Do negative advertisements demobilize voters? The answer to this question is crucial. If negativity is systematically turning people away from the polls, then the very nature of American political legitimacy is in question. Existing literature on the topic has produced a series of conflicting results. While some scholars show that negativity leads to a lower likelihood of turnout (Ansolabehere and Iyengar 1995), others find that negativity is actually mobilizing the electorate (Goldstein and Freedman 2002). Still others, however, show that the overall effect of negativity is null (Finkel and Geer 1008). In this dissertation I argue that we have reached this empirical stalemate because we have not adequately addressed the conditions under which individuals are exposed to negativity during a campaign. I claim that two factors shape the effect of negativity on turnout: (1) most individuals select which candidate they prefer long before Election Day and (2) negativity has a different effect before and after an individual has made this selection. Combining these factors, I use five independent empirical tests to show that negative advertising is most likely to demobilize at a certain time: after an individual has selected which candidate he prefers, but before an individual has had the chance to implement this selection with a vote.