Simulating the EMergent Impact of Regulations Across cultures (SEMIRA)

The SEMIRA project studies the interaction between emerging normative behaviour and new regulations in different cultures and different countries. The goal is to support the preparation of new regulations of different government bodies. A theoretical framework will be developed based on work done on emerging norms on the one hand and formal frameworks on norm enforcement on the other hand. Two types of agent-based simulation tools will be developed based on it. One tool will validate the theoretical framework with actual people. The other will do large scale simulations based on emergent behaviour. The novelty in this approach is twofold. We will combine simulations with a few rich agents and simulations with many simple agents. Based on this rich (but computationally expensive) model we will build simulations with simple agents that will abstract away many of the details needed for the virtual environment simulation. The second novelty is that we will look at a combination of emergent patterns of behaviour that come from individual decisions in an environment with (new) imposed regulations and rules of behaviour in certain circumstances that are inherent in the culture of the society. We use a combination of top down and bottom up change of behavioural patterns. As a methodology we use a combination of empirical and simulation experiments. As use case the new smoking legislation in bars and restaurants is taken. Because the regulation is already promulgated it is possible to validate the results from our experiments against real data from practice. First, cultural influences on individual decision making are modelled. The resulting model will form the basis for a virtual reality simulation environment situated in a virtual bar. Based on this simulation we will build an agent based simulation platform that will contain hundreds of simplified agents. The simplified agents will, however, still exhibit the main characteristics for their behaviour as the agents of the simulated bar. We will populate it with the data from the different countries and validate the outcomes of the simulations against what happened in reality.