Cognitive Supplement Guide: What The Evidence Shows

Walk along it and the labels fall into a small number of groups, which is more useful to know than any individual ingredient.

There is a botanical group, built on plants with long traditional use and a modern literature of varying quality. There is a phospholipid and choline group, sold on membrane and neurotransmitter arguments. There is a vitamin and mineral group, generally arguing from deficiency. And there is an amino acid group, which is where this guide spends most of its time, because it is the family that reaches cognition labels from somewhere else entirely.

That last point is the one worth carrying around. Arginine, citrulline and beta alanine built their reputations in exercise science, where the outcomes are power output, time to exhaustion and muscle perfusion. Their appearance on a memory label is an argument by extension: better blood flow, better-perfused brain tissue, better thinking.

The extension is plausible and it is not the same thing as evidence. This guide covers the families where the underlying trials can actually be pointed at, and it gives the quantity every time, because the quantity is where nearly every one of these arguments comes apart.

Arginine is the substrate an enzyme in the vessel lining turns into nitric oxide, and nitric oxide is the signal that relaxes the smooth muscle around a small artery. That chain is textbook physiology and nothing in this guide disputes it.

The difficulty is oral arginine. Swallowed, it meets arginase in the gut wall and the liver and a substantial share never arrives in circulation as arginine. The catabolism review sets out what the body does with it, and it is the reason the pooled human evidence starts so high: the 2022 dose-response meta-analysis found blood pressure moving at four grams a day and above, and not at all beyond nine.

Citrulline is the pharmacological answer. It bypasses that first-pass step and is converted to arginine in the kidney, so gram for gram it raises plasma arginine further. The dose-ranging pharmacokinetic study climbed 2, 5, 10 and 15 grams in the same volunteers, which tells you what scale the researchers considered worth testing.

And then the outcome evidence disappoints. The pooled citrulline blood pressure analysis found no significant change in brachial or central pressure. The exercise literature is friendlier: a meta-analysis of citrulline malate and strength, a twelve-gram crossover with caffeine and a head-to-head asking whether the malate matters at all all work at gram scale.

The friendliest perfusion result in this whole family is worth knowing as a benchmark. A contrast-ultrasound crossover did move muscle perfusion measurably, using three grams of arginine hydrochloride and eight grams of citrulline malate in a single serving. Eleven grams of amino acids, in one dose, to shift blood flow in a leg.

Beta alanine is the rate-limiting precursor of carnosine, a dipeptide stored in muscle and in brain tissue. Unlike the vascular family it does nothing acutely: the effect, if any, comes from accumulation across weeks.

The sports nutrition position stand describes loading as four to six grams daily for at least four weeks. A carnosine loading study compared two protocols and measured washout, and a formulation comparison ran six grams a day for twenty-eight days to separate sustained from rapid release.

What makes this family interesting on a cognition label is that it carries the most direct evidence in the whole aisle for the outcome those labels are sold on. A 2024 meta-analysis of histidine-containing dipeptides pooled the trials and reported improved delayed recall.

The underlying trials are specific and they are not large. A ten-week randomised study in older adults used 2.4 grams a day and measured cognition, mood and physical function; an imaging companion reported structural changes in the hippocampus and amygdala in the same cohort. An executive-function trial used the same 2.4 grams in middle-aged participants after endurance exercise.

And the negative result belongs beside them. A fourteen-week study at 6.4 grams a day moved neither serum BDNF nor Stroop test performance, and a four-week military trial at the same dose tested physical and cognitive performance together. A family with a positive pooled finding and a clean null at a higher dose is a family where the question is open.

For a label reader the arithmetic is the whole story: the positive results sit at 2.4 grams a day for ten weeks. A row printed in tens of milligrams is a different proposition, however encouraging the meta-analysis.

Niacin is the precursor of NAD, the coenzyme that moves electrons through nearly every energy-releasing reaction a cell runs. The pathway is not in dispute and neurons run it hard.

The established link between this vitamin and thinking runs through deficiency. Pellagra produces confusion alongside its skin and digestive signs, which is why B3 appears in any discussion of nutrition and cognition at all. It is a disease of absence rather than a lever.

The modern research uses a different compound. Nicotinamide riboside is a B3 form given specifically to raise NAD, and the trials are serious: a randomised placebo-controlled study gave one gram a day for ten weeks to older adults with mild cognitive impairment, measuring cerebral blood flow alongside cognitive scores, and a crossover in subjective cognitive decline and mild impairment gave the same amount for eight weeks. The second found no significant between-group difference on its primary outcome.

None of that transfers to a niacin row on a label. A gram a day of nicotinamide riboside and a few milligrams of niacin equivalents are different compounds at different scales, and a label carrying the second is providing a nutrient rather than running the experiment.

One practical note for readers. The flushing reaction people associate with niacin needs immediate-release nicotinic acid in the hundreds of milligrams. Nobody should treat a warm face as evidence a capsule is working, and nobody should expect one from a nutritional amount.

Minerals turn up on these panels in two very different roles and it is worth telling them apart.

Occasionally the mineral is the point: magnesium sold for sleep, zinc sold on deficiency, iron in a population where deficiency is genuinely common. Those rows carry amounts in the tens or hundreds of milligrams and a percentage of a daily reference beside them.

More often the mineral is there for the tablet or the capsule itself. Calcium carbonate is dense, cheap and flows well, and a small amount improves how a powder behaves in a filling machine. A row of that kind will carry a single-figure percentage of a daily value, and that percentage is the tell.

The useful thing about the second sort is what it rules out. The whole body of advice about spacing calcium away from thyroid medicine comes from real supplement doses: 1,200 mg of elemental calcium in the original study and 500 mg in the follow-up comparing salts. A row at a few per cent of a daily value cannot reach that mechanism, and a reader can stop worrying about it.

US panels declare elemental mineral content with the salt named in brackets, so 18 mg of calcium as calcium carbonate means 18 mg of calcium inside roughly 45 mg of carbonate. That convention is worth knowing before comparing two labels.

Every section above ends in the same place, and it is not a coincidence. Across this aisle the ingredients are real, the mechanisms are real, and the amounts on the labels are a fraction of the amounts in the trials.

There are two honest reasons for that and one dishonest one.

The honest reasons are physical. A capsule somebody will swallow once a day holds somewhere around 600 to 800 mg of powder. Four grams of arginine is five or six capsules on its own, and a formula carrying six ingredients at trial dose would be a scoop rather than a capsule. Cost is the second: these are commodity materials, but gram-scale servings across six ingredients multiply quickly.

The dishonest reason is that a small amount of a well-known ingredient sells almost as well as a large one. A label can name citrulline and borrow every headline the citrulline literature ever generated, and nothing on the panel obliges it to print the quantity the headline came from.

Which is why the scorecard in the companion guide puts the dose question second, and why the only labels that can be assessed at all are the ones that print a figure against every row.

Anybody researching citrulline will meet a 2018 meta-analysis in the Journal of Human Hypertension reporting that citrulline supplementation lowered blood pressure. It is the most-quoted paper in this corner of the aisle.

It carries a retraction notice. The record in PubMed is flagged as a retracted publication and the notice itself was published in the same journal in 2021. Neither fact reaches the marketing pages that cite the original, because a citation copied from another marketing page is never checked against the record.

The un-retracted replacement points the other way. A meta-analysis in Current Hypertension Reports pooled the clinical trials and found no significant effect on brachial or central blood pressure. That is the result this website cites, on several pages, including the ones about the product it sells.

The general lesson is more useful than the specific one. A citation in this category is worth exactly as much as the last person who checked it, and checking is a thirty-second job: open the PubMed record, read the flags at the top, and look at what the abstract says the dose was.

That is the rule this website applies to its own reference list, and it is the reason the list is shorter than it could be.

Dietary supplements in the United States may make structure and function claims: statements that an ingredient supports or helps maintain the normal working of some part of the body. They may not claim to diagnose, treat, cure, prevent or mitigate a disease, and that boundary is what separates a supplement from a drug.

In practice the strongest permissible sentence in this aisle is something close to “supports cognitive function”. It sounds like a promise and it is a category description, and it requires no trial behind it.

Any statement about memory loss, about decline, about dementia or about treating anything is outside the boundary. A label carrying one has a compliance problem rather than a strong product, and a reader who spots one has learned something about the seller.

The FDA's questions and answers on dietary supplements sets out what the agency does and does not do here, which is less than most buyers assume: these products are not approved before sale and their claims are not reviewed.

The practical consequence for a reader is that the claim on the front of a bottle carries almost no information. Everything that does carry information is on the back.

This website sells one product in this aisle, so the guide would be incomplete without saying where it lands.

It carries three of the four families: the vascular amino acids as five of its seven rows, a carnosine precursor, and a B vitamin. It prints a figure against every row and carries no blend total, which puts it in the minority of labels that can be assessed at all.

And the amounts sit below their own trial quantities, on every amino acid row, by factors from about twenty upwards. That is the finding the whole of this guide predicts, and it applies here as it applies across the aisle.

The full scoring is on the scorecard, where this product fails three checks of six, and every printed figure is in the panel page.