Predicting the dynamic interplay between epidemiology, behavior, society, and policy that unfolds during a pandemic.
To contain and mitigate a novel pathogen threat, we need to act promptly, strategically, and cooperatively to mitigate the threat while minimizing negative socioeconomic and health impacts. Such resilient strategies should be based on a predictive understanding of the cascading behavioral and societal dynamics. They also require a broad, integrated, granular, and reliable surveillance system, reflecting the interacting states of our biological, social, and environmental systems.
Given the perceived and actual socioeconomic, physical and mental health costs of non-pharmaceutical interventions (NPI’s), they are often impeded by low levels of adherence, societal pushback, and community opposition. Tracking individual and communal perceptions and willingness to adhere to recommended countermeasures – medical and behavioral – and understanding the origins and drivers of such behaviors are vitally important to effective pathogen responses. COVID-19 has spurred advances in tracking, characterizing, and predicting pandemic-related behavior, including pandemic fatigue, risk perception, vaccine uptake, levels of trust in public institutions, and more. These studies leverage diverse data, such as self-reported behaviors in global surveys, natural language from social media, mobility statistics at the aggregate or individual level, and databases of NPIs enacted by governments. They have provided retrospective insight into behavioral dynamics over the first two years of the COVID-19 pandemic, and serve as a critical step towards: (i) fully elucidating the complex interplay between pathogen dynamics, public policy, and social behavior, (ii) creating predictive intelligence tools to provide situational awareness and forecasting of social behavior during future pathogen threats, and (iii) designing adaptive intervention strategies that anticipate and harness individual and collective behavior to mitigate pathogen risks.
Pilot Study 2.1 – Hackathon: Predicting mass gatherings from anonymized mobility data and their impact on infection dynamics. We will host a prediction challenge with the goal of forecasting the timing, location, size and catchment of mass gatherings from anonymized GPS tracking. We will specify allowable data and groups will be free to use any statistical approach and integrate other non-mobility behavioral, health, policy, or socioeconomic data. The contest will be open to students and postdocs and advertised on all collaborating campuses.
Pilot Study 2.2 – A testbed for investigating the interdependent dynamics of pathogens, social media, and precautionary behavior. We will develop an online testbed for investigating the interdependent dynamics of pathogens, social media, and precautionary behavior, which human players and bots can play.