August 2014
Volume 14, Issue 10
Vision Sciences Society Annual Meeting Abstract  |   August 2014
Why do failures of lightness constancy take the form of gamut compression?
Author Affiliations
  • Alan Gilchrist
    Psychology Dept, Rutgers University
  • Stephen Ivory
    Psychology Dept, Rutgers University
Journal of Vision August 2014, Vol.14, 79. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Alan Gilchrist, Stephen Ivory; Why do failures of lightness constancy take the form of gamut compression?. Journal of Vision 2014;14(10):79.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

The pattern of failures of constancy provides a kind of signature of the visual software serving constancy. In scenes containing two or more regions of illumination (i.e., almost always) the range of lightness values is compressed relative to the actual range. To exploit this important clue, we measured gamut compression for a row of 5 target squares, of various gray shades, standing in a spotlight (30 X ambient) suspended inside a vision tunnel lined with checkerboard walls. By varying the luminance range of both the 5 squares and the checkerboard walls we produced 6 conditions (n=15, between subjects) used to test 5 stimulus metrics potentially underlying the compression. The amount of compression was predicted by the ratio of highest target luminance to highest checkerboard luminance (equivalent to perceived illumination difference), but not by overall luminance range nor by the formula in anchoring theory nor by two other metrics. Two additional experiments were run to test the hypothesis that the compression results from a lack of information about the spotlight/ambient illumination difference. To exclude any assumption that the ambient illumination applies to the 5 squares, they were moved into a second room and seen through an aperture, with luminances and sizes held constant. Compression was identical, suggesting that border ownership at the occlusion boundary enclosing the squares is not critical. To reveal exactly the spotlight/ambient illumination difference we placed the row within a rectangular beam of light projected onto the far wall of the original room. Rather than reducing the compression, this produced significantly more compression, suggesting that an occlusion boundary segregates frameworks of illumination better than a cast illumination boundary. Overall our results appear to require a major overhaul of anchoring theory, perhaps with both an anchor and a coefficient of scaling established within each framework.

Meeting abstract presented at VSS 2014


This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.