Today, we are all Babushkas: Introducing the Bayesian model of identity integration

Roxane de la Sablonnière
Professor, Department of Psychology, University of Montreal

Samuel Mérineau
PhD Student, University of Montreal

Abstract: We live in a new, contagious and threatening world. Covid-19 continues to spread across borders and nations. In response to this threat, people need to modify their current ways of living. These changes are profoundly interwind with our concept of who we are. As such, changing our lives will modify deep aspects of ourselves: our identities. Several theories explained the process by which people change their identities through integrating a new identity. However, our current knowledge on the literature on identity integration rests on static theories and methods. We still struggle to understand the iterative dynamic internal process people go through each time they face a situation calling upon a new identity. In response to such challenges, we propose a probabilistic approach, the Bayesian model of identity integration. We test our model in two longitudinal studies – in the context of cannabis legalization in Canada (Study 1, N= 1 682) and in the context of COVID-19 pandemic (Study 2, N=3 617). In each study, our model produced predicted scores of identity integration based on Bayesian operations. Next, we linked these scores with responses of identity integration reported by participants at Time 3. Results show significant and positive relationships between our predicted and observed scores of identity integration. In addition to studies 1 and 2, we propose a third study using video games to deepen the understanding of identity integration. Our Bayesian model of identity integration has the potential to explain the dynamic process behind identity integration and help millions of people around the world dealing with identity change in times of dramatic social change.