Learning affects top down and bottom up modulation of eye movements in decision making
Files
(Published version)
Date
2013
Authors
Orquin, J.L.
Bagger, M.P.
Mueller Loose, S.
Editors
Advisors
Journal Title
Journal ISSN
Volume Title
Type:
Journal article
Citation
Judgment and Decision Making, 2013; 8(6):700-716
Statement of Responsibility
Conference Name
Abstract
Repeated decision making is subject to changes over time such as decreases in decision time and information use and increases in decision accuracy. We show that a traditional strategy selection view of decision making cannot account for these temporal dynamics without relaxing main assumptions about what defines a decision strategy. As an alternative view we suggest that temporal dynamics in decision making are driven by attentional and perceptual processes and that this view has been expressed in the information reduction hypothesis. We test the information reduction hypothesis by integrating it in a broader framework of top down and bottom up processes and derive the predictions that repeated decisions increase top down control of attention capture which in turn leads to a reduction in bottom up attention capture. To test our hypotheses we conducted a repeated discrete choice experiment with three different information presentation formats. We thereby operationalized top down and bottom up control as the effect of individual utility levels and presentation formats on attention capture on a trial-by-trial basis. The experiment revealed an increase in top down control of eye movements over time and that decision makers learn to attend to high utility stimuli and ignore low utility stimuli. We furthermore find that the influence of presentation format on attention capture reduces over time indicating diminishing bottom up control.
School/Discipline
Dissertation Note
Provenance
Description
Access Status
Rights
Copyright 2013 Jacob L. Orquin, Martin P. Bagger and Simone Mueller Loose. The authors license this article under the terms of the Creative Commons Attribution 3.0 License.