A software architecture for mechanism-based social systems modelling in agent-based simulation models

Area of Science:

• Social Sciences
• Computational Social Science
• Sociology

Background:

• Analytical sociology emphasizes middle-range theories to explain social phenomena.
• Agent-based modeling is a key tool for simulating complex social dynamics.
• Integrating diverse social theories into computational models presents significant challenges.

Purpose of the Study:

• To introduce the Mechanism-Based Social Systems Modelling (MBSSM) software architecture.
• To provide a unified framework for expressing and implementing social theory mechanisms within agent-based models.
• To facilitate the comparison and integration of different social theories.

Main Methods:

• The MBSSM architecture is designed using object-oriented programming principles and Unified Modelling Language (UML) diagrams.
• It supports the representation of individual behavior components that drive social mechanisms.
• Worked examples demonstrate its application in modeling population-level alcohol use dynamics using norm theory and combined norm-role theories.

Main Results:

• The MBSSM architecture successfully integrates individual behavior mechanisms derived from social theories into agent-based simulations.
• It provides a computational environment for representing, comparing, and integrating social theories.
• The architecture enables testing families of theories for their explanatory power regarding concrete social phenomena.

Conclusions:

• MBSSM offers a powerful computational framework for advancing analytical sociology and agent-based social simulation.
• It facilitates abductive reasoning by allowing systematic testing of theoretical explanations for social dynamics.
• The architecture supports the development of more sophisticated and integrated models of social behavior and systems.