Moskotina R. Agent-based modeling as a tool for sociological theory formalization (on the example of the model of protest behavior)

This thesis is about agent-based modeling, a method allows us to figure out how actions and interactions of agents at the micro-level can lead to the emergence of macro-level social phenomena. Agent-based modeling is a method of theoretical analysis of complex systems. There are verbal theories (theories are expressed in natural language) and formal theories. Formal theory is the verbal theory is written in one or more formal languages (for example, a programming language). It makes sense to use formal theories in order to explain the behavior of complex systems. First, complex system is emergent. The properties of behavior of complex system cannot be reduced to the properties of the components of this system. Second, complex system is dynamic and its state changes over time. Third, complex system usually is nonlinear. Causation in the complex system cannot be described by a linear function. Agent-based modeling allows us to create agent-based models (there are formal theories presented as computer programs) and conduct computer experiments (simulations) with them. Building an agent-based model requires the formalization procedure. Formalization assumes that sociological theory expressed in natural language is presented as an agent-based model. However, it raises a number of questions. First, what sociological theories can be presented as agent-based models and how they should be formulated. Second, how to formalize sociological theory. Third, how to conduct computer experiments with an agent-based model. The thesis aims to answer these questions.