Purchase this article with an account.
D. Cao, J. Pokorny, V.C. Smith; Matching rod percepts with cone stimuli . Invest. Ophthalmol. Vis. Sci. 2004;45(13):2319.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: At mesopic light levels, rod input enhances brightness and decreases saturation. Literature reports on hue are contradictory. To address this issue, we matched the appearance of changes in rod activation with changes in cone activation. We used a 2–channel photostimulator with 4 primaries in each channel that allowed independent control of the stimulation of each photoreceptor type of the human retina. Methods: We used two stimulus patterns: (1) A 2° central field within a 10° surround, located at 6° in the temporal retina. The cone signals in the center and surround were kept identical but the rod signal in the center was incremented in a 1 Hz temporal square–wave. The effect of rod stimulation on color was measured by a temporal matching technique, in which the observer adjusted cone signals of the center during a matching epoch to match the rod percept seen during a stimulus epoch. (2) A 5°– 10° annulus was presented with a fixation point at the center. The rod signal was modulated with a step function profile with a cycle of 2 s: the first 0.5 s served as the stimulus epoch in which the rod signal was higher than the rod signal in the matching epoch; the following 1.5 s served as the matching epoch in which the observer adjusted cone signals to match the percept of the annulus in the stimulus epoch. For both stimulus patterns, there were 6 chromaticities at 7 retinal illuminances (1–160 photopic td). The rod signal Weber contrast was fixed at 0.3. Observers also characterized perceived rod hue with matches to Munsell chips. Results: For both stimulus patterns, the matching L/(L+M) was smaller than, and was linearly related to, the stimulus L/(L+M) at all light levels, implying greater M–cone contribution. With the center–surround pattern only, at the lowest light levels, 1–2 td, the matching S/(L+M) was larger than, and was linearly related to, the stimulus S/(L+M). To match the 0.3 rod Weber contrast, at the lower light levels, the matching luminance (L+M) Weber contrast was 0.15–0.3. Matching cone Weber contrast decreased exponentially with luminance level. Matches of rod percepts to Munsell chips indicated an increase of "greenness" or decrease in "redness". The 1–2 td condition with Stimulus Pattern (1) also showed an increase in "blueness". Conclusions: The experiment showed that observers can match rod percepts with cone stimuli. Effects are most pronounced at 1–2 tds and suggest that the color percepts are "greenish" and "bluish".
This PDF is available to Subscribers Only