Seen but Ignored: Understanding User Disengagement from Emergency Alerts in High-Frequency Contexts — A Case Study of South Korea

Paper Title

Seen but Ignored: Understanding User Disengagement from Emergency Alerts in High-Frequency Contexts — A Case Study of South Korea

Publication Info

• Topic area: User disengagement from high-frequency emergency alert systems
• Keywords: Public Warning Systems, alert fatigue, disengagement, Protective Action Decision Model, high-frequency alerts, trust erosion, Cry Wolf Effect, information overload, user typologies, South Korea

Background and Problem

• Problem / challenge: Existing emergency alert systems fail to account for user disengagement caused by high-frequency alerts, leading to psychological detachment, alert fatigue, and reduced protective actions.
• Significance: Understanding disengagement is critical for ensuring public safety, as disengaged users are less likely to act during actual emergencies, leaving them vulnerable.
• Motivation and related work: Previous research has focused on improving alert dissemination technologies and message design but has largely overlooked disengaged users. This study addresses this gap by exploring disengagement as a dynamic, adaptive process influenced by cumulative exposure and trust erosion.

Solution

• Proposed approach: A qualitative study examining user disengagement from South Korea’s high-frequency Cell Broadcast System (CBS) alerts, using the Protective Action Decision Model (PADM) and Extended Parallel Process Model (EPPM) as frameworks.
• Novelty:
  1. Introduces a three-part typology of user responses: Responders, Ignorers, and Blockers.
  2. Identifies structural barriers like actionless search and differential adaptations to the Cry Wolf Effect.
  3. Extends PADM to account for cognitive anchors and schemas in high-frequency contexts.
  4. Proposes design implications for adaptive, user-sensitive alert systems.
• Procedure and key techniques:
  1. Conducted a qualitative study with 37 participants in South Korea, categorized into Responders, Ignorers, and Blockers based on alert engagement patterns.
  2. Participants reviewed 16 representative CBS alerts and participated in semi-structured interviews.
  3. Data were analyzed using thematic coding, integrating PADM and EPPM frameworks to map disengagement pathways.

Results

• Concrete findings:
  • 83.8% of participants rated alerts with clear location, hazard type, and guidance positively.
  • Repeated exposure led to disengagement mechanisms like actionless search (information seeking without action) and the Cry Wolf Effect (trust erosion from false alarms).
  • User types exhibited distinct disengagement pathways: Responders maintained trust, Ignorers delayed action, and Blockers opted out entirely.
• Advantage over baselines: Highlights the inadequacy of current message-centric models by showing how cumulative exposure reshapes user engagement, offering a more nuanced understanding of disengagement in high-frequency contexts.
• Experiments / evaluation:
  • Participants reviewed CBS alerts and evaluated message components (e.g., hazard, location, guidance).
  • Semi-structured interviews explored cognitive and emotional pathways of engagement.
  • Data analysis revealed user-specific trajectories and systemic barriers to action.
• Limitations and future work:
  • Findings are based on South Korea’s unique high-frequency alert context, limiting generalizability.
  • User typologies were derived from qualitative data and require validation through large-scale surveys.
  • Future studies should explore hazard-specific dynamics and test proposed design interventions in real-world settings.

Summary

This study investigates user disengagement from high-frequency emergency alerts in South Korea, identifying three user types—Responders, Ignorers, and Blockers—each with distinct disengagement pathways. Key findings include the role of cognitive anchors, actionless search, and differential adaptations to the Cry Wolf Effect. The study extends PADM to account for cumulative exposure and proposes design strategies like tiered information architectures and context-aware framing to improve user engagement. These insights inform the design of adaptive, user-sensitive Public Warning Systems that sustain trust and resilience in high-frequency environments.