1. Treat hearing loss
Top of the list and almost never acted on. Among modifiable factors, untreated hearing loss carries one of the heaviest published associations with later cognitive decline. The typical gap between noticing a hearing problem and doing something about it is measured in years.
The immediate mechanism is simpler than the long-term one: effort spent decoding speech is effort not available for storing it. If you are asking people to repeat themselves, get a hearing test before you buy anything else on this page.
2. Walk
Aerobic exercise has the strongest evidence base of any intervention in this field, and the useful version of it is ordinary walking rather than anything that requires equipment or a membership. Regular, moderate, most days.
Nobody has a convincing single mechanism for why it works this well. It works anyway, consistently, across many studies.
3. Sleep, and specifically breathing while asleep
Memory consolidation happens during sleep. That is not a metaphor; it is when the day's material gets filed. Chronic short sleep degrades it directly.
The specific thing worth chasing is sleep apnoea. If anyone has told you that you snore heavily, or that you stop breathing at night, that is worth a proper investigation. It is common, routinely missed, and treating it improves next-day function measurably.
4. Control blood pressure
Midlife hypertension is among the sturdier associations with later cognitive outcomes. It is also silent, which is why it goes unmanaged. A cuff at home costs very little and tells you more than any brain-health article will.
5. Talk to people
Social contact outperforms puzzle books, and the reason is structural: a person cannot be anticipated and a grid of clues can. Conversation demands attention, prediction, memory for what was just said, and emotional reading, all at once and in real time.
If you are choosing between a daily crossword and a weekly conversation with somebody you find slightly difficult, choose the conversation.
6. Eat in a way that protects your blood vessels
The dietary patterns with the best evidence here — Mediterranean-style eating, broadly — are also the ones that are good for the heart, which is probably not a coincidence. Vegetables, fish, olive oil, nuts, less processed meat, less sugar.
No single food does anything. Anybody selling you one is selling you a story.
7. And then, if you want, a supplement
Last. Deliberately. A daily capsule cannot replace any of the six above, and taking one while an untreated sleep disorder or uncorrected hearing loss continues is how a year goes by with nothing improving.
If you do want one, the standard worth holding it to is whether the label publishes amounts rather than hiding them in a blend, and whether anyone will tell you which ingredients have evidence and which do not. In VapoCept's formula, four of six have randomised human trials attached and two have none at all — and we would rather name them than let the label imply otherwise.
What to ignore
- Brain training apps. People get better at the app. Transfer to daily life is much less clear.
- Single superfoods. No individual food has been shown to do anything meaningful for memory.
- Anything promising to reverse memory loss. Nothing sold as a supplement does this, and the claim itself tells you what you are dealing with.
- Mega-dose vitamins without a deficiency. If your B12 is low, correcting it matters. If it is not, more does nothing.