"I See Me Here": Mental Health Content, Community, and Algorithmic Curation on TikTok

Title of the Paper

“I See Me Here”: Mental Health Content, Community, and Algorithmic Curation on TikTok

Citation Information

• Subject Area: Mental health content and communities on social media platforms
• Keywords: Mental health, community, social media, interviews, TikTok, algorithmic recommendation, social computing, online communities, information quality, data insights

Research Background and Issues

• Identified Problems or Challenges:
  • TikTok hosts a vast amount of user-generated mental health content, which, while raising awareness about mental health issues, may also spread misinformation, encourage self-diagnosis, and potentially harm users' mental well-being.
  • TikTok's core algorithm (FYP, For You Page), while facilitating content dissemination, may lead to information overload, inaccurate information, and identity-related impacts.
• Significance:
  • Social media has become a critical source for young people to access mental health information and mutual support, with TikTok as a prominent example. Investigating how it shapes users' mental health knowledge and emotional states holds significant social and academic value.
• Research Motivation and Related Work:
  • The authors build upon prior studies on online mental health communities (OMHC) to explore how TikTok fosters a unique “sense of community” despite lacking formal community structures.
  • Existing research on mental health dissemination via social media has primarily focused on content analysis or public health perspectives, with limited exploration from the user experience angle.

Solution

• Main Methods:
  • Conducted semi-structured in-depth interviews with 16 TikTok users to explore their perceptions of mental health content and community on the platform.
  • Applied constructivist grounded theory to analyze users' identification and evaluation of community boundaries, algorithmic recommendations, and mental health topics.
• Innovation and Features:
  • Introduced the concept of “Permeable Communities” to describe how TikTok communities accommodate users' browsing and participation through flexible boundaries.
  • Highlighted the dual role of algorithmic recommendations (FYP) in shaping community formation and mental health content dissemination, revealing both positive impacts and potential risks.
• Implementation Steps:
  1. Recruit active TikTok users for in-depth interviews;
  2. Include video review tasks and visual-guided creation tasks in the interviews to explore users' understanding of content and community interactions;
  3. Conduct qualitative data analysis through coding, induction, and thematic extraction to summarize commonalities and differences in user experiences.

Research Findings

• Specific Findings:
  • Content Types: Users encounter diverse mental health content, including informational/clinical content, practical everyday content, and comforting content.
  • Ambiguity in Community Definition: Users have differing perceptions of whether TikTok constitutes a “community.” Some feel a sense of community, while others view it merely as a collection of content. Regardless, the “permeable community” characteristic facilitates cross-participation among users.
  • Tension Between Authenticity and Information Credibility: The authenticity and relatability of content attract users but may also obscure the spread of misinformation and creators' pursuit of “virtual fame.”
  • FYP's Runaway Effect: For long-term TikTok users, the FYP algorithm can become a “runaway train,” hindering personalized content adjustments and, in some cases, causing psychological discomfort.
• Advantages Over Existing Solutions:
  • Innovatively combines community theory with mental health content research to uncover the mechanisms of community interaction and mental health dissemination on TikTok.
• Experiments and Evaluation:
  • Collected diverse user perspectives, reflecting the complex ecosystem of TikTok users' mental health content consumption.
  • Compared to previous quantitative and content analysis studies, this research provides a more contextualized and in-depth understanding of user experiences.
• Limitations and Future Directions:
  • Limitations: The participant sample is skewed toward female and white demographics, limiting the ability to fully represent the diverse user perceptions on the platform; the algorithmic impact on mental health content was not further quantified.
  • Future Directions: Employ participatory design methods to develop more user-friendly recommendation algorithms and community management mechanisms; extend research to other social media platforms to compare the similarities and differences in community and mental health content ecosystems.

Conclusion

This study reveals the unique ecosystem of mental health content and communities on TikTok, proposing new approaches to improve social media platform design to reduce algorithmic harm and enhance mental health support for users. This framework not only helps understand how emerging social platforms shape users' sense of community but also provides insights for creating more inclusive and health-oriented social spaces.