Causal Analysis of an Agent-Based Model of Human Behaviour

This article investigates causal relationships leading to emergence in an agent-based model of human behaviour. A new method based on nonlinear structural causality is formulated and practically demonstrated. The method is based on the concept of a causal partition of a model variable which quantifies the contribution of various factors to its numerical value. Causal partitions make it possible to judge the relative importance of contributing factors over crucial early periods in which the emergent behaviour of a system begins to form. They can also serve as the predictors of emergence. The time-evolution of their predictive power and its distribution among their components hint at the deeper causes of emergence and the possibilities to control it.

[1]  L. Hluchý,et al.  Causal Analysis of the Emergent Behavior of a Hybrid Dynamical System , 2014 .

[2]  Joseph Y. Halpern,et al.  Causes and Explanations: A Structural-Model Approach. Part II: Explanations , 2001, The British Journal for the Philosophy of Science.

[3]  Ken R. McNaught,et al.  Design classes for hybrid simulations involving agent-based and system dynamics models , 2012, Simul. Model. Pract. Theory.

[4]  L. Hluchy,et al.  Towards causal analysis of data from human behaviour simulations , 2012, 2012 4th IEEE International Symposium on Logistics and Industrial Informatics.

[5]  Peter Spirtes,et al.  Introduction to Causal Inference , 2010, J. Mach. Learn. Res..

[6]  Joseph Y. Halpern,et al.  Causes and Explanations: A Structural-Model Approach. Part I: Causes , 2000, The British Journal for the Philosophy of Science.

[7]  Christoph Urban,et al.  PECS: A Reference Model for the Simulation of Multi-Agent Systems , 1997, Tools and Techniques for Social Science Simulation.

[8]  Dolores Cañamero,et al.  Modeling motivations and emotions as a basis for intelligent behavior , 1997, AGENTS '97.

[9]  Joachim von Puttkamer.  Editors , 2018, Journal of Modern European History.

[10]  Michal Wrzeszcz,et al.  A Novel Way of Using Simulations to Support Urban Security Operations , 2015, Comput. Informatics.

[11]  Kathleen M. Carley,et al.  Balancing the criticisms: Validating multi-agent models of social systems , 2008, Simul. Model. Pract. Theory.

[12]  L. Hluchy,et al.  Summarizing the behaviour of complex dynamic systems , 2013, 2013 IEEE 11th International Symposium on Applied Machine Intelligence and Informatics (SAMI).

[13]  Michael Luck,et al.  Agent technology, Computing as Interaction: A Roadmap for Agent Based Computing , 2005 .

[14]  R. W. Rogers,et al.  Deindividuation and the self-regulation of behavior. , 1989 .

[15]  J. Pearl.  Causality: Models, Reasoning and Inference , 2000 .

[16]  L. Hluchy,et al.  Exploring human behaviour models through causal summaries and machine learning , 2013, 2013 IEEE 17th International Conference on Intelligent Engineering Systems (INES).

[17]  D. Cliff,et al.  Complexity and emergent behaviour in ICT systems , 2004 .

[18]  Bibb Latané,et al.  Dynamic Social Impact , 1996 .

[19]  B. Latané.  Dynamic Social Impact: The Creation of Culture by Communication. , 1996 .