Candidate uncertainty, mental models, and complexity: Some experimental results

Since the work of Downs (1957), spatial models of elections have been a mainstay of research in political science and public choice. Despite the plethora of theoretical and empirical research involving spatial models, researchers have not considered in great detail the complexity of the decision task that a candidate confronts. Two facets of a candidate’s decision process are investigated here, using a set of laboratory experiments where subjects face a fixed incumbent in a two-dimensional policy space. First, we analyze the effect that the complexity of the electoral landscape has on the ability of the subject to defeat the incumbent. Second, we analyze the impact that a subject’s “mental model” (which we infer from a pre-experiment questionnaire) has on her performance. The experimental results suggest that the complexity of a candidate’s decision task and her perception of the task may be important factors in electoral competition.