“If Someone Downvoted My Posts—That’d Be the End of the World”: Designing Safer Online Spaces for Trauma Survivors

Title of the Paper

"If Someone Downvoted My Posts—That’d Be the End of the World": Designing Safer Online Spaces for Trauma Survivors

Paper Information

• Topic Area: Human-Computer Interaction (HCI), Trauma Survivors, Online Community Design
• Keywords: Trauma Survivors, Information Disclosure, Social Media, Social Support, Anonymity, Content Persistence, Filter Bubbles, True Crime Communities

Research Background and Problem

• Identified Challenges: Trauma survivors face difficulties disclosing their experiences on online platforms, particularly the "silent group" lacking the language to describe their trauma. Many choose to remain hidden rather than participate due to fears of negative feedback or other social risks.
• Significance: Information disclosure is considered a critical step in trauma recovery, associated with positive outcomes such as improved mental health and a stronger sense of self. However, this process may be hindered rather than supported by existing online technology designs.
• Research Motivation and Related Work:
  • While there is extensive research on information disclosure, few studies address the complexities of silent groups or the human-computer interaction factors influencing disclosure.
  • The authors identified that existing models and theories (e.g., information disclosure decision models) lack a comprehensive perspective on how trauma survivors interact within social and technological environments.

Solution

• Method or Model:
  • Introduced a metaphor of a "spinning top" to illustrate the decision-making process of trauma survivors regarding information disclosure, emphasizing emotional activation, toppling, and steadying.
  • Conducted 29 in-depth interviews to understand trauma survivors' behavioral patterns on social media and their interactions with platform design.
• Innovations:
  • Proposed two new concepts: indirect feedback and transportability, highlighting the potential support of technological design for trauma survivors' recovery.
  • Systematically explored for the first time the impact of social media's technical features, such as anonymity, content persistence, and filter bubbles, on information disclosure.
• Implementation Steps and Techniques:
  • Performed qualitative interview analysis combined with a theoretical framework (Trauma-Informed Computing Framework).
  • Iteratively coded the data during analysis and referenced related literature to construct a new theory.

Research Findings

• Specific Findings:
  • Identified three effects influencing trauma survivors' disclosure decisions: emotional activation, toppling effect, and steadying effect.
  • Confirmed that anonymous design is critical for some trauma survivors but has limitations, such as the "anonymous post" feature in Facebook groups failing to fully protect survivors' privacy.
  • Highlighted that indirect feedback on social media can help some survivors build confidence and gain social support.
  • Found that filtering algorithms (filter bubbles) can counteract societal filter bubbles, offering a positive role in challenging mainstream narratives for trauma survivors.
• Advantages: Compared to existing solutions, this study deepens the understanding of trauma survivors' disclosure behaviors in technological environments and provides specific, practical recommendations for platform design.
• Experimental or Evaluation Results: By mapping interview data to the "spinning top model," the study revealed how trauma survivors behave when technological designs align with their adaptive needs.
• Limitations and Future Directions:
  • The study sample was limited to the U.S., and cultural differences may not have been fully considered.
  • Did not deeply analyze how different stages of trauma recovery influence disclosure behavior.
  • Suggested future research to include quantitative experiments to validate the spinning effect model's applicability to other trauma survivor groups.

Design Recommendations

1. Improve anonymity features, such as using unique IDs in Facebook groups to protect user identities while allowing administrators to perform moderate monitoring.
2. Provide more flexible anti-personalization features, enabling users to control the familiarity of algorithm-recommended content.
3. Social media platforms should offer effectiveness data for trauma care tools and incorporate human moderation mechanisms to optimize their use.
4. Encourage the creation of virtual training or workshops to enhance the professional capabilities of group administrators and content moderators.
5. Introduce semi-automated trigger warnings and targeted filtering systems (e.g., FilterBuddy) to provide a safer environment for community management.

Conclusion

This study expands the understanding of trauma survivors' information disclosure behaviors in the field of Human-Computer Interaction, emphasizing the importance of social media technology design in supporting silent groups. Through an innovative theoretical model, new technological features, and specific design recommendations, the authors contribute practical suggestions for creating safer and more inclusive online spaces.