Facing the Illusion and Reality of Safety in Social VR

Qingxiao Zheng, Tue Ngoc Do, Lingqing Wang, Yun HuangarXiv 2022 · DOI · arXiv

Abstract

As social Virtual Reality (VR) transitions from a niche hobby to a mainstream medium, the psychological and physical stakes of user safety have escalated. Unlike traditional 2D social media, the immersive, multi-sensory nature of VR removes the physical buffer between users, making harassment feel significantly more personal and traumatic. This paper investigates the disconnect between user safety concerns and current platform implementations. By analyzing over 18,000 social media entries and stress-testing four major VR platforms, the research identifies emerging psychological phenomena like "phantom touch" and advocates for a shift toward "boundary-aware" design to protect users in the evolving Metaverse.

Introduction

The promise of the "Metaverse" lies in its ability to provide presence—the feeling of truly being in a space with others. However, this same presence creates a unique vulnerability. In 2D environments, harassment is mediated by a screen; in VR, actions such as virtual sexual harassment or "teabagging" are experienced as violations of personal physical space. As the user base expands, the industry faces a critical challenge: our understanding of safety on 2D platforms does not fully translate to the 3D, embodied experiences of social VR.

This study addresses the gap in comprehensive safety frameworks for immersive spaces. By examining how users perceive risks and how industry leaders like Meta and VRChat attempt to mitigate them, the authors highlight the urgent need for safety protocols that respect the psychological realities of virtual embodiment.

Methodology

The researchers employed a two-pronged exploratory approach to capture both user sentiment and platform functionality:

  1. Social Media Discourse Analysis: Using Python-based tools (Tweepy and PRAW), the team collected 18,274 posts from Twitter and Reddit over a three-month period. A thematic analysis was performed on this corpus to categorize the primary safety concerns voiced by the VR community.
  2. Platform Stress-Testing: The researchers selected four dominant social VR platforms—Rec Room, VRChat, Horizon Worlds, and AltspaceVR—for direct investigation. Using an Oculus Quest 2, a researcher "stress-tested" these environments by intentionally triggering safety protocols to observe their efficacy during onboarding, exploration, and active social play.

The Psychology of Virtual Presence

The study's findings reveal that the immersive nature of VR leads to unexpected psychological outcomes that complicate safety design.

The "Phantom Sense" Phenomenon

One of the most striking findings is the report of "phantom touch" or "phantom sense." Users frequently reported feeling physical sensations—such as a tingly feeling during a "headpat"—despite the absence of haptic hardware. While these sensations are often positive in a consensual context, they raise significant alarms regarding negative interactions. If a user can feel a friendly touch, the psychological impact of being "stabbed," "shoved," or otherwise assaulted in VR is far more damaging than in traditional gaming.

The Sleeping Phenomenon

An emerging trend involves users remaining in VR while sleeping, often to stay close to friends or to escape noisy real-world environments. This behavior introduces unique safety risks, including physical hazards from headset cables and a total inability to react to or report virtual harassment while unconscious. Current safety features are largely designed for active users, leaving "sleepers" entirely vulnerable.

Current Safety Mechanisms and Their Efficacy

The research synthesized existing safety tools into three primary categories, noting varying degrees of success across platforms:

  • Social Boundary Settings: Platforms have implemented "Personal Space Bubbles" (e.g., Horizon Worlds’ 4-foot default) and "Trust Ranks" (VRChat) to allow users to control proximity. Rec Room utilizes private "Dorm Rooms" as a safe haven.
  • Quick Reactions: To avoid the friction of navigating menus during a crisis, platforms have introduced gestural controls. Rec Room’s "talk to the hand" gesture (which mutes and blocks a user) and VRChat’s "Safe Mode" (which instantly hides all nearby avatars) are cited as effective rapid-remediation tools.
  • Visual and Audio Filtering: Users can hide specific avatars or adjust "screen-space shaders" to mitigate visual nuisances. Furthermore, "junior accounts" are used to protect minors by preventing the disclosure of identifiable information.

Discussion: Toward Boundary-Aware Design

Despite these features, the study identifies a significant level of corporate distrust. Many users expressed skepticism that platforms prioritizing rapid growth would effectively moderate "creeps" or prioritize the safety of minors over engagement metrics.

The authors argue that safety in the Metaverse cannot be achieved through simple "blocking" mechanisms alone. Instead, they advocate for "boundary-aware" design. This framework suggests that systems should understand the nuances of human interaction, moving beyond binary "on/off" safety switches to more granular controls. This includes systems that recognize different layers of interaction—from personal space bubbles to trust-based permissions that dictate how much a stranger can interact with a user's avatar.

Conclusion and Future Work

This exploratory study highlights that the psychological implications of VR, particularly "phantom sense," are not yet addressed by current safety designs. Future research must focus on redefining social norms for 3D spaces and developing community-based reporting systems. Rather than relying solely on individual "block" buttons, the next generation of social VR should empower bystanders to intervene, moving toward a model of collective community safety. As we build the Metaverse, our safety protocols must be as immersive and nuanced as the environments they are designed to protect.

Summary updated

Abstract

The ethical design of social Virtual Reality (VR) is not a new topic, but "safety" concerns of using social VR are escalated to a different level given the heat of the Metaverse. For example, it was reported that nearly half of the female-identifying VR participants have had at least one instance of virtual sexual harassment. Feeling safe is a basic human right - in any place, regardless in real or virtual spaces. In this paper, we are seeking to understand the discrepancy between user concerns and designs in protecting user safety in social VR applications. We study safety concerns on social VR experience first by analyzing Twitter posts and then synthesize practices on safety protection adopted by four mainstream social VR platforms. We argue that future research and platforms should explore the design of social VR with boundary-awareness.
BibTeX
@misc{zheng2022facing,
  author = {Qingxiao Zheng and Tue Ngoc Do and Lingqing Wang and Yun Huang},
  title = {{Facing the Illusion and Reality of Safety in Social VR}},
  year = {2022},
  eprint = {2204.07121},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2204.07121},
}
Figure