ECVP 2005 Abstract
doi:10.1068/v050444

Cite as:
Lovell P G, Tolhurst D J, Párraga C A, Baddeley R J, Leonards U, Trościanko J, Trościanko T, 2005, "Opponent channel responses to changes in the illuminant of natural scenes for primates and birds" Perception 34 ECVP Abstract Supplement

Opponent channel responses to changes in the illuminant of natural scenes for primates and birds

P G Lovell, D J Tolhurst, C A Párraga, R J Baddeley, U Leonards, J Trościanko, T Trościanko

Evidence suggests that the red - green opponent system is optimised for the task of detecting fruit against a dappled background of leaves. One reason the red - green opponent (RG) system aids the detection of fruit is its relative insensitivity to shadows and other variations in the illuminant. This is because the peak sensitivities of the L and M cones are close together; where the light becomes redder or bluer, the response of the RG system tends to vary only a little in comparison with the blue - yellow opponent (BY) channel. Using time-lapse recording with an LMS calibrated camera and a spectrometer, we examined the opponent responses to a scene containing foliage and fruit from dawn until dusk. Opponent responses were computed for primates (human) and birds (starling) using the Macleod - Boynton RG and BY equations. The contrast between the fruit and the surrounding region was assessed with d'. As expected, responses of RG opponent channels were more stable than those of BY channels under varying illuminants, and d' scores were higher. Furthermore, primate RG responses were more reliable than bird responses; this was due to the fact that bird cones are not only more widely spaced but also because the action spectra of the L and M cones have a narrower bandwidth.

[This Research was supported by BBSRC (ref S11501) and EPSRC/Dstl (ref RG/S56405/01).]

These web-based abstracts are provided for ease of seaching and access, but certain aspects (such as as mathematics) may not appear in their optimum form. For the final published version of this abstract, please see
ECVP 2005 Abstract Supplement (complete) size: 2356 Kb