varying depression and social anxiety symptoms participated in a 14-day EMA protocol (12,950 completed EMA surveys) measuring momentary interpretations and emotions. Preregistered multilevel analysis showed that higher depressive and social anxiety symptoms were each associated with greater momentary persistence of negative interpretations. At the within-person level, persistence of negative interpretations predicted subsequent increases in negative emotions but did not predict decreases in positive emotions. In the reverse direction, higher negative emotions and lower positive emotions predicted greater subsequent persistence of negative interpretations. These mood-congruent emotion-to-cognition effects were stronger at higher symptom severity. Complementary dynamic structural equation models revealed reciprocal cross-lagged associations between persistence of negative interpretations and negative emotions, and small reciprocal effects involving positive emotions. Across models, emotion-to-cognition pathways were stronger than cognition-to-emotion pathways. Together, these findings indicate that persistence of negative interpretations is both a symptom-linked individual-difference marker and a momentary process that may contribute to self-reinforcing cognition-emotion cycles in daily life. The DL-IIT provides a scalable method for identifying high-risk moments that could be targeted with just-in-time interventions aimed at enhancing cognitive flexibility. (PsycInfo Database Record (c) 2026 APA, all rights reserved).</p>","PeriodicalId":73914,"journal":{"name":"Journal of psychopathology and clinical science","volume":" ","pages":"953-968"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148802575","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Self-reports of personality functioning and depression are practically fungible: Commentary on Kerber et al. (2026).","authors":"Joshua D Miller, Colin E Vize, Nathaniel L Phillips, Donald R Lynam","doi":"10.1037/abn0001165","DOIUrl":"10.1037/abn0001165","url":null,"abstract":"<p><p>This article comments on Kerber et al. (see record 2027-04247-002) examining the prospective correlates of depression (DEP) and personality functioning (PF) in a very large sample of individuals followed via an ecological momentary assessment framework. Using sophisticated bifactor modeling to control for the problematically large latent correlation between the two constructs (the authors report correlations of .75 and .87, depending on measurement approach), the authors suggest that PF and DEP have distinct affective signatures with important ramifications for how treatment is tailored. We argue in this commentary that the very strong latent correlations between PF and DEP, along with nearly identical regression weights in predicting various outcomes (profile agreement across outcomes, <i>r</i> = .95), indicates that there is no practical difference between these two constructs as typically measured. The authors believe the bifactor models allow them to disentangle PF and DEP, but this raises the \"perils of partialing.\" The empirical profiles manifested by the original and residualized latent scores for PF bear no relation to one another (profile agreement, <i>r</i> = -.06), and 53% of their regression weights change direction when comparing the two. We also raise concerns with the viability of their bifactor models, particularly for PF. The residualized PF factor is relatively weak, changes fairly dramatically across analyses, and is unlikely to replicate across samples. Ultimately, we believe Kerber et al. (2026) suggestion that PF and DEP are distinct with differing treatment implications is unsupported by their data. (PsycInfo Database Record (c) 2026 APA, all rights reserved).</p>","PeriodicalId":73914,"journal":{"name":"Journal of psychopathology and clinical science","volume":" ","pages":"903-907"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148521593","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Contextual insights from large language models in ambulatory assessment: Differential reactivity of affect and self-esteem to daily events in borderline personality disorder.","authors":"David Levi Tekampe, Tobias D Kockler, Ulrich W Ebner-Priemer, Angelica Mendes, Martin Bohus, Philip S Santangelo","doi":"10.1037/abn0001143","DOIUrl":"10.1037/abn0001143","url":null,"abstract":"<p><p>Ambulatory assessment (AA) research often relies on closed-ended context checklists, whereas open-ended event descriptions preserve richer idiographic context but are rarely analyzed at scale. Borderline personality disorder (BPD) is characterized by marked instability, providing a clinically informative use case for testing scalable approaches to contextual dynamics in daily life. We investigated the association of event context and sentiment (i.e., emotional connotation) derived using a large language model (LLM) with self-reported momentary calmness, affective valence, and self-esteem in a sample of 265 women (131 individuals with BPD and 134 healthy controls) undergoing a 4-day high-frequency AA with 12 hourly assessments per day. We evaluated an automated LLM pipeline for classifying 10,844 free-text entries and for rating event-related sentiment, enabling quantitative modeling of context and appraisal in AA. The free-text entries were classified into five categories: daily routines and household, leisure and recreation, social and personal relationships, work and professional engagements, health and well-being, plus an indeterminate category. Results showed the BPD group had lower values across outcomes and reacted with significantly lower calmness and self-esteem during social/personal events, as well as significantly lower self-esteem during health- and well-being-related activities, compared with controls. When sentiment was modeled alongside event categories, more positive-than-usual sentiments were associated with higher calmness, valence, and self-esteem, whereas more negative-than-usual sentiments were associated with lower states, with larger sentiment reactivity in BPD for calmness, valence and self-esteem. By employing LLMs for automated text categorization and sentiment analysis of free-text entries within AA research, we have effectively incorporated contextual information into our analysis. (PsycInfo Database Record (c) 2026 APA, all rights reserved).</p>","PeriodicalId":73914,"journal":{"name":"Journal of psychopathology and clinical science","volume":" ","pages":"1007-1021"},"PeriodicalIF":4.6,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148594879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using large language models to create lexicons for interpretable text models with high content validity: The suicide risk lexicon.","authors":"Daniel M Low, Osiris Rankin, Daniel D L Coppersmith, Kate H Bentley, Matthew K Nock, Satrajit S Ghosh","doi":"10.1037/abn0001152","DOIUrl":"10.1037/abn0001152","url":null,"abstract":"<p><p>Researchers often want to measure a variety of constructs such as anxiety, discrimination, or loneliness in text data from surveys, interviews, social media, and electronic health records. Using large language models (LLMs)-although optimal for text classification-remain infeasible for some researchers due to concerns around computational expertise, cost, privacy, and compute requirements. Therefore, some researchers prefer to use lightweight models for large data sets or interpretable models to avoid mistakes in high-stakes scenarios. Lexicons offer simple baselines to LLMs by searching for relevant phrases and can be used together with LLMs to guarantee capturing specific keywords. However, building new lexicons is resource intensive. In this study, we found that GPT-4 Turbo was able to automatically create a lexicon for 49 known risk factors for suicidal thoughts and behaviors, which we release as the Suicide Risk Lexicon. Generating a lexicon with LLMs quickly measures most constructs relevant for suicide risk detection, resulting in high content validity. This lexicon was able to accurately predict risk in crisis counseling conversations. After validating the lexicon with clinician ratings, the lexicon modestly outperformed the linguistic inquiry and word count lexicon, which has low content validity for mental illness, and performed similarly to some black-box deep learning models. We discovered that active suicidal ideation and direct self-injury were stronger indicators of imminent risk than passive suicidal ideation and depressed mood in this ecological setting of crisis counseling conversations. To simplify creating new lexicons for other research domains, we introduce a Python package, construct-tracker. (PsycInfo Database Record (c) 2026 APA, all rights reserved).</p>","PeriodicalId":73914,"journal":{"name":"Journal of psychopathology and clinical science","volume":" ","pages":""},"PeriodicalIF":4.6,"publicationDate":"2026-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13614521/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149042267","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A framework for evidence-based psychotherapy with AI (EBP-AI).","authors":"Elizabeth C Stade, Philip Held, H Andrew Schwartz, Shannon Wiltsey Stirman, Johannes C Eichstaedt","doi":"10.1037/abn0001148","DOIUrl":"https://doi.org/10.1037/abn0001148","url":null,"abstract":"<p><p>Artificial intelligence (AI) systems and large language models offer substantial potential to augment or even fundamentally change elements of psychological assessment and treatment. However, current AI technologies have yet to demonstrate the capacity to effect meaningful and sustained clinical change. This gap reflects both the limited integration of clinical science knowledge into language models and applications built using them, as well as the mismatch between the brief, minutes-long nature of most AI interactions and the months-long course of most evidence-based treatments. Here we introduce the evidence-based psychotherapy with AI framework, which articulates a set of principles for developing effective clinical AI applications: (a) psychodiagnostic assessment, (b) longitudinal case conceptualization, (c) appropriately dosed intervention planning, (d) meaningful progress evaluation, (e) rigorous validation with clinical populations, (f) attention to real world implementation and use, (g) clinically appropriate style, and (h) understanding clinical psychology as a living science. We introduce a set of key technical questions for the development and evaluation of clinical large language models and AIs aligned with these principles. Despite their potential, current clinical AIs fall short, in part due to issues with memory, sycophancy, and prioritizing short-term helpfulness over long-term clinical impact. Responsible and ethical design of effective, clinical-science-based AI systems will require understanding their limitations and strategically extending their capabilities. (PsycInfo Database Record (c) 2026 APA, all rights reserved).</p>","PeriodicalId":73914,"journal":{"name":"Journal of psychopathology and clinical science","volume":" ","pages":""},"PeriodicalIF":4.6,"publicationDate":"2026-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13483332/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148802559","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fine-tuning large language models for detecting negative self-referential language in smartphone text among youth with depression.","authors":"Isaac N Treves, Lilian Y Li, Libby Hod, Adela Schwartz, Paul A Bloom, Jamaal Spence, Alexander Saul Rosenberg, Katherine Durham, Xuhai Xu, Esha Trivedi, David Pagliaccio, Nicholas B Allen, Stewart A Shankman, Randy P Auerbach","doi":"10.1037/abn0001159","DOIUrl":"https://doi.org/10.1037/abn0001159","url":null,"abstract":"<p><p>Adolescent smartphone language provides a lens into negative self-referential thinking, which is central to major depressive disorder (MDD). Prior studies have linked language features, including negative sentiment and first-person pronouns, to mood and depression, suggesting that naturalistic language may identify who is at risk and when that risk is greatest. However, studies of adolescent smartphone social communication have typically relied on rule-based models not validated for heterogeneous, context-sensitive language. To address this gap, we determined whether transformer models, including large language models, optimized detection of depression risk in extensive adolescent smartphone text data. In this study, 223 adolescents (Mage = 16.43 years, current MDD = 37, remitted MDD = 103, healthy controls = 83) installed a smartphone app, which prompted participants to provide mood ratings once a day and acquired all keyboard inputs over 12 months (mean text entries per participant = 17,683). Ten thousand text entries were double-coded for entry-level sentiment (positive, neutral, negative) and self-reference for training and testing traditional rule-based approaches (VADER, pronoun counts) and transformer approaches, including GPT-4-mini. In the full data set, between-participant associations with depressive symptoms and within-participant relations to daily mood and depressive episodes were examined. Fine-tuned transformer models best aligned with human-coded sentiment labels (F1GPT4-MINI = .84, F1VADER = .59) and accurately detected self-reference (F1T5-BASE = .97). Adolescents with current and remitted MDD exhibited more transformer-based negative self-referential language than healthy controls (ORs = 1.38, 1.26). Increased negative self-referential language predicted worse next-day mood (β = -.033, p < .001) and a higher likelihood of next-week depressive episodes (OR = 1.66, 95% confidence interval [1.06, 2.62], p = .028), although the association with depressive episodes was not robust to sensitivity analyses. Transformer models may be integrated into digital mental health care to detect when youth are at risk for depression. (PsycInfo Database Record (c) 2026 APA, all rights reserved).</p>","PeriodicalId":73914,"journal":{"name":"Journal of psychopathology and clinical science","volume":" ","pages":""},"PeriodicalIF":4.6,"publicationDate":"2026-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13483330/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148802591","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}