Source-linked AI summary
Beyond Harassment: Exploring the Harm Experienced by People with Disabilities in Social Virtual Reality
Xinran Adeline Li, Kexin Zhang, Yuhang Zhao, Yaxing Yao
TL;DR
The paper examines harms experienced by people with disabilities in social VR, where embodied interactions and disability disclosure can intensify negative experiences. It combines a literature review with a survey of 67 PWD, identifying 19 harm types across 5 categories and reporting their perceived severity. The findings characterize disability-linked and social-VR-specific harms and inform design implications for mitigating them.
Problem
Harms experienced by PWD in social VR, including their connections to disability identity and perceived severity, have not been systematically investigated.
Method
The study conducted a literature review followed by an online survey of 67 PWD about harassment experiences and resulting harms in social VR.
Results
The study identified 19 harm types in 5 categories and reported perceived severity for each type.
Takeaways & Limitations
The findings reveal unique harms in social VR and provide design implications for mitigating harms during or after harassment.
Takeaways & Limitations
The sample comprised 67 participants exclusively from the United States, with concentrated representation of mental and mobility-related disabilities and limited incident detail from the survey format.
Abstract
from arXiv · showhide
People with disabilities (PWD) are increasingly engaging in social virtual reality (VR) platforms, where immersive and embodied interactions can intensify negative experiences. While prior work has examined harassment in VR, little is known about the harms experienced by PWD and the perceived severity associated with different harassment and disability types. Unlike harassment, which represents behaviors, harm is more critical to designing effective protections, as it reflects the consequences and impact; the realism of VR and the vulnerability resulting from disability identity can further amplify such impact. To characterize and model harms for PWD, we conducted a literature review, followed by an online survey with 67 PWD to understand participants' harassment experiences and resulting harms in social VR. We identified 19 types of harm in 5 categories, and reported the severity perception of each type of harm. Finally, we analyzed our results from the critical disability theory perspective, summarized the uniqueness of harm in social VR, and discussed design implications for specialized safety mechanisms that mitigate harm for PWD.
1 Introduction
Social VR’s immersive, embodied interactions can intensify safety risks and make harassment feel more realistic, especially for people with disabilities who disclose disability identities. This paper addresses the gap in systematic knowledge about the resulting harms and their perceived severity.
- Social VR combines full-body tracking, synchronous voice chat, and interactive features that blur virtual and physical presence.
- Disability disclosure can improve social connection but also increase risks of disability-targeted harassment, including bullying, ableist language, and insults.
- Existing research has identified harassment behaviors and broad harm categories, but has not systematically investigated harms experienced by PWD in social VR.
- The study asks what harms PWD experience from social VR harassment, how disability identity and the environment relate to those harms, and how severe participants perceive them.
- A literature review and survey of 67 PWD produced a taxonomy of 19 harms across 5 categories and reported perceived severity for each type.
2 Related Work
Prior research documents harassment and broad online harms affecting PWD, but social VR’s embodied features and disability-related vulnerabilities create underexamined forms of harm. The paper positions a structured, PWD-centered investigation as necessary.
- PWD experience stigma, discrimination, cyberbullying, and online ableist microaggressions across offline and online settings.
- Social VR’s embodied avatars, real-time interaction, and immersive features can make harassment more destructive and complicate its negative impacts.
- Harassment risks may increase when disability, gender, sexual orientation, or ethnicity is visible through avatars or voice.
- Existing harassment frameworks emphasize attacks, risks, companies, or infrastructures, while harm taxonomies mainly use broad categories such as psychological, social, and reputational harm.
- Prior taxonomies do not capture deeper links between harms, disability identities, disability-related symptoms, and social VR embodiment.
3 Method
The study first reviewed literature to structure disability-related harassment categories, then surveyed PWD about social VR experiences and resulting harms. Researchers combined qualitative thematic analysis with descriptive severity statistics.
- Literature review: The literature review searched HCI and AR/VR databases, incorporated relevant real-world and online incidents, and identified 8 disability-related harassment types from 32 papers.
- Survey design: The survey collected demographics, social VR use and disability disclosure, harassment incidents, and harms associated with each reported incident.
- Participants and recruitment: 67 participants with diverse disabilities completed the survey, which averaged 50 minutes and provided $10 compensation after response validation.
- Survey refinement: The survey was piloted with 12 participants to refine navigation, wording, and branching logic before the formal study.
- Analysis: Researchers cleaned responses, coded original descriptions inductively through Reflexive Thematic Analysis, finalized a 37-code codebook, and recategorized codes into broader categories.
- Analysis: Quantitative analysis used descriptive statistics, including counts, percentages, and minimum, maximum, mean, and variance severity ratings for each harm type.
- Ethics: The study received IRB approval and used anonymous responses, sensitive survey language, and participants’ options to skip questions or withdraw.
4 Results
The study identifies five categories of harm and documents both established and disability-specific consequences of harassment in social VR. Reported impacts span emotional distress, internalized psychological disruption, social disconnection, and embodied symptom activation.
- Harassment Incidents: Participants reported 152 harassment experiences, including eight literature-derived types and two new types: intentionally triggering disability symptoms and indirect harassment.The new types involved deliberately provoking disclosed disability symptoms and being harmed by witnessing or overhearing disability-targeted harassment.
- Five Categories of Harm: The study identified five harm categories: emotional, internalized, social, somatic recall, and sensational harm.Many harms were distinctive or more intense for people with disabilities in disability-related contexts.
- Social and Somatic Harm: Harassment also produced social and embodied consequences, including exclusion-related loneliness, reduced trust, and reactivated disability symptoms.Somatic recall harm describes the body reactivating prior trauma or disability-related symptoms in response to harassment.
- Internalized Harm: Internalized harm involved sustained disruptions such as sleep problems, eroded self-worth, dissociation, impaired focus, and self-harm desire.Cognitive intrusion could persist from a week to several months, with harassment-related thoughts and memories repeatedly replaying.
5 Discussion
The discussion frames harm in social VR as extending beyond immediate harassment, with embodied environments and disability-related vulnerabilities shaping consequences. It connects these findings to platform design, prevention, awareness, and post-harassment support.
- 5.1 Rethinking the Harms in Social VR: A Critical Disability Theory Perspective: Harassment in social VR can lead some PWD to avoid VR and withdraw from offline social activities when platform systems fail to create safe spaces.The authors interpret this through Critical Disability Theory as evidence that VR platforms can become disabling environments.
- 5.2 Unique Harm in Social VR: Participants described disrespect, humiliation, and isolation tied to disability targeting, with emotional experiences resembling physical-world harassment.The discussion suggests social VR harms may feel closer to real-world experiences because of its embodied, immersive, and interactive nature.
- 5.2 Unique Harm in Social VR: Social withdrawal after harassment may extend from social VR into offline social life, a pattern rarely discussed in prior online harassment research.Participants reported avoiding VR platforms, sometimes for weeks, and reduced willingness to engage in real-world social activities.
- 5.2 Unique Harm in Social VR: The study did not observe financial or physical harms among participants, although the authors recommend precautions if such harms migrate to social VR.Prior work cited in the discussion reports financial harms for PWD in social media and physical-world contexts.
- 5.3 Design Implications: The proposed interventions span AI detection, platform features, educational nudges, and mental-health support across different stages of harassment.AI mechanisms are proposed to monitor voice, text, avatar interactions, and environmental object placements; virtual agents could provide assistance after harassment.
- 5.4 Limitations: The study is limited by its 67-person United States sample, concentration of mental and mobility-related disabilities, and survey-based incident detail.The authors state that interviews could provide deeper understanding of harms and that findings may not apply to other cultural contexts or countries.
6 Conclusion
The conclusion presents a harm framework for PWD in social VR and combines literature review, survey findings, severity analysis, Critical Disability Theory, and design implications. It identifies five categories and 19 harm subtypes.
- 6 Conclusion: The paper identifies 19 harm subtypes in five categories centered on PWD’s experiences with harassment in social VR.The framework describes how harassment leads to negative impacts.
- 6 Conclusion: The study uses a systematic literature review and a survey with 67 PWD to analyze harms and severity ratings.The conclusion also reports analysis from a Critical Disability Theory perspective.
- 6 Conclusion: The paper discusses unique harms in social VR and offers design implications for mitigating harm during or after harassment.These implications are presented as part of the paper’s broader contribution.
A.1 Harassment Definitions
Table 3 defines harassment types, distinguishing eight types drawn from prior work from newly defined types introduced in this study.
- A.1 Harassment Definitions: Table 3 lists harassment type definitions, with the first eight types drawn from prior work and bolded types newly defined in this study.The caption specifies the table’s provenance distinction and formatting convention.
A.2 Survey Protocol
The survey protocol collected participant demographics, social VR use and disclosure practices, harassment incidents, resulting harms, severity ratings, and comparisons with real-world experiences.
- Participant and usage information: The protocol asked about age, diagnosed disability types, gender, ethnicity, social VR applications, usage duration, and usage frequency.Participants selected from predefined demographic, disability, platform, duration, and frequency options.
- Disability disclosure: Participants reported disability disclosure methods in social VR, including avatar appearance, profiles, name tags, and verbal conversation.The protocol also allowed participants to indicate that they had not disclosed their disability or provide another method.
- Harassment experiences: The harassment section collected participant definitions, selected experiences from eight categories, incident descriptions, frequency, location, platform, offender count, and relationship.Participants could add a category when the predefined list did not cover their experience.
- Harm assessment: For physical, phantom-sensation, psychological, economic, and other harms, participants described effects, rated severity, and explained their reasoning.The protocol included examples such as bodily injury, property damage, phantom touch or pain, depression, anxiety, sleep problems, scams, and account hacks.
- Cross-context comparison: The survey asked whether harassment and its harms in social VR were consistent with real-world experiences and requested explanations.Response options ranged from very similar to very different.
Other Types of Harassment
The questions used for one harassment type are repeated for other types, including Treating Disability as Inability and Sexual Harassment.
- The same question structure is applied to other harassment types, such as Treating Disability as Inability and Sexual Harassment.