The number plate eye test is not a night-driving test

The clocks go back on Sunday 25 October and the night-driving conversation starts in every practice. The standard drivers are held to is the number plate eye test: a plate read in good daylight at 20 metres, plus acuity of at least 6/12. When Anglia Ruskin cut acuity and contrast together with simulation glasses in young volunteers, a Snellen 6/12 cut-off caught 61% of plate failures near the cut-off, against 97% in their earlier study of uncorrected blur.

The daylight in that standard is not decoration: in a night-driving simulator, contrast sensitivity tracked how early older drivers saw a hazard, and high-contrast acuity fell short of significance.

What the number plate eye test actually measures

DVLA's guidance sets the plate in good daylight at 20 metres, with Snellen 6/12 and a minimum field of vision on top. It writes the gap down itself, under cataract:

Glare may counter an ability to pass the number plate test of the minimum requirements, even when cataracts allow apparently appropriate acuities.

Where the chair stops predicting the kerb

Anglia Ruskin ran it twice, in two groups. First, in a study in the British Journal of Ophthalmology, 120 staff and students read charts indoors uncorrected, then a real plate outdoors: in the uncertain band, 6/9 to 6/36, a 6/12 cut-off caught 97% of the failures. Then, in the study published in Eye, 62 adults of mean age 25 wore simulation glasses that cut acuity and contrast sensitivity together, closer to cataract than to an old prescription. In its own band, 6/12 to 6/24, sensitivity fell to 61% and the false negatives, 6/12 in the chair and a fail at the kerb, rose from 2%, one person in 59, to 15%. Each is a proportion of its own band, not of all drivers.

What predicts hazard detection in the dark

The night-driving evidence comes from a simulator. Fifteen ophthalmologically healthy drivers aged 55 to 81 drove at obstacles of three contrasts while LED arrays on cable robots reproduced an oncoming car's dipped beams. Contrast sensitivity tracked the median distance at which a hazard was picked up, and low-contrast acuity at least as strongly. High-contrast acuity was not significantly associated (r = 0.37, P = 0.10), a null too small to trust. The headlights cut contrast sensitivity in 20 of 22 instances, a relative fall of 41%.

The study that cuts the other way

The largest of the driving studies, and the only one on real roads, found no significant association with daytime driving. In Alabama, 144 drivers aged 70 and over had a vision battery, then a 15-mile on-road drive scored by a masked driving rehabilitation specialist. None had acuity worse than 20/40, or 6/12, and 18.8% were impaired on Pelli-Robson, below 1.5 log sensitivity. That was not significantly associated with how they drove: the age-adjusted odds ratio for a worse score was 1.56 (95% CI 0.59 to 4.09), wide enough to hold a real effect. Only processing speed under divided attention and motion perception were associated. Every drive was in daylight, the authors say.

Why it matters in practice

There is no contrast-sensitivity threshold for car drivers in UK law, and the College of Optometrists' position is that the evidence is not there to say which tests and cut-offs would be better. The same statement says blur, glare and reduced contrast sensitivity, at night and in cataract, are known to affect driving and to raise the risk of at-fault and fatal collisions.

• Use a logMAR chart such as an ETDRS, or a Snellen scored by full line. At 6/7.5 or better every participant in the blur study read the plate, but one in the contrast study failed it at +0.08 logMAR, so in cataract ask for a plate check too.
• Ask anyone between 6/9 and 6/12 to read a plate at 20 metres themselves, in daylight. They pass the chart, and some cannot.
• Measure contrast sensitivity where it changes the conversation: early cataract, or the patient whose acuity is fine and whose nights are not. It is information, not a legal test.
• College guidance says an anti-reflection coating improves transmission, and that a sunglass tint passing under 75% of the light is not for road use in twilight or at night.
• The DVLA, or the DVA in Northern Ireland, decides medical fitness and the patient must tell it. Advise them not to drive, put it in writing, record it, and if they will not stop, tell them you will notify the agency, then do so and tell them you have.

That last step is the one optometrists are least comfortable with. In the GOC's 2026 registrant survey, 91% of optometrists had told a patient in the past year that they were no longer fit to drive, but only 32% would be comfortable telling the DVLA themselves if the patient would not, and 96% of all registrants who answered had not done so that year. Of all those uncomfortable, 72% named breaching confidentiality, which the College's guidance says the public interest can override when a driver rejects advice not to drive.

The caveats

• Simulation glasses on 62 adults of mean age 25 are not cataract in older drivers.
• The night-driving result is 15 healthy volunteers in one simulator. Several authors declare industry ties, and the German Social Accident Insurance funded it.
• The Alabama drivers came from eye clinics, so 18.8% is not a population figure, and vision was tested six months before.
• The night-driving authors are explicit that their own data do not speak directly to the real-world value of routinely testing night vision.
• The GOC figures are self-reported by the 3,451 registrants who answered, an 11% response rate.

The number plate eye test is not a bad test. It is a daylight test, and it was never meant to answer the question that arrives in the last week of October.

Sources

1. 1.Meeting the UK driving vision standards with reduced contrast sensitivity · Eye, 16 October 2015
2. 2.Contrast sensitivity and night driving in older people: quantifying the relationship between visual acuity, contrast sensitivity, and hazard detection distance in a night-time driving simulator · Frontiers in Human Neuroscience, 29 July 2022
3. 3.Advising patients on visual fitness to drive: implications of revised DVLA regulations · British Journal of Ophthalmology
4. 4.Visual functions associated with on-road performance by older drivers evaluated by a certified driving rehabilitation specialist · Ophthalmic and Physiological Optics, 31 March 2022
5. 5.Visual disorders: assessing fitness to drive · DVLA
6. 6.Position on vision standards for driving · College of Optometrists
7. 7.Examining patients who drive: Guidance for Professional Practice · College of Optometrists
8. 8.Registrant workforce and perceptions survey 2026: research report · General Optical Council, 2 September 2026