Purchase this article with an account.
Dingcai Cao, Hao Sun, Barry B. Lee; The Weighting Of Rod And Cone Inputs To Primate Magnocellular Ganglion Cells. Invest. Ophthalmol. Vis. Sci. 2011;52(14):4568.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Recently we reported that rod and cone inputs to magnocellular (MC) ganglion cells were linearly summed prior to a saturation site [Cao, Lee & Sun, JOV, 10(11):4, 1-15, 2010]. Here we used a newly developed technique [Sun, Smithson, Zaidi & Lee, J. Neurophysiol, 95(2), 837-849, 2006] to further estimate the magnitudes of rod inputs to retinal MC ganglion cells relative to cone inputs at various retinal illuminance levels.
The responses of parafoveal MC ganglion cells were recorded from macaque retina using an in vivo preparation. Isolated rod or cone modulation was generated using a four-primary photostimulator that can control rod and cone stimulations independently. The stimulus was an 8-deg diameter uniform field of which the photoreceptor excitation was modulated around circumferences of a rod-cone plane in clockwise and counterclockwise directions at different temporal frequencies (1.22, 2.44, 4.88, 9.76 and 15.2 Hz) and retinal illuminance levels (0.2, 2, 20 and 200 Td). The preferred response vector in the rod-cone plane was used to estimate the relative weights of rod and cone inputs for each cell, with the relative weight 0 for complete cone dominance and 1 for complete rod dominance.
The relative weights of rod inputs to MC ganglion cells increased as the retinal illuminance level decreased: the averaged rod weight was 0.41 (SD = 0.22) at 200 Td, 0.60 (SD = 0.23) at 20 Td, 0.81 (SD = 0.13) at 2 Td or 0.82 (SD = 0.08) at 0.2 Td, respectively. The rod weights did not vary with temporal frequencies up to 15.2 Hz, where cells showed little rod response and estimation of rod inputs became unreliable. At higher retinal illuminance levels (20 and 200 Td), the weights of rod inputs to MC ganglion cells tend to increase with cell eccentricity, but this trend disappeared at low retinal illuminance (0.2 and 2 Td) where rod inputs were dominant regardless of cell eccentricities.
Relative rod inputs increased continuously as the retinal illuminance levels decreased, suggesting a smooth transition from cone vision to rod vision.
This PDF is available to Subscribers Only