Agent Based Model of Anti-Vaccination Movements: Simulations and Comparison with Empirical Data

Area of Science:

• Sociology
• Computational Social Science
• Public Health

Background:

• Understanding vaccine hesitancy is complex due to psychological and social factors.
• The internet and social media play a crucial role in the spread of anti-vaccination (AV) sentiment.
• Analyzing AV movement dynamics requires understanding online information dissemination.

Purpose of the Study:

• To analyze long-term internet activity related to the AV movement in the US and Poland.
• To develop an Agent-Based Model (ABM) simulating AV movement dynamics.
• To evaluate strategies for combating the AV movement.

Main Methods:

• Big Data analysis of multiyear internet activity preceding the COVID-19 pandemic.
• Development of an Agent-Based Model (ABM) incorporating factors like medical contact, public campaigns, interpersonal communication, and infosphere influence.
• Simulation of AV movement dynamics, contrasting rational pro-vaccination information with emotional AV propaganda.

Main Results:

• Internet activity shows nonlinear, short-lived, high-intensity peaks characteristic of the AV movement.
• ABM simulations closely replicate observed AV interest patterns.
• Small changes in parameters like willingness to post AV content can cause significant drops in vaccination support due to nonlinear feedback.
• Standard pro-vaccination campaigns have limited impact; targeted, emotional online campaigns and content removal are more effective but carry social costs.

Conclusions:

• The AV movement's dynamics are nonlinear and driven by emotional appeals amplified by social media.
• Effective AV counter-strategies require emotionally resonant online engagement or content moderation.
• Potential social and political costs of AV counter-measures, such as polarization and martyrdom narratives, must be considered.