see the faint patterns your eye chart never asks about.
A calibrated contrast sensitivity test that runs in your browser. Your first test is free. Unlimited testing, saved history, and PDF reports are $20 — once. A screening signal, not a diagnostic test.
Every column is one spatial frequency, every row one contrast. The boundary where the pattern just disappears traces your own contrast sensitivity function.
three steps. about three minutes.
No printed chart, no clinic visit. A short calibration teaches the test about your screen and viewing distance, then an adaptive staircase finds your threshold at each frequency.
calibrate your screen
Hold a credit card to the screen to set pixel pitch, then a quick blind-spot check for viewing distance. About 60 seconds.
see faint patterns
Striped Gabor patches appear at five spatial frequencies. Tap which way they tilt. A 2-down-1-up staircase finds the faintest contrast you can still resolve.
get your CSF score
Your contrast sensitivity function, plotted against a normative band — with a plain-English read of what each frequency means.
contrast sensitivity isn’t acuity.
A 20/20 chart asks one question: can you read tiny, very high-contrast letters? Contrast sensitivity asks the whole map — how faint a pattern you can see, across a range of pattern sizes measured in cycles per degree (cpd).
We build on the same psychophysics behind the Pelli-Robson chart and the FACT grating test, but use an adaptive staircase instead of a fixed printed page — so a screen can replace the ceiling-limited chart without losing the measurement.
A sample CSF curve against the typical healthy-adult range. Human sensitivity peaks near 3 cpd and falls off at both ends.
conditions associated with contrast sensitivity loss.
Contrast sensitivity is a non-specific signal. A dip can have many causes — uncorrected refraction, fatigue, dry eye, lighting, or a real change in visual processing. The conditions below are commonly discussed in the research; none are diagnosed by this test.
vision after a concussion
Letter-chart acuity often reads 20/20 after concussion while contrast sensitivity is degraded.
MS visual symptoms
A sensitive probe of optic-pathway involvement — often abnormal when acuity is still 20/20.
early signs
Can pick up an early signal before standard tests register it — not a substitute for an ophthalmologist.
cataract & night driving
Why an eye exam can say 20/20 while glare and dim-light contrast make night driving hard.
brain fog with a vision layer
Early reports describe contrast changes; the literature is small and still open.
the Shoemaker pattern contested
A mid-band loss has been associated with biotoxin exposure in some literature — many other conditions produce the same pattern.
the science we build on.
The measurement and the staircase aren’t ours — they’re standard psychophysics, decades deep. These four papers are the load-bearing ones.
Pelli D, Robson J, Wilkins A. The design of a new letter chart for measuring contrast sensitivity. Clinical Vision Sciences. 1988;2(3):187–199.
The Pelli-Robson chart. Established the clinical standard for measuring contrast sensitivity with letters of fixed size and decreasing contrast — the lineage our test sits in.
Mäntyjärvi M, Laitinen T. Normal values for the Pelli-Robson contrast sensitivity test. Journal of Cataract and Refractive Surgery. 2001;27(2):261–266.
What “typical” means. Defined the normative band by age, which is what lets a single curve be read as within, at the edge of, or below the typical range.
Lesmes LA, Lu ZL, Baek J, Albright TD. Bayesian adaptive estimation of the contrast sensitivity function: the quick CSF method. Journal of Vision. 2010;10(3):17.
The quick CSF method. Showed the whole curve can be estimated efficiently with adaptive sampling — the principle behind measuring on a screen in minutes.
Watson AB, Pelli DG. QUEST: a Bayesian adaptive psychometric method. Perception & Psychophysics. 1983;33(2):113–120.
How thresholds converge. The adaptive procedure that places each trial near your threshold instead of wasting trials — why three minutes is enough.
The full protocol — stimulus, frequencies, staircase rule, calibration — lives on the methodology page.
$20. once. no subscription.
Try it free. If it’s useful, twenty dollars unlocks unlimited testing for life — no recurring charge, no upsell, no “pro” tier.
first test
free, no account
- One full calibrated test
- Your CSF curve & plain-English read
- Result saved locally on your device
- No saved history or PDF
unlimited
everything, forever
- Unlimited tests — track change over time
- History saved with a comparison view
- Printable PDF report & shareable link
- Bring an actual curve to your clinician
One-time payment. No subscription, no auto-renew. 30-day refund if it isn’t useful.
get your CSF report, and track the trend.
Enter your email and we'll send your printable report and remember your history, so the next test compares against this one. One test is a snapshot; a trend is a signal.
No marketing churn. We email your results and the occasional Notes post. Unsubscribe anytime.
a pre-visit screen your patients can run from their phone.
A calibrated CSF curve, a PDF, and a per-frequency read your patient brings to the visit. A screening signal that supports the conversation — never the basis for a decision. A provider tier is in the works.
questions people ask.
No. It's a screening signal, not a diagnostic test. A low result means something is affecting your visual processing — not what. If something feels off about your vision, see a clinician with a calibrated chart.
Contrast is relative to a background. The test surface is held at mid-gray so the patterns are measured against a known reference, and the calibration steps teach the test your screen's pixel pitch, your viewing distance, and your display's gamma — so a result on your laptop means the same thing as one on your phone.
About three minutes after a 60-second calibration. The full CSF curve across all five frequencies runs a little longer.
Yes. $20 one-time, lifetime unlimited. There's no recurring charge, no auto-renew, and no "premium" tier above it. Your first test is free so you can decide before paying.
The free test keeps results on your device — nothing is sent to a server. If you buy unlimited and give an email to save history, we store the minimum needed to do that and nothing else. We don't sell data and we don't run third-party analytics.
You get a log CS value at each frequency and an AULCSF summary, plotted against the normative band. Each point is tagged within, at the edge of, or below the typical range. A single isolated dip is usually noise; a coherent low-mid-band loss is worth attention.
As many times as you like with unlimited. Repeated tests during recovery or treatment let you watch the curve move rather than relying on one snapshot — which is the whole point of saving history.
ready to see your CSF?
Your first test is free. Unlimited is $20, once. About three minutes — no account needed to try.
take your first test free