Media-driven adaptive behavior in pandemic modeling and data analysis

1

Human behavior and public attitudes towards preventive control measures, such as personal protection, screening, isolation, and vaccine acceptance, play a crucial role in shaping the course of a pandemic. These attitudes and behaviors are often influenced by various information sources, most prominently by social media platforms.

The primary information usually comes from government bodies, e.g. CDC, responsible for public health mandates. However, social media can amplify, modify, or distort this information in numerous ways. The dual nature of social media can either raise awareness and encourage protective behaviors and reduce transmission, or have the opposite effect by spreading misinformation and fostering non-compliance.

To analyze the interplay between these components, we have developed a coupled SIR-type dynamical model that integrates three essential components: (i) disease spread, as reported by official sources; (ii) the response of social media to this information; and (iii) the subsequent modification of human behavior, which directly impacts the spread of disease.

To calibrate and validate our model, we utilized available data sources on the Covid-19 pandemic from a one-year period (2021-2022) in the United States, as well as data on social media responses, particularly tweets. By analyzing the data and conducting model simulations, we have identified significant inputs and parameters, such as initial compliance levels and behavioral transition rates. These factors enable a quantitative assessment of their contributions to disease outcomes, including cumulative outbreak size and its dynamic trajectory.

This modeling approach gives some valuable insights into the relationship between public attitudes, information dissemination, and their impact on the progression of the pandemic. By understanding these dynamics, we can inform policy decisions, public health campaigns, and interventions to effectively combat the spread of Covid-like pathogens and future pandemics.