Who Does Not Benefit from Fact-checking Websites? A Psychological Characteristic Predicts the Selective Avoidance of Clicking Uncongenial Facts

Title of the Paper

Who Does Not Benefit from Fact-checking Websites? A Psychological Characteristic Predicts the Selective Avoidance of Clicking Uncongenial Facts

Paper Information

• Research Domain: Human-Computer Interaction and Psychology, focusing on misinformation and fact-checking
• Keywords: Fact-checking, Misinformation, Selective Avoidance, Psychological Factor, Click Bias, Interface Design, Effortful Thinking, Cognitive Reflection, Selective Exposure

Research Background and Problem

• Issues and Challenges:
  • In the context of rampant misinformation, even though fact-checking websites provide extensive accurate information, some users still avoid clicking on fact-checking content that contradicts their pre-existing beliefs.
  • This avoidance prevents certain users from correcting their erroneous beliefs, potentially fostering the continued spread of like-minded misinformation on the internet.
• Significance:
  • Understanding user behavior and designing decision-supportive interactive interfaces are critical for mitigating the spread and impact of misinformation.
  • Addressing the issue of selective avoidance is a key challenge in enhancing the effectiveness of fact-checking websites.
• Research Motivation and Related Work:
  • This study is based on the "selective exposure" theory, focusing on the impact of selective avoidance on user behavior.
  • While most previous studies have explored selective exposure in the context of political topics, this paper extends the investigation to the fact-checking environment and proposes methods to quantify avoidance and exposure behaviors.

Solution

• Proposed Method and Innovation:
  • Introduced the "Fact Avoidance/Exposure Index (FAEI)" to distinguish between selective avoidance and selective exposure, enabling a more precise measurement of user click behavior.
  • Conducted experiments to examine how psychological traits (e.g., reactive closed-mindedness) influence clicking and avoidance behaviors.
• Implementation Steps and Key Techniques:
  • Experimental Design: Conducted in two phases, involving online surveys with 506 participants to measure psychological traits and clicking behaviors.
  • Psychological Trait Measurement: Assessed participants using five psychological questionnaires (BSR, GCB, AOT, NFC, CRT).
  • Click Behavior Analysis: Provided a fact-labeled list of short and long information items, observing users' clicking and avoidance behaviors.
  • Data Evaluation: Calculated individual FAEI scores, grouped users into "fact-avoidance" and "fact-exposure" categories, and statistically analyzed the relationship between psychological traits and clicking behaviors.

Research Findings

• Specific Findings:
  • Approximately 43% of users tend to selectively avoid clicking on fact-checking content that contradicts their pre-existing beliefs, resulting in 93% of erroneous beliefs remaining uncorrected.
  • Psychological Trait Influence: Bullshit Receptivity (BS) significantly predicts users' selective avoidance behaviors, revealing that reactive closed-mindedness is an important psychological factor.
  • Design Evaluation: The proposed FAEI not only distinguishes user behaviors but also more accurately reflects individual differences compared to existing metrics.
• Advantages over Existing Solutions:
  • FAEI better differentiates between selective avoidance and selective exposure, avoiding the floor effect present in existing metrics.
  • Provides directions for interface design improvements tailored to different user characteristics.
• Experimental or Evaluation Results:
  • Structured experiments revealed that passive conflict detection provided by the interface triggered avoidance behaviors in some users.
  • The fact-exposure group was more likely to click on contradictory fact links compared to the fact-avoidance group.
• Limitations and Future Directions:
  • Limitations include constraints in the experimental design (e.g., requiring users to click on at least five links), which may affect the generalizability of the results.
  • The large number of items in the psychological questionnaires may impact user focus and require further optimization.
  • Future research should explore the impact of additional design elements on clicking behavior, such as "link position ranking bias" or "social media user interactions."

Summary and Design Guidance

• Practical Contributions:
  • Demonstrates the varying effectiveness of general fact-checking interface designs for different user types.
  • Highlights the necessity of targeted interventions in design, particularly to reduce reactive closed-mindedness in high-BS users.
• Future Design Directions:
  • Investigate how interface design can counteract selective avoidance behaviors, such as reordering links, adding social importance prompts, or implementing lightweight interventions.
  • Continuously integrate user psychological traits with design optimizations to enhance the overall effectiveness of fact-checking websites.