Using Agent-Based Modeling to Predict the Diffusion of Safe Teenage Driving Behavior Through an Online Social Network

Abstract

Objective: To use agent-based modeling to examine how peer influence in a social network affects behavior change among teenage drivers. Background: Teenagers are involved in more fatal driving crashes than any other age group. A solution to the problem is to utilize both feedback systems and online social networks to promote safe driving behavior. Methods: Using a combination of intention (from the Theory of Planned Behavior) and habit, agent-based modeling was used to predict the spread of safe driving behavior through an online social network after the implementation of a feedback system Results: The initial conditions of the social network (e.g., the number of agents to initially engage in safe driving behavior and the number of opinion leaders who use the feedback system), as well as the threshold used to determine intention drive safely had a significant effect on the final number of agents to engage in safe driving behavior. Conclusion: Agent-based modeling suggests that leveraging the power of feedback systems and social networks can lead to a positive change in teenage driving behavior. Application: Agent-based modeling can be a viable tool in predicting teenage driver behavior given the correct choice of parameters.