Vision loss can make a standard dementia screen misleading when the test depends on seeing, copying, or drawing. Clinicians may need to adapt the assessment or use a validated vision-independent version so eyesight is not mistaken for cognitive decline. A cognitive screen is a brief check for possible impairment, not a diagnosis. Results must be interpreted alongside sensory limitations, daily functioning, medical history, and further evaluation when needed.
Medical information disclaimer: This article is for general educational purposes only and does not provide medical advice, diagnosis, or treatment. Always consult a physician or other qualified health professional about symptoms, medications, tests, or treatment decisions.
Table of Contents
- Why can a standard test give the wrong impression?
- Why not simply remove the visual questions?
- What does a better adaptation look like?
- How should the result be used?
- What should patients and families ask before testing?
Why can a standard test give the wrong impression?
The standard MoCA, or Montreal Cognitive Assessment, includes tasks that require functional vision. A person who cannot see those items may lose points because of inaccessible test materials, even when the underlying cognitive skill is intact. A striking example comes from a low-vision sample of 14 people.
Thirteen failed the full MoCA, while an occupational-therapy assessment identified cognitive loss in only one, according to a 2010 Journal of Visual Impairment & Blindness report. This small sample does not establish a general false-positive rate, but it shows how strongly sight loss can distort results. The problem is not limited to the final score. An examiner may also misinterpret hesitation, skipped tasks, or incorrect drawings unless the person's visual access is considered from the start.
Why not simply remove the visual questions?
Deleting inaccessible items changes what the test measures. It may leave some cognitive abilities underrepresented and make the original cutoff inappropriate. In a reanalysis of original MoCA validation data, removing visual items and applying one adjusted cutoff reduced sensitivity for mild cognitive impairment from 90% to 44%.
Sensitivity for alzheimer dementia fell from 100% to 87%, as detailed in the same vision-related MoCA analysis. Sensitivity means the proportion of people with the condition whom the screen correctly flags. That reanalysis did not use a visually impaired sample, so its percentages should not be treated as expected results for people with sight loss. Its practical lesson is narrower: subtraction is not automatically a valid adaptation.
What does a better adaptation look like?
A stronger approach replaces visually dependent tasks with accessible tasks designed to assess the same cognitive domains. This preserves more of the test's intended scope than simply dropping questions. MoCA-VI uses alternative spoken forms for the visual components. In a 2025 validation study of 128 adults with distance vision worse than 6/12, a score of 24 or below detected possible dementia with 95.9% sensitivity and 92.4% specificity.
The Psychological Assessment study included 79 cognitively healthy participants and 49 with dementia. Test–retest reliability was 0.84 among 26 people who repeated it. Those findings support MoCA-VI as a promising option for the population studied. They do not make 6/12 a universal eligibility threshold or prove that the same cutoff performs equally well in every language, education group, or care setting.
How should the result be used?
Whether standard or adapted, a brief screen estimates whether further investigation may be needed. The Alzheimer's Association advises further evaluation or specialist referral after a failed brief cognitive test to establish whether mild cognitive impairment or dementia is present. A normal score does not settle the question either.
The NICE dementia recommendations say clinicians should consider sensory impairment, including sight loss, among potentially reversible causes when investigating suspected dementia. NICE also says dementia should not be ruled out solely because a cognitive-instrument score is normal. The next step should address any mismatch between the score and the wider picture. Changes in everyday functioning, information from someone who knows the person well, sensory barriers, and the conditions under which testing occurred can all affect how much weight the score deserves.
What should patients and families ask before testing?
Tell the examiner about sight loss before the assessment begins, including difficulty seeing print, diagrams, objects, or drawing spaces. Ask how the test will separate inaccessible visual tasks from cognitive performance.
Useful questions include: Cutoffs cannot be assumed to transfer unchanged between populations. A 2024 PLOS One study of visually impaired residential-care residents in Hong Kong identified limits involving education adjustment, mild cognitive impairment assessment, reliability testing, and generalizability to community-dwelling adults. Before acting on a score, ask which population validated that test version and cutoff.
- Does this version require seeing, copying, or drawing?
- Will inaccessible items be replaced with validated alternatives?
- Is the proposed cutoff supported for people like the person being tested?
- Will the report document the visual limitation and every adaptation?
- What further evaluation will follow if the score conflicts with daily functioning?
