Story
Cognitive Aging & Cognitive Reserve
Last updated
In one pass Growing old is not the same as developing dementia.
Educational content, not medical advice — consult a clinician.
Story path
Chapter 1
How normal aging differs from dementia
Growing old is not the same as developing dementia. The saying "the brain just goes with age" lumps three different things together: normal aging; mild cognitive impairment (MCI), an in-between state in which a person is clearly worse than their peers but still runs daily life on their own; and dementia.
Normal cognitive aging is selective, not a collapse of the whole system (Harada 2013, a review). The earliest and most common change is reacting half a beat later: finding words, doing sums and making decisions all take a little longer. This is a decline in processing speed. Holding a string of digits in mind, juggling several tasks, or solving an unfamiliar problem on the spot drifts down slowly after midlife. Vocabulary, general knowledge, and the judgment built up from years of reading people and situations often hold steady until about 70, and some keep growing. So a 70-year-old who cannot retrieve a name is common, yet their judgment about people and events may be better than it was at 30. Slow is not broken.
Dementia is not the inevitable end point of aging; many people live to a great age without it. Treating a slightly worse memory as an early sign of dementia is the first fear you can set down.
One kind of change has nothing to do with aging: speech that suddenly slurs, sudden weakness in the arm and leg on one side, or a drooping mouth should be treated as a stroke. Go to the emergency department immediately.
Normal cognitive aging is selective, not a collapse of the whole system (Harada 2013, a review). The earliest and most common change is reacting half a beat later: finding words, doing sums and making decisions all take a little longer. This is a decline in processing speed. Holding a string of digits in mind, juggling several tasks, or solving an unfamiliar problem on the spot drifts down slowly after midlife. Vocabulary, general knowledge, and the judgment built up from years of reading people and situations often hold steady until about 70, and some keep growing. So a 70-year-old who cannot retrieve a name is common, yet their judgment about people and events may be better than it was at 30. Slow is not broken.
Dementia is not the inevitable end point of aging; many people live to a great age without it. Treating a slightly worse memory as an early sign of dementia is the first fear you can set down.
One kind of change has nothing to do with aging: speech that suddenly slurs, sudden weakness in the arm and leg on one side, or a drooping mouth should be treated as a stroke. Go to the emergency department immediately.
Mechanism · Why speed is the first thing to slow
"Slow is not broken" should be more than a comfort. The most common explanation for it is a concrete physical one (it is the mainstream account in neuroscience, not the only one). Once it is clear, you can work out many later observations yourself.1. Brain cells are wired together by cables
Every nerve cell sends out a long, thin process, the axon, and the signal runs along it to the next cell. Wrapped around the axon is an insulating sheath called myelin, rolled on in layers by glial cells. The large pale area on a slice of brain, the white matter, is pale because of that sheath. White matter is the wiring district, packed with insulated cables; gray matter is where the cell bodies cluster.
The insulation lets the electrical signal jump. Myelin leaves a bare gap at regular intervals, so the signal does not have to crawl inch by inch along the whole axon; it leaps from one gap to the next. The thicker and more intact the wrap, the farther and faster each leap.
2. Age acts on exactly this insulation
With age, myelin thins and breaks up, repair falls behind wear, and small injuries build up in the white matter. Autopsy and imaging studies both find that white matter shrinks more with age than gray matter does (Harada 2013, a review). Once the insulation thins, the signal leaps along the axon less cleanly, and every line runs a little slower.
Notice what is not happening: nerve cells are not dying off in batches. The axons are still there and so are their destinations; only the transmission along the way has slowed.
3. Why a slightly slower line shows up as reacting half a beat late
Almost nothing you do is finished by a single brain region. Seeing a written word, recognizing it, retrieving its meaning, and turning it into a spoken sentence is a relay among several regions, and every handoff has to cross white matter. One slightly slower handoff goes unnoticed; a chain of them adds up to what you feel: finding a word half a beat late, doing sums less quickly than before, reacting a little less sharply behind the wheel.
This also explains why fluid intelligence (the capacity to handle new problems on the spot: holding a string of digits, juggling tasks, solving a problem you have never seen) is affected first. These tasks bounce between several regions several times. They have the most handoffs, so they are the most sensitive to conduction speed. Timed tests make this especially plain: often an older person is not getting the answers wrong, they are running out of time.
4. The same mechanism explains why vocabulary declines more slowly
Turn it around. Saying a word you have used for decades, recalling a piece of general knowledge, or judging whether someone is trustworthy means retrieving something already stored. That relies on connections set in place by years of repeated use, and it needs far fewer cross-region handoffs. So crystallized intelligence (accumulated knowledge and experience) can hold steady until about seventy, and in some people keep growing: it runs on stock more than on speed.
The line is now clear. Abilities that need on-the-spot calculation are sensitive to conduction speed and slow first; abilities that are already stored depend less on speed and decline much more slowly. These are not two unrelated observations but two faces of one mechanism.
5. Slowness is a property of the wiring, not a breakdown
Thinner insulation slowing the signal is one thing; neurons dying in large numbers and synapses wrecked by plaques and tangles is another. The first is gradual and even, and everyone has some of it; the second is dementia. Reading my reactions have slowed as I am getting dementia is like mistaking aging cables for a burned-out machine.
Follow the chain one step further. Anything that damages white matter, such as high blood pressure injuring the brain's small vessels, high blood glucose impairing the microcirculation, or smoking cutting the blood supply, will make this slowing come earlier and hit harder. That is why the risk list in the chapter How much dementia risk is modifiable is almost entirely cardiovascular, and why what is good for the heart is broadly good for the brain.
Clinical · Where MCI ends and dementia begins
Mild cognitive impairment (MCI)Clearly a step below people of the same age: the person or their family notices it and objective tests pick it up, yet daily life is still managed independentlyAn in-between state: some people progress to dementia, some stay stable, and some improve back to normal
The last half of that sentence is worth remembering on its own: MCI is not a waiting room for dementia. Among those who improve, some were never being held back by a degenerative process at all, but by something reversible, such as poor sleep, low mood, a thyroid problem, or certain medications. Remove the cause and their performance returns. So if you are told you have a touch of MCI, the first job is to look for reversible causes with your doctor, not to buy a supplement.
Dementia
Cognitive decline is severe enough to undermine independent living: cannot manage money, gets lost in familiar places, cannot keep their own medicines straightAlzheimer's disease is the most common form but not the only one; vascular dementia, Lewy body dementia, and frontotemporal dementia each have their own pattern
The line is drawn at independent living, not at a test score
Why draw it at whether you can still run your own life rather than at a number? Scores on cognitive tests are strongly swayed by education, native language, and that day's mood and sleep, so the same person can score quite differently on different days. Whether someone takes the wrong pills or gets lost on a familiar street is a functional outcome, and those factors do not scramble it so easily.
There is a more important reason too. The amount of disease in the brain and the symptoms a person shows do not match one to one; some people live normally with brains full of plaques (the chapter Why some brains cope with damage is about exactly this). Since scores and lesions do not line up, drawing the line at whether daily life still holds together is the more honest choice.
Chapter 2
Why some brains cope with damage
Two brains can carry the same amount of damage, and one person falls apart while the other carries on normally. Part of the difference is cognitive reserve.
Pathologists noticed something odd long ago. Some older people die with brains full of the plaques and tangles of Alzheimer's disease (the two hallmark lesions it leaves in the brain), yet their thinking was normal in life and they were never diagnosed with dementia.
The neuroscientist Yaakov Stern explains this with cognitive reserve (Stern 2012, a review): under the same injury, some brains can switch to a more efficient pathway, or a different one altogether, to get the job done, so symptoms appear later. It should be kept apart from brain reserve. Brain reserve is closer to hardware: brain volume and the number of neurons and synapses, which physically absorb more damage. Cognitive reserve is closer to software: how many ways you have of doing the same thing.
Reserve is built over a lifetime: years of schooling, whether your work demanded thought, whether you kept reading, learning new things and mixing with people. None of this stops plaques from forming. What it does is let the brain keep functioning in the face of more disease.
One important limit: reserve delays symptoms, not the disease itself. Once the damage breaks through the reserve, decline can actually be faster.
Pathologists noticed something odd long ago. Some older people die with brains full of the plaques and tangles of Alzheimer's disease (the two hallmark lesions it leaves in the brain), yet their thinking was normal in life and they were never diagnosed with dementia.
The neuroscientist Yaakov Stern explains this with cognitive reserve (Stern 2012, a review): under the same injury, some brains can switch to a more efficient pathway, or a different one altogether, to get the job done, so symptoms appear later. It should be kept apart from brain reserve. Brain reserve is closer to hardware: brain volume and the number of neurons and synapses, which physically absorb more damage. Cognitive reserve is closer to software: how many ways you have of doing the same thing.
Reserve is built over a lifetime: years of schooling, whether your work demanded thought, whether you kept reading, learning new things and mixing with people. None of this stops plaques from forming. What it does is let the brain keep functioning in the face of more disease.
One important limit: reserve delays symptoms, not the disease itself. Once the damage breaks through the reserve, decline can actually be faster.
Mechanism · Reserve is backup routes, not a warehouse
Reserve is not a warehouse; it is a way of getting thereThe word reserve suggests a storage tank in the brain that, once full, can take a hit. A closer description: it is how many ways you have of doing the same thing.
Imagine driving across a city every day. Someone who has only ever used one main road cannot get through once that road is closed. Someone who has lived there for years and driven every street turns down a side lane at once, arriving a little later, but arriving. High reserve is not a wider road; it is more backup routes.
What this corresponds to in the brain is quite concrete. Repeat a task the same way and the few pathways involved get stronger. Do the same task in different ways (thinking in words, thinking in images, judging from experience) and different regions get drawn in. The same task ends up stored as several sets of wiring. Plaques and tangles destroy some of those connections, but as long as one route still gets through, your performance still looks normal.
Where it comes from, and why these experiences count
In Stern's model, reserve comes from three kinds of experience:
Years of educationThe cognitive complexity of your job (work that demands thought and decisions)Lifelong mental and social engagement (reading, learning, socializing, challenging hobbies)
What they share is not what knowledge was gained but years of being pushed to solve new problems in different ways, which is exactly how backup routes are laid. Social life is the same: dealing with people is one of the most varied and least automatic mental loads there is, calling on language, memory, emotion, and guessing what the other person is thinking all at once.
Reassuring, but read it precisely
Cognitive reserve is not a gift fixed at birth. The evidence for it comes mainly from observation: people with more education, more mentally demanding jobs, and more social and mental activity show symptoms later for the same amount of disease. That shows an association; on its own it cannot prove that these activities caused the difference, and no trial has yet answered which activities, in what amount, build how much reserve. So lifelong learning and social connection have a plausible mechanism behind them, more than a pep talk, but they are not a prescription with a tested dose.
Why decline can be steeper once reserve is breached
Go back to the backup routes and this is not hard to work out. A person with high reserve is held up by one spare route after another. Nothing shows on the surface, while the disease underneath keeps building. By the time the last routes fail and symptoms finally appear, the damage is often already heavy, so the stretch from still looking fine to clearly disabled is short and steep.
What reserve buys, then, is symptoms that appear later, not disease that grows more slowly. That does not change the conclusion (pushing symptoms back by years is a large gain), but it explains why families so often say it all fell apart at once.
Chapter 3
How much dementia risk is modifiable
Dementia is not simply fate: a sizable share of the risk lies in factors that can be changed.
The Lancet Commission on dementia is an international panel of dementia researchers that regularly pools the world's research into a report and estimates what share of dementia could, in theory, be prevented or delayed by removing modifiable risk factors. The 2020 report (Livingston 2020) counted 12 such factors, together accounting for about 40% of dementia worldwide. The 2024 update (Livingston 2024) added two more, high (low-density lipoprotein cholesterol, often called bad cholesterol) and uncorrected vision loss, for 14 in all, raising the share to about 45%.
Read that 45% carefully. It is a population-level estimate: if the whole population were free of these factors, this is the most by which dementia could fall in theory. It is not a guarantee that any one person who does everything right will be spared. It is calculated mainly from observational data; the factors overlap, so they cannot simply be added up; and the remainder is driven by things like age and genes that cannot be changed. The direction is still clear: a large part of dementia lies within reach of what you and your doctor can do together.
The Lancet Commission on dementia is an international panel of dementia researchers that regularly pools the world's research into a report and estimates what share of dementia could, in theory, be prevented or delayed by removing modifiable risk factors. The 2020 report (Livingston 2020) counted 12 such factors, together accounting for about 40% of dementia worldwide. The 2024 update (Livingston 2024) added two more, high (low-density lipoprotein cholesterol, often called bad cholesterol) and uncorrected vision loss, for 14 in all, raising the share to about 45%.
Read that 45% carefully. It is a population-level estimate: if the whole population were free of these factors, this is the most by which dementia could fall in theory. It is not a guarantee that any one person who does everything right will be spared. It is calculated mainly from observational data; the factors overlap, so they cannot simply be added up; and the remainder is driven by things like age and genes that cannot be changed. The direction is still clear: a large part of dementia lies within reach of what you and your doctor can do together.
In practice · The 14 factors, and where to start
The 14 factors, laid out by stage of lifeEarly life: less educationMidlife: hearing loss, high (bad cholesterol), depression, traumatic brain injury (TBI), physical inactivity, diabetes, smoking, high blood pressure, obesity, heavy drinkingLater life: social isolation, air pollution, vision loss
First, look at the shape of the list
Read it end to end: blood pressure, blood lipids, blood glucose, hearing, vision, exercise, smoking, social contact. Almost all of it is heart and blood-vessel health, the senses, and lifestyle, and not one item is a miracle supplement. That is no coincidence; the chapter How blood vessels affect the brain explains why the list looks like this.
Three to start with
These three are common, can be changed in midlife, and each has its own benefit for the heart, mood or daily life:
Get blood pressure under control: it damages exactly the brain's small vessels, and the benefit of lowering it for the heart and for preventing stroke has long been settledDo not leave hearing loss untreated: it is one of the largest factors on this listGet moving: it acts on blood pressure, blood glucose, sleep and mood at the same time
The chapters that follow take each one in turn: how to do it and how strong the evidence is.
Why split the list by stage of life
The same factor weighs differently at different ages, which is often overlooked. The midlife column is the longest because damage in those years builds up slowly. Blood pressure has to run high for ten or twenty years before the brain's small vessels are pressed into structural change, and those changes take a long time more to show up as cognitive symptoms you can notice. The later you start, the harder it is to undo what has already built up.
Turned around, that is the best news on the list: what you do in midlife pays out decades later. The feeling that it is still early is exactly the condition that lets it work.
Chapter 4
How blood vessels affect the brain
That risk list is almost all cardiovascular because the brain is extremely hungry for blood: it is about 2% of body weight yet uses about 20% of the body's blood flow and oxygen. When the supply system fails, the brain pays first.
What is good for the heart is broadly good for the brain. This is one of the few conclusions in brain-health research that points consistently in one direction. High blood pressure, diabetes, high (bad cholesterol) and smoking damage the brain's vessels large and small, and the damage builds up as vascular injury: small strokes (tiny, often unnoticed areas of dead brain tissue) and white-matter lesions (damage to deep brain tissue seen on scans). These cause vascular cognitive decline directly, and they also leave the brain less able to withstand the changes of Alzheimer's disease.
So controlling blood pressure, blood glucose and blood lipids is one effort with two payoffs: every mechanism described in the stories on hypertension, type 2 diabetes and dyslipidemia carries an extra meaning here, protecting the brain. Exercise is on the list too, but the strength of its evidence is different: its benefits for blood pressure and blood glucose rest on many trials, while trial evidence that it directly prevents dementia is still lacking.
What is good for the heart is broadly good for the brain. This is one of the few conclusions in brain-health research that points consistently in one direction. High blood pressure, diabetes, high (bad cholesterol) and smoking damage the brain's vessels large and small, and the damage builds up as vascular injury: small strokes (tiny, often unnoticed areas of dead brain tissue) and white-matter lesions (damage to deep brain tissue seen on scans). These cause vascular cognitive decline directly, and they also leave the brain less able to withstand the changes of Alzheimer's disease.
So controlling blood pressure, blood glucose and blood lipids is one effort with two payoffs: every mechanism described in the stories on hypertension, type 2 diabetes and dyslipidemia carries an extra meaning here, protecting the brain. Exercise is on the list too, but the strength of its evidence is different: its benefits for blood pressure and blood glucose rest on many trials, while trial evidence that it directly prevents dementia is still lacking.
Mechanism · Why bad vessels hit white matter first
The brain's blood is delivered from the surface inward. Large arteries branch across the surface, finer and finer, and finally plunge into the interior. Deep white matter, the wiring district inside the brain packed with insulated cables, sits at the far end of this supply system, where several vessels' territories meet and none has much to spare. Medicine calls such a place a watershed area.Year after year, high blood pressure pounding through these tiny penetrating arteries thickens their walls, narrows their channels and stiffens them; high blood glucose, high blood lipids and smoking each worsen the same thing from a different angle. The result is deep white matter that lives, for years, somewhere between just enough and not quite enough.
On current understanding, when supply falls short, the first thing to suffer is myelin, the insulation around the axon, which takes a lot of energy to maintain. Once the insulation is damaged, signals travel more slowly along the axon. Anything that needs several regions to hand off (understanding a sentence, doing a sum, deciding whether to brake while driving) must cross white matter at every handoff, so the slowing comes in chains.
So the white-matter lesions seen on a scan, the vascular cognitive decline doctors describe, and the half-a-beat-slower reactions you notice yourself are often three layers of one process.
This chain also shows why vascular risk and Alzheimer's changes stack. A person with both has fewer backup routes left, so the same load of plaques brings symptoms sooner.
Exercise: worth doing, but state the evidence accurately (Brasure 2018)
The idea that exercise prevents dementia is everywhere. Here is what the evidence actually says:
A systematic review done for a US national panel found that the existing (RCTs, which randomly split people into an exercise group and a comparison group) are not enough to conclude that exercise prevents cognitive decline or dementia. Most trials were small, most followed people for only about six months, and the exercise programs and measuring tools varied widely.Some results in the same body of data leaned toward benefit, but not strongly enough to conclude anything. Multidomain programs that bundled exercise, diet and cognitive training improved several cognitive measures, but that evidence, too, was rated only low in strength.The benefits of exercise for blood pressure, blood glucose, blood lipids, sleep and mood are backed by many trials, and all of those are brain-protective factors on the list.
The honest version, then: exercise is one of the most worthwhile things you can do for your brain, because it helps indirectly through several proven pathways, not because a trial has shown that running directly prevents dementia. For how to train, see Exercise as medicine.
The calibration here matters: neither "exercise is useless" nor "exercise guarantees protection from dementia." The claim should match the evidence.
While you are here, learn to read sentences like this
The evidence is not enough to conclude and it has been shown not to work are two different statements that often get mixed up. The first means we do not know yet; the second means we know it does not work. The exercise section above is the first kind; the large ginkgo trial in the chapter Do brain supplements work? is the second. Once you can tell these two apart, you can judge for yourself how much weight a study shows deserves.
Chapter 5
Do brain supplements work?
No supplement has been shown by reliable evidence to prevent dementia or make people with normal cognition sharper.
Fear of losing your memory and the wish to protect your brain are real, and they have turned brain and memory supplements into a big business. The most thoroughly studied is ginkgo. In a large, long, adequately dosed randomized double-blind trial (neither the participants nor the researchers knew who got the real pill; DeKosky 2008, the GEM study), it neither lowered the rate of dementia nor slowed the move from mild cognitive impairment (MCI) to dementia. This is not too small a sample to see an effect; it is tested properly, and no effect. A systematic review that went through the common brain supplements one by one (Butler 2018) reached an equally sober conclusion: for people with normal cognition or MCI, the evidence is not enough to recommend any of them.
Keep two things separate. A true deficiency should be treated: a lack of vitamin B12 can cause reversible cognitive problems, and older adults absorb it less well, so it is worth testing (see Vitamin B12). And eating well overall, which observational studies link with better cognition, is not the same as any single capsule working (see Fats & Omega-3).
Move that money and attention to unglamorous things with evidence behind them, such as controlling blood pressure and looking after your hearing, and the return is far higher.
Fear of losing your memory and the wish to protect your brain are real, and they have turned brain and memory supplements into a big business. The most thoroughly studied is ginkgo. In a large, long, adequately dosed randomized double-blind trial (neither the participants nor the researchers knew who got the real pill; DeKosky 2008, the GEM study), it neither lowered the rate of dementia nor slowed the move from mild cognitive impairment (MCI) to dementia. This is not too small a sample to see an effect; it is tested properly, and no effect. A systematic review that went through the common brain supplements one by one (Butler 2018) reached an equally sober conclusion: for people with normal cognition or MCI, the evidence is not enough to recommend any of them.
Keep two things separate. A true deficiency should be treated: a lack of vitamin B12 can cause reversible cognitive problems, and older adults absorb it less well, so it is worth testing (see Vitamin B12). And eating well overall, which observational studies link with better cognition, is not the same as any single capsule working (see Fats & Omega-3).
Move that money and attention to unglamorous things with evidence behind them, such as controlling blood pressure and looking after your hearing, and the return is far higher.
Evidence · The ginkgo trial and the supplement review
Ginkgo: the GEM study (Ginkgo Evaluation of Memory)Design: more than 3000 community-dwelling adults aged 75 or older, with normal cognition or MCI (mild cognitive impairment), randomized and double-blind. One group took a standardized ginkgo extract, EGb761, 120 mg twice a day; the other took a placebo. Follow-up was about 6 years.Result: ginkgo did not lower the rate of all-cause dementia or of Alzheimer's disease, and in people who already had MCI it did not slow progression to dementia.This is a weighty negative result: not a sample too small to see anything, but a large sample, long follow-up and a full dose showing that it really does not work.
Why this negative result is especially valuable
Most trials that found no effect leave a back door open: too few people, too little time, too low a dose, so someone can still say maybe it works and we just could not see it. GEM shut all three at once: enough people, long enough follow-up, a full standard dose. With every door shut there was still no effect, and that is what moves the conclusion from we do not know to it does not work.
The next time you read that a study shows it works, turn the same ruler around and ask three questions: How many people? Followed for how long? Was the dose the one you can actually buy?
The whole category of over-the-counter brain supplements
A second systematic review done for the same US national panel (Butler 2018) went through the common over-the-counter (OTC) brain supplements one by one:
It covered ω-3 (fish oil), soy, ginkgo, folic acid and B vitamins, β-carotene, vitamin C, vitamin D with calcium, multivitamins, and multi-ingredient formulasThe conclusion was uniform and sober: for people with normal cognition or MCI, current evidence is not enough to recommend any over-the-counter supplement for preventing cognitive declineThe only flicker was daily folic acid plus B12: its improvement on some memory tests was statistically significant, but its clinical importance is doubtfulThe ω-3 finding was specific: in a study of about 884 people followed for 4 years, ω-3 supplements did no better than a B-vitamin comparison group on overall cognition and memory
Why eating well overall is linked to better cognition, yet a capsule does nothing
These look contradictory but are not, and you can work out why:
Observational studies can only tell you that two things show up together. People who eat fish, vegetables and olive oil for years also tend to sleep well, move more, keep their blood pressure in check and stay socially active, and each of those is on the brain-protection list in its own right. Crediting the whole result to one food books a whole way of life to the wrong account.Even if one component truly helps, several steps separate it in food from it in a capsule: whether the dose is right, whether the absorbed form is the same, whether other components of the food come with it. Break any one step and the capsule cannot reproduce what happens at the table.More fundamentally, the white-matter-and-vessels chain in the chapter How blood vessels affect the brain is worn in over many years. A capsule cannot undo decades of vessel damage, while a way of eating is repeated every day for decades, and only that can reach the chain.
So nutrition matters to the brain and supplements help the brain are not the same sentence. The first holds; the second, for now, does not.
Chapter 6
What helps most, and warning signs
What to do for your brain is actually clear. The biggest levers are exactly the unglamorous things that take persistence and that each pay off for health in their own right. From the strongest evidence to the weakest:
Tier 1 · Look after your blood vessels: blood pressure, blood glucose, blood lipids, quitting smoking. How these damage the brain's vessels is the best understood, and the benefit of controlling them for the heart and for stroke has long been settled.Tier 2 · Do not neglect your senses: not hearing well is not "just aging, put up with it." Get your hearing tested and wear a hearing aid if you need one; do not put off the glasses you need or the cataract surgery you have been advised to have.Tier 3 · Use your brain, stay connected, sleep well: learning new things, keeping a challenging hobby and staying social are where cognitive reserve comes from; sleep and exercise each bring benefits of their own.Tier 4 · Do them together: a randomized trial in Finland suggests that doing several of these together may beat relying on any one.
Tier 1 · Look after your blood vessels: blood pressure, blood glucose, blood lipids, quitting smoking. How these damage the brain's vessels is the best understood, and the benefit of controlling them for the heart and for stroke has long been settled.Tier 2 · Do not neglect your senses: not hearing well is not "just aging, put up with it." Get your hearing tested and wear a hearing aid if you need one; do not put off the glasses you need or the cataract surgery you have been advised to have.Tier 3 · Use your brain, stay connected, sleep well: learning new things, keeping a challenging hobby and staying social are where cognitive reserve comes from; sleep and exercise each bring benefits of their own.Tier 4 · Do them together: a randomized trial in Finland suggests that doing several of these together may beat relying on any one.
Evidence · The trial behind each tier
Tier 1: look after vascular risk (the most complete case)Blood pressure: midlife high blood pressure damages the brain's small vessels, the chain described in the chapter How blood vessels affect the brain. Bringing blood pressure into the target range your doctor sets has long-settled benefits for the heart and for preventing stroke, and the brain gets a share on the side. For the mechanism, see Hypertension.Blood glucose and lipids: controlling diabetes and high (bad cholesterol) protects heart and brain together. See Type 2 diabetes and Dyslipidemia.Quitting smoking: the benefit for the brain's vessels is as certain as it is for the heart and lungs.
Tier 2: do not neglect your senses, hearing above all (Lin 2023)
Hearing loss is one of the largest factors on the list. ACHIEVE was a large testing whether a hearing intervention can slow cognitive decline: 977 adults aged 70–84 with untreated hearing loss, one group given a hearing intervention (including hearing aids), the other given health education.Read the result carefully. Across all participants, the difference in 3-year cognitive decline was not statistically significant. In the subgroup at higher risk of cognitive decline, decline slowed by about 48%. The benefit was concentrated in people at higher risk, and a subgroup result like this still needs to be repeated in other trials.What this means in practice: do not treat poor hearing as normal aging to put up with. Get your hearing tested and wear hearing aids if you need them. In the same way, do not put off glasses you need or cataract surgery (the 2024 list added vision loss).
How are hearing and cognition actually connected?
Readers ask this more than anything else, so here are the current explanations and how firm each one is:
Less input: the sound signal coming up from the inner ear weakens, the patch of cortex that processes sound has nothing to do for years, and pathways that are used less fade.Understanding borrows other resources: when you cannot hear clearly, you are really guessing, filling in missing syllables from context, lip movements and experience. That guessing uses attention and memory, so in the same conversation someone with poor hearing spends extra mental effort, leaving less for understanding and remembering. It also explains a very common experience: in a noisy restaurant, a person with poor hearing not only cannot hear clearly but also cannot remember what was just said.Less social life: when you cannot hear, conversation is tiring, so you go out less and drift into social isolation, and social isolation is itself on the risk list.
All three are still explanatory hypotheses, with no winner yet. What ACHIEVE showed is that the intervention may help people at higher risk; it did not show which pathway does the work. The calibration is here so that you know to get your hearing checked without treating any one mechanism as settled.
Tier 3: use your brain, stay connected, sleep well (a plausible mechanism, and each has its own benefits)
Lifelong mental and social engagement: the source of cognitive reserve, through learning new things, keeping a challenging hobby and staying social. Social isolation is itself on the list.Sleep: long-term sleep disruption is linked with cognitive risk (an observed association); breaking the insomnia cycle is worth doing in its own right. See Insomnia and Chronic stress.Exercise: as described earlier, its indirect pathways are solid, making it one of the most worthwhile habits there is. See Exercise as medicine.
Tier 4: doing several things together may beat relying on one (Ngandu 2015)
The FINGER trial recruited older adults in Finland whose risk of cognitive decline was raised and randomly split them into two groups. One received a 2-year multidomain program (diet, exercise, cognitive training and monitoring of vascular risk); the other received standard health advice. On the change in total score across a battery of cognitive tests, the program group did better than the comparison group, and the difference was statistically significant.It measured test scores and did not show that dementia was prevented. What it suggests is that doing the things above together may work better than betting on one.
Why doing them together can beat picking one
This is not a vague more is always a bit better. It follows from reasons the earlier chapters already gave:
These factors feed the same road. Blood pressure, blood glucose, blood lipids and smoking all damage the same small vessels in the brain, and sleep and exercise in turn affect blood pressure and blood glucose. Control only one and the others keep doing damage in the same place.Cognitive reserve and vascular health, though, run on two different roads: the first decides how many backup lines you have left, the second how fast the lines themselves wear out. Working on both at once means destroying fewer roads on one side and building more on the other.
So the answer to which one should I do first is often the one you can keep doing, adding the others gradually. They are not substitutes for one another; they stack.
Red flag · When worse memory needs a doctor
My memory has been worse lately. Should I manage it myself or see a doctor?Most likely normal aging or a reversible cause (start with lifestyle)
Now and then you cannot recall a name, or walk into a room and forget why, but a cue brings it backDaily life, work, money and medicines all carry on as usualOften tied to reversible factors such as poor sleep, stress, low mood, certain medicines, or a thyroid or B12 problem; check and deal with these first (see Vitamin B12, and the stories on insomnia and chronic stress)
When to have a doctor assess it
You or your family feel it is clearly worse than before, and it has lasted several months or moreGetting lost repeatedly, asking the same question over and over, mistakes with medicines or money, changes in judgmentThis deserves a proper cognitive assessment, not scaring yourself or buying supplements blindly; an early assessment also means reversible causes get treated sooner
Red flags (seek care promptly)
Sudden confusion, slurred speech, weakness on one side, facial droop or loss of part of the field of vision → treat as a stroke and go to the emergency department immediately (time is brain)Cognitive decline that progresses quickly over days to weeks, plus an unsteady walk or muscle jerks (myoclonus) → look for a treatable acute causeA change in thinking plus fever, a severe headache or a recent head injury → seek care immediately
These are true red flags, not "just getting old." Do not wait.
Related stories
Cognitive aging pulls several threads together:
The vascular-risk trio: hypertension, type 2 diabetes and dyslipidemia, each of which also protects the brainBackgrounds you can change: insomnia and chronic stressThe brain-protecting habit with the most solid indirect pathways: Exercise as medicineA reversible nutritional cause of cognitive problems: Vitamin B12The difference between an overall diet and a single capsule: Fats & Omega-3Depression is on the risk list and is also often mistaken for age-related cognitive decline: Depression & Anxiety
Normal slowing is not dementia, and dementia is not inevitable. The evidence says a sizable share of dementia risk lies in factors that can be changed, and the things truly worth doing first are controlling blood pressure, looking after your hearing, moving, sleeping well, and keeping your mind busy and your ties to people alive, not supplements. Understanding the mechanism means you neither panic nor get fleeced. This page is not a diagnosis; any clear or lasting change in thinking should be assessed by a doctor.
References · 9
- Harada, C. N., Natelson Love, M. C., & Triebel, K. L. (2013). Normal cognitive aging. Clinics in Geriatric Medicine, 29(4), 737-752. Processing speed and fluid intelligence decline with age while crystallized intelligence (vocabulary, accumulated knowledge) is largely preserved. 10.1016/j.cger.2013.07.002
- Stern, Y. (2012). Cognitive reserve in ageing and Alzheimer's disease. The Lancet Neurology, 11(11), 1006-1012. Defines cognitive reserve (vs brain reserve) as the brain's capacity to use efficient/alternative networks, explaining why equal pathology produces unequal clinical impairment. 10.1016/S1474-4422(12)70191-6
- Livingston, G., Huntley, J., Sommerlad, A., et al. (2020). Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. The Lancet, 396(10248), 413-446. Twelve modifiable risk factors account for ~40% of worldwide dementia (population attributable fraction). 10.1016/S0140-6736(20)30367-6
- Livingston, G., Huntley, J., Liu, K. Y., et al. (2024). Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission. The Lancet, 404(10452), 572-628. Adds high LDL cholesterol and untreated vision loss for 14 modifiable factors totaling ~45% of dementia risk. 10.1016/S0140-6736(24)01296-0
- Brasure, M., Desai, P., Davila, H., et al. (2018). Physical activity interventions in preventing cognitive decline and Alzheimer-type dementia: a systematic review. Annals of Internal Medicine, 168(1), 30-38. Evidence insufficient to conclude that physical activity prevents cognitive decline/dementia, though trends favored benefit. 10.7326/M17-1528
- DeKosky, S. T., Williamson, J. D., Fitzpatrick, A. L., et al. (2008). Ginkgo biloba for prevention of dementia: a randomized controlled trial (GEM study). JAMA, 300(19), 2253-2262. EGb761 120 mg twice daily in >3000 adults aged 75+ over ~6 years did not reduce incidence of dementia or Alzheimer's disease. 10.1001/jama.2008.683
- Butler, M., Nelson, V. A., Davila, H., et al. (2018). Over-the-counter supplement interventions to prevent cognitive decline, mild cognitive impairment, and clinical Alzheimer-type dementia: a systematic review. Annals of Internal Medicine, 168(1), 52-62. Evidence insufficient/low-strength for omega-3, ginkgo, B vitamins, vitamins C/D, multivitamins; none recommended for cognitive protection. 10.7326/M17-1530
- Lin, F. R., Pike, J. R., Albert, M. S., et al. (2023). Hearing intervention versus health education control to reduce cognitive decline in older adults with hearing loss in the USA (ACHIEVE): a multicentre, randomised controlled trial. The Lancet, 402(10404), 786-797. Among 977 adults aged 70-84, hearing intervention slowed 3-year cognitive decline by ~48% in the higher-risk subgroup (non-significant in total population). 10.1016/S0140-6736(23)01406-X
- Ngandu, T., Lehtisalo, J., Solomon, A., et al. (2015). A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (FINGER): a randomised controlled trial. The Lancet, 385(9984), 2255-2263. Multidomain intervention significantly improved overall cognitive performance vs control. 10.1016/S0140-6736(15)60461-5