JMIR Publications

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:

Reflecting on clinical experience and input from people with lived experience, we generated 54 representative synthetic user messages systematically varying in length and communicative register, ranging from raw venting to pragmatic requests, and presented each to three frontier LLMs (GPT-5.5, Claude Opus 4.7, Gemini 2.5 Pro).

Methods:

Two independent clinical coders qualitatively coded the 162 resulting responses using an inductive approach

Results:

Coding yielded eight higher-order categories. Acknowledging and validating segments appeared in 89.5% of responses, emotion-focused suggestions in 84.6%, and interpretation in 67.3%. 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.