The cognitive science view of humans as "cognitively limited information processors" has spread to many of the sister behavioral sciences, including political science. Few political scientists have thoroughly embraced this approach, however, to the extent of adopting its methodologies or critically examining its implications for the discipline. Even fewer political scientists are conducting the type of research that will advance our basic understanding of how people think about, feel, and process social information. One of the difficulties for political scientists pursuing an information processing perspective is the dearth of serious research methodologies that are well-suited to gathering detailed information processing data in the types of complex social environments that political scientists like to study. Moreover, few political scientists are trained in the type of experimental methods that are utilized in basic cognitive research. This project seeks to rectify this situation.
With prior support from the National Science Foundation, the principal investigators developed one such methodology, which they call dynamic process tracing, and software to implement the methodology, Dynamic Process Training Environment Program. This task has taken longer than was originally anticipated. Pretesting and error checking of "Version 1.0" of the software has already begun, as has a nascent dissemination program. This project will result in the completion of the software, its final testing, and its distribution.
Once the software has been completed and thoroughly tested, it will be publicized, making it and an associated programming manual freely available to others in the scientific community. The principal investigators will also provide training sessions at professional meetings to encourage other researchers to use the program.
??This project was designed to build an test a web-enabled portal for researchers interested in studying mass and elite decision making. The portal, publicly available at www.processtracing.org, allows researchers to design and field experiments in which information is provided to decision makers and the processes they use to examine and integrate the information are to be observed. The primary task of this project was the development of the software that operates the Dynamic Process Tracing Environment and extending its features to allow a wide range of research across multiple disciplines. Features of the system include the ability to track study participants information search and acquisition, to design studies that manipulate the flow of information and the number of alternatives and attributes from which decision makers can select, and to mimic social networking environments including the sharing and tagging of information. Data collected through experiments using this system can be used to help us better understand how people make decisions, leading to possible improvements in the way information is provided to decision makers. In collaboration with Profs. Jennifer Merrolla of the Claremont Graduate School, and Cindy Kam and Elizabeth Zechmeister of Vanderbilt University, as part of this project we also conducted a demonstration Multi-Investigator Multi-Site (MIMS) experiment using the new software. A MIMS project gathers together a number of collaborators with different ideas that can be tested simultaneously in the same experimental setting. A single experiment is developed with random manipulations designed to test all of those ideas, and then each collaborator runs an agreed-up number of subjects through the experiment at their home institution. The data from the different research sites are subsequently combined into a single dataset which is available to all of the multiple collaborators. This particular experiment involved a mock presidential election including both a primary election and a subsequent general election campaign between the two winning candidates from the primary. The experiment manipulated the emotional experience "setting ste stage" for the election campaign, the gender, ethnicity/race, and ideology of the candidates in each primary, the viability of the minority candidates, and the candidate pairing for the general election campaign. A 2-minute demonstration of the primary campaign can be observed at http://dpte.polisci.uiowa.edu/dpte/action/player/launch/616/?pass=JEPSDemo. Click on the SUBMIT button to view the demo. [Your computer must have Flash 9 Player, and the browser must be set to "allow pop-ups" (or at least not block them).] At least one publication has resulted from this MIMS project, and several additional papers are currently "under review." At this writing about 270 researchers have requested and been given free and open access to the system. Studies underway or concluded using it have looked at how voters choose between candidates, how leaders process information during a decision making task, and how the different ways of presenting political information influence the ways in which people respond to to the political environment.
