// learn.shawon.ch / evidence-desk / magnesium-l-threonate VERDICT
← Evidence desk

Evidence desk · Supplement

Magnesium L-threonate

The verdict

Skip it. Magnesium L-threonate is sold as the form that crosses into the brain and sharpens memory and cognition. That story rests on rodent work plus a handful of small, short, often company-linked human trials — the human cognition evidence in healthy adults is weak.

It also delivers very little elemental magnesium per large dose and costs far more than plain forms. If you actually want magnesium, use glycinate for repletion. The "brain magnesium" benefit is not established.

Before you read on

A $40 bottle of "brain magnesium" promises a real cognitive edge — and points to published trials to prove it. One question separates a real edge from a marketing story. Which question, and which way does the answer here cut? Decide before you scroll.

The claims, sorted

L-threonate is pitched on cognition first and magnesium second. Separate the two: the cognition claim is the whole reason to pay the premium, and it's the one that doesn't hold.

What's claimed vs. what holds
ClaimBest evidenceHolds up?
Raises brain magnesium → better memoryRodent studies (Neuron 2010)Animal only
Improves cognition in healthy adultsFew small RCTs, often industry-linkedWeak
Efficient way to correct magnesiumElemental-content mathNo — low yield
Worth the price over plain formsCost per gramNo

Where the evidence actually comes from

This is the funding filter again — the same trap as the collagen card, just dressed in neuroscience. The headline mechanism comes from a 2010 Neuron paper (Slutsky, Liu, et al.) showing that a then-new compound, magnesium L-threonate, raised magnesium in rat brains and improved learning and memory. Genuinely interesting science. But it is rats, and the lead investigator went on to commercialize the compound as the patented form "Magtein."

So when you reach the human trials, ask Chapter 14's question — who paid, and who designed it? The human RCTs are few, small (roughly 100 people each), short (6–12 weeks), and several are tied to the maker or test a multi-ingredient "Magtein-based formula" rather than threonate alone. Trials run or funded by the patent-holder reporting that the patent-holder's product works is exactly the pattern the toolkit teaches you to discount. Strip those out and the independent human cognition evidence in healthy adults is close to empty.

WEIGHT OF SUPPORTIVE EVIDENCE a lot rodent + sponsored ~none independent human
Schematic: the case is tall where it's animal or sponsored, near-empty where it's independent and human.

And note what's being measured: short cognitive-test scores over a few weeks are a surrogate, not durable real-world cognition or any protection against decline. Even if the small effects were real, they'd be a long way from "sharper brain."

The other catch: you're barely getting magnesium

Even setting cognition aside, threonate is an inefficient way to take magnesium. The molecule is mostly the threonate carrier — elemental magnesium is only about 8% of its weight. A typical ~2,000 mg dose therefore delivers roughly 144 mg of actual magnesium, well under a third of the ~400 mg daily target — at several times the price of plain forms. You're paying a premium for the patented carrier and the brain story, not for magnesium.

ELEMENTAL MG PER TYPICAL DOSE (mg) 0 100 ~200 glycinate ~144 threonate (2,000 mg)
Threonate carries a smaller magnesium payload per dose and costs more — a worse repletion tool.

To be fair: this isn't dangerous. Magnesium is well tolerated and threonate won't hurt you. It's a value and honesty problem, not a safety one — a cheap commodity (magnesium) repackaged as a premium, patented brain product on thin, conflicted evidence.

By phase

Unlike most cards, phase doesn't change the answer here. Threonate's pitch is cognition, which is phase-invariant — so the verdict is "skip" throughout. If a magnesium goal does come up (sleep, repletion), that's the glycinate card's job, in any phase.

Phase 1 · Aggressive cut (Zepbound)

Skip. A cut doesn't create a cognition need that threonate is shown to meet. If anything, this is the phase to be ruthless about cost-per-benefit on the stack — and threonate is the kind of low-evidence, high-price line item to cut, not add. Want magnesium for sleep on a cut? Glycinate.

Phase 2 · Building muscle

Skip. Magnesium matters for muscle and nerve function, but you get that from food and, if needed, glycinate — far cheaper and with more elemental magnesium. Threonate adds no muscle-building edge.

Phase 3 · Maintenance

Skip. Maintenance is when you keep only what's earned its place. Threonate's cognition claim hasn't earned it in healthy adults, so there's nothing to maintain. Repletion, if you want it, stays with glycinate.

For you

You already think in funding filters — this card is mostly a clean worked example of one. Your current stack runs the well-evidenced levers; threonate would be the opposite of that, a story-driven add with conflicted backing.

My call — skip; if you want magnesium, use glycinate

Don't add magnesium L-threonate. The cognition pitch (more so given Vyvanse already handles your focus need) rests on rodent work and small, maker-linked human trials — exactly the evidence you'd discount in anyone else's stack. And as a magnesium source it's the worst-value form: low elemental yield, high price.

If you ever want magnesium — for sleep, or to nudge that flagged HOMA-IR 3.65 insulin-resistance picture where the glycinate card weighs the (modest) glucose evidence — route it to magnesium glycinate: cheaper, more elemental magnesium per dose, gentler on the gut. Treat threonate as a textbook case of the funding filter you already use: a real commodity sold as a premium brain story on thin, conflicted proof.

Profile used: 41, 5′9″, Phase 1 cut on Zepbound, on Vyvanse / Lisinopril / tretinoin, sophisticated about evidence and cost-conscious about the stack, flagged HOMA-IR 3.65. Tell me if any of that moves and I'll update.

Recall check

  1. What single question turns "there are published trials" into "but discount them" here?
  2. Which species is the famous "elevates brain magnesium & improves memory" result in, and who commercialized it?
  3. About how much elemental magnesium does a ~2,000 mg threonate dose deliver, and why is that a problem?
  4. If you wanted magnesium for sleep or repletion, which form should you use instead, and why?

Explain it back

In one sentence, tell a friend why "brain magnesium" L-threonate is a worse buy than plain magnesium glycinate — covering both the evidence and the elemental-magnesium math.

Check it yourself

This verdict is just the toolkit applied. Re-derive it with the evidence ladder, publication bias & funding, surrogate vs hard outcomes, and is it strong enough to act? — then compare against the sensible form on the magnesium glycinate card, and see how it fits the protocol and the operating filters.

Sources · Slutsky, Liu et al., "Enhancement of learning and memory by elevating brain magnesium," Neuron 2010 (rodent) · Magtein-based formula in healthy Chinese adults, RCT (Nutrients 2022) — note funding/affiliations · Magtein cognition & sleep RCT (2026) — small, short, industry-linked. Evidence summaries, not medical advice — confirm anything that touches your medication with your prescriber.

Learn · Shawon Chowdhury · an evidence verdict, kept rough on purpose · not medical advice