Purchase this article with an account.
Rick Gurnsey, Sharon L. Sally, Jeffrey A. Ball; Equating the “visibility” of luminance- and contrast-modulations. Journal of Vision 2002;2(7):210. doi: https://doi.org/10.1167/2.7.210.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose. Form discrimination thresholds are known to change with stimulus contrast. Therefore, care must be taken when comparing discrimination thresholds for the same property (e.g., orientation) conveyed via different media (e.g., luminance, chromatic or contrast modulations). In recognition of this issue, it is common practice to set stimuli to equal multiples of their respective detection thresholds when the discrimination task is performed. Our purpose is to demonstrate that the multiple of threshold technique embodies a flawed theory of perceptual scaling. We compared orientation discrimination for gratings defined by luminance and contrast modulations.
Method. (i) Detection thresholds were established for luminance and contrast modulations (referred to generally as contrasts). (ii) Contrast discrimination thresholds (JNDs) were determined for the first JND above detection threshold and many suprathreshold contrasts. (iii) A function was fit to the detection threshold and increment threshold data, in order to specify one JND above any given contrast. (iv) Orientation discrimination thresholds were determined for the first 10 JNDs above detection threshold for both luminance and contrast modulations. (v) Results were obtained for two experienced psychophysical observers.
Results. Luminance and contrast modulations that were equal numbers JNDs above their respective detection thresholds were not generally at equal multiples of detection threshold, indicating that the multiple of threshold technique embodies a flawed theory of perceptual scaling. Orientations discrimination thresholds decreased as stimulus contrast increased but the conclusions drawn about the differences between luminance and contrast modulations depended on the scale used (i.e., JNDs vs multiples of threshold).
Conclusions. Setting stimuli to equal multiples of detection threshold does not ensure that they have the same “visibility”, “salience”, “perceptual impact”, or perceptual contrast.
This PDF is available to Subscribers Only