Digital Humans for Depression Assessment and Intervention Support: Scoping Review
Bibliographic record
Abstract
Background: The growing global burden of mental health disorders has intensified the search for scalable, accessible, and cost-effective interventions. Conversational agents in the form of digital humans have emerged as promising tools to deliver mental health support across diverse populations and settings. Objective: This scoping review aimed to analyze the role of digital humans in depression management, identifying their specific applications in both diagnostic processes and therapeutic interventions. Additionally, it aimed to evaluate the design choices implemented in digital human systems, including their appearance, interaction modalities, back-end intelligence systems, and the various roles they assume. Methods: Following the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines, we systematically searched peer-reviewed literature across major databases, including ACM Digital Library, IEEE Xplore, Web of Science, and PubMed, to capture both psychological and technological perspectives. The search query included a wide variety of synonyms for digital humans and depression: ("avatar" OR "virtual agent" OR "embodied conversational agent" OR "relational agent" OR "digital human" OR "virtual human" OR "virtual character") AND ("Major Depressive Disorder" OR "Depression"). Studies were included if they described the development, implementation, or evaluation of digital humans designed to support mental health outcomes. Data were charted on agent design, therapeutic approach, target population, delivery context, and reported effectiveness. Results: In total, 20 studies (2010-2024) were included. Depression assessment studies comprised 35% (n=7), interventions 55% (n=11), and combined approaches 10% (n=2). Assessment protocols included the questionnaires Patient Health Questionnaire-9 and Very Short Visual Analog Scale of the Center for Epidemiologic Studies Depression Scale - Visual Analog Scale - Very Short version, semistructured interviews based on Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition criteria, and interactive tasks designed to elicit emotional responses. Intervention approaches used cognitive behavioral therapy, psychoeducation, compassion-focused therapy, and avatar therapy. Digital humans assumed 5 distinct roles: interviewer (n=6), facilitator (n=3), counselor (n=3), educator (n=3), and actor (n=5). Interviewers primarily appeared in assessment studies, presenting structured questions. Counselors engaged in therapeutic dialogues, while educators delivered psychoeducational content. Facilitators assisted participants in achieving system goals. Actors portrayed specific emotions or dysfunctional beliefs to facilitate therapeutic processes. Studies highlighted digital humans' utility in enhancing diagnostic processes and therapeutic interventions, noting the potential for transformation through physiological data integration. Conclusions: This study demonstrates that digital humans represent a transformative advancement in depression management, offering innovative applications across both assessment and intervention phases. The evidence reveals digital humans' effectiveness in replicating traditional therapeutic roles while providing unique advantages, including 24/7 accessibility, reduced stigma, consistent care delivery, and personalized support. Digital humans can successfully function to establish therapeutic alliances and elicit meaningful engagement comparable with human providers. Findings underscore the need for continued research to fully realize digital humans' potential in addressing depression-specific needs, advocating for expansion into diverse therapeutic scenarios, and exploration of unexplored digital human applications.
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 | 0.021 | 0.089 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.007 | 0.008 |
| Bibliometrics | 0.021 | 0.018 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.006 | 0.004 |
| Open science | 0.003 | 0.004 |
| Research integrity | 0.004 | 0.002 |
| Insufficient payload (model declined to judge) | 0.009 | 0.001 |
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".