Two people walk into a dim restaurant. One reads the menu without thinking. The other tilts it toward the candle, squints, and gives up. They have similar Snellen acuity in the daytime. They wear similar prescriptions. Outside on the sidewalk five minutes ago they read the same street sign at the same distance. Inside, with the lights down, one of them suddenly can't see.
That gap — the one that opens up as light falls — is real, measurable, and predictable. It is not "I have bad night vision," in the loose sense people sometimes mean it. It is the visual system operating in a regime it is less well-suited to, and the amount by which any given eye degrades in that regime varies enormously across people. This post is about what changes physiologically when light drops, why contrast sensitivity falls with it, and why two eyes that look the same at noon can be unrecognisable at dusk.
Three regimes of vision
Human vision does not work the same way across the range of light levels it encounters. Engineers and vision scientists carve that range into three regimes.
Photopic is bright daylight and well-lit indoor environments — roughly above 3 cd/m² of mean luminance. The cone photoreceptors run the show. Acuity is at its best, colour vision is intact, the pupil is small (2–3 mm in a healthy young adult), and the contrast sensitivity function (CSF) peaks at its highest point, somewhere around 3–6 cycles per degree.
Scotopic is moonless-night dark — below about 0.003 cd/m². The cones have effectively shut off and the rods carry the load. There is no colour. The fovea, which is rod-free at the very centre, has a literal blind spot for the dimmest stimuli; you can sometimes see a faint star better by looking slightly to the side of it. Spatial resolution drops by an order of magnitude.
Mesopic is everything in between: dusk, candle-lit interiors, a parking garage at midnight, the dashboard glow of a car on a dark highway. This is the regime most "I can't see at night" complaints actually live in. Both rods and cones contribute, the visual system is in a transitional state, and small individual differences in optics and physiology get amplified.
The clinical literature is clear that the CSF is not a fixed signature of an eye; it is a function of the eye and the light it is operating in. Reduce the mean luminance and two things happen at once: the curve drops, and its peak shifts toward lower spatial frequencies. A review of the spatial CSF and its neurophysiological bases summarises the pattern: at scotopic and low-mesopic luminance the system loses sensitivity to fine detail first, and the whole curve sits below its daylight position (Joshi & Sivaprasad, 2023). The eye chart you would draw for the same person at noon and at dusk are different charts.
The dashed curve is the same eye in a dim room. The peak is lower, the curve is narrower, and the system has less to work with everywhere — but especially at the higher spatial frequencies that fine detail depends on.
Why dim light makes contrast harder
Three things change together as the light drops. None of them on its own is dramatic; the combination is.
The pupil dilates. A dim room is a signal to let more light in, and the iris dilates from a daytime 2–3 mm to a mesopic 4–6 mm in a healthy young adult — and as much as 7–8 mm in younger people in true darkness. Letting more light in is the right move for the rod-and-cone trade-off. But a large pupil also exposes the periphery of the lens and cornea — exactly the parts that are least well-corrected by your glasses and most likely to be slightly imperfect. Higher-order aberrations (spherical aberration, coma) scale with pupil size; a 6 mm pupil produces measurably more retinal blur than a 3 mm pupil through the same optics. The same eye, optically speaking, is a different optical instrument in a dim room than in a bright one.
Photon noise rises. This one is statistical and surprisingly important. The visual system is a photon counter at low light, and the noise on a photon count grows in proportion to the square root of the count. Halve the mean luminance and your signal-to-noise ratio degrades. The neural side of the visual system has to make do with a noisier image, and the part it gives up first is fine detail — which is why the high-frequency arm of the CSF compresses faster than the rest as the lights go down.
The system shifts from cone-driven to rod-driven. Rods outnumber cones roughly 20-to-1 in the retina, but they pool their signals together in clusters that share ganglion cells. That pooling is what gives rods their sensitivity in the dark — and also what limits their spatial resolution. As the cones lose authority, you trade away resolution for sensitivity. The peak of the CSF migrates toward lower spatial frequencies because that is what the rod-dominated system can deliver.
These three together explain why the dashed curve in the diagram above looks the way it does. The mesopic eye is operating with a larger pupil exposing more aberration, a noisier photon signal, and a substrate that has traded fine-detail resolution for sensitivity to absolute light level. Even a perfectly healthy young eye does worse in a dim room. Owsley, Sekuler and Siemsen's foundational adult-lifespan study, run at controlled mean luminance, established the shape of normal CS across age; subsequent work has shown that dropping the luminance level shifts every age band downward, with the steepest impact on the high-frequency end (Owsley, Sekuler & Siemsen, 1983).
Why the gap between people widens
Two eyes that perform similarly in bright light can diverge sharply as the lights go down. The mechanical reason is that the modifiers of low-light vision — pupil dilation, intraocular scatter, tear-film stability, optical aberration, retinal sensitivity, aging cones — vary far more across people than the daytime equivalents do. In bright light the small pupil masks a lot of optical imperfection. In dim light, the larger pupil reveals it.
Several things make some eyes more affected than others.
Pupil-size differences. People are born with different iris architectures, and the maximum dilation in dim light varies meaningfully across healthy adults. A large dilator pulls in more peripheral lens optics and exposes more aberration; a smaller dilator gets less light through, which is its own problem.
Senile miosis. With age, the pupil's maximum dilation diminishes — the so-called senile miosis. Older adults end up with less retinal illuminance for any given ambient light level, on top of whatever else is happening with their optics. The net effect: an older eye in a dim room is operating on less light than a younger eye in the same room. This compounds with the normative age-related contrast decline of roughly 10% per decade after age 20 (Owsley, Sekuler & Siemsen, 1983; Mäntyjärvi & Laitinen, 2001).
Intraocular scatter. The lens of the eye gradually loses some transparency across life. Even before anything would be called a cataract, light scatter increases — and scatter is the enemy of contrast. The scattered light fills in the dark parts of the image, washing out the differences the visual system needs to detect. Intraocular scatter is barely visible in bright light (where a clean signal dominates) and dominant in dim light (where every washed-out percent matters). This is the mechanism behind glare disability — the long recovery time after a passing headlight that many older drivers report — and it is also what we describe in the cataract and night-driving piece, where scatter is the same mechanism applied to a specific condition.
Tear-film quality. The very first refractive surface of the eye is the tear film — and a stable, smooth tear film does a lot of work that nobody notices until it stops. A dry eye, a slow blink rate, a recent contact-lens shift, or just hours staring at a screen can degrade tear-film optical quality. In bright light with a small pupil this barely matters. In dim light with a large pupil the disturbed tear film scatters more light through more of the optical system, and contrast drops disproportionately.
Uncorrected refractive error. Especially astigmatism. A small uncorrected astigmatism can pass unnoticed at small pupil sizes and become symptomatic at larger ones. The blur an astigmatism produces is direction-dependent — vertical edges may look fine while horizontal edges go fuzzy, or the other way around — and the symptom is most obvious on low-contrast tasks in dim light.
Early ocular disease. Glaucoma, early macular change, mild diabetic retinopathy, and early cataract all degrade low-light contrast before they degrade daytime acuity, for reasons that overlap with the mechanisms above. Joshi and Sivaprasad's 2023 review of the spatial CSF outlines how these condition-related changes layer on top of luminance-dependent shifts of the curve (Joshi & Sivaprasad, 2023). A normal-acuity person with an early disease process may be unremarkable in the clinic at full chart luminance and meaningfully impaired at home in a dim hallway.
What this means for how you experience low-light environments
Three patterns are common in the wild, and each maps onto the physiology above.
The first is the dim-restaurant-menu experience. The mean luminance is comfortable for conversation but well below office levels; the print on the menu is at moderate contrast on paper that is itself reflecting some scattered candlelight. A small uncorrected astigmatism, a tired tear film, or a slightly hazy lens that would be invisible in office light is suddenly the difference between reading and not.
The second is the dark-parking-garage experience. Walking from a bright street into a dim concrete enclosure, the visual system has to dark-adapt — cones decay first, rods come up slowly over twenty minutes. During the transition, contrast sensitivity is at its worst. People with smaller maximum-dilated pupils, older eyes, or any of the scatter-promoting conditions above have a longer transition and a lower steady-state in the dim space.
The third is the screen-at-night experience. A bright screen in a dark room is a contrast-mismatch problem — the eye is partly adapted to the dark surround and partly to the bright stimulus, and the contrast of any low-contrast feature on the screen is set against a moving adaptation point. People with otherwise unremarkable daytime vision can have real, repeatable trouble with low-contrast UI elements at night. (This is a separate question from how a vision test itself should be lit — if you take a contrast sensitivity test in a dim room, you will likely score lower than the same eye would score in a normally-lit room. The calibration and screen-settings post discusses how to set up the test environment to avoid biasing the result.)
What an at-home test can — and cannot — tell you
Note. A contrast sensitivity test is a screening signal of overall visual function, not a diagnosis of any specific condition. A lower-than-typical result is a reason to take a better question to your eye doctor — not a verdict. Many of the mechanisms above (uncorrected refraction, dry eye, scatter, early lens change, age) can be diagnosed and addressed by a routine eye exam; some of them require dedicated testing. If your day-to-day low-light experience has changed, the right next step is a clinician, not an online tool.
A well-calibrated at-home contrast sensitivity test gives you something a Snellen chart cannot: a curve. The shape and position of that curve, taken at a known mean luminance on a known screen, is the closest you will get to a personal map of how your eyes work in a regime that the eye chart does not probe. The clinical primer in our overview of what contrast sensitivity measures walks through what the curve looks like and why it matters. The pattern of low-light difficulty in this post — high-frequency arm pulling down faster than the rest, peak migrating leftward — is exactly the pattern the test is designed to detect.
The honest framing: at-home testing is not a substitute for a clinical exam. It is a way to bring numbers to a conversation that, today, is usually held in symptoms alone. A patient who can say I scored below age-typical at 6 and 12 cycles per degree, and dim-light tasks have become harder is in a different conversation from a patient who can only say I think my eyes have changed.
Practical next steps
- Take the test in normally-lit conditions first to establish your daytime baseline — see the how-to-take-the-test guide for setup.
- Note the conditions where you struggle — restaurants, parking garages, screens at night, reading in a dim room.
- Bring both pieces to an eye exam if the dim-light difficulty is new or has worsened. The number is concrete; the symptom story is concrete; together they are a much sturdier conversation than either alone.
- Mention specifically that you have noticed a difference in low light, even if your acuity has been fine. That gives the clinician a reason to look at parts of vision — refraction at large pupil, tear-film stability, lens transparency, optic nerve health — that a standard chart does not directly probe.
Take the test
Take the test now. Three minutes, a normally-lit room, a saved result. You will have a number for how your eyes handle the contrast tasks that the chart on the wall does not measure — and a starting point for whether the dim-room difficulty you have noticed is age and physiology, something a clinician should look at, or both.
References
- Pelli, D. G., Robson, J. G., & Wilkins, A. J. (1988). The design of a new letter chart for measuring contrast sensitivity. Clinical Vision Sciences, 2, 187–199. Foundational paper introducing the Pelli-Robson chart and the argument that contrast sensitivity is a clinically meaningful complement to visual acuity.
- Owsley, C., Sekuler, R., & Siemsen, D. (1983). Contrast sensitivity throughout adulthood. Vision Research, 23(7), 689–699. The reference dataset for how the CSF changes across the adult lifespan; foundation for the roughly 10%-per-decade age-related decline.
- Mäntyjärvi, M., & Laitinen, T. (2001). Normal values for the Pelli-Robson contrast sensitivity test. Journal of Cataract and Refractive Surgery, 27(2), 261–266. Age-stratified normative Pelli-Robson values used in clinical practice; baseline for what "typical" looks like across decades.
- Owsley, C. (2003). Contrast sensitivity. Ophthalmology Clinics of North America, 16(2), 171–177. Clinical review of CSF including age-related changes, band-specific losses, and the conditions that move the curve.
- Joshi, M. R., & Sivaprasad, S. (2023). Spatial contrast sensitivity function and its neurophysiological bases. Progress in Retinal and Eye Research, PMC10527080. Review of how the CSF depends on retinal and cortical processing, and how luminance regime and disease shift the curve.