Beyond the Single Turn: Reframing Refusals as Dynamic Experiences Embedded in the Context of Mental Health Support Interactions with LLMs

2026-02-02

AI Summary

This study addresses the frequent refusal responses triggered by safety mechanisms in large language models during mental health support interactions, which are often treated as isolated incidents despite their potential for psychological harm. Through a sequential mixed-methods approach—comprising a survey (N=53) and follow-up interviews (N=16)—the research integrates perspectives from both users and mental health professionals to propose the first multi-stage framework of refusal experiences in mental health contexts. The framework delineates five phases: expectation formation, trigger, information framing, resource referral, and downstream impact. Moving beyond conventional binary, policy-compliance–centered evaluation paradigms, it reconceptualizes refusal mechanisms as embedded within the broader interaction flow, offering empirical insights and design guidance for developing more humane, context-aware response strategies that mitigate secondary harm to help-seekers.

Technology Category

• Natural Language Processing: Safety and Robustness
• Multiagent Systems: Mechanism Design
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction

Application Category

• Responsible Web: Social and technical mechanisms of refusal for web technologies and applications
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Security and Privacy: Large-scale security measurements

Abstract

Content Warning: This paper contains participant quotes and discussions related to mental health challenges, emotional distress, and suicidal ideation. Large language models (LLMs) are increasingly used for mental health support, yet the model safeguards -- particularly refusals to engage with sensitive content -- remain poorly understood from the perspectives of users and mental health professionals (MHPs) and have been reported to cause real-world harms. This paper presents findings from a sequential mixed-methods study examining how LLM refusals are experienced and interpreted in mental health support interactions. Through surveys (N=53) and in-depth interviews (N=16) with individuals using LLMs for mental health support and MHPs, we reveal that refusals are not isolated, single-turn system behaviors, but rather constitute dynamic, multi-phase experiences: pre-refusal expectation formation, refusal triggering and encounter, refusal message framing, resource referral provision, and post-refusal outcomes. We contribute a multi-phase framework for evaluating refusals beyond binary policy compliance accuracy and design recommendations for future refusal mechanisms. These findings suggest that understanding LLM refusals requires moving beyond single-turn interactions toward recognizing them as holistic experiential processes embedded within the entire LLM design pipeline and the broader realities of mental health access.

Problem

• LLM refusals
• mental health support
• safeguard mechanisms
• user experience
• emotional distress

Innovation

• LLM refusals
• mental health support
• dynamic interaction framework
• user experience
• safeguard design