Exploring human behaviour models through causal summaries and machine learning

This paper is a case study meant to demonstrate the relevance of causal summaries for exploratory analysis of human behaviour models. We broadly define a causal summary as a partition of the significant values of the analyzed variables (in our case the simulated motives fear and anger of human beings) into separate contributions by various “causing” factors, such as social influence or external events. We demonstrate that such causal summaries can be processed by machine learning techniques (e.g. clustering and classification) and facilitate meaningful interpretations of the emergent behaviours of complex agent-based models.