Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Practicing Stress Relief for the Everyday: Designing Social Simulation Using VR, AR, and LLMs

Anna Fang, Hriday Chhabria, Alekhya Maram, Haiyi Zhu

TL;DR

This paper investigates whether social simulation using VR, AR, and LLMs can help people practice stress relief in everyday situations. The authors developed eight low-fidelity prototypes across three stress scenarios and conducted prototype-driven semi-structured interviews with 19 participants. Key findings show that AR is preferred for real-world transfer, but non-immersive text-based tools offer accessible alternatives; realism must balance potential trauma; LLM-driven guidance raises questions about agency. The work provides design considerations for safe, transferable self-care simulations and argues for reframing HCI toward everyday self-care interventions.

Abstract

Stress is an inevitable part of day-to-day life yet many find themselves unable to manage it themselves, particularly when professional or peer support are not always readily available. As self-care becomes increasingly vital for mental well-being, this paper explores the potential of social simulation as a safe, virtual environment for practicing stress relief for everyday situations. Leveraging the immersive capabilities of VR, AR, and LLMs, we developed eight interactive prototypes for various everyday stressful scenarios (e.g. public speaking) then conducted prototype-driven semi-structured interviews with 19 participants. We reveal that people currently lack effective means to support themselves through everyday stress and found that social simulation fills a gap for simulating real environments for training mental health practices. We outline key considerations for future development of simulation for self-care, including risks of trauma from hyper-realism, distrust of LLM-recommended timing for mental health recommendations, and the value of accessibility for self-care interventions.

Practicing Stress Relief for the Everyday: Designing Social Simulation Using VR, AR, and LLMs

TL;DR

This paper investigates whether social simulation using VR, AR, and LLMs can help people practice stress relief in everyday situations. The authors developed eight low-fidelity prototypes across three stress scenarios and conducted prototype-driven semi-structured interviews with 19 participants. Key findings show that AR is preferred for real-world transfer, but non-immersive text-based tools offer accessible alternatives; realism must balance potential trauma; LLM-driven guidance raises questions about agency. The work provides design considerations for safe, transferable self-care simulations and argues for reframing HCI toward everyday self-care interventions.

Abstract

Stress is an inevitable part of day-to-day life yet many find themselves unable to manage it themselves, particularly when professional or peer support are not always readily available. As self-care becomes increasingly vital for mental well-being, this paper explores the potential of social simulation as a safe, virtual environment for practicing stress relief for everyday situations. Leveraging the immersive capabilities of VR, AR, and LLMs, we developed eight interactive prototypes for various everyday stressful scenarios (e.g. public speaking) then conducted prototype-driven semi-structured interviews with 19 participants. We reveal that people currently lack effective means to support themselves through everyday stress and found that social simulation fills a gap for simulating real environments for training mental health practices. We outline key considerations for future development of simulation for self-care, including risks of trauma from hyper-realism, distrust of LLM-recommended timing for mental health recommendations, and the value of accessibility for self-care interventions.
Paper Structure (50 sections, 4 figures, 1 table)

This paper contains 50 sections, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. HCI for Mental Health and Self-Care
  4. Simulation in HCI
  5. Immersive Technology for Mental Health
  6. Virtual and Augmented Reality Technology for Mental Health and Self-Care
  7. Large Language Models (LLMs) for Mental Health
  8. Methods
  9. Key Design Dimensions
  10. Design Prototypes
  11. Prototype Development
  12. Semi-Structured Interview
  13. Participants and Recruitment
  14. Exposure Condition.
  15. Interview Protocol.
  16. ...and 35 more sections

Figures (4)

  • Figure 1: Examples of different prototypes, shown here for the public speaking scenario.
  • Figure 2: Onboarding system of box breathing guidance. The user is asked to inhale, hold, exhale, and hold for 4 seconds each. The bubble expands and contracts following the breath, with a countdown timer for 4 seconds for each stage.
  • Figure 3: Example of interaction flow in the VR prototype for roommate conflict. The user can select a topic (left) then begin a dialogue with the generated scene (middle). At any point, the user may trigger the breathwork guidance and conduct box breathing while the simulation is paused (right).
  • Figure 4: SUDS score rating among participants for their exposure condition