The Cognitive Supplement Guide

The complaint that sends someone to this shelf is rarely dramatic. A name that will not come in the middle of a sentence. A second trip upstairs for the thing the first trip was for. A paragraph read twice because the first pass did not land.

Most of that is ordinary and most of it is attention rather than memory. Attention narrows when sleep is short, when a week is loud, when there is more to hold than there used to be, and it narrows in a way that feels exactly like forgetting.

There is a second group of causes that matters more and gets looked at less. Hearing loss makes conversation effortful and effort looks like forgetfulness. Thyroid function, blood pressure, alcohol and several common medicines do the same. The National Center for Complementary and Integrative Health puts the sensible order plainly: talk to the person who already knows what else you take before you add something to it.

Four families, and the third is the one whose reputation on this shelf is borrowed rather than earned.

Most of the money on this shelf is in the first family and most of the interesting recent work is in the last one. The third is where labels get confusing, because its literature is enormous and almost none of it is about memory.

Niacin is vitamin B3 and the body converts it into the two coenzymes that carry electrons through energy metabolism. Every cell uses them and nervous tissue shows a shortage early, which is the whole of the mechanism and the whole of the honest claim.

The Institute of Medicine chapter sets the adult requirement at 16 mg NE a day for men and 14 for women, with a tolerable upper intake of 35 mg a day set on flushing. Those three numbers are the yardstick for any niacin row on any label, and they are unusual on this shelf in being settled.

The cognition research is weaker than the adequacy research and it is important to say which is which. The Chicago Health and Aging Project analysis followed thousands of older adults and compared quintiles of niacin from FOOD. It is observational, it is about diet, and it is not a trial of a supplement. A 2025 observational study of B-vitamin intake reaches a similar kind of conclusion with similar limits, and the MedlinePlus entry on niacin is the plain-language version of the same position.

Arginine is the amino acid the body builds nitric oxide from, and nitric oxide relaxes blood vessels. That single sentence is why five amino acids appear on cognition labels all over this shelf, and the rest of this section is about how far it actually carries.

Taken by mouth, arginine is heavily metabolised in the gut wall and the liver before it reaches the circulation, which is why trial protocols reach for grams. The head-to-head pharmacokinetic study found citrulline the more reliable route to the same end, and it still took 3 g twice daily to raise trough plasma arginine. The dose-ranging study worked with single doses of 2 to 15 g.

The vascular results are real and they are conditional. A meta-analysis of short-term oral arginine found the effect on flow-mediated dilation concentrated in people whose baseline was low. A 2018 systematic review reached a similar conclusion in patients with cardiovascular or metabolic disease, and a 2021 meta-analysis set arginine against ordinary tree nuts, which is a useful sense of scale.

The cognition evidence is one trial. A 2026 randomised, placebo-controlled pilot gave 6.4 g of oral L-arginine a day for 30 days; 38 were randomised and 35 finished, with sixteen in the arginine arm. It is a pilot, its authors call it one, and it is the only randomised cognition trial on oral arginine this guide could find. A 2026 study of arginine metabolites in small-vessel disease is a different kind of paper about the same neighbourhood.

The step from “nitric oxide relaxes blood vessels” to “so this capsule improves thinking” has a name in the literature, and the people who work on it are candid that it is a hypothesis.

A 2024 review states it plainly: the nitric-oxide-dependent neurovascular coupling hypothesis, which is the proposal that raising nitric oxide availability improves the local blood-flow response that supports neural activity.

It has been tested. A 13-week randomised pilot gave overweight and obese older adults incremental doses of nitrate-rich beetroot juice and measured both cognitive function and cerebral blood flow. Neither was modified. A 2026 pilot in older adults with metabolic syndrome is the most recent attempt at the same question from a different angle.

Two things follow. A plausible mechanism is not a finding. And a route that produced no measurable change at trial doses is not more likely to at milligram ones.

Beta-alanine is the rate-limiting half of carnosine, a dipeptide that accumulates in muscle and in brain tissue over weeks rather than hours. Its reputation is built on exercise performance, and the cognition work is recent, small and genuinely worth reading.

The International Society of Sports Nutrition position stand sets the working threshold at 4 to 6 g a day for four weeks. A trial comparing two protocols ran 6 g a day for four weeks against 12 g a day for two. A 2025 systematic review of dosing strategies covers the same ground for strength and power. Those are the amounts this compound is studied at.

The cognition work is a different literature. A ten-week imaging trial in 60-to-80-year-olds reported improved fractional anisotropy scores in the hippocampus and amygdala at 2.4 g a day — with six supplemented and four control participants analysed, which is ten people in total. A meta-analysis of histidine-containing dipeptides found an improvement in delayed recall and closed by asking for the doses to be standardised. The NEAT trial gave 2 g of carnosine daily for up to twelve weeks and reported better outcomes in its younger participants.

Set against that, a 2026 multilevel meta-analysis found no ergogenic effect on repeated sprint ability at all, and a 2023 review of the proposed brain mechanisms is a collection of arguments rather than a conclusion. A systematic risk assessment describes paraesthesia as the only reported effect, tied to single doses around 800 mg and above.

Not every row on a supplement panel is there to do something in the body. Calcium carbonate is the industry’s standard dry filler and flow agent, and it appears on labels at weights that have nothing to do with the calcium conversation.

The adult requirement is 1,000 to 1,200 mg a day, and the MedlinePlus entry on dietary calcium is the plain-language version. Supplement research is run at hundreds of milligrams: an absorption crossover trial in postmenopausal women is typical of the design. The two large cardiovascular meta-analyses, one of supplements generally and one restricted to randomised trials, are conversations about gram-scale products.

So a calcium row at eighteen or twenty milligrams is not a small version of that conversation. It is a different subject, and the honest reading is that the row is doing a manufacturing job. The same is true of silicon dioxide, magnesium or vegetable stearate, and of the capsule shell itself.

One practical rule comes out of this: count the rows that are actives before you count the rows on the panel.

A dietary supplement in the United States may carry a structure or function claim and may not claim to treat or prevent a disease. That is why the language on this shelf is so uniform: supports memory, supports cognitive function, supports focus.

Five things on almost every label on this shelf, and the gap between what each one says and what it gets read as.

The second row is the valuable one: a printed milligram figure is the only thing on a label that can be checked against a published trial. The third row is why it matters, because a blend total tells a reader nothing about any component.

Every section above ends in the same place: a number on a label set beside a number in a trial. Doing that comparison takes a minute and it is the single most useful thing anyone can do on this shelf.

Every figure was read out of that paper’s own abstract. They describe the trials, not any particular capsule.

Hold that table beside any label on this shelf and most of the marketing falls away. A row at a fraction of a per cent of its own trial dose is not a weak version of the trial; it is a different quantity of the same molecule, and the trial is silent about it.

That is not a reason to buy nothing. It is a reason to know what you are buying, and to pay attention to the rows that are at real amounts. The Mind Honey Pro ingredients page runs this exact comparison on the bottle this website sells, row by row, including the five rows where it comes out badly.