Detection of colour signals in objects defined by luminance contrast

Abstract

AbstractPurpose The aim was to compare chromatic detection thresholds using two methods that are often employed to isolate the use of colour signals: a. Colour thresholds measured in dynamic luminance contrast (LC) noise when a spatially structured object is defined only by colour signals b. Thresholds for detection of colour changes in objects defined by luminance contrast (i.e., the luminance pedestal (LP) technique that is often used in visual psychophysics to isolate the use of colour signals) Earlier findings in patients with cerebral achromatopsia (Proc.R.Soc.Lond.B. 258 :327‐334 (1994)) suggest that different neural mechanisms handle the use of colour signals in the two conditions. It is therefore of interest to establish whether the corresponding colour thresholds in (a) and (b) are also different in normal subjects.Methods Colour detection thresholds were measured using both LP and LC noise methods for discrete LP contrast levels and background luminances. The same four‐alternative, forced‐choice staircase procedure was used to measure thresholds for every combination of target and background luminance.Results The amplitude of dynamic LC noise does not affect colour thresholds. In general, technique (a) yields smaller thresholds when compared to the LP technique (b) for all pedestal LC levels and background luminances. Negative luminance contrast pedestals yield increased colour thresholds for both the red/green and yellow/blue axes.Conclusion When chromatic signals are added to objects defined by negative luminance contrast, the corresponding colour thresholds are no longer processed independently and increase monotonically with luminance contrast.