Identities have shaped the history of the recent centuries. Especially, identities bound to a territory and to political values have been a central force to shape politics. The mobilization of national identities has often led to the emergence of nation states. In contrast, the lack of a European identity among citizens of the European Union has lead to the exit of a member state from the political community. As many studies have focused on the importance of identities for politics, this dissertation contributes to answering the question of how these political identities develop, highlighting the role of political actions and social interactions for identity development. In the first paper I show that voting in the Brexit referendum has increased Remainers’ European identity but not at the cost of national identity. This shows that actions matter, bringing the abstract into concrete actions habitualizes an identity, bringing it more to the forefront. Additionally, this case demonstrates that a European identity does not need to come at the cost of a national identity. Even in situations where European and national identities are most likely been seen as incompatible, European identity growth does not result in the reduced identification with the nation state. These compatibility of transnational and national identities shows that even people who subscribe to a more cosmopolitan values can still have an attachment to their nation. What really makes the difference is whether a person has a European identity or not. However, are actions or information gain the driver of identity change in the aftermath of such highly salient events? In the second paper I show that most likely social influences rather than cognitive factual information gain is the driver of identity change. In this paper, I have investigated whether the rejection of the EU Constitution in France in 2005 has led to an increase knowledge on the contents of the EU Constitution in the European public. I show that even though all context conditions were in place for information gain to happen and the European public was relatively interested in the referendum the European public did not know more about the Constitution after the rejection than before. Even the Dutch, who had the most to gain by informing themselves, as they voted immediately after the French, did not increase their issue-specific knowledge. The same is true for the French, even though their own country has been the one who first rejected the Constitution. This paper demonstrates that even the rejection of a government backed proposal for an EU Constitution does not motivate people to get the facts right. This study renders an informational effect for the change in identities after the Brexit vote less likely. As a referendum unveils who is, from a socio-demographic perspective, the “typical” European or not. In paper 3, I investigate how the identification with the “typical” European is associated with a person‘s own European identity. To show that the more a person identifies with the social groups she perceives to support the EU, the more European the person feels herself, I introduce a new survey measures — the Group Closeness Score — that captures the social identities for the socio-demographic groups who sort into political camps. And in fact, my empirical findings suggest a strong empirical association between identification with the “typical” European and a person‘s own Europeanness. Therefore, my findings provide a mechanism for how the transnational-nationalist cleavage, the dividing line between more cosmopolitan orientated citizens and people who oppose further transnational integration, becomes contagious and might spread through society. People do not only care about the policy but also which fellow citizens are supporting (or opposing) the policy. However, as big, salient referenda are not a regular event, Dominik Schraff I show in the fourth paper co-authored with Dominik Schraff how the local context is positively associated with a person’s European identity. We argue that in neighborhoods in which there is a large share of a more mixed immigrant population, natives become much more likely to hold a more inclusive European identity next to national identity. We test this hypothesis with geo-coded survey data in the Netherlands. We find empirical evidence that natives in more diverse neighborhoods hold more stronger inclusive identities than people who live in less diverse neighborhoods. This pattern is driven by people having higher European identities rather than weaker their national identities. This paper shows that not only big events have a huge impact but daily small interactions also influence the boundaries and degrees of identities.