Revealing the benefit of eye motion for acuity under emulated cone loss
- Department of Electrical Engineering & Computer Sciences, University of California, Berkeley, United States
- Herbert Wertheim School of Optometry & Vision Science, University of California, Berkeley, United States
Figures
Figure 1
Acuity threshold experiment.
Acuity is plotted versus the fraction of cones remaining and the equivalent dropout percentage under the cone dropout condition (blue) and the pixel dropout condition (red) for four subjects. Dashed lines show the fitted regression lines to individual subject data, while the solid lines show the fitted regression line across all four subjects for the two conditions, where is the logMAR acuity and is the logarithm of the fraction of cones remaining in the mosaic. The bold points show the average acuity across subjects with error bars indicating the standard error of the mean.
Figure 2
Stimulus duration experiment.
The difference in fraction correct between the cone and pixel dropout conditions on a 4-alternative forced choice (4AFC) Landolt C task is plotted versus stimulus duration. Individual subject Δ Performance is plotted with faint lines. The average Δ Performance across those four subjects is plotted with dark lines and error bars indicating the standard error of the mean.
Figure 3
Sampling coverage analysis.
(a) The average sampling coverage of the Landolt C stimulus across all trials in the acuity threshold experiment is plotted versus the fraction of cones remaining and the equivalent dropout percentage for the cone dropout condition (cyan) and the pixel dropout condition (orange). The dashed lines show the fits to each individual subject’s data, and the solid lines show the fits to the data averaged across subjects, where is the sampling coverage in decimal form and is the logarithm of the fraction of cones remaining in the mosaic. (b) The measured acuity for each subject at each dropout percentage is plotted versus the calculated sampling coverage from the corresponding trials. Each point represents one subject at one dropout percentage, either under the cone dropout (cyan) or pixel dropout (orange) condition. The data is fit by a linear relationship (black), where is the logMAR acuity and is the sampling coverage in decimal form.
Figure 4
Eye motion analysis.
Iso-density contour area (ISOA) comparison between dropout conditions. (a) Eye motion traces are plotted for the cone dropout condition (blue) and pixel dropout condition (red) in the experiment varying stimulus duration for four subjects as indicated in the upper-right corner of each column. The iso-density contour line containing 68% of the eye motion data is plotted on top of the eye motion traces. The apparent cutoff for some traces in the horizontal dimension arises when the subject’s eye moves outside of the trackable region. (b) The ISOA enclosed by the contours is plotted for the cone dropout (blue) and pixel dropout (red) conditions. Each symbol indicates a different subject. No significant difference was found between ISOA under the cone dropout and pixel dropout conditions.
Videos
Video 1
Dynamic comparison between cone and pixel dropout stimulus conditions.
The leftmost panel shows a live retinal video from an example trial. The top and bottom panels show visualizations of the cone dropout and pixel dropout conditions, respectively. The live retinal image is tracked against a premade retina map where cone locations have been labeled. In the cone dropout condition, a subset of these cone locations have been removed from the map, while the Landolt C stimulus is left unchanged. In the pixel dropout condition, pixels have been removed from the stimulus, while the map remains intact. In each case, the stimulation pattern is the result of sampling the Landolt C stimulus with the underlying retina map as the retinal mosaic moves across the letter. Simulated stimulation patterns are shown for each condition. The final column shows the accumulated information over the course of both example trials. In the cone dropout case, more information is gathered over time due to fixational eye movements, which allow ‘surviving’ cones to sample more of the letter’s area.
Additional files
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Revealing the benefit of eye motion for acuity under emulated cone loss
eLife 15:RP111289.
https://doi.org/10.7554/eLife.111289.3
