Therapeutic Functions in Large Language Model Responses to Emotional Disclosures: Qualitative Content Analysis
- 1Division of Digital Psychiatry, Department of Psychiatry, Beth Israel Deaconess Medical Center, 330 Brookline Avenue, Boston, MA, 02215, United States, 1-617-667-6700
Abstract
Background: Millions of people now use general-purpose large language models (LLMs) for mental health support, yet how these models respond to emotional disclosures and whether their responses functionally resemble established therapeutic practices remain uncharacterized.
Objective: This study aimed to characterize how frontier LLMs respond to emotionally charged messages and examine whether their response patterns resemble therapeutic practices associated with self-awareness.
Methods: We generated 54 simulated user inputs representing mild-to-moderate everyday emotional difficulties. These simulated user inputs were generated using 3 LLMs (GPT-5.5, Claude Opus 4.7, and Gemini 2.5 Pro). The simulated user inputs systematically varied across 3 lengths and 5 communicative registers, ranging from raw venting to pragmatic requests. Then, each simulated user input was presented as a standalone user turn to the same 3 frontier LLMs, yielding 162 responses. Two independent clinical coders qualitatively coded the resulting responses using an inductive approach.
Results: The codes were compared and consolidated through iterative consensus, which yielded 8 higher-order categories. Acknowledging and validating segments appeared in 89.5% (145/162) of responses, emotion-focused suggestions in 84.6% (137/162), and interpretation in 67.3% (109/162). These patterns converged on emotional self-awareness, a construct associated with adaptive emotion regulation, suggesting that LLMs may repeatedly enact a recognizable micro-process of validation, affect labeling, interpretation, reflection, and regulation support. Models differed in stance, with some assigning emotional labels and interpretations directly rather than prompting users’ own reflection.
Conclusions: Frontier LLMs respond to emotional disclosures in patterned, largely convergent ways that approximate therapeutic techniques for building emotional awareness, though delivered more didactically than in clinical practice. Whether these responses build genuine self-awareness, reduce symptoms, or provide therapeutic benefit requires controlled and naturalistic investigation.
Related
- Exploration of emotional and psychosocial responses following initial HIV Diagnosis: A qualitative study
- Mothers' emotional and caregiving experiences during the first month after a child's acute lymphoblastic leukemia diagnosis: A qualitative study
- Emotional and social burden of hidradenitis suppurativa: a qualitative multi-centre narrative medicine project in Italy
- Positive emotional responses while walking along a forest route compared with a city route: evidence from wearable facial electromyography
- Pregnant Women's Experiences of Obstetric Abdominal Ultrasound: A Qualitative Study Exploring Emotional, Cognitive, and Relational Dimensions.
BibTeX
@article{notsu2026therapeutic,
title = {{Therapeutic Functions in Large Language Model Responses to Emotional Disclosures: Qualitative Content Analysis}},
author = {Haruka Notsu and Sean Ryan and Phuong Anh Nguyen and Jill Noorily and Matthew Flathers and John Torous},
journal = {JMIR Formative Research},
year = {2026},
month = oct,
doi = {10.2196/104513},
url = {https://doi.org/10.2196/104513},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Therapeutic Functions in Large Language Model Responses to Emotional Disclosures: Qualitative Content Analysis
AU - Haruka Notsu
AU - Sean Ryan
AU - Phuong Anh Nguyen
AU - Jill Noorily
AU - Matthew Flathers
AU - John Torous
PY - 2026
DA - 2026/10/01
JO - JMIR Formative Research
DO - 10.2196/104513
UR - https://doi.org/10.2196/104513
AB - Background: Millions of people now use general-purpose large language models (LLMs) for mental health support, yet how these models respond to emotional disclosures and whether their responses functionally resemble established therapeutic practices remain uncharacterized. Objective: This study aimed to characterize how frontier LLMs respond to emotionally charged messages and examine whether their response patterns resemble therapeutic practices associated with self-awareness. Methods: We generated 54 simulated user inputs representing mild-to-moderate everyday emotional difficulties. These simulated user inputs were generated using 3 LLMs (GPT-5.5, Claude Opus 4.7, and Gemini 2.5 Pro). The simulated user inputs systematically varied across 3 lengths and 5 communicative registers, ranging from raw venting to pragmatic requests. Then, each simulated user input was presented as a standalone user turn to the same 3 frontier LLMs, yielding 162 responses. Two independent clinical coders qualitatively coded the resulting responses using an inductive approach. Results: The codes were compared and consolidated through iterative consensus, which yielded 8 higher-order categories. Acknowledging and validating segments appeared in 89.5% (145/162) of responses, emotion-focused suggestions in 84.6% (137/162), and interpretation in 67.3% (109/162). These patterns converged on emotional self-awareness, a construct associated with adaptive emotion regulation, suggesting that LLMs may repeatedly enact a recognizable micro-process of validation, affect labeling, interpretation, reflection, and regulation support. Models differed in stance, with some assigning emotional labels and interpretations directly rather than prompting users’ own reflection. Conclusions: Frontier LLMs respond to emotional disclosures in patterned, largely convergent ways that approximate therapeutic techniques for building emotional awareness, though delivered more didactically than in clinical practice. Whether these responses build genuine self-awareness, reduce symptoms, or provide therapeutic benefit requires controlled and naturalistic investigation.
SN - 2561-326X
ER -