Purchase this article with an account.
Michael Barnett, Geoffrey Aguirre; A spatial model of human retinal cell densities and solution for retinal ganglion cell displacement. Journal of Vision 2018;18(10):23. doi: 10.1167/18.10.23.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Retinal ganglion cells (RGCs) are radially displaced from their receptive fields within ~20°r (retinal degrees) of the fovea, with the magnitude varying both by eccentricity and polar angle. Correction for this displacement is needed to relate measurements of the RGCs to measurements of the cones, perception, or cortex. Theoretically, displacement magnitude may be derived by relating the number of midget RGCs in a retinal patch to the number of midget receptive fields (RFs) at a corresponding visual field location (Drasdo et al., 2007). We have developed a spatial model of retinal cell populations that solves for RGC displacement at any arbitrary retinal position. We begin with empirical measurements of cone and RGC densities (Curcio et al., 1990). These values are transformed to midget RF and RGC density through parameterized, eccentricity-independent linking functions. We then use a constrained, non-linear search over the linking parameters to enforce convergence of the midget RF and RGC cumulative functions beyond 20°r. The difference in spatial position of equivalent values of the cumulative functions yields the degree of RGC displacement. The modeling code is available (https://github.com/gkaguirrelab/rgcDisplacementMap). The output of our model resembles empirical measurements of RGC displacement (Drasdo et al., 2007). We find a peak RGC displacement of 3.45°r, compared to an empirical value of 3.2. The calculated end of the displacement zone varies from 16°r in the nasal retina to 22°r on the temporal retina, consistent with empirical findings. The linking function parameters imply a midget RGC fraction as a function of eccentricity that is intermediate to prior models (Dacey 1993; Drasdo, 2007) and matches recent empirical measurements (Liu et al., 2017). Finally, we demonstrate that the model may be applied to individual subject data, making possible studies that link non-invasive measurements of cone and RGC density to visual function and cortical organization.
Meeting abstract presented at VSS 2018
This PDF is available to Subscribers Only