The sentence it is standing in for
“Supports memory” belongs to a category of wording known as structure-function. A dietary supplement is free to talk about affecting how some part of the body is built or how it behaves. What it may never say is that it diagnoses, treats, cures or heads off an illness — print that and the bottle has become a drug, with an entirely different approval route sitting behind it.
Hence “supports memory” passes and “treats memory loss” does not. This whole aisle lives in the space between those two sentences.
Read it, then, as a statement about what a seller may lawfully put on a box. It says nothing about what anybody has demonstrated. Only the second of those is worth your time.
Five memories, not one
Memory is a set of systems. A claim that will not say which system it means is not a claim you can weigh.
| Type | Everyday example | Affected by age? |
|---|---|---|
| Working | Keeping a code in mind while you hunt for a pen | A little |
| Episodic | Who said what over Tuesday's lunch | Hit hardest |
| Semantic | Words, dates, how an engine works | Level or rising |
| Procedural | Clutch control, touch typing, scales | Nearly permanent |
| Prospective | Posting the letter on your way out | Fails most visibly |
“My memory is going” nearly always points at the episodic and prospective rows. Those two also happen to be the pair the years press on hardest, which is why the complaint is near-universal and hardly ever a sign of anything out of the ordinary.
Nothing was lost, so nothing gets restored
Past fifty, the trouble is usually the search rather than the file. Material sits there intact while the route in has slowed, which is the whole reason a word you gave up on surfaces in the parking lot with nobody waiting for it.
A memory supplement is therefore working, at absolute best, on the speed of a lookup. Restoration never comes into it, because nothing went missing. The full version of that distinction sits here — it also happens to mark the boundary between ordinary and worth-a-visit.
Count what the studies actually counted
The ginkgo finding propping up this phrase concerned appointments volunteers had undertaken to keep — part episodic, part prospective, and conveniently close to the thing people grumble about. Six weeks, 188 healthy volunteers somewhere between forty-five and fifty-six.
Recognition memory in that same group: no advantage whatsoever. Neighbouring task, one domain, flat result.
Curcumin was tested on composite scores for memory and attention, forty people followed for eighteen months. Lutein was attention, fifty-one people, one year. That is the entire body of work standing behind the phrase as applied to this formula: real findings, narrow ones, and a long way short of “improves memory” said flatly.
Three questions and a label becomes readable
- Which single ingredient is carrying this claim?
- What did that ingredient's trial measure, over what period, in how many people?
- Does the milligram figure here match the figure in that trial?
Amounts pooled into a blend leave question three unanswerable, and a phrase you cannot check is packaging rather than information.