How our vision simulations were built
The research basis and rendering conventions behind every frame
Medically reviewed by Stephen Johnston, MD. Reviewed 2026-10-23.
Every view on our vision simulations page is drawn from a single computer-rendered park scene, changed only by masks and filters. This page records why each view looks the way it does. Each rendering decision is tied to a published source; the exact parameter values (degrees of field, blur radius, luminance) are the illustrators' judgment and are stated frame by frame in the clinician page for each condition, under "How the simulation frames were parameterized".
Cross-cutting rendering conventions
Convention: Central scotomas are not black. Evidence: Patients with bilateral central scotomas "reported a blank white area with the same color and luminance as the background" when viewing a page, and "are not fully aware of their scotomas" (Pratt, Stevenson and Bedell, Optom Vis Sci 2017, PMC5321850). The AAO lists Stargardt central vision as "blurry, distorted or have dark areas" (AAO).. How it is drawn: Scotoma region is filled with the heavily averaged surrounding color and luminance (slightly grayed in later stages), never a black disc.
Convention: Lost peripheral field as absence, not black paint. Evidence: Patients describe "blind spots in the peripheral vision causing them to bump into things" and a field that constricts "like seeing through a tunnel" (Gene Vision); no source describes the lost field as black. Black masks are the convention in simulation glasses and tape simulations, which the AMD-simulation review notes do not realign with gaze and lack realism (Frontiers in Neuroscience 2021).. How it is drawn: Lost field is a dim (about 25 percent), desaturated, featureless color-averaged smear. Reads as "nothing there" rather than a black wall, while remaining unmistakable to a sighted viewer.
Convention: Soft edges for RP, sharp edges for choroideremia. Evidence: In choroideremia "visual sensitivity falls precipitously at the atrophic border" with transitions "over a distance smaller than the diameter of the Goldmann III stimulus" (PMC6778501). RP scotomas begin as "a group of isolated scotomas around 20 degrees from fixation" that coalesce gradually (EyeWiki RP).. How it is drawn: Choroideremia masks feathered 2 px; RP/Usher/rod-cone masks feathered 1.5 to 2.5 degrees with low-frequency irregularity.
Convention: Photopsias. Evidence: 93 percent of RP respondents report photopsias; the commonest shapes are "crescents, arcs or semi-circles" (60 percent), mostly peripheral in earlier disease and more central in later disease (PMC2783747).. How it is drawn: Faint pale arcs, peripheral in mid-stage frames, more central and more numerous in advanced frames.
Convention: Photophobia and glare. Evidence: Photophobia and reduced contrast sensitivity are listed complications of RP (StatPearls RP); in achromatopsia bright light turns the scene "into a bright haze and all structured vision is lost" (Nordby, quoted at achromatopsia.info).. How it is drawn: Highlight bloom plus a lifted black floor, scaled per condition and stage.
Convention: Nystagmus. Evidence: Listed for achromatopsia (Moorfields), LCA (EyeWiki LCA) and some cone dystrophy (Macular Society).. How it is drawn: Horizontal smear of 4 to 16 px in those frames only.
Convention: Acuity to blur. Evidence: Illustrators' assumption: frame treated as 90 degrees wide; Gaussian sigma about three times the minimum angle of resolution in frame pixels (20/40 about 2 px, 20/100 about 5, 20/200 about 10, 20/400 about 20), because the image is viewed at roughly a third of its real angular size.. How it is drawn: Stated per file.
Stage models per condition
Retinitis pigmentosa. Night blindness and mid-peripheral field difficulty in adolescence; isolated scotomas about 20 degrees from fixation coalesce into a mid-peripheral ring whose "outer edge expands relatively quickly to the periphery, while the inner edge constricts slowly toward fixation"; tunnel vision late, often with a small central island and good central vision into the 50s or 60s; 93 percent of autosomal dominant patients under 20 keep a central field radius of 10 degrees or more, falling to 60 percent over 40 (EyeWiki RP). Central blur and color loss come when cones degenerate later (Gene Vision). Stages drawn: early (patchy ring), canonical mid (complete ring, 26 degree island, temporal crescent), advanced (10 degree island, cone involvement).
Night blindness. First symptom of RP: inability to see in low light or "adapt rapidly from bright to dim environments" (StatPearls RP). Stages drawn: early (slow adaptation, unlit scene at a third of normal), moderate (unlit areas reduce to shapes), canonical severe (only direct lamp light).
Stargardt disease. Progressive central scotoma; Group 1 within 10 degrees, Group 2 within 10 to 35 degrees; presenting acuity 20/20 to 20/400; abnormal color vision and delayed dark adaptation; peripheral vision preserved (StatPearls Stargardt). Onset first or second decade. Stages drawn: early (blur and mild distortion, no visible hole), canonical (10 degree filled-in scotoma), advanced (22 degree, grayed).
Choroideremia. Night blindness in the first decade, "patchy loss of midperipheral vision", "irregular scotomas", good acuity to the fifth to seventh decade, then rapid central decline leaving "small islands of vision... in the fovea" with light perceived in the far periphery (EyeWiki Choroideremia). Sharp borders per PMC6778501. Stages drawn: early (sharp patches), canonical (scalloped ring with far islands), advanced (foveal island plus far light perception).
Cone dystrophy. Reduced central clarity, color and light tolerance; progressive form develops in late childhood or early adulthood and can reach 20/200 or worse; peripheral vision usually unaffected and night vision usually good because rods are spared (NORD). Stages drawn: early, canonical, advanced, plus a dim-light frame showing the preserved rod vision.
Achromatopsia. Not progressive; absence of color, reduced acuity, "vision decreases as light levels increase", nystagmus (Retina UK). First-person: in sunlight "the scene I am gazing at is quickly washed out, turned into a bright haze" (Nordby) and night is the best time (achromatopsia.info). Because it does not progress, the variants are by light level: bright sun, canonical shade, dusk.
Best disease. Six stages: previtelliform vision normal (20/20); vitelliform normal or mildly reduced; pseudohypopyon and vitelleruptive mild loss; atrophic "greatly reduced", 20/40 to 20/200; CNV often 20/200 or worse and can be sudden; lesion seen mostly ages 3 to 15 (UCL Institute of Ophthalmology). Stages drawn: vitelliform, canonical atrophic, CNV.
Usher syndrome. Type 1: night blindness in the first decade, then "peripheral vision loss (tunnel vision), and, eventually, central vision loss", and the classic description puts severe loss in the third decade, though the course varies widely from person to person; the temporal sector often survives longest; type 2 notices dim-light trouble in adolescence or early adulthood with slower progression (PMC11942519). Stages drawn on the type 1 timeline: early (childhood), canonical (teens to 20s), advanced (third decade).
Leber congenital amaurosis. Profound impairment from birth, acuity "20/400 and below" with rare exceptions, about one third with no light perception, nystagmus and photophobia, course "generally stable or very slowly progressive" (EyeWiki LCA). Because it is stable, the variants are a severity spectrum: canonical 20/400, light perception only, no light perception.
Rod-cone dystrophy, advanced. The RP family at its late end: one quarter of RP and Usher patients over 45 have 20/200 or worse in both eyes, and 0.5 percent are completely blind (EyeWiki RP). Stages drawn: canonical advanced (12 degree tunnel with cone involvement), end-stage (island under 4 degrees, far-peripheral light perception).
Known limits of still images
Scotomas move with the eye; a still cannot show the preferred retinal locus people with central loss adopt (the review above rates gaze-contingent displays the most realistic). Captions on the hub should say the frame shows a person looking at the center of the scene.
Nystagmus, photopsia flicker and slow dark adaptation are time-based; the stills freeze them.
All frames are binocular approximations; many of these conditions are asymmetric between eyes (notably Best disease CNV).
Where the per-frame parameters are stated
The clinician page for each condition ends with the parameters used for that condition's frames, so a reader can check a frame against the evidence it rests on:
Retinitis pigmentosa (not yet published)
Night blindness (not yet published)
Stargardt disease (not yet published)
Choroideremia (not yet published)
Cone dystrophy (not yet published)
Achromatopsia (not yet published)
Best disease (not yet published)
Usher syndrome (not yet published)
Leber congenital amaurosis (not yet published)
Rod-cone dystrophy and Bardet-Biedl syndrome (not yet published)
How to cite this page
A Race Against Blindness. How our vision simulations were built: research basis and rendering conventions. Education Hub, Oct 2026. https://araceagainstblindness.org/education/vision-simulations/methodology