About Engaging and Governing Strategies: A Thematic Analysis of Dark Patterns in Social Networking Services

Title of the Paper

Exploring User Engagement and Governance Strategies: A Thematic Analysis of "Dark Patterns" in Social Networking Services

Bibliographic Information

• Field of Study: Research on design ethics and "dark patterns" in user interfaces within the Human-Computer Interaction (HCI) domain
• Keywords: Social Networking Services (SNS), social media, interface design, dark patterns, user well-being, ethical interface design, user decision-making guidance, time management

Research Background and Issues

• Identified Problems or Challenges:

  1. Existing literature pays limited attention to dark patterns in Social Networking Services (SNS), with most research focusing on the e-commerce domain.
  2. User autonomy on social networks, particularly regarding personal data control and time management, is constrained by design strategies.
  3. Current laws and regulations primarily focus on data and privacy protection, often overlooking the harmful consequences of interface design.

• Significance: A deeper investigation into the use of dark patterns in social networks and industry-specific patterns can support the development of future user protection regulations and guidelines.

• Research Motivation:
  The authors aim to address the following questions:

  1. What types of dark patterns are currently employed by the four major SNS platforms (Facebook, Instagram, TikTok, and Twitter)?
  2. Are there dark patterns unique to the social networking domain?

Proposed Solution

• Proposed Approach:

  • Conducted detailed video recordings (16 hours) of the mobile applications of the four major SNS platforms, combined with user interaction logs, analyzed by six HCI experts.
  • Leveraged existing dark pattern classification frameworks to perform a thematic analysis of the collected interface elements and user interaction data, exploring SNS-specific or previously undescribed dark design patterns.

• Innovations:

  • Proposed two SNS-specific categories of design strategies: "Engagement Strategies" and "Governance Strategies."
  • Defined five new dark patterns: Interactive Hooks, Social Brokering, Decision Uncertainty, Labyrinthine Navigation, and Redirective Conditions.

• Implementation Steps and Key Techniques:

  1. Data Collection: Six HCI researchers interacted with the interfaces of the four platforms, recording videos and conducting concurrent think-aloud sessions.
  2. Coding and Analysis: Extracted existing dark pattern classifications; created new inductive codes for uncovered phenomena; used thematic analysis to derive final patterns and strategies.
  3. Categorization and Summary: Summarized dark patterns into two high-level strategy categories and five specific patterns, providing detailed example analyses.

Research Findings

• Specific Findings:

  • Identified 44 known dark patterns widely present in SNS (selected from 80 existing types).
  • Defined five previously undescribed dark patterns:
    1. Interactive Hooks: Attract users to engage and extend their time on the platform through reward mechanisms (e.g., infinite scrolling, autoplay content).
    2. Social Brokering: Interface features designed to guide users to expand their social network and establish more connections (e.g., friend suggestions, ambiguous content personalization settings).
    3. Decision Uncertainty: Weakening user autonomy in decision-making through design ambiguity and interference.
    4. Labyrinthine Navigation: Complex interface structures that hinder users from quickly finding desired features or settings.
    5. Redirective Conditions: Increasing the threshold for natural operations to push users toward actions intended by the designers.

• Advantages Over Existing Solutions:

  • A first attempt to systematically analyze dark patterns in the SNS domain.
  • Provides a high-level strategic framework ("Engagement" and "Governance") to understand the operational logic of dark patterns.

• Experiment and Evaluation Results:

  • Facebook and Instagram exhibited the most types of dark patterns, with 41 and 39 types respectively, followed by Twitter (35 types) and TikTok (37 types).
  • Case analyses revealed that these patterns in SNS are evidently aimed at enhancing engagement and restricting user decision-making.

• Limitations and Future Directions:

  1. Limited to the mobile applications of Facebook, Instagram, TikTok, and Twitter, without covering other social platforms.
  2. Data collection occurred during the COVID-19 pandemic, lacking in-person supervision during the study.
  3. Participants were restricted to the HCI research field; future studies could involve experts from cognitive science or psychology for joint analysis.
  4. Future research directions:
     • Extend the study to more SNS platforms and different devices (e.g., desktop versions).
     • Explore the relationship between dark patterns and user psychological biases.
     • Conduct interdisciplinary collaborations on user protection in interface design under broader regulatory contexts.

The above provides a structured summary of the key points from the paper A Thematic Analysis of Dark Patterns in Social Networking Services.