Understanding Empathy and Bystander Intervention in Cyberbullying: An Experiment Design of AI System Intervention

Cyberbullying has become a pervasive and severe issue on social networking sites (SNSs), necessitating effective intervention strategies. While prior research has focused extensively on technical solutions to tackle cyberbullying, The role of interface design in motivating bystander reporting behavior through empathy mechanisms has received limited attention. To address this gap, we propose an experimental design to examine the effects of AI-generated prompts on bystander empathy and their reporting behavior. Drawing on the perspective-taking view, we investigate how AI prompt-other and -self influence different components of empathy, i.e., emotional recognition and emotional resonance. Additionally, we explore the relationship between empathy components and bystander reporting behavior. For contributions, our study is expected to untangle the conflict assumptions in empathy-help relationships by proposing the coexistence of altruism and egoism. Our findings are expected to inform SNSs of the effective intervention strategy that leverages AI-generated prompts to motivate bystanders' proactive behavior of reporting cyberbullying.