NEWS Making decisions that affect large groups of people can be challenging as we all relate to norms in different ways. In his doctoral work, Christian Kammler at Umeå University has studied how social simulations can help decision-makers understand how their decisions affect people, specifically based on individuals' perspective on norms.
Christian Kammler, Doctoral student at the Department of Computing Science, Umeå University
ImagePrivate"It is challenging for decision-makers to make well-founded decisions because some people follow the norm while others try to circumvent it. There are, of course, also those who completely break the norm," says Christian Kammler, doctoral student at the Department of Computer Science, Umeå University.
Christian’s doctoral work has led to an agent-based decision system, where virtual agents—simulated individuals or groups—interact within social simulations to model real-world behaviors. This system helps explore how people respond to policies from different perspectives, providing insights into societal dynamics.
"Decision-makers need intuitive tools that help them change and create new norms, understand how people behave, and analyze the potential impacts of their intended policies", says Christian.
Christan’s system relies on three key elements to determine how individuals react to a norm: needs, which drive what we do right away; values, which guide our behavior on a larger scale; and social affordances, which help individuals understand what actions are possible based on how they see the world.
Models using Christian’s agent-based decision system allow decision-makers to see the consequences of their decisions before they are implemented and can be adjusted accordingly. In this way, potentially negative consequences can be detected in good time.
"A central aspect of this system is the ability for decision-makers to modify norms in real-time within the simulation. This feature makes it possible to test and adjust policies dynamically and observe potential effects before they are put into practice," says Christian.
Christian Kammler is a researcher within the national research program Wallenberg AI, Autonomous Systems and Software Program – Humanity and Society (WASP-HS). WASP-HS enables cutting-edge research, expertise, and capacity building in the humanities and social sciences on how artificial intelligence affects humanity and society and vice versa. The program is coordinated from Umeå University. Read more about WASP-HS.
Christian Kammler, Department of Computer Science, Umeå University, will defend his doctoral thesis "Modeling Norms for Social Simulations: Increasing Realism in Social Simulations to Support Decision-Makers in Their Decision-Making" on Friday, April 4, 2025. The faculty opponent is Javier Vázquez-Salceda from Universitat Politècnica de Catalunya, Spain.
Read the full thesis.