Predicting Detection Times with Perceptual Euclidian Distance. (September 2016)

• Record Type: Journal Article
• Authors: Steelman, Kelly S.; North, Hannah
• Is Part Of: Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting. Volume 60:Part 1(2016)
• Page Start: 1585
• Page End: 1589
• Publication Date: 2016-09
• Subjects: Human engineering -- Congresses; 620.8205
• DOI: 10.1177/1541931213601366
• Languages: English
• ISSNs: 1541-9312

Abstract

How should we select a set of symbol colors to optimize detection times? Here we suggest a simple, easy- to-calculate technique for predicting symbol detection times in cued and uncued visual search tasks. We used Perceptual Euclidian Distance (PED) to measure color similarity among symbols within the set (PEDset) and between each symbol color and the background color (PEDbg). Post hoc analyses of data from two previous change-detection experiments indicated that PEDbg was negatively correlated with detection time, but only in uncued visual search. PEDset, in contrast, was negatively correlated with detection time in cued search. In the current experiment, we designed a new symbology set that included three symbols that were equidistant in PED color space and a fourth symbol that was farther away in PED color space. We used this symbol set in a change detection experiment using the flicker paradigm. Consistent with the results of our previous analyses, we found that cued detection time was correlated with PEDset. Finally, we present a summary of data from seven experiments demonstrating that this pattern of effects holds over a variety of background colors and symbol sets. The overall results suggest that the PED may serve as an easy-to-use technique for selecting symbols that will facilitate particular performance objectives.