Non-technical Description of the Project
In nearly every decision a person makes, they are required to combine multiple, sometimes conflicting sources information. Thus, understanding how people use these multiple sources of information is fundamental to understanding and predicting choices. Established approaches within psychology allow examination these processes in some situations. For example, to explore decision stages, a decision-maker may be asked to describe her thought process when making a choice, and that description is used to inform the researcher. In many situations, direct information about the choice processes is not accessible. For example, a decision maker may search her memory to find the information that could be used for choosing between options. When memory search is fast, people rarely have clear insight into how the memories were accessed and used. To allow for investigating decisions based on multiple sources of information across the widest range of situations, the principal investigators will integrate a powerful methodology based on mathematical cognitive modelling with standard decision-making research methodologies. The research will accomplish three objectives: (1) to gain an important understanding how multiple sources of information are combined during decision making, (2) conduct new empirical tests of the fundamental process across different decision making tasks, and (3) rigorously compare among the top decision making theories based on the new empirical findings. These results will inform decision makers, allow assessment of decision making efficiency, and enable better models of choice behavior.
Technical Description
Over the years, several decision making models were proposed that can account for many of the typical findings in human decision making, using both choice accuracy and response time output measures. However, no clear agreement about the underlying cognitive mechanism has been reached. These models can often be distinguished by strong implications about the processes that lead to a decision. Specifically, these models imply (a) how information is searched for, (b) when this information search is stopped, and (c) how the acquired information is integrated to reach the decision. Nonetheless, the disagreement remains. In part this is due to model mimicking: Without appropriate experimental design and analysis, different theories can make equivalent predictions. For example, inferences based on traditional mean response time (RT) analysis are limited due to model mimicking, as different decision-making models can imply exactly the same mean response time patterns. To overcome this limitation, we will integrate the powerful systems factorial technology (SFT) with traditional decision-making methods. SFT has been successfully applied in a wide range of perceptual and cognitive tasks to identify processing order, stopping rule and process dependency (a, b, and c from above), but has only been applied to decision making in very limited way. Our research will link SFT to the classical methods in judgment and decision making to improve our understanding of the cognitive processes underlying decision-making. Our project will also contribute to improving the professional academic development of undergraduate and graduate students at two state universities that traditionally have fewer research opportunities for students. All aspects of this research project, from methodology to theoretical development, will be communicated at national and international professional Furthermore, the research will be submitted for publication in high-quality, peer-reviewed journals. One of the aims of the proposed research is to disseminate the findings to a broader audience through our respective university press offices
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.