Behind the Meme: Understanding User Experiences with Memes on Social Media

Paper Title

Behind the Meme: Understanding User Experiences with Memes on Social Media

Publication Info

• Topic area: User experiences and privacy concerns related to memes on social media.
• Keywords: memes, social media, privacy, user comfort, tagging, photo-based memes, contextual integrity, consent, meme governance.

Background and Problem

• Problem / challenge: Existing research on memes has focused on detecting harmful content (e.g., hate speech, misinformation) but has overlooked privacy and contextual concerns arising from seemingly safe memes. There is limited understanding of how meme interactions affect user comfort and privacy.
• Significance: Memes are a prevalent form of online communication, but inappropriate sharing or tagging can lead to discomfort, privacy violations, and reputational harm. Addressing these issues is crucial for creating safer online environments.
• Motivation and related work: Prior studies have examined meme toxicity, face-based meme sharing, and contextual privacy in photos but have not explored broader meme content types or the perspectives of meme recipients. This paper addresses these gaps by focusing on user experiences with safe memes.

Solution

• Proposed approach: A mixed-methods study combining Reddit data analysis and a large-scale survey to investigate user comfort, privacy concerns, and expectations for meme management technologies.
• Novelty:
  1. Systematic exploration of privacy and discomfort caused by safe memes.
  2. Identification of contextual factors (e.g., sender identity, sharing scope) shaping user comfort.
  3. Recommendations for consent-based tagging, proactive photo protection, and context-aware meme governance tools.
  4. Public release of a contextual meme dataset for future research.
• Procedure and key techniques:
  1. Analyzed 2,317 Reddit posts to identify meme interaction patterns and user concerns.
  2. Conducted a survey with 324 participants, presenting 1,269 scenarios combining meme attributes, sender types, and actions.
  3. Used statistical modeling (CLMM) to analyze comfort levels and qualitative coding to explore user expectations for technological interventions.

Results

• Concrete findings:
  • Comfort levels were significantly influenced by sender identity, sharing scope, and meme attributes.
  • Face-based meme creation caused the greatest discomfort, especially when shared publicly or by strangers.
  • Memes with sensitive themes (e.g., emotions, drinking) heightened privacy concerns.
  • Demographic factors (e.g., education, gender, age, region) affected comfort, with European participants showing greater sensitivity to relational context.
• Advantage over baselines:
  • Unlike prior work focused on harmful memes, this study highlights privacy risks in safe meme interactions and provides actionable insights for platform design and policy.
• Experiments / evaluation:
  • Reddit data analysis achieved >93% classification accuracy for interaction types.
  • Survey scenarios covered diverse sender-action combinations and meme attributes, with responses analyzed using CLMM and qualitative coding.
• Limitations and future work:
  • Limited demographic diversity in the survey sample (predominantly Western, educated participants).
  • Focused on static memes; future work should explore video memes and broader cultural contexts.
  • Potential biases from self-reported data and scenario design.

Summary

This study investigates user experiences with memes on social media, focusing on privacy and contextual concerns. Analysis of 2,317 Reddit posts and a survey with 324 participants revealed that even safe memes can cause discomfort, particularly when they involve face-based content, sensitive themes, or inappropriate sharing contexts. Demographic factors and individual privacy preferences further shaped comfort levels. The findings highlight the need for consent-based tagging, proactive photo protection, and context-aware meme governance tools. Future research should expand to video memes, diverse cultural contexts, and cross-platform protections. The publicly released dataset supports further exploration of meme-related privacy issues.