Purchase this article with an account.
Randall S. Birnkrant, Jeremy M. Wolfe, Melina A. Kunar, Matthew Sng; Is shininess a basic feature in visual search?. Journal of Vision 2004;4(8):678. doi: 10.1167/4.8.678.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
In visual search for a target among distractors, search is efficient (shallow slope of the function relating reaction time to set size) when attention is guided to the target by a basic feature like a unique color, size, or orientation. Does apparent shininess, produced by specular reflection, serve as a guiding feature? In Exp. 1, 14 Os searched static displays for a sphere made to appear “shiny” by an appropriately placed highlight. Matte distractors had a similarly located dark patch. A separate experiment obtained shininess ratings of these stimuli on a 9-pt scale from naïve observers. Shiny spheres were rated 5.2; matte, 1.4. Each shiny sphere was presented with an adjacent black dot. Matte spheres had adjacent white dots so that search could not be based on the presence of a particular dot polarity in the display. The task was easy, but both searches were inefficient (shiny target: 23 ms/item, matte target: 36 ms/item). The search rate asymmetry was reliable (p<0.01). In Exp. 2, 16 Os searched for shiny spheres taken from the work of Fleming, Dror, & Adelson, (2003) (Avg. rating 6.9). Distractors contained the same specular reflections ‘cut-up’ and replotted (Avg. Rating 3.4). Search remained inefficient (shiny targets: 26 ms/item, matte targets: 41 ms/item). The search asymmetry was marginally significant (p=0.052). These results suggest that shininess is not a guiding feature, but negative findings must be interpreted cautiously. Prior claims of feature status for shininess involved binocular luster (Wolfe & Franzel, 1988). Binocular luster is a dynamic stimulus (like binocular rivalry). Recent demonstrations of shininess from moving reflections are compelling (Hartung & Kersten, 2002). Perhaps such stimuli will produce efficient search for shiny targets. However, for the present, the conclusion must be that search for a shiny target is not efficient.
This PDF is available to Subscribers Only