A novel cognitive behavioral therapy–based generative AI tool (Socrates 2.0) to facilitate Socratic dialogue: protocol for a mixed methods feasibility study
Open accessPublished version of a preprint
Abstract
BACKGROUND: Digital mental health tools, designed to augment traditional mental health treatments, are becoming increasingly important due to a wide range of barriers to accessing mental health care, including a growing shortage of clinicians. Most existing tools use rule-based algorithms, often leading to interactions that feel unnatural compared with human therapists. Large language models (LLMs) offer a solution for the development of more natural, engaging digital tools. In this paper, we detail the development of Socrates 2.0, which was designed to engage users in Socratic dialogue surrounding unrealistic or unhelpful beliefs, a core technique in cognitive behavioral therapies. The multiagent LLM-based tool features an artificial intelligence (AI) therapist, Socrates, which receives automated feedback from an AI supervisor and an AI rater. The combination of multiple agents appeared to help address common LLM issues such as looping, and it improved the overall dialogue experience. Initial user feedback from individuals with lived experiences of mental health problems as well as cognitive behavioral therapists has been positive. Moreover, tests in approximately 500 scenarios showed that Socrates 2.0 engaged in harmful responses in under 1% of cases, with the AI supervisor promptly correcting the dialogue each time. However, formal feasibility studies with potential end users are needed. OBJECTIVE: This mixed methods study examines the feasibility of Socrates 2.0. METHODS: On the basis of the initial data, we devised a formal feasibility study of Socrates 2.0 to gather qualitative and quantitative data about users' and clinicians' experience of interacting with the tool. Using a mixed method approach, the goal is to gather feasibility and acceptability data from 100 users and 50 clinicians to inform the eventual implementation of generative AI tools, such as Socrates 2.0, in mental health treatment. We designed this study to better understand how users and clinicians interact with the tool, including the frequency, length, and time of interactions, users' satisfaction with the tool overall, quality of each dialogue and individual responses, as well as ways in which the tool should be improved before it is used in efficacy trials. Descriptive and inferential analyses will be performed on data from validated usability measures. Thematic analysis will be performed on the qualitative data. RESULTS: Recruitment will begin in February 2024 and is expected to conclude by February 2025. As of September 25, 2024, overall, 55 participants have been recruited. CONCLUSIONS: The development of Socrates 2.0 and the outlined feasibility study are important first steps in applying generative AI to mental health treatment delivery and lay the foundation for formal feasibility studies. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/58195.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Compared to an earlier single-agent version, the addition of two supervisory AI agents appears to improve dialogue quality by reducing redundant, off-topic, overly lengthy answers and repetitive conversational loops.
“Compared to an earlier single-agent version of the tool, the addition of the 2 supervisory AI agents appears to improve dialogue quality by reducing the risk of redundant, off-topic, or overly lengthy answers and repetitive conversational loops (ie, segments in which similar questions or statements are repeated without progressing the dialogue) [13].”
From AI-Facilitated Cognitive Reappraisal via Socrates 2.0: Mixed Methods Feasibility Study. (2025) · cites this paper as [13]
The Introduction’s “Socrates 1.0” section describes the earlier single-agent tool as verbose, prone to getting off task, and caught in conversational loops. The “Socrates 2.0” section says the added agents appeared to improve dialogue quality and reduce looping, with supervisor feedback helping keep conversations focused. Our copy does not establish that the added agents reduced redundant or overly lengthy answers specifically. (LLM, conf=0.84)
Experimental — accuracy not yet measuredLLM-based digital tools that emulate therapists' Socratic dialogue style may be more engaging than static worksheets and thereby increase adherence to out-of-session practice and help users engage in cognitive reappraisal exercises.
“LLM-based digital tools that emulate therapists’ Socratic dialogue style may be more engaging than static worksheets and thereby increase adherence to out-of-session practice and ultimately help users engage in cognitive reappraisal exercises [13].”
From AI-Facilitated Cognitive Reappraisal via Socrates 2.0: Mixed Methods Feasibility Study. (2025) · cites this paper as [13]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Narrative Methods for the Human Sciences
(2008) · Gender in Management An International Journal · doi:10.1108/17542410810897562 · Cited by 3,784 · see citing works
The Efficacy of Cognitive Behavioral Therapy: A Review of Meta-analyses
Hofmann, S. G., Asnaani, A., Vonk, I. J. J., Sawyer, A. T., & Fang, A. (2012) · Cognitive Therapy and Research · doi:10.1007/s10608-012-9476-1 · Cited by 3,445 · cited 1× · see citing works
“CBTs have been shown to be effective for a multitude of mental health disorders, including posttraumatic stress disorder, depression, and anxiety, among others [11].”
BackgroundPsychometric assessment of three newly developed implementation outcome measures
Weiner, B. J., Lewis, C. C., Stanick, C., Powell, B. J., Dorsey, C. N., Clary, A. S., Boynton, M. H., & Halko, H. (2017) · Implementation Science · doi:10.1186/s13012-017-0635-3 · Cited by 2,819 · cited 2× · see citing works
“The Acceptability of Intervention Measure, Intervention Appropriateness Measure, and Feasibility of Intervention Measure [32] is a validated and widely used 12-item measure scored on a 5-point scale (completely disagree to completely agree) that will be administered to determine the perceived…”
MethodInfluentialLöwe, B., Wahl, I., Rose, M., Spitzer, C., Glaesmer, H., Wingenfeld, K., Schneider, A., & Brähler, E. (2010) · Journal of Affective Disorders · doi:10.1016/j.jad.2009.06.019 · Cited by 2,473 · cited 1× · see citing works
“To assess depression and generalized anxiety, participants will be asked to complete the Patient Health Questionnaire-4 [26].”
BackgroundThe System Usability Scale: Past, Present, and Future
Lewis, J. R. (2018) · International Journal of Human–Computer Interaction · doi:10.1080/10447318.2018.1455307 · Cited by 1,958 · cited 2× · see citing works
“Following their 2-week access to Socrates 2.0, clinicians will complete similar measures to the other users, including the System Usability Scale [31], Acceptability of Intervention Measure, Intervention Appropriateness Measure, Feasibility of Intervention Measure [32], and Mobile Health App…”
MethodInfluentialCognitive processing therapy for PTSD a comprehensive manual
Resick, P. A., Monson, C. M., & Chard, K. M. (n.d.) · Cited by 868 · see citing works
The mHealth App Usability Questionnaire (MAUQ): Development and Validation Study
Zhou, L., Bao, J., Setiawan, I. M. A., Saptono, A., & Parmanto, B. (2019) · JMIR mHealth and uHealth · doi:10.2196/11500 · Cited by 675 · cited 2× · see citing works
“Participants will also be asked to complete the Mobile Health App Usability Questionnaire [33], which is a validated and widely used 18-item measure scored on a 7-point scale (strongly disagree to strongly agree) that assesses ease of use, satisfaction, and usefulness of mobile apps.”
MethodInfluentialMunder, T., Wilmers, F., Leonhart, R., Linster, H. W., & Barth, J. (2010) · Clinical Psychology & Psychotherapy · doi:10.1002/cpp.658 · Cited by 538 · cited 1× · see citing works
“Finally, the Working Alliance Inventory-Short Revised [34], which is a validated 12-item measure scored on a 5-point scale (seldom to always or always to seldom) will be used to examine participants’ experiences with their AI therapist.”
MethodPhilippe, T. J., Sikder, N., Jackson, A., Koblanski, M. E., Liow, E., Pilarinos, A., & Vasarhelyi, K. (2022) · JMIR Mental Health · doi:10.2196/35159 · Cited by 436 · cited 1× · see citing works
“Researchers have suggested that digital interventions can increase accessibility and reduce barriers to mental health care, offering flexible and scalable solutions [4,5].”
BackgroundD’Alfonso, S. (2020) · Current Opinion in Psychology · doi:10.1016/j.copsyc.2020.04.005 · Cited by 408 · cited 1× · see citing works
“The ability of large language models (LLMs) to mimic human language provides the opportunity to build digital mental health tools that may feel more natural and engaging [7,8].”
BackgroundStade, E. C., Stirman, S. W., Ungar, L. H., Boland, C. L., Schwartz, H. A., Yaden, D. B., Sedoc, J., DeRubeis, R. J., Willer, R., & Eichstaedt, J. C. (2024) · npj Mental Health Research · doi:10.1038/s44184-024-00056-z · Cited by 391 · cited 5× · see citing works
“The ability of large language models (LLMs) to mimic human language provides the opportunity to build digital mental health tools that may feel more natural and engaging [7,8].”
BackgroundMethodInfluentialCoombs, N. C., Meriwether, W. E., Caringi, J., & Newcomer, S. R. (2021) · SSM - Population Health · doi:10.1016/j.ssmph.2021.100847 · Cited by 333 · cited 1× · see citing works
“The use of digital mental health tools to augment traditional psychotherapies is becoming increasingly important due to a growing shortage of clinicians and other barriers to care (eg, cost and accessibility) [1-3].”
BackgroundThe Relationship Between Homework Compliance and Therapy Outcomes: An Updated Meta-Analysis
Mausbach, B. T., Moore, R., Roesch, S., Cardenas, V., & Patterson, T. L. (2010) · Cognitive Therapy and Research · doi:10.1007/s10608-010-9297-z · Cited by 309 · see citing works
Perceptions and opinions of patients about mental health chatbots: scoping review
Abd-Alrazaq, A. A., Alajlani, M., Ali, N., Denecke, K., Bewick, B. M., & Househ, M. (2021) · Journal of Medical Internet Research · doi:10.2196/17828 · Cited by 301 · cited 1× · see citing works
“To date, most of the existing digital mental health tools have relied on rule-based algorithms to provide mostly prescripted responses, at times resulting in these tools feeling less natural compared with dialogues with human therapists [6].”
BackgroundKazantzis, N., Whittington, C., Zelencich, L., Kyrios, M., Norton, P. J., & Hofmann, S. G. (2016) · Behavior Therapy · doi:10.1016/j.beth.2016.05.002 · Cited by 280 · see citing works
Developing an optimal short-form of the PTSD Checklist for DSM-5 (PCL-5)
Zuromski, K. L., Ustun, B., Hwang, I., Keane, T. M., Marx, B. P., Stein, M. B., Ursano, R. J., & Kessler, R. C. (2019) · Depression and Anxiety · doi:10.1002/da.22942 · Cited by 206 · cited 1× · see citing works
“Traumatic stress symptoms and potential manic or psychotic symptoms will be assessed via the Posttraumatic Stress Disorder Checklist-4 [29] and Washington Early Recognition Center Affectivity and Psychosis screen [30], respectively.”
MethodSupporting Homework Compliance in Cognitive Behavioural Therapy: Essential Features of Mobile Apps
Tang, W., & Kreindler, D. M. (2017) · JMIR Mental Health · doi:10.2196/mental.5283 · Cited by 120 · cited 2× · see citing works
“…better understand the true feasibility, as well as the potential risks and problems of generative AI tools as complements to CBT. Future studies should examine whether using LLMs would make out-of-session practice, such as examining one’s thoughts, more engaging for patients than worksheets [21].”
BackgroundHomework Compliance Counts in Cognitive-Behavioral Therapy
LeBeau, R. T., Davies, C. D., Culver, N. C., & Craske, M. G. (2013) · Cognitive Behaviour Therapy · doi:10.1080/16506073.2013.763286 · Cited by 105 · cited 1× · see citing works
“Although research demonstrates that learning and practicing skills via repeated practice and homework is important for symptom improvement [16-20], therapists frequently observe low homework completion rates in practice [21,22].”
BackgroundBraun, J. D., Strunk, D. R., Sasso, K. E., & Cooper, A. A. (2015) · Behaviour Research and Therapy · doi:10.1016/j.brat.2015.05.004 · Cited by 91 · see citing works
Fidelity to the Cognitive Processing Therapy Protocol: Evaluation of Critical Elements
Farmer, C. C., Mitchell, K. S., Parker-Guilbert, K. S., & Galovski, T. E. (2016) · Behavior Therapy · doi:10.1016/j.beth.2016.02.009 · Cited by 90 · cited 1× · see citing works
“A key intervention in CBTs is Socratic dialogue [9,10,12-15].”
BackgroundMeasuring Homework Compliance in Cognitive-Behavioral Therapy for Adolescent Depression
Gaynor, S. T., Lawrence, P. S., & Nelson‐Gray, R. O. (2006) · Behavior Modification · doi:10.1177/0145445504272979 · Cited by 81 · cited 1× · see citing works
“Although research demonstrates that learning and practicing skills via repeated practice and homework is important for symptom improvement [16-20], therapists frequently observe low homework completion rates in practice [21,22].”
BackgroundA Practical Survey on Zero-shot Prompt Design for In-context Learning
Columbia University, N. Y. C., & Li, Y. (2023) · Proceedings of the Conference Recent Advances in Natural Language Processing - Large Language Models for Natural Language Processings · doi:10.26615/978-954-452-092-2_069 · Cited by 68 · cited 1× · see citing works
“To create the first version of the tool, Socrates 1.0, the team initially used a zero-shot approach [23], prompting a single-agent model to engage the user in Socratic dialogue.”
MethodInfluentialMulti-Agent Collaboration: Harnessing the Power of Intelligent LLM Agents
Talebirad, Y., & Nadiri, A. (2023) · arXiv · doi:10.48550/arxiv.2306.03314 · Cited by 65 · cited 2× · see citing works
“Through the inclusion of multiple AI agents [24], CBT experts on our team felt as though Socrates 2.0 produced higher quality Socratic dialogue which seemed to resemble responses from a human therapist more closely.”
BackgroundMethodInfluentialEzawa, I. D., Pfeifer, B. J., & Strunk, D. R. (2019) · Journal of Consulting and Clinical Psychology · doi:10.1037/ccp0000392 · Cited by 55 · cited 3× · see citing works
“Building on recent advances relating to LLMs, our goal was to create and examine the feasibility of a generative artificial intelligence (AI) tool that can complement traditional cognitive behavioral therapies (CBTs) by facilitating a core therapeutic intervention: Socratic dialogue [9,10].”
BackgroundVittorio, L. N., Murphy, S. T., Braun, J. D., & Strunk, D. R. (2022) · Behaviour Research and Therapy · doi:10.1016/j.brat.2022.104035 · Cited by 40 · cited 3× · see citing works
“Building on recent advances relating to LLMs, our goal was to create and examine the feasibility of a generative artificial intelligence (AI) tool that can complement traditional cognitive behavioral therapies (CBTs) by facilitating a core therapeutic intervention: Socratic dialogue [9,10].”
BackgroundMamah, D. T., Owoso, A., Sheffield, J. M., & Bayer, C. (2014) · Comprehensive Psychiatry · doi:10.1016/j.comppsych.2014.07.004 · Cited by 38 · cited 1× · see citing works
“Traumatic stress symptoms and potential manic or psychotic symptoms will be assessed via the Posttraumatic Stress Disorder Checklist-4 [29] and Washington Early Recognition Center Affectivity and Psychosis screen [30], respectively.”
MethodDoes Homework “Matter” in Cognitive Behavioral Therapy for Adolescent Depression?
Simons, A. D., Marti, C. N., Rohde, P., Lewis, C. C., Curry, J. F., & March, J. S. (2012) · Journal of Cognitive Psychotherapy · doi:10.1891/0889-8391.26.4.390 · Cited by 38 · cited 1× · see citing works
“Although research demonstrates that learning and practicing skills via repeated practice and homework is important for symptom improvement [16-20], therapists frequently observe low homework completion rates in practice [21,22].”
BackgroundThe OCI-4: An ultra-brief screening scale for obsessive-compulsive disorder
Amitai Abramovitch, A., Abramowitz, J. S., & McKay, D. (2021) · Journal of Anxiety Disorders · doi:10.1016/j.janxdis.2021.102354 · Cited by 37 · cited 1× · see citing works
“Obsessive-compulsive tendencies will be assessed using the Obsessive-Compulsive Inventory-4 [28].”
MethodDigital mental health care: five lessons from Act 1 and a preview of Acts 2–5
Insel, T. R. (2023) · npj Digital Medicine · doi:10.1038/s41746-023-00760-8 · Cited by 35 · cited 1× · see citing works
“Researchers have suggested that digital interventions can increase accessibility and reduce barriers to mental health care, offering flexible and scalable solutions [4,5].”
BackgroundStade, E. C., Stirman, S. W., Ungar, L., et al. (2023) · doi:10.31234/osf.io/cuzvr · Cited by 35 · see citing works
Hara, K. M., Aviram, A., Constantino, M. J., Westra, H. A., & Antony, M. M. (2017) · Cognitive Behaviour Therapy · doi:10.1080/16506073.2016.1253605 · Cited by 31 · see citing works
Carmel, A., Villatte, J., Rosenthal, M. Z., Chalker, S. A., & Comtois, K. (2016) · Cognitive and Behavioral Practice · doi:10.1016/j.cbpra.2015.08.001 · Cited by 24 · cited 1× · see citing works
“This process is similar to the bug-in-ear supervision method where human supervisors observe therapists’ sessions and provide feedback in real time via a device in the therapists’ ear [25].”
MethodNichols, A. L., & Webster, G. D. (2015) · Personality and Individual Differences · doi:10.1016/j.paid.2015.01.043 · Cited by 22 · see citing works
A Practical Survey on Zero-shot Prompt Design for In-context Learning
Li, Y. (2023) · arXiv · doi:10.48550/arxiv.2309.13205 · cited 1×
“To create the first version of the tool, Socrates 1.0, the team initially used a zero-shot approach [23], prompting a single-agent model to engage the user in Socratic dialogue.”
MethodInfluential
Flagged first, then by citations
Cited by
Sorted by citations
“It’s Not Only Attention We Need”: Systematic Review of Large Language Models in Mental Health Care
Bucher, A. M., Egger, S., Vashkite, I., Wu, W., & Schwabe, G. (2025) · JMIR Mental Health · doi:10.2196/78410 · Cited by 37 · see citing works
“Some studies have gone a step further by developing custom web-based applications tailored to specific use cases, for example, to conduct exercises with help-seekers and peers [90,93] or to display analyses for medical professionals [109], typically integrating LLMs through APIs.”
InfluentialBoit, S., & Patil, R. (2025) · JMIR Mental Health · doi:10.2196/75078 · Cited by 24 · see citing works
“In practice, PE has emerged as a state-of-the-art technique for adapting powerful general-purpose LLMs to specialized domains without requiring resource-intensive retraining processes [68].”
Voultsiou, E., & Moussiades, L. (2026) · Electronics · doi:10.3390/electronics15030524 · Cited by 16 · see citing works
Sun, D., Zheng, Y., Xu, J., & Yang, Z. (2026) · Journal of Computer Assisted Learning · doi:10.1002/jcal.70210 · Cited by 10 · see citing works
Held, P., Stade, E. C., Dondanville, K. A., & Stirman, S. W. (2025) · Journal of Traumatic Stress · doi:10.1002/jts.23188 · Cited by 8 · see citing works
“During development and early feasibility work, the combination of multiple AI agents in Socrates 2.0 was perceived to have improved response quality and improved safety, such as an increased ability to stay on track (Held et al., 2024).”
InfluentialAI-Facilitated Cognitive Reappraisal via Socrates 2.0: Mixed Methods Feasibility Study.
Held, P., Pridgen, S. A., Szoke, D. R., Chen, Y., Akhtar, Z., & Amin, D. (2025) · JMIR Mental Health · doi:10.2196/80461 · Cited by 7 · see citing works
“As described in a study by Held et al [13], the AI interface also displayed safety notices, reminding participants of the nonclinical nature of the tool and providing information on how to seek professional help if needed.”
InfluentialAligning large language models for cognitive behavioral therapy: a proof-of-concept study
Kim, Y., Choi, C.-H., Cho, S., Sohn, J.-Y., & Kim, B.-H. (2025) · Frontiers in Psychiatry · doi:10.3389/fpsyt.2025.1583739 · Cited by 9 · see citing works
Leveraging Artificial Intelligence for Advancements in Mental Disorders: A Short Review
Kasereka, S. K., Tshibangu, K. N., Nyembo, M. M., Tshitenge, L. K., Muzindusi, M. K., Ilunga, G. W. K., Amor, S. B., Bope, C. D., Tashev, T., & Kyamakya, K. (2025) · Procedia Computer Science · doi:10.1016/j.procs.2025.03.087 · Cited by 3 · see citing works
Mao, L., Dong, L., Li, W., Hu, X., Lee, K.-P., & Zhou, Z. (2026) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3791696 · Cited by 3 · see citing works
You, D., Monteiro, D., & Seo, K. (2026) · Interactive Learning Environments · doi:10.1080/10494820.2026.2662461 · Cited by 2 · see citing works
Redesigning trauma‐focused interventions to address capacity challenges in mental health systems
McLean, C. P., Holder, N. D., Holliday, R. P., & Rosen, C. S. (2026) · Journal of Traumatic Stress · doi:10.1002/jts.70090 · Cited by 2 · see citing works
Yang, Y., Qi, J., & Ji, X. (2026) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2026.2683905 · Cited by 2 · see citing works
Shereefdeen, H., & MacKay, M. (2026) · Digital Health · doi:10.1177/20552076261416713 · Cited by 1 · see citing works
Schuster, R., Plessen, C. Y., Carlbring, P., & Walther, A. (2026) · JMIR Mental Health · doi:10.2196/91746 · Cited by 2 · see citing works
“Furthermore, early multiagent frameworks have been introduced [61,62] to model therapists or to simulate clinical dialogues.”
Ridings, L. E., Held, P., Kuhn, E. R., & Stirman, S. W. (2026) · Journal of Traumatic Stress · doi:10.1002/jts.70092 · Cited by 1 · see citing works
“…application to trauma-focused treatment has been supported by innovative research more recently, with goals to enhance therapist training and increase patient access to and engagement with PTSD treatments (e.g., Held et al., 2024; Lenton-Brym et al., 2025; Stade, Eichstaedt, Kaysen, et al., 2025).”
Beyond Compliance: How AI Could Help Creative Writers by Refusing Them
Qin, H. X., Zhu, G., Fan, M., & Hui, P. (2026) · Creativity and Cognition · doi:10.1145/3803784.3807571 · Cited by 1 · see citing works
Artificial Intelligence in Cognitive Behavioral Therapy
Tzioumaki, N., Arvanitaki, E., & Xirakia, E. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-5072-1.ch010 · Cited by 1 · see citing works
AI-Facilitated Cognitive Reappraisal via Socrates 2.0: A Mixed Methods Feasibility Study (Preprint)Preprint
Held, P., Pridgen, S., Szoke, D. R., Chen, Y., Akhtar, Z., & Amin, D. (2025) · doi:10.2196/preprints.80461
Islam, F., Zhao, K., Knijnenburg, B. P., Chen, M.-H., Leow, A., Muthukanagaraj, P., Zhan, L., & Shi, C. (2025) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw69685.2025.00243
“For instance, to improve monitoring and enhance interaction quality, [14] used AI agents and developed a multi-agent-based LLM platform that simulates dialogues.”
Namaziandost, E., & Teng, M. F. (2026) · International Journal of Applied Linguistics · doi:10.1111/ijal.70268
Clinical Risks of Generative AI in Psychotherapy: A Systematic Review
Ech-Cheikh, S., Kharmoum, N., Hamdouni, Y. E., & Ziti, S. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-30563-3_50
Held, P., Smith, D. L., Pridgen, S., Hutzler, I., & Szoke, D. R. (2026) · doi:10.2196/preprints.108739
Haber, Y., Refoua, E., Yirmiya, K., Rafaeli, E., Yinon, D., Atzil–Slonim, D., Fonagy, P., Meinlschmidt, G., Hadar‐Shoval, D., Simon, T., et al. (2026) · doi:10.2196/preprints.110192
The influence of socratic dialogue within GenAI on critical thinking and technology perception
Bereznikova, A., Theophilou, E., & Hernández‐Leo, D. (2026) · Educational Technology Research and Development · doi:10.1007/s11423-026-10693-0
Toward a Framework for Assessing Patient-Facing Gen AI Tools in Mental Healthcare
Durneva, P., & Ghorbanie, H. (2026) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2026.106741
Liu, Y., Mu, J., Li, Y., Ke, Y., & Zhang, Y. (2026) · Frontiers in Psychology · doi:10.3389/fpsyg.2026.1926092
“…to test LLMs in cognitive restructuring, cognitive reappraisal, CBT-style tasks, and Socratic dialogue (Sharma et al., 2023; Ayers et al., 2023; Lee et al., 2024; Gabriel et al., 2024; Sharma et al., 2024; Zhan et al., 2024; Held et al., 2024, 2025; Hodson and Williamson, 2024; Xiao et al., 2024).”
InfluentialLiu, Y., Xu, K., Zhao, S., Li, D., Zhang, Y., Chen, J., & Wu, Y. (2026) · doi:10.2196/preprints.113358
Depressive Belief Rigidity in Language Models: Representation and Limits of Activation-Based Control
Xie, S. (2026) · International Artificial Intelligence and Data Processing Symposium (IDAP) · doi:10.1109/idap71869.2026.11727911
“In contrast, multi-agent systems comprise several LLMs that interact and converse to achieve the desired outcome [28], [29].”
- Authors
- Held, P., Pridgen, S. A., Chen, Y., Akhtar, Z., Amin, D., & Pohorence, S.
- Affiliations
- Rush University Medical Center (US)
- Corresponding author
- Held, Philip
- Venue
- JMIR Research Protocols
- Publisher
- JMIR Publications Inc.
- Published
- 2024-10-10
- Volume
- 13
- Pages
- e58195
- ISSN
- 1929-0748 (ISSN-L)
- PMC
- PMC11502974
- OpenAlex
- W4401788886
- Cited by
- 28
- References
- 32
- Citations by year
- 2026: 19 · 2025: 9
- FWCI
- 11.06
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- Locations
- JMIR Research Protocols (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- U.S. Department of Defense (W81XWH2210739, W81XWH)
- Agency for Healthcare Research and Quality (R21 HS028511)
- Topics
- Social Sciences › Psychology › Applied Psychology › Digital Mental Health Interventions
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Anxiety, Depression, Psychometrics, Treatment, Cognitive Processes
- Keywords
- Socratic dialogue, cognitive behavioral therapy, generative AI, large language models, multi-agent systems, digital mental health tools, feasibility, acceptability, usability, user satisfaction, chatbots, access to mental health care, hallucination mitigation
- MeSH terms
- Artificial Intelligence, Mental Disorders, Feasibility Studies, Humans, Cognitive Behavioral Therapy
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
