9321236 Lau The vote decision is made in the context of a dynamic ever- changing information environment, and yet the chief methodology with which political scientists study elections, surveys, is a static one. This investigation explores the information acquisition and choice strategies employed by voters during political campaigns with a new research technique, a dynamic process tracing methodology, that allows the vote decision to be studied while it is being made. The essence of this methodology is to provide information about (fictitious) candidates in a manner that allows the researchers to record and study the way in which that information is gathered. This is accomplished by the use of modern, multi-media computer technology. Information about candidates (e.g., policy positions; background/personal characteristics) is stored in the computer; however, only brief "labels" for this information (Smith' stand on issue "A") are visible and the information must be physically "accessed" (by clicking a mouse) for it to be learned. The labels "scroll" down the computer screen so that only a limited amount of information is available at any given time. This scrolling emulates the dynamic, ongoing nature of political campaigns. The implications of this research for models of the vote decision and for basic theories of democracy are significant. A new methodology is developed that, for the first time, will allow a direct assessment of how voters, with more or less personal resources (i.e., expertise), gather information about candidates involved in political campaigns in more or less demanding situations. With this methodology scientists can explore how expertise and broad campaign strategies affect the ways information is gathered and combined into a decision, and how those factors affect the normative "correctness" of the vote choice. These are new questions for political scientists, and this methodology promises to open many new fertile research qu estions for study. ***
