About

Faith Gunning is an Associate Professor of Psychology in Psychiatry at Weill Cornell Medical College. Her research focuses on neuroimaging, neuroplasticity, and the neurobiological underpinnings of depression and anxiety, particularly in late-life populations. She has contributed extensively to understanding the structural and functional neuroanatomy associated with mood disorders, utilizing advanced neuroimaging techniques such as fMRI and TMS coil placement optimization. Her work also explores the heterogeneity of depression, social reward processing, and the development of personalized interventions, including digital and mobile health technologies, to improve treatment outcomes for mental health conditions.

Research topics

• Computer Science
• Internal medicine
• Psychiatry
• Medicine
• Psychology
• Biology
• Neuroscience
• Pediatrics
• Computational biology
• Physical therapy
• Clinical psychology

Selected publications

• Generative AI Use and Depressive Symptoms Among US Adults

  JAMA Network Open · 2026-01-21 · 3 citations

  articleOpen access

  Importance: Generative artificial intelligence (AI) has rapidly entered mainstream use in the US, but its association with mental health has not been characterized. Objective: To examine the associations of the extent and type of generative AI use among US adults with negative affective symptoms in a large, nationally representative sample. Design, Setting, and Participants: This survey study used data from a 50-state US internet nonprobability survey conducted between April and May 2025. Survey respondents were aged 18 years and older. Data were analyzed in August 2025. Exposure: Participants self-reported generative AI and social media use. Main Outcomes and Measures: The outcome of interest, negative affect, was measured using the Patient Health Questionnaire 9-item (PHQ-9). Results: There were 20 847 unique participants, with mean (SD) age 47.3 (17.1) years and 10 327 (49.5%) female, 10 386 (49.8%) male, and 134 (0.6%) nonbinary participants; 2152 participants (10.3%) reported using AI at least daily, including 1053 participants (5.1%) who reported daily use and 1099 participants (5.3%) who reported use multiple times per day. Among participants who used daily or more frequently, 1033 (48.0%) reported use for work, 246 (11.4%) for school, and 1875 (87.1%) for personal applications. In survey-weighted regression models, daily or more frequent AI use was significantly more common among men, younger adults, those with higher education and income, and those in urban settings. Greater AI use was associated with greater levels of depressive symptoms in sociodemographic-adjusted regression models: (daily use: β = 1.08 [95% CI, 0.55-1.62]; multiple times per day: β = 0.86 [95% CI, 0.35-1.37]) compared with nonuse, and with greater likelihood of reporting at least moderate depressive symptoms (odds ratio [OR], 1.29 [95% CI, 1.15-1.46]); similar patterns were observed for anxiety and irritability. The highest estimates were observed among individuals using AI for personal use (β = 0.31 [95% CI, 0.10-0.52]) and those aged 25 to 44 years (β = 1.22 [95% CI, 0.70-1.74]) or 45 to 64 years (β = 1.38 [95% CI, 0.72-2.05]). Conclusions and Relevance: This survey study found that AI use was significantly associated with greater depressive symptoms, with magnitude of differences varying by age group. Further work is needed to understand whether these associations are causal and explain heterogeneous effects.

  PublisherOA PDFDOI

• Emulated trial of artificial intelligence use and subsequent depressive outcomes in a survey of US adults

  BMJ Mental Health · 2026-05-01

  articleOpen access

  BACKGROUND: Generative artificial intelligence (AI) use has been suggested to have adverse mental health consequences but a causal relationship has not been examined. OBJECTIVE: To simulate a randomised controlled trial of AI use in a work, school or personal context by applying target trial emulation to multiple waves of data from a nationally representative survey. METHODS: We conducted a target trial emulation using non-probability survey data from three waves of a nationally representative survey conducted between 18 June 2024 and 8 January 2025. Participants aged ≥18 years reported generative AI use frequency at baseline. High-frequency use was defined as multiple times per week or more. The primary outcome was depressive symptom severity measured using the Patient Health Questionnaire 9-item (PHQ-9) at follow-up. Generalised causal forests assessed heterogeneity of treatment effects. FINDINGS: Among 19 099 participants assessed at baseline, 2862 (15.0%) reported AI use at least multiple times per week. A subset of 3109 (16.3%) returned for follow-up. In the primary weighted analysis, high-frequency use was not significantly associated with change in PHQ-9 score at follow-up (mean difference -0.18, 95% CI -0.94 to 0.59; p=0.65). Multiple sensitivity analyses using alternate outcome definitions also did not identify significant causal effects. Generalised causal forests yielded no significant evidence of heterogeneity of effect (p=0.81). CONCLUSIONS: In an emulated randomised trial among US adults, generative AI use was not associated with subsequent depressive symptoms. This result does not support the premise that AI use causes greater depressive symptoms, although adverse outcomes among vulnerable individuals cannot be excluded. CLINICAL IMPLICATIONS: AI use is unlikely to cause increased depressive symptoms among most US adults. Continued monitoring should clarify potential risks among vulnerable populations.

  PublisherDOI

• Sociodemographic disparities, healthcare system trust, and social support in mental health treatment among U.S. adults with depressive or anxiety symptoms

  Journal of Mood and Anxiety Disorders · 2026-01-13 · 1 citations

  articleOpen access

  Large and persistent sociodemographic disparities in rates of mental health treatment in the United States have been reported, but whether these differences reflect institutional mistrust or limited social support remains unclear. This study described current treatment use among American adults with moderate-to-severe depressive or anxiety symptoms and examined whether trust in health care institutions and availability of emotional support were associated with lack of treatment. A cross-sectional analysis was conducted using data from a nationally distributed, web-based opinion survey of 9733 American adults with moderate-to-severe depressive or anxiety symptoms (Patient Health Questionnaire-9 score ≥10 and/or Generalized Anxiety Disorder-2 score ≥3). The survey was fielded April 10th-28th, 2025, using quota sampling for age, gender, race, ethnicity, education, U.S. census region, and urbanicity; post-stratification weights approximated the U.S. adult population. The primary outcome was no current mental health treatment (neither antidepressant nor psychotherapy use). Weighted logistic regression estimated odds ratios for treatment absence by sociodemographic characteristics, trust in physicians and hospitals, scientists and researchers, the Centers for Disease Control and Prevention, pharmaceutical companies, and emotional support. Among 9733 adults with elevated symptoms, 66.3 % reported no current treatment. Racial and ethnic minority groups, men, and those born outside the United States had higher odds of being untreated, while public insurance predicted lower odds. Lower trust in doctors and hospitals, lower trust in science, and lack of emotional support each independently predicted treatment absence, but inclusion of these variables did not meaningfully attenuate sociodemographic disparities.

  PublisherOA PDFDOI

• 626. Loneliness and Suicidal Ideation and Behavior in Schizophrenia and Depression

  Biological Psychiatry · 2026-04-25

  article
  PublisherDOI

• Predictors of dropout from psychotherapy in community settings: a large-scale electronic health records study

  2026-03-29

  articleOpen access

  Importance: Psychotherapy dropout poses a significant public health problem, predicting depression relapse and illness persistence. Identifying who is at risk of psychotherapy dropout can guide personalized and scalable strategies to mitigate risk and improve outcomes. Objective: We estimated the prevalence and predictors of psychotherapy dropout in a large community setting. Design: In this prognostic study, we analyzed electronic health record data collected between 2008-2022. We trained classifiers using logistic regression (LR) and random forest machine learning (RF) to identify predictors of dropout.Setting: Two large academic medical centers, 6 community hospitals, and their affiliated outpatient networks in Massachusetts. Participants and Exposures: Out of 423,636 individuals with depression, we included 40,732 patients (aged 18-80) who had ≥1 individual psychotherapy session and ≥1 any documented visit in the year before and after the session.Main Outcomes and Measures: Predictors included demographics, medical conditions, medical/psychiatric services (e.g. prescription, diagnostic, procedural codes). We assessed the number of individuals who stopped psychotherapy after ≤3 sessions. We trained LR and RF models to maximize predictive accuracy (Area Under the Curve; AUC) and extracted top predictors of dropout. Results: Psychotherapy dropout rate was 28.4% (n=11,571). AUC values were 0.64 for LR (95% CI: 0.63 - 0.65) and 0.66 for RF (95% CI: 0.65 - 0.67). A prior mental health encounter (e.g. group psychotherapy, psychiatric evaluation) and white and/or non-Hispanic self-identified background predicted decreased dropout likelihood. A mental disorder due to a physiological condition and a prior medical admission predicted increased likelihood of dropout.Conclusions and Relevance: Nearly a third of patients who begin psychotherapy drop out, underscoring the need for rapid and scalable risk detection and mitigation strategies. Brief mental health encounters may protect against dropout, whereas psychotherapy referrals during or following medical admissions may increase risk. Integrated care models with brief interventions embedded in medical settings may reduce dropout risk. Brief risk detection based on readily available EHR data can inform the likelihood of psychotherapy dropout and guide clinical recommendations. Together, our findings can inform targeted interventions to reduce risk of psychotherapy dropout among vulnerable populations in the community.

  PublisherDOI

• 474. Negative Urgency and Delayed Discounting in Schizophrenia-Spectrum and Depressive Disorders: A Preliminary Study

  Biological Psychiatry · 2026-04-25

  article
  PublisherDOI

• 199. Discrepancy Between Implicit Association and Explicit Suicidal Ideation and Relation With Emotion Regulation

  Biological Psychiatry · 2026-04-25

  article
  PublisherDOI

• Predictors of dropout from psychotherapy in community settings: a large-scale electronic health records study

  PsyArXiv (OSF Preprints) · 2026-03-30

  preprintOpen access

  Importance: Psychotherapy dropout poses a significant public health problem, predicting depression relapse and illness persistence. Identifying who is at risk of psychotherapy dropout can guide personalized and scalable strategies to mitigate risk and improve outcomes. Objective: We estimated the prevalence and predictors of psychotherapy dropout in a large community setting. Design: In this prognostic study, we analyzed electronic health record data collected between 2008-2022. We trained classifiers using logistic regression (LR) and random forest machine learning (RF) to identify predictors of dropout. Setting: Two large academic medical centers, 6 community hospitals, and their affiliated outpatient networks in Massachusetts. Participants and Exposures: Out of 423,636 individuals with depression, we included 40,732 patients (aged 18-80) who had ≥1 individual psychotherapy session and ≥1 any documented visit in the year before and after the session. Main Outcomes and Measures: Predictors included demographics, medical conditions, medical/psychiatric services (e.g. prescription, diagnostic, procedural codes). We assessed the number of individuals who stopped psychotherapy after ≤3 sessions. We trained LR and RF models to maximize predictive accuracy (Area Under the Curve; AUC) and extracted top predictors of dropout. Results: Psychotherapy dropout rate was 28.4% (n=11,571). AUC values were 0.64 for LR (95% CI: 0.63 - 0.65) and 0.66 for RF (95% CI: 0.65 - 0.67). A prior mental health encounter (e.g. group psychotherapy, psychiatric evaluation) and white and/or non-Hispanic self-identified background predicted decreased dropout likelihood. A mental disorder due to a physiological condition and a prior medical admission predicted increased likelihood of dropout. Conclusions and Relevance: Nearly a third of patients who begin psychotherapy drop out, underscoring the need for rapid and scalable risk detection and mitigation strategies. Brief mental health encounters may protect against dropout, whereas psychotherapy referrals during or following medical admissions may increase risk. Integrated care models with brief interventions embedded in medical settings may reduce dropout risk. Brief risk detection based on readily available EHR data can inform the likelihood of psychotherapy dropout and guide clinical recommendations. Together, our findings can inform targeted interventions to reduce risk of psychotherapy dropout among vulnerable populations in the community.

  Publisher

• Objective Quality Assessment for Precision Functional MRI Data

  bioRxiv (Cold Spring Harbor Laboratory) · 2026-02-11

  articleOpen access

  Precision functional mapping (PFM) enables individual-level characterization of brain network organization but requires substantially more and higher-quality fMRI data than is standard. Despite its growing use, objective criteria for data sufficiency and quality needed to ensure interpretable and replicable individual-level results remain unclear. Here, we introduce the Network Similarity Index (NSI), an objective measure of the extent to which functional connectivity (FC) patterns in an individual dataset express the large-scale network structure required for PFM. NSI captures the integrity of low-spatial-frequency, coherent network organization and denoising fidelity, and aligns closely with blinded expert assessments of PFM usability. NSI also accounts for variability in the rate at which FC becomes reliable across individuals. Here, we provide an open-source framework for NSI-based data quality evaluation and models for linking NSI values with expert-judged PFM suitability. This framework can also inform expected returns from additional data collection, enabling principled decisions about data sufficiency and replication in precision fMRI research.

  PublisherOA PDFDOI

• Precision Assignment to Psychosocial Interventions for Late-Life Depression

  JAMA Psychiatry · 2025-09-17 · 2 citations

  articleOpen access

  Importance: Most older adults with depression lack access to efficacious psychotherapies due to a critical clinician shortage. Even when treated, response rates are limited to approximately 50%. A treatment decision rule (TDR) may maximize treatment efficacy and resources by assigning patients to their optimal intervention. This is the first study to propose a TDR for late-life depression designed for community settings. Objective: To develop a scalable TDR for assignment to a psychotherapy or usual care intervention for late-life depression that can be delivered easily in community settings. Design, Setting, and Participants: In this prognostic study, adults 60 years or older with major depression participated in randomized controlled trials comparing psychotherapy with usual care. Participants were recruited from outpatient and community settings of Weill Cornell Medicine and the University of California San Francisco between 2002 and 2011. Data were analyzed from May 2023 to May 2025. Interventions: Participants received either psychotherapy (problem-solving therapy, psychotherapy for late-life depression and medical burden) or usual care (supportive therapy, treatment as usual, or case management). Main Outcomes and Measures: The primary outcome was mean reduction in depression severity (measured by the Hamilton Depression Rating Scale [HAM-D]). A generated effect modifier TDR was applied to identify the optimal intervention for each patient based on baseline characteristics (demographics, depression severity, social support, cognition, and disability). The TDR maximized depression severity reduction and the proportion of patients treated with the usual care intervention. Results: In 427 older adults with late-life depression (mean [SD] age, 72.7 [8.7] years; 70% female), the predicted HAM-D score reduction with TDR-based intervention was a mean of 49.1% (95% CI, 47.4%-51.0%). The TDR improved expected depression severity reduction by 34% compared with usual care (HAM-D reduction, 36.6% [95% CI, 34.5%-38.7%]) and the TDR was somewhat superior to assigning all patients to receive psychotherapy (HAM-D reduction, 46.7% [95% CI, 44.2%-48.8%]). Older adults with higher depression severity, stronger social support, and lower cognitive functioning should receive psychotherapy; those with lower depression severity, higher cognitive functioning, and low social support would benefit from usual care. Conclusions and Relevance: In this study of older adults with depression, pending prospective testing, the automatic TDR may be used in community settings to inform treatment assignment. The TDR has the potential to increase precision, cost-effectiveness, and response rates among older adults with depression. Trial Registration: ClinicalTrials.gov Identifiers: NCT00601055, NCT00151372, NCT00052091, NCT00540865.

  PublisherOA PDFDOI

Recent grants

• Efficacy of biomarker-guided rTMS for treatment resistant depression

  NIH · $6.2M · 2019–2026

• NIH Grant R56MH114976

  NIH · $724k · 2019

• NIH Grant K23MH074818

  NIH · $676k · 2011

• Research Fellowship in Geriatric Mood Disorders

  NIH · $4.5M · 1990–2026

• NIH Grant R01MH097735

  NIH · $1.9M · 2019

Frequent coauthors

• Conor Liston

  234 shared

• George S. Alexopoulos

  Cornell University

  152 shared

• Abhishek Jaywant

  Cornell University

  147 shared

• Marc J. Dubin

  Cornell University

  127 shared

• Lindsay W. Victoria

  118 shared

• Joan Toglia

  Mercy University

  100 shared

• Roy H. Perlis

  98 shared

• Nili Solomonov

  Weill Cornell Medicine

  90 shared

Awards & honors

• George Alexopoulos, M.D. Honorary Directorship, Psychiatry,…