Influence or Deception? Evaluating Social Suggestions with Persuasive Statements for Security and Privacy Settings

Paper Title

Influence or Deception? Evaluating Social Suggestions with Persuasive Statements for Security and Privacy Settings

Publication Info

• Topic area: The impact of social and authority-based suggestions, combined with persuasive statements, on user decision-making in security and privacy settings.
• Keywords: Security and privacy settings, persuasive design, social proof, authority-based suggestions, deceptive patterns, user decision-making, transparency, ethical design, Elaboration Likelihood Model, user resilience.

Background and Problem

• Problem / challenge: Configuring security and privacy (S&P) settings is cognitively demanding for non-expert users. While suggestions based on social proof or authority can assist users, they can also be exploited as deceptive patterns to steer users toward less-protective settings.
• Significance: Misleading S&P suggestions can compromise user autonomy and privacy, raising ethical concerns and practical risks in digital environments.
• Motivation and related work: Previous studies have examined the effects of social proof, authority-based suggestions, and persuasive statements independently, but their combined effects, particularly in deceptive contexts, remain underexplored. This study addresses this gap by analyzing how these techniques influence user decision-making and are perceived when deception is revealed.

Solution

• Proposed approach: An empirical study using a 2 × 2 × 2 factorial design to investigate the combined effects of suggestion sources (Public vs. Experts), persuasive statements (Present vs. Absent), and suggestion purposes (Honest vs. Deceptive) on user decisions in S&P settings.
• Novelty:
  1. Analysis of combined effects of social proof, authority-based suggestions, and persuasive statements in both honest and deceptive contexts.
  2. Examination of user evaluations of deceptive suggestions after their nature is revealed.
  3. Empirical grounding in the Elaboration Likelihood Model (ELM) to interpret persuasion mechanisms.
  4. Recommendations for transparent and ethical S&P interface design.
• Procedure and key techniques:
  • Conducted an online survey with 1,433 U.S. participants.
  • Participants configured 13 S&P settings for a fictional social media service under different experimental conditions.
  • Suggestions were framed as coming from either "Public" or "Experts," with or without accompanying persuasive statements, and were either honest or deceptive.
  • Participants rated the helpfulness of suggestions and their acceptability post-debriefing.

Results

• Concrete findings:
  • Honest suggestions with persuasive statements increased adherence rates (e.g., 83.40% adherence in Experts-Honest-Statement condition vs. 60.18% in the control group).
  • Deceptive suggestions reduced adherence but were partially mitigated by persuasive statements (e.g., 68.83% adherence in Experts-Deceptive-Statement condition).
  • Perceived helpfulness strongly correlated with adherence, with expert-framed suggestions rated as more helpful than public-framed ones.
• Advantage over baselines:
  • Expert suggestions had a greater influence than public suggestions (e.g., 18.95% higher adherence for Experts-Honest compared to control).
  • Persuasive statements amplified the impact of both honest and deceptive suggestions.
• Experiments / evaluation:
  • 2 × 2 × 2 factorial design with eight experimental groups and one control group.
  • Metrics included adherence rates, perceived helpfulness, and acceptability ratings.
  • Logistic regression analyses identified significant effects of suggestion source, persuasive statements, and deception.
• Limitations and future work:
  • Limited ecological validity due to the use of a fictional platform.
  • Lack of qualitative insights into participants' reasoning.
  • Results may not generalize across cultures; future studies should explore cross-cultural differences.
  • Need to examine long-term effects and interactions with default settings.

Summary

This study investigates how social proof- and authority-based suggestions, combined with persuasive statements, influence user decisions in security and privacy settings under honest and deceptive conditions. Findings show that persuasive statements amplify adherence to suggestions, even when deceptive, and that expert-framed suggestions are perceived as more helpful than public-framed ones. However, participants consistently judged deceptive practices as unacceptable. The study highlights the need for transparent and rational S&P interface designs and complementary user education to foster resilience against manipulative designs. Future research should explore cross-cultural generalizability and long-term effects of persuasive techniques.