Work

MOZAIC: A mathematical approach to formalizing behavioral theories – applied to an agent-based model of mobility

Year: 2026

Type: article

Abstract: The representation of human behavior in social simulation models, such as agent-based models (ABM), holds significant potential for addressing current societal challenges, particularly given the profound ecological impact humans have in the Anthropocene. However, implementing social science theories of human decision making in ABM poses challenges, such as the lack of standards for formalizing a given social science theory into computer code from its inevitably ambiguous natural-language formulation. This paper suggests drawing on the long tradition of formalization in mathematics and introduces the MOZAIC framework, “Mathematical fOrmaliZAtIon for deCision models,” as a stepping stone toward a formalized theory that can be exchanged, discussed, and contextualized across various cases and models. In a sequence of four steps—(1) setting the context, (2) defining agents, environment, and interaction structures, (3) defining decision model elements, and (4) formalizing the theory based on these—it guides a formalization process for arriving at a mathematical definition of a social science theory in a given model context. To illustrate the usefulness of the MOZAIC framework, we apply it to the case of mobility tool ownership decision making and formalize five example theories—rational actor, bounded rationality, habitual decision making, prospect theory, and descriptive norms. The resulting Mathematical Building Blocks (MBB) offer mathematical descriptions of the elements and structures of the theories. These descriptions allow for comparison of the theories and aim to be reusable in other ABMs by filling in context-specific details. Providing the perspective of mathematical language, our objective is to contribute to a discussion of community standards for formalizing social science theories in ABM.

Source: Frontiers in Social Psychology

DOI: 10.3389/frsps.2026.1772740

Authors: Stefanie Schutera, Geeske Scholz, Sarah Wolf

Institutions: Technische Universität Berlin, Freie Universität Berlin, University of Bremen, Global Climate Forum

Language: English

Topic: Multi-Agent Systems and Negotiation

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $2,353

Funders: Deutsche Forschungsgemeinschaft, Bundesministerium für Bildung und Forschung