Complexity economics and agent-based modelling have great potential to represent and analyse processes of societal change. At the beginning of the 21st century, it is obvious that the advances in information technology and artificial intelligence will fundamentally change the economy and society. It is also clear that there must be a societal transformation towards a sustainable economy. These transformations are the result of the complex interactions between societal groups and individuals. A key problem in models of societal transformation is how to model the behaviour of individuals. Neither rational choice models nor models based on simple heuristics are convincing. I propose a behavioural model based on psychological research on values and attitudes in which attitudes towards an object determine an individual's behavior. It is an attractive feature of the model that attitudes can change due to social interaction. The change in attitudes and the resulting changes in behaviour are both prerequisites for and consequences of societal transformations.