Health and Habit: an Agent-based Approach

Veronika Kurchyna; Stephanie Rodermund; Jan Ole Berndt; Heike Spaderna; Ingo Timm

In: KI 2022: Advances in Artificial Intelligence. German Conference on Artificial Intelligence (KI), Springer, 2022.


Data-driven models with weak theoretical foundations for the examination of interventions and concepts to improve and maintain health lack explainability of results and suggestions. The use of agent-based models is a possible approach to remedy this issue. Modelling behaviour and the formation of habits using established theoretical psychological frameworks is a way of improving the utilisation of agent-based models when researching health-related questions. This paper proposes a concept implementing the Health Action Process Approach and the Social Cognitive Learning Theory to model the process of behaviour change within the Beliefs-Desires-Intentions Framework. The concept illustrates how an agent workflow can incorporate these psychological models and explain how social influence contributes to the formation of habits.

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence