One test tells you less than you think

A memory score on its own is a snapshot. It compares you to other people of your age and education, which is useful, but the range of normal is wide. Two people can both score comfortably within it and be on completely different trajectories.

What actually matters is the direction of travel, and you cannot see a direction from a single point.

Why a baseline is worth more than a verdict

If you take the exam today and again in a year, the second result is compared against something far more relevant than a population average. It is compared against you.

That matters because people vary enormously. A retired professor scoring at the population average may have declined significantly. Someone who left school at fifteen and scores below it may be doing exactly what they have always done. Your own earlier result removes most of that noise.

It also changes the conversation with your doctor. "I am worried about my memory" is hard to act on. "Here are two results a year apart, and here is what changed" is not.

A couple looking at verbal memory test results

When to establish a baseline

The best time is while you are not worried. A baseline taken when everything feels fine is the most valuable one you will ever have, because it is the cleanest.

Reasonable prompts to start:

  • Turning fifty or sixty, when other routine health checks begin
  • A family history of Alzheimer's disease or another dementia
  • Starting a medication that can affect cognition
  • After a head injury, once you have recovered
  • Any time you have caught yourself wondering

How often to repeat it

For most people once a year is sensible. It is frequent enough to catch a trend and infrequent enough that you are not reacting to noise.

Take it under similar conditions each time. Same sort of time of day, somewhere quiet, not exhausted and not immediately after a bad night. Memory performance moves around with sleep, stress, illness and medication, and you want to measure the trend rather than the weather.

If something changes noticeably in between, do not wait for the anniversary.

What a change does and does not mean

A lower score next year is not a diagnosis. Many causes of memory trouble are treatable, including thyroid problems, vitamin B12 deficiency, sleep apnoea, depression, alcohol and the side effects of common medications.

What a measured change gives you is a reason to go and find out which, at a point where finding out is still useful. That is the whole argument for tracking rather than waiting.

Where to go next

The best baseline is the one you take before you are worried.

If something has already changed, the warning signs checklist will help you describe it. To see how this compares with a clinic screen, read our comparison.