Short answer: No magnesium form is proven to improve thinking in people. Magnesium L-threonate raised brain magnesium in rats, and one small, company-funded trial in 44 adults found better cognitive scores. Citrate and amino-acid chelates such as glycinate are absorbed better than oxide. Eat magnesium-rich foods first, and keep supplements within the 350 mg/day upper limit.
The evidence that a magnesium supplement helps the brain is preliminary: rat studies plus one small human trial. Magnesium L-threonate is the only form tested for cognition in people, and the company developing it funded and wrote up that trial.
For most people, food is the better first step. Pumpkin seeds, almonds, leafy greens, beans and whole grains can supply much of the 420 mg (men) or 320 mg (women over 30) recommended each day. If you add a supplement, the upper limit for magnesium from supplements is 350 mg a day for adults. Ask a doctor or pharmacist first, especially if you have kidney disease or take medication.
What magnesium does in the brain
Magnesium is a cofactor for more than 300 enzyme systems. In the brain, its best-known job is guarding the NMDA receptor, a gate for glutamate (the brain’s main “go” signal) that is central to learning and memory.
At rest, a magnesium ion sits inside the receptor’s channel and blocks calcium from entering. When a neuron is strongly activated, the magnesium is pushed out, calcium flows in, and the synapse is strengthened. This strengthening, called long-term potentiation, is widely considered the cellular basis of learning.
Without enough magnesium, the gate works less well. The receptors respond more to background glutamate, which leads to excess calcium, oxidative stress and neuron damage. This process has been implicated in neurodegenerative diseases, anxiety and chronic pain.
Magnesium also affects the balance between glutamate and GABA (the main calming signal). When magnesium is low, the brain shifts toward a more excitable state. That may show up as anxiety, trouble relaxing, insomnia or, in extreme cases, seizures.
Low magnesium is also associated with higher C-reactive protein and more inflammatory cytokines, both part of neuroinflammation. Mazur (2007) reviewed rat studies in which magnesium deficiency triggered an inflammatory syndrome within a few days. The authors note that magnesium’s role in human immune responses still needs study.
How common is low magnesium intake?
About 48 percent of the US population got less magnesium from food than the Estimated Average Requirement (EAR) in 2005–2006, according to a review of national survey data (Rosanoff 2012). The EAR is the level that meets the needs of half the population, so it sits below the Recommended Dietary Allowance (RDA). The RDA is 420 mg/day for men and 320 mg/day for women over 30 (400 and 310 mg/day at ages 19–30).
In Spain, the ANIBES survey of 2,009 people (Olza 2017) found that most participants reported intakes below 80 percent of national and European recommendations.
Several things push intake down or losses up:
- Processed food: Magnesium is concentrated in whole grains, nuts, seeds and leafy greens, the foods that ultra-processed products displace. Refining grain removes most of the bran and germ, where much of the magnesium in wheat is found.
- Food supply: An analysis of UK food tables (Thomas 2007) reported falling mineral content in many foods between 1940 and 2002. Whether that reflects soil, crop varieties or measurement methods is debated.
- Stress: Cortisol and adrenaline-type hormones increase magnesium loss in urine. Seelig (1994) described a feedback loop in which stress depletes magnesium and low magnesium amplifies stress reactions.
- Medications and alcohol: Long-term proton pump inhibitors (PPIs) for acid reflux are well-established causes of depletion. Diuretics, certain antibiotics and some diabetes medications can also lower magnesium. Heavy drinking increases loss in urine and reduces absorption.
A standard blood test can miss low stores. Serum magnesium reflects less than 1 percent of body magnesium, because 99 percent sits inside cells in bone, muscle and soft tissue. Red blood cell magnesium, ionized magnesium and loading tests are more sensitive but rarely ordered.
Best food sources of magnesium
| Food | Serving | Magnesium |
|---|---|---|
| Spinach, cooked | 1 cup | ~157 mg |
| Pumpkin seeds | 1 oz (28 g) | ~156 mg |
| Swiss chard, cooked | 1 cup | ~150 mg |
| Black beans, cooked | 1 cup | ~120 mg |
| Quinoa, cooked | 1 cup | ~118 mg |
| Brown rice, cooked | 1 cup | ~84 mg |
| Almonds | 1 oz (28 g) | ~80 mg |
| Cashews | 1 oz | ~74 mg |
| Lentils or chickpeas, cooked | 1 cup | ~70–80 mg |
| Dark chocolate, 70%+ cacao | 1 oz (28 g) | ~65 mg |
| Avocado | 1 medium | ~58 mg |
| Salmon, cooked | 3 oz (85 g) | ~26 mg |
Dark leafy greens are consistently the best vegetable sources, because magnesium is part of chlorophyll. Our pages on leafy greens and dark chocolate cover those foods in more detail. Milk chocolate is not comparable, and refined grains lose most of their magnesium.
Magnesium forms compared
Supplements differ in how much elemental magnesium they deliver, how well it is absorbed, and whether there is any evidence it reaches the brain.
| Feature | L-Threonate | Glycinate | Taurate | Citrate | Oxide |
|---|---|---|---|---|---|
| Elemental Mg content | ~8% | ~14% | ~9% | ~16% | ~60% |
| Absorption | Not well studied in humans | Good | Not well studied | Good | Poor (about 4%) |
| Brain penetration | Shown in rats | Not shown | Not shown | Not shown | Not shown |
| Cognitive evidence | Rats + one small human RCT | None direct | None direct | None | None |
| GI tolerance | Good | Good | Good | Fair (laxative effect) | Poor |
| Cost | High | Medium | Medium | Low | Very low |
L-threonate: About 2,000 mg of the compound gives 144 mg of elemental magnesium. It is expensive and not an efficient way to correct low total body magnesium. Its case rests on rat studies and one small trial (below). For general repletion, citrate or glycinate cost less and give more magnesium per tablet. For a closer look at the two forms, see magnesium threonate vs glycinate.
Glycinate (bisglycinate): In a 60-day trial, an amino-acid chelate of magnesium was absorbed better than magnesium oxide (Walker 2003). Glycine is a calming neurotransmitter and also helps NMDA receptors work. In a small study of sleep-restricted volunteers, 3 g of glycine before bed reduced next-day fatigue and improved a vigilance test (Bannai 2012). A typical glycinate dose contains less glycine than that, and magnesium glycinate itself has not been tested for sleep in trials we could find. There is no evidence it raises brain magnesium.
Taurate: Not well studied in humans. Taurine, the carrier, affects GABA signaling in lab studies, but we found no cognition trials. Some cardiologists favor it for heart health. No specific use is supported by trials.
Citrate: Walker (2003) ran a 60-day randomized trial in 46 healthy adults taking 300 mg of elemental magnesium a day. Citrate gave the highest blood magnesium of the forms tested (citrate, amino-acid chelate, oxide), and oxide did no better than placebo. At higher doses it acts as a laxative, which also makes it a common choice for constipation.
Oxide: It has the most magnesium by weight but is poorly absorbed. Firoz and Graber (2001) measured fractional absorption of about 4 percent in healthy volunteers, lower than magnesium chloride, lactate or aspartate. It is cheap, mostly used as a laxative or antacid, and more likely to cause loose stools. Even so, the insomnia trial below used magnesium oxide.
The threonate research: rats and one human trial
The laboratory of Guosong Liu at MIT and Tsinghua University developed magnesium L-threonate (MgT). According to the authors, earlier work suggested that raising brain magnesium could improve synaptic plasticity, but ordinary supplements raised blood magnesium without a matching rise in the brain. They tested several magnesium compounds in nerve-cell cultures and chose L-threonate, a breakdown product of vitamin C, as the carrier.
In a 2010 paper in Neuron (Slutsky 2010), rats given oral MgT showed:
- higher magnesium in the fluid around the brain and spinal cord;
- a higher density of synaptic markers in the hippocampus (the dentate gyrus and CA1), which correlated with memory improvement;
- better learning, working memory, and short- and long-term memory, with aged rats also improving on a pattern-completion task;
- stronger NMDA receptor (NR2B) signaling and long-term potentiation.
These are animal results. They show what is possible in rats, not what a supplement does in people.
The one human trial (Liu 2016) was randomized, double-blind and placebo-controlled. It enrolled 44 adults aged 50 to 70 with subjective cognitive complaints: 23 took the MgT formulation MMFS-01 (marketed as Magtein) and 21 took placebo. Over 12 weeks, the MgT group improved more than placebo on a composite score across four cognitive domains (p = 0.003), most clearly in executive function and working memory. Using a model based on normative test data, the authors estimated that “brain age” improved by about 9 years. Effects on sleep and anxiety could not be separated from strong placebo effects.
The trial was small, and Neurocentria, the company developing the compound, funded it and wrote the paper. The “brain age” figure is a modeled estimate, not a biological measurement. We found no independent replication.
Magnesium and sleep
Two small trials suggest magnesium may help sleep in older adults. Neither used glycinate or threonate, so claims that one of these forms is best for sleep go beyond the evidence.
In a randomized, double-blind trial, Abbasi (2012) gave 46 older adults with insomnia 500 mg/day of elemental magnesium (as magnesium oxide) or placebo for 8 weeks. The magnesium group improved more on insomnia severity, sleep time, sleep efficiency and time to fall asleep. They also had higher melatonin and lower cortisol. This dose is above the 350 mg/day upper limit for supplemental magnesium and was given under study supervision.
In a crossover study of 12 adults aged 60–80, Held (2002) found that magnesium (as effervescent tablets) increased slow-wave (deep) sleep and lowered night-time cortisol compared with placebo. Deep sleep is important for memory consolidation. For other ways food affects sleep, see sleep and diet.
Who may need more magnesium
- Older adults: Absorption falls and urinary loss rises with age, and older adults tend to eat less magnesium. They are also more likely to take medicines that lower magnesium or interact with supplements, so supplements are best discussed with a doctor or pharmacist.
- People under chronic stress: Stress increases magnesium loss (Seelig 1994). A systematic review of 18 studies (Boyle 2017) suggested a benefit on subjective anxiety in anxiety-prone people, but rated the evidence quality as poor.
- Athletes: Intense exercise raises magnesium needs through sweat and higher metabolic demand. A review by Nielsen and Lukaski (2006) linked exercise-induced depletion with impaired performance.
- People on PPIs or diuretics: Long-term omeprazole, esomeprazole and related drugs are one of the most common medical causes of depletion. Loop and thiazide diuretics also increase urinary loss. Discuss magnesium monitoring with your doctor.
- Heavy drinkers: Chronic heavy drinking is a well-established cause of depletion and contributes to the nerve problems seen in alcohol use disorder.
Doses used in studies, and safety
The RDA covers magnesium from food and supplements together, and most people can get much of it from food. Doses used in the studies above:
- Magnesium L-threonate: Liu (2016) used 1.5 g (for people weighing 50–70 kg) or 2 g (70–100 kg) of the compound a day, about 108–144 mg of elemental magnesium.
- Magnesium oxide: Abbasi (2012) used 500 mg of elemental magnesium a day.
- Absorption trial: Walker (2003) used 300 mg of elemental magnesium a day.
Safety:
- The Tolerable Upper Intake Level (UL) for magnesium from supplements and medicines (not food) is 350 mg/day for adults, set by the US National Academy of Medicine. Above this, diarrhea, nausea and cramping become more likely. Take doses above the UL only under medical supervision. If you combine products, add up the elemental magnesium in each.
- People with kidney disease should not take magnesium supplements without a doctor’s advice. Impaired kidneys cannot clear the excess, and magnesium can build up to dangerous levels.
- Magnesium interacts with several medicines (see the FAQ). Ask a doctor or pharmacist before starting any supplement, especially if you are pregnant, breastfeeding, taking medication or have a health condition.
- Supplements can be taken with or without food. Taking them with a meal may reduce stomach upset.
Frequently Asked Questions
Can magnesium help with anxiety?
Possibly, but the evidence is weak. Boyle (2017) reviewed 18 studies in anxiety-prone people (mild anxiety, premenstrual syndrome, postpartum, high blood pressure). About half reported a benefit, and the authors rated evidence quality as poor. Magnesium is not a substitute for anxiety treatment prescribed by a doctor. See also diet and anxiety.
How long does magnesium take to work?
The insomnia trial (Abbasi 2012) measured outcomes after 8 weeks. The threonate trial (Liu 2016) measured them at six and twelve weeks, and the composite cognitive score had improved by week six.
Can I take too much magnesium?
Yes. The first sign is usually diarrhea, nausea or cramping. Very high intakes, such as large doses of magnesium laxatives or antacids, can cause dangerous blood levels (hypermagnesemia), with low blood pressure, muscle weakness and irregular heartbeat. The risk is much higher with kidney disease. Magnesium from food is not a concern, because healthy kidneys excrete the excess.
Does magnesium interact with any medications?
Yes. Taken at the same time, it can reduce absorption of some antibiotics (tetracyclines, fluoroquinolones) and bisphosphonates (osteoporosis drugs); a pharmacist can advise on spacing. Some diuretics increase magnesium loss, while potassium-sparing diuretics reduce it. Long-term PPI use can lower magnesium.
Sources
Slutsky, I., Abumaria, N., Wu, L. J., Huang, C., Zhang, L., Li, B., … & Liu, G. (2010). Enhancement of learning and memory by elevating brain magnesium. Neuron, 65(2), 165–177. doi:10.1016/j.neuron.2009.12.026
Liu, G., Weinger, J. G., Lu, Z. L., Xue, F., & Sadeghpour, S. (2016). Efficacy and safety of MMFS-01, a synapse density enhancer, for treating cognitive impairment in older adults: a randomized, double-blind, placebo-controlled trial. Journal of Alzheimer’s Disease, 49(4), 971–990. doi:10.3233/JAD-150538
Rosanoff, A., Weaver, C. M., & Rude, R. K. (2012). Suboptimal magnesium status in the United States: are the health consequences underestimated? Nutrition Reviews, 70(3), 153–164. doi:10.1111/j.1753-4887.2011.00465.x
Abbasi, B., Kimiagar, M., Sadeghniiat, K., Shirazi, M. M., Hedayati, M., & Rashidkhani, B. (2012). The effect of magnesium supplementation on primary insomnia in elderly: a double-blind placebo-controlled clinical trial. Journal of Research in Medical Sciences, 17(12), 1161–1169. PMID:23853635
Held, K., Antonijevic, I. A., Künzel, H., Uhr, M., Wetter, T. C., Golly, I. C., Steiger, A., & Murck, H. (2002). Oral Mg(2+) supplementation reverses age-related neuroendocrine and sleep EEG changes in humans. Pharmacopsychiatry, 35(4), 135–143. doi:10.1055/s-2002-33195
Boyle, N. B., Lawton, C., & Dye, L. (2017). The effects of magnesium supplementation on subjective anxiety and stress — a systematic review. Nutrients, 9(5), 429. doi:10.3390/nu9050429
Seelig, M. S. (1994). Consequences of magnesium deficiency on the enhancement of stress reactions; preventive and therapeutic implications (a review). Journal of the American College of Nutrition, 13(5), 429–446. doi:10.1080/07315724.1994.10718432
Mazur, A., Maier, J. A., Rock, E., Gueux, E., Nowacki, W., & Rayssiguier, Y. (2007). Magnesium and the inflammatory response: potential physiopathological implications. Archives of Biochemistry and Biophysics, 458(1), 48–56. doi:10.1016/j.abb.2006.03.031
Bannai, M., Kawai, N., Ono, K., Nakahara, K., & Murakami, N. (2012). The effects of glycine on subjective daytime performance in partially sleep-restricted healthy volunteers. Frontiers in Neurology, 3, 61. doi:10.3389/fneur.2012.00061
Firoz, M., & Graber, M. (2001). Bioavailability of US commercial magnesium preparations. Magnesium Research, 14(4), 257–262. PMID:11794633
Walker, A. F., Marakis, G., Christie, S., & Byng, M. (2003). Mg citrate found more bioavailable than other Mg preparations in a randomised, double-blind study. Magnesium Research, 16(3), 183–191. PMID:14596323
Nielsen, F. H., & Lukaski, H. C. (2006). Update on the relationship between magnesium and exercise. Magnesium Research, 19(3), 180–189. PMID:17172008
Olza, J., Aranceta-Bartrina, J., González-Gross, M., Ortega, R. M., Serra-Majem, L., Varela-Moreiras, G., & Gil, Á. (2017). Reported dietary intake, disparity between the reported consumption and the level needed for adequacy and food sources of calcium, phosphorus, magnesium and vitamin D in the Spanish population: findings from the ANIBES study. Nutrients, 9(2), 168. doi:10.3390/nu9020168
Thomas, D. (2007). The mineral depletion of foods available to us as a nation (1940–2002) — a review of the 6th edition of McCance and Widdowson. Nutrition and Health, 19(1-2), 21–55. doi:10.1177/026010600701900205