In the case of infectious diseases, the interaction between human behavior and disease dynamics can greatly influence the resulting outbreak size and characteristics. People behave differently depending on the information available, so the influence of local and global information about a virus outbreak is studied. Agent-based modeling provides a tool for representing individual behavioral differences, but its calibration is complex. Several factors influence model results, such as the network structure representing social relationships. Using an exploratory methodology called “Iterative Exploratory Modeling”, the assumptions used for a particular agent-based model are tested and verified, demonstrating the benefits of said approach. While genuine disease dynamics can be achieved with the used graph based on the Barabási-Albert model, the specific graph attributes have a much larger impact on the results than the local and global information. This emphasizes the need for testing assumptions at a fundamental level, which “Iterative Exploratory Modeling” can provide.

Keywords: Information - Dynamics - Simulation - Agent-Based Modeling - Multi-Agent Simulation - Epidemiology - Social Networks - Graphs - Network Aspects - Social Influence - Local vs. Global Information