Synthetic Information Environments for Policy Informatics: A Distributed Cognition Perspective

Socially-coupled policy domains, such as public health, are influenced by the collective behavior of millions of individuals who participate in them and respond to policy plans and interventions. Policy-making for these systems is challenging because their complexity and interdependencies that can have cascading consequences. The promise of policy informatics has been to provide computational tools that make managing information easier and to facilitate more efficacious policy planning by incorporating methodologies such as simulation. Here, we argue that we need to do more. Successful policy informatics has to take into account the differing perspectives and motivations of different stakeholders in the policy-making process. We describe synthetic information environments as a systematic solution to this problem by presenting a distributed cognition perspective. Synthetic information is constructed by combining information from multiple unstructured data sources. A synthetic information environment provides models for forecasting and for in-silico experimentation with intervention policies, while also maintaining and propagating constraints between different stakeholder views. When viewed as cognitive augmentation technology, we see that such an environment draws multiple stakeholders into a single distributed cognitive system, thereby enabling efficacious interaction and planning. We describe two different case studies in the context of pandemic preparedness planning which illustrate the use of our system.