What Is a Normal Hippocampal Volume Percentile?

A hippocampal volume percentile shows where the size of your memory center falls compared with healthy people your own age. It doesn’t say whether your brain is healthy or sick. That leaves a wide middle band of percentiles that is simply ordinary, and only the very bottom of the range usually draws attention. The number is also an estimate that shifts with the software, the scanner and even how hydrated you were, so doctors read it alongside your symptoms rather than on its own.¹

How the Percentile Is Calculated

Percentiles come from large reference groups. Charts have been built from almost 20,000 generally healthy adults in the UK Biobank, so an automated tool can place your scan within that normal spread in seconds.² The charts are drawn separately for women and men and for the left and right hippocampus, and they’re corrected for head size.³ That’s why your report may show two percentiles, one for each side.

A percentile is a rank, not a score. At the 30th percentile, about 30 in 100 healthy people your age have a smaller hippocampus and 70 have a larger one. The 50th percentile is simply average for your age, and the band from the 25th to the 75th holds half of healthy people. Because volume shrinks with normal aging, a 75-year-old at the 50th percentile has less actual volume than a 50-year-old at the 50th.

On the report itself, you’ll often see a raw volume in cubic millimetres next to a percentile. The percentile is the part that accounts for your age. A percentile is not a percentage: a 20th percentile doesn’t mean 80% of your hippocampus is missing.

Many automated reports show both a raw volume and a volume expressed as a share of head size, and the percentile is usually based on the adjusted one, so look for which version the percentile refers to. A person with a large head can have a big raw volume yet a modest percentile, while a person with a small head can have a small raw volume yet a perfectly average one. If the report doesn’t say whether head size was corrected for, ask the radiologist, because the answer changes how to read the number.

Which Percentiles Count as Low

The usual flag sits at the 5th percentile, the common threshold for calling the hippocampus abnormally small.⁴ A result between about the 5th and 95th percentile is usually reported as within normal limits. The raw volume at that cutoff falls with age. For one popular program, it drops from about 3,200 cubic mm on the left at age 56 to about 2,500 at age 90.⁵ A different program put the same age-56 cutoff above 4,500 cubic mm, so raw numbers can’t be compared across tools.⁶

The 5th percentile is a convention, not a biological line. By definition, about 1 in 20 healthy people fall below it, so landing there doesn’t prove anything is wrong. A result at the 6th percentile isn’t meaningfully different from one at the 4th, even if the report labels them differently.

In people already being evaluated for suspected Alzheimer’s, researchers tested a higher cutoff. The best dividing line was near the 29th percentile, but even there the scan sorted patients only modestly better than chance.⁷ So a percentile between 5 and 30 is a grey zone that needs other tests, not a verdict.

Reports sometimes call this range “borderline” or “low normal,” often for roughly the 5th to 25th percentile. Many healthy people land there. Doctors weigh it with memory testing and your history, and a repeat scan in a year or two can show a trend.

Your doctor will ask whether you or your family have noticed changes in memory, because a low-normal number in someone with no symptoms usually calls for little more than awareness. If you do have symptoms, the same number carries more weight and may lead to memory testing or a repeat scan. Doctors also look at the scan by eye.

Why Age, Sex and Head Size Change Your Number

Men’s hippocampi are larger on average at every age, because men’s heads and brains are larger overall. Once head size is accounted for, the sex difference largely disappears.⁸ Because the charts are sex-specific, a woman’s 40th percentile and a man’s 40th percentile correspond to different raw volumes. A family member’s report should be read against their own chart, never against yours.

The adjustment method itself matters. Different ways of measuring and correcting for head size can change how hippocampal volume relates to sex and memory.⁹ Adjusted volumes also sort people with and without dementia slightly better than raw volumes.¹⁰

Ethnicity can explain as much variation in hippocampal size as age, and most reference charts are based on mostly white populations.¹¹ A percentile from a chart that doesn’t match your background may be less exact. If you’re not of European ancestry, ask what population the reference chart came from. A borderline result may reflect chart mismatch, and a trend across your own scans avoids the issue.

The left and right sides differ too. The right hippocampus is usually slightly larger, even at birth, and the gap is smaller in left-handed people. A small difference between sides is normal; a large one draws attention.

Does a Low Percentile Mean Alzheimer’s

In people with memory complaints, low volume does predict progression. Among older adults with mild cognitive impairment, a stage between normal aging and dementia where daily life is largely intact, about 1 in 3 developed dementia within about three years. Those with smaller hippocampi were more likely to progress, even after accounting for age, genes and memory tests.¹² Not everyone with mild memory problems progresses, and hippocampal size is one of several risk clues.

Most of the research studied people who already had memory complaints, so the risk numbers don’t transfer to someone with no symptoms. A healthy person with a low-normal percentile is far more likely to be at the lower end of normal variation. If you were scanned for headaches or after a fall, a low percentile is a reason to mention it at a routine visit, not a reason to expect dementia.

Automated reports can mislead. Radiologists using them sometimes grew more confident in wrong calls of shrinkage, so the reports are meant as a support, not a replacement for an expert’s reading.¹³ Ask your doctor how the percentile fits your symptoms and the visual read of your scan.

A workup for memory concerns usually includes memory and thinking tests and blood tests to rule out treatable causes. Sometimes it adds PET scans or spinal fluid tests, and follow-up scans to check for change.

How Precise the Measurement Is

Software and scanners shift the number. A widely used program measures hippocampi larger than careful hand tracing does.¹⁴ Volumes from software and from published hand-traced studies aren’t interchangeable, and software can struggle with motion blur or unusual anatomy.

Repeating a scan on the same machine changes volume by about 1.5% on average. Switching between 1.5 and 3 Tesla scanners changes it by about 2.7%.¹⁵

Near a cutoff, a few percent can move you from one label to another. A report that says “low” on one scan and “normal” on the next may reflect no real change in your brain. The change in raw volume, in percent, measured on the same scanner and software, tells you more than the label.

Your body on scan day matters as well. Mild dehydration in older adults was linked with a smaller measured hippocampus,¹⁶ and time of day can affect measured volume.¹⁷ For follow-ups, ask for the same scanner, software and a similar time of day, and be well hydrated. Look for a steady trend over several years, and ask the radiologist whether the change exceeds normal variation.

Other Reasons the Hippocampus May Be Small

Epilepsy scarring often shows as a very low, one-sided result. In people with scarring of the hippocampus from temporal lobe epilepsy, about 9 in 10 had volume below the 5th percentile.¹⁸ Comparing sides correctly identified the seizure side about 88% of the time.¹⁹ Alzheimer’s tends to affect both sides, though not always evenly, and epilepsy scarring usually comes with a history of seizures, sometimes starting in childhood. That’s why a large left-right gap on a report prompts questions about seizures.

Mood and stress conditions are also linked to smaller volume. People with major depression had a left hippocampus about 19% smaller in one study.²⁰ That was a group average, and many people with depression have a hippocampus in the normal range. It isn’t a sign of dementia. In PTSD, shrinkage in part of the hippocampus reversed with a brain-training treatment.²¹

It helps to tell your doctor about any seizures, even old ones, and past depression or trauma. Mention alcohol use and medications too. All of these change how a low result is read.

Can You Protect Your Hippocampal Volume

Exercise has the strongest evidence. A year of walking-based training grew the hippocampus about 2% in older adults, roughly reversing 1 to 2 years of aging. Those who stayed inactive lost volume, though fitter people lost less.²² Across many trials the benefit was clear in people 65 and older but not in younger groups.²³ Exercise mainly prevents the decline rather than building extra volume.²⁴ Don’t expect it to push a low percentile up much, but it may help keep it from sliding, and it’s worth pursuing even if your number is average.

Aim for brisk walking most days, such as 150 minutes a week. Brisk means your heart rate and breathing rise but you can still talk. Cycling, swimming and dancing work the same way at a similar effort. Light strolling is better than nothing but less likely to match the trial’s training.

If you have heart issues, check with your doctor before starting. Build up from short walks; consistency over months matters most.

Genes set part of your baseline. Hundreds of common gene variants influence hippocampal size across ancestries,²⁵ so some people are naturally lower on the chart without disease.