Jeffrey B. Vancouver - US grants

"This award is funded under the American Recovery and Reinvestment Act of 2009
(Public Law 111-5)."


Most formal models of decision-making treat it as a static process (i.e., how the current state of some set of variables affects the probability of a particular choice). This approach ignored the fact that each day individuals are faced with numerous decisions regarding the allocation of resources (e.g., time, attention, and effort) across competing tasks in order to meet or maintain many goals. It also ignores that fact that, over time, circumstances (both internal and external) change and learning occurs both of which may result in changes in the choices one makes to attain or maintain one?s goals. Finally, contemporary models of decision making are largely silent with regards to the underlying psychological and information processes that are responsible for observed decision patterns over time.
Under this grant, a group of researchers composed of an engineer and three psychologists use a simple information processing structure ? the weighted difference model common in engineering control theories ? to explain numerous decision-making, motivational, and economic phenomena. For example, this structure is used to construct models that explain why 1) individuals work longer on days when they are paid a relatively lower wage; 2) beliefs in one?s capacities to perform are sometimes negatively and sometimes positively related to the resources one allocates to the task; 3) individuals tend to switch back and forth between two tasks with the same deadline, but then focus on only one as the deadline draws near; 4) individuals change their risk attitudes when making choices as they approach their goal, and 5) individuals? change their choices over time despite no change in the options or their attributes. Several empirical studies are designed to test the proposed models. If found valid, the research will demonstrate that models based on this structure not only allow predictions over time (i.e., are dynamic), but also are consistent with empirically confirmed elements of well-accepted static models of decision making and motivation. Moreover, they integrate aspects of economic, psychomotor, and learning models, providing a general, comprehensive, yet parsimonious approach to understanding human behavior in naturalistic contexts.

Nontechnical Description of the Significance and Importance

Mathematics is the language of science, yet in the social sciences computational representations of the processes hypothesized to explain phenomena are rare. A primary reason for this is a lack of understanding of the value of formal, computational approaches and a lack of training in computational modeling strategies. This lack of appreciation and training is particularly true at the intersection of psychology and organizational science, which includes the disciplines of industrial-organizational psychology and organizational behavior. The workshop at the Ohio University brings together an international group of scholars from several sub-disciplines of psychology and organizational science to highlight the value and training a new set of scholars in computational modeling methodology.

Technical Description

The intellectual merit of the project arises from a series of presentations by computational modelers of difference types that will focus on the value of computational modeling for addressing specific phenomena and for facilitating the construction of a cumulative science focused on understanding human behavior at work. That is, computational modeling provides researchers with the means to explore dynamic psychological and contextual concepts mathematically and with simulations, including information regarding what study designs can be used to investigate the causal theories properly. Moreover, quantitative modeling techniques provide a vehicle for integrating levels of theory ranging from neurological to macro processes (e.g., economic activity). The broader impact of this project will be achieved via training of a cadre of new computational modelers and the publication of an edited collection of the papers presented by the established modelers and newly trained modelers. That is, as part of the training element of the workshop the trainees will develop and evaluate new computational models of specific phenomena. The best of these models will be submitted to journals for special issue consideration on computational modeling.