Visual Acuity Chart Design & Scoring

The Snellen Chart

The Snellen chart is standardized to a 6-metre (20-foot) testing distance. At that distance, accommodation is effectively relaxed for a patient with normal focusing ability. The optotype on the "6/6" (20/20) line is sized so the whole letter subtends 5 arc minutes of visual angle, the classical benchmark for normal resolving power.

A Snellen fraction is read as a ratio of distances, not a percentage: 6/60 means the patient can just resolve, at 6 metres, an optotype that a person with normal acuity could resolve from 60 metres away. The larger the denominator, the worse the acuity. 6/60 corresponds to 20/200 in US notation: the same finding, expressed in feet rather than metres.

ETDRS Design and logMAR Scoring

The ETDRS (Early Treatment Diabetic Retinopathy Study) chart was designed to fix the Snellen chart's shortcomings for research use. Snellen charts vary in how many letters sit on each line and space letters unevenly from line to line, which makes "lines gained or lost" a fuzzy, non-linear unit. ETDRS charts use a logarithmic size progression instead: exactly five letters per line, letters of matched legibility, and consistent spacing. The result is interval-level data that behaves well under ordinary statistics, which is why virtually every modern acuity trial reports outcomes in ETDRS letters rather than Snellen lines.

LogMAR (log of the Minimum Angle of Resolution) is the score this design enables. MAR is 1 arc minute at 6/6, so log₁₀(1) = 0.0: a logMAR of 0.0 is normal acuity. At 6/60, MAR is 10 arc minutes, so log₁₀(10) = 1.0. Each full ETDRS line change moves the logMAR score by exactly 0.1. Since each line holds five letters, each individual letter read correctly is worth 0.02 logMAR, which lets acuity be scored to a precision no Snellen chart can offer. A common scoring convention: 55 letters read correctly on the standard 4-metre ETDRS chart equals logMAR 0.0, and every 5 letters fewer adds 0.1. A patient who reads 40 letters at 4 metres is 15 letters (3 lines) below that mark, for a logMAR of 0.30, roughly equivalent to 6/12.

Near Acuity and N-Notation

Near acuity is tested separately from distance acuity, at the patient's habitual reading distance (typically 30-40 cm) rather than at a fixed standardized distance the way Snellen and ETDRS distance charts are. UK practice records the result in N-notation: N5 is roughly equivalent to small, dictionary- or footnote-sized print, while N8 approximates standard newspaper body text. These are practical, real-world reference sizes rather than a logarithmic scale, so N-notation and logMAR near equivalents don't map onto each other as cleanly as their distance-acuity counterparts.

American practice more often uses M-notation instead: a letter's size is defined by the distance at which a patient with 6/6 (20/20) distance acuity can just resolve it, so a 1M letter is legible at 1 meter and a 2M letter at 2 meters. Decimal near acuity follows directly from that definition: divide the actual test distance in meters by the letter's M size. A patient who reads a 2M letter at a 1-meter test distance has a near decimal acuity of 0.5.

Vernier Acuity and Hyperacuity

Not every visual threshold is limited by the eye's optical resolving power. Vernier acuity, the ability to detect a small offset between two abutting line segments, has thresholds as fine as 5-10 arcseconds: an order of magnitude finer than the eye's actual optical resolution limit. This form of "hyperacuity" doesn't depend on resolving two separate points optically. It relies instead on cortical spatial interpolation, where a population of adjacent photoreceptors and downstream cortical neurons encode relative position differences across many receptors at once rather than any single receptor resolving a gap. Because the mechanism sits downstream of image formation, vernier acuity can remain surprisingly well-preserved even when the retinal image itself is blurred, as in early cataract.

Clinical Significance: The MCID Threshold

ETDRS letter scoring gives fine-grained, statistically well-behaved data, so clinical trials needed a threshold for how large a letter change actually matters to a patient rather than reflecting routine test-retest noise. In landmark neovascular AMD trials, a gain or loss of at least 15 ETDRS letters (3 lines, or 0.3 logMAR) became the standard benchmark for a minimally clinically important difference (MCID): a change large enough to represent a real, perceptible functional effect rather than measurement variability. Changes of 10 letters or fewer generally fall within normal test-retest variability for most patients, which is part of why the 15-letter bar became the field's accepted threshold for a clinically meaningful endpoint.

Related Topics

Continue exploring Clinical Examination Skills: