This study is concerned with the effects of party system polarization on citizens' voting behavior. Following a top-down perspective, polarization is understood here as a feature of the style of the political competition. In this framework, the main focus of my research is on its implications on the democratic connection, i.e. the link between the societal demands on the one hand, and the legislative and governmental action on the other. This work is structured as a collection of studies that tackle three different but interrelated effects. The first study is interested in the mechanical effects of a polarized political supply has on voters' propensity to support more than one party over time. The second study follows the same rationale, but is rather interested in whether polarization prevents voters to withdraw their support from the party they voted for in the past, even if they evaluate negatively its past performance. Finally, the third study is focused on what effects polarization have on voters' evaluation criteria, i.e. if it decreases or increases the extent to which voters consider ideology, competence and
partisan attachment as they evaluate political objects. Findings show that polarization reduces electoral competitiveness by making voters less willing to change their party support over time, even when they evaluate negatively the party they voted for previously. Moreover, higher party polarization is positively related to citizens' propensity to have a partisan attachment, and the way in which partisanship frames their evaluation of party ideology and competence.
