Motivated Reasoning, Affect, and the Role of Memory in Voter Decision Making

Abstract

For much of the history of voting research, citizens have been viewed as less than well informed, paying little attention to politics and maintaining only a limited grasp on issues. When election day rolls around, votes are cast either in accordance with a party identification uninformed by issue content or simply based on group affiliations or the nature of the times (Campbell et al. 1960). This lack of competence is clearly evident in voters’ inability to respond effectively to open-ended survey questions designed to elicit opinions on the candidates of the day. More recently, however, political psychology has challenged this conventional wisdom. Presaged by Fiorina’s (198 1) idea of party identification as a rational retrospective evaluation of past party performance, and supported by a well-developed psychology literature on person-perception (Wyer and Srull 1980, 1986; Hastie and Park 1986), Lodge, McGraw, and Stroh (1989; Lodge, Steenbergen, and Brau 1995) argue that candidate evaluation is an online process, during which candidate affect is developed as information is encountered over time. Affect—represented by the online tally—is maintained in memory while the information that informs it can be safely discarded. When the time comes to vote, citizens need merely query the tally, with no memory search for information about the candidates. Thus, it should not be surprising that efforts to find extensive issue content in the memories of citizens should fail. Indeed, if voters process campaigns online, they could be considering far more information than they can recall for survey researchers. Voters could even be well-informed democratic citizens.