Files in this item



application/pdfBALLEW-THESIS-2020.pdf (3MB)
(no description provided)PDF


Title:Exploring the roles of display properties and display memorability in the contextual cueing task
Author(s):Ballew, Kirk Michael
Advisor(s):Lleras, Alejandro
Contributor(s):Simons, Daniel J
Department / Program:Psychology
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):contextual cueing
simulation study
visual search
Abstract:An underlying assumption of contextual cueing experiments is that repetition of a display will elicit contextual cueing, irrespective of the particular arrangement of items in the display. However, there is reason to doubt this assumption; a number of display properties have been shown to predict response times in visual search experiments, including target eccentricity, target quadrant, visual crowding, and visual clutter. The effects of display properties may therefore limit the interpretability of contextual cueing effects that do not control for differences in these variables. Further, display properties may be linked to the memorability of a display. If this is the case, then it may be possible for displays with certain properties to facilitate or inhibit context learning, because such displays are more or less memorable. The present study sought to estimate the impact of the confounding influence of display properties on RTs in the contextual cueing task. A multiple regression analysis of a large contextual cueing data set confirmed that target eccentricity, target quadrant, and visual clutter impact response times in the contextual cueing task. The first simulation study showed that significant group-level contextual cueing effects are unlikely to arise in the absence of context learning. However, display properties can influence the size of effects analyzed at the level of individual displays. A second simulation study investigated a variety of parameters that may influence RTs in the contextual cueing task. Surprisingly, the strength of association between display properties and display memorability was not predictive of contextual cueing. However, the mean of the memorability score distribution, the probability of a learned display showing an effect, and the magnitude of the effect were all predictive of contextual cueing. Crucially, interactions between these variables suggest that contextual cueing may reflect small effects that occur somewhat frequently or larger effects that occur infrequently, but not somewhere in between.
Issue Date:2020-07-08
Rights Information:Copyright 2020 Kirk Ballew
Date Available in IDEALS:2020-10-07
Date Deposited:2020-08

This item appears in the following Collection(s)

Item Statistics