Construct Validity of Agent-based Simulation:Normative Behaviour

We aim to assess the construct validity of agent-based simulations of normative behaviour with respect to social psychology. Although individual psychology is of great importance for the explanation of normative behaviour, social psychology is more appropriate as a theoretical backdrop of agent-based modelling.  Whilst individual psychology focuses on norm internalisation, social psychology focuses on the behavioural aspects and behavioural consequences of norms in a social setting, as do most agent-based simulations. We will take up this focus and see what agent-based simulation can contribute to the field of normative behaviour. In so doing, we aim to assess the extent to which current simulations are theoretically embedded and to discuss the potential contribution of social psychological theories in this theoretical embedding.

This is very much work in progress rather than a finished product. We will give a short introduction to relevant theories in social psychology followed by a discussion of two famous simulations of social norms, Axelrod (1986) and Epstein (1999). We will give an outline of the social psychology theories that are related to those models. We then discuss briefly one of the most recent normative behaviour simulations resulting from the ‘EMIL: Emergence in the Loop’ project. We finish with the most important part of this talk: YOUR FEEDBACK.