Best Supplements for Cognitive Function (2026): Our Pick, and the Tests Behind the Research

Last updated: November 2026 · By Tyler, cofounder of Avantera Health · Medically reviewed by Jill Ingenito, DO, FACOG

The short answer: Our pick for cognitive function is Avantera Elevate, chosen on one criterion you can check yourself: each of its six focus and memory ingredients (caffeine from green tea leaf, L-theanine, bacopa, CDP-choline, rhodiola and lion's mane) has been tested in healthy adults on a named thinking test, and this page tells you which test, what it measured and what was found, including the results that came up empty.* We make Elevate. The decoder table below works for any brand: when a label or an ad says an ingredient was "clinically studied," it tells you what the people in that study were actually asked to do.

"Cognitive function" isn't one thing in research. Studies split it into parts, such as attention, processing speed, working memory, and learning and recall, and they measure each part with a specific test. A supplement can help on one test and do nothing on the next, so the test is the detail that matters.

Key takeaways

• Supplement studies measure cognitive function with named tests, such as the Stroop test for attention or a word-list test for learning and recall. A result on one test doesn't carry over to the others.
• The same ingredient can pass one test and fail another. Bacopa improved word-list learning after 12 weeks in one trial and showed no difference on any test in another 12-week trial.
• Some tests are measured an hour after one dose; others only after weeks of daily use. The table below says which.
• Caffeine with L-theanine has a meta-analysis behind its same-day results, which found moderate gains in attention switching and self-rated alertness in the first two hours after a dose (Camfield 2014).
• Our pick for cognitive function you can check test by test is Avantera Elevate: six focus and memory ingredients, each with a named test on record, and all nine doses on the label.* We make it.

What does "cognitive function" mean in a supplement study?

In a study, cognitive function means the scores people get on a battery of short tasks, usually on a computer, taken before and after the supplement or a placebo. Each task targets one part of thinking:

• Attention: staying on a task, ignoring distraction and catching what you're watching for.
• Processing speed: how fast you see something and respond.
• Working memory: holding a few things in mind while you use them, like a phone number while you dial.
• Learning and recall: taking in new information and getting it back later.
• Alertness: how awake and clear-headed people say they feel, rated on a scale.

That's why two claims about the same ingredient that sound opposite can both be true: "improved attention" might come from one test an hour after a dose, and "no effect on memory" from a different test after a month. The decoder below lists the tests behind the research on six common focus and memory ingredients.

What do the tests in cognitive function studies actually measure? The decoder

Ten tests that show up in the research on caffeine, L-theanine, bacopa, CDP-choline, rhodiola and lion's mane. For each one: what the person does, the part of cognitive function it measures, and what the placebo-controlled trials in healthy adults found, including the trials that found nothing. We checked every study against its PubMed record on November 9, 2026.

Test	What you do	What it measures	What the trials found, by ingredient
Stroop test	Name the ink color of a color word printed in a different color (the word RED printed in blue), as fast as you can	Attention and ignoring distraction, plus speed	L-theanine: faster responses after one dose in adults 50 to 69, no difference from placebo after 12 weeks (Baba 2021). Bacopa: better 1 and 2 hours after one dose (Benson 2014); no difference after 12 weeks in another trial (Sathyanarayanan 2013). Rhodiola: better scores after 7 days in resistance-trained adults (Koozehchian 2025); no change after one dose in a cycling trial (Noreen 2013). Lion's mane: faster an hour after one serving of a lion's mane blend (Docherty 2023); no change after 4 weeks in another trial (Grozier 2022)
Rapid visual information processing (RVIP)	Watch a fast stream of single digits and press a key whenever three odd or three even digits appear in a row	Sustained attention (vigilance) and speed	Caffeine: every dose tested changed performance (Smit and Rogers 2000). Caffeine with L-theanine: better accuracy (Haskell 2008). L-theanine alone: a pooled analysis of 5 trials found a benefit that grew with the dose (Mátyus 2025), while a dose-by-dose trial found no gain after accounting for placebo (Dassanayake 2023). Bacopa: fewer false alarms after 90 days (Stough 2008); no difference after 12 weeks in another trial (Sathyanarayanan 2013)
Attention switching	Switch back and forth between two simple sorting rules as a cue on screen changes	Shifting attention between tasks, speed and accuracy	Caffeine with L-theanine: faster and more accurate switching 60 minutes after one dose (Owen 2008); better switching accuracy in two more trials (Giesbrecht 2010; Einöther 2010). A meta-analysis found moderate effects on switching accuracy in the first 2 hours (Camfield 2014)
Continuous performance test (CPT-II)	Press a key for every letter that flashes on screen, except one target letter you have to hold back on	Sustained attention and holding back a response: misses and false presses	CDP-choline: healthy women aged 40 to 60 who took citicoline daily for 28 days made fewer errors than women on placebo (McGlade 2012)
Simple and choice reaction time	Press a key the moment a signal appears, or press the key that matches which of several signals appears	Processing speed	Caffeine: faster simple reaction time 30 minutes after a dose, in regular coffee drinkers and non-drinkers alike (Haskell 2005). L-theanine alone: a meta-analysis of 31 trials found faster choice reaction time after a single dose (Gerolymos 2026). Rhodiola: faster simple and choice reaction time after 4 weeks in healthy young men (Jówko 2018). Bacopa: shorter choice reaction time across trials of 12 weeks or longer (Kongkeaw 2014)
Word-list learning (Auditory Verbal Learning Test)	Hear a list of words, say back as many as you can over several tries, then recall them again after a delay	Learning and recall of new information	Bacopa: faster learning and less forgetting after 12 weeks, with no difference at 5 weeks (Stough 2001); better delayed recall after 12 weeks in adults over 55 (Morgan 2010); no difference after 12 weeks in adults 35 to 60 (Sathyanarayanan 2013). Lion's mane: word recall was lower an hour after one serving of a lion's mane blend, and the placebo group improved more on delayed recall over 28 days (Docherty 2023)
Paired associates	Learn which items go together, then recall the pairs later	Episodic memory, the memory for specific things you learned	CDP-choline: improved more than placebo after 12 weeks in healthy adults 50 to 85 with ordinary age-related memory complaints. The trial's main measure, a spatial memory test, didn't differ between groups (Nakazaki 2021)
Digit span backward	Hear a string of digits and say them back in reverse order	Working memory	Bacopa: improved after 6 weeks in medical students (Kumar 2016); no difference after 12 weeks in another trial (Stough 2001). CDP-choline: no difference from placebo on digit span after 12 weeks (Nakazaki 2021)
Inspection time	Two lines flash for a split second; say which one was longer. The test finds the shortest flash you can still judge	Visual processing speed	Bacopa: improved after 12 weeks (Stough 2001); no difference after 12 weeks in another trial (Sathyanarayanan 2013)
Bond-Lader alertness scale	Mark how you feel on a set of lines between opposites, such as alert and drowsy	Self-rated alertness, how awake you say you feel	Caffeine with L-theanine: a meta-analysis found moderate gains in self-rated alertness in the first 2 hours after a dose (Camfield 2014); higher self-rated alertness and less tiredness in a separate trial (Giesbrecht 2010)

Bacopa appears on six of these ten tests, more than any other ingredient here, and it also has the clearest null: in one 12-week trial of 66 adults aged 35 to 60, bacopa didn't differ from placebo on any test measured (Sathyanarayanan 2013). That's the pattern across this research: an ingredient moves some tests in some trials, which is why the test name matters more than the word "clinically" on a label.

How to read a "clinically studied" claim with the decoder

When a label or an ad says an ingredient supports cognitive function, three questions get you to what the research measured:

1. Which test? A faster Stroop time is a result about attention and distraction. It isn't a result about memory.
2. After one dose, or after weeks? Caffeine and L-theanine results in the table were measured within an hour or two. Bacopa's learning and recall results came after about 12 weeks of daily use and weren't seen at 5 weeks.
3. Who was studied? Healthy young adults, adults over 55 and athletes in training are different groups, and a result in one may not carry to another.

Then look for the trials that found nothing. In the decoder, five of the six ingredients have at least one.

How did we choose our pick for cognitive function?

We chose on four criteria, each one checkable against the decoder and the label table on this page:

1. Every focus and memory ingredient has a named test on record. Each one has been tested in healthy adults, against a placebo, on at least one named thinking test listed in the decoder, with the study linked under Sources.
2. The tests cover more than one part of cognitive function. Between them, the ingredients have been tested on attention, processing speed, working memory, learning and recall, and alertness.
3. Every dose on the label. No proprietary blends, so you can see the amount of each ingredient you're taking.
4. Caffeine you can count. The amount is printed, so you can add it to your coffee and tea for the day.

Avantera Elevate meets all four, which is why it's our pick. Nine ingredients. Every dose on the label. About $1.78 a day. We make Elevate, so this is one brand's pick, and we've kept every criterion to something you can verify. If you want no caffeine, or a single ingredient, the label table below names products for each.

For the other products, we read each label on the brand's own website on November 9, 2026, and we list what the label and product page say rather than what we think of them. For the decoder, every finding comes from a published human study or review, listed under Sources, and we kept to trials in healthy adults. The studies cited here tested individual ingredients, at the amounts and in the groups described. They aren't studies of Avantera products. The full research on each ingredient, with doses and evidence grades, is in our grades for 20 nootropic ingredients.

Our pick for cognitive function you can check test by test: Avantera Elevate

Avantera Elevate is our pick for cognitive function you can check test by test.* We make Elevate. Six of its nine ingredients carry the focus and memory side, and each of the six appears in the decoder above with the test it was studied on. We read Elevate's Supplement Facts panel on avanterahealth.com on November 9, 2026.

Elevate supports acetylcholine, your brain's memory messenger, two ways: CDP-choline and bacopa.* 200 mg of CDP-choline (citicoline) is a source of choline the body uses to make acetylcholine, and 300 mg of bacopa monnieri (standardized to 50% bacosides) is an herb studied for how it supports acetylcholine activity.* Bacopa is the ingredient that appears most often in the decoder, on six of the ten tests, with its learning and recall results measured after weeks of daily use.

A cup of coffee's caffeine, with twice the L-theanine for calm, steady focus.* Each serving pairs 95 mg of caffeine (from green tea leaf) with 200 mg of L-theanine, the pair tested on attention switching, RVIP and the Bond-Lader alertness scale. The rest of the label is 300 mg of rhodiola root extract standardized to 3% salidrosides (tested on the Stroop test and reaction time), lion's mane, turmeric, ginger for stomach comfort, and black pepper extract, which helps absorption. The lion's mane line reads 50 mg: it is a 10:1 extract made from 500 mg of lion's mane fruiting body, standardized to 30% polysaccharides (details here).

The directions are two capsules within the first hour of waking, on an empty stomach or with a light meal. If you're sensitive to caffeine, the product page suggests starting with one capsule.

Avantera Elevate lists all nine ingredients and every dose on the label, with no proprietary blends, and every batch is third-party tested. Elevate costs $49.95 for a 28-day supply, about $1.78 a day, with a 30-day money-back guarantee on your first order. Elevate ships as a subscription every 4 weeks, and you can pause, skip or cancel anytime from your account. The full label is on the Avantera Elevate product page, what customers say about it is on our Avantera Elevate reviews page, and certificates of analysis are on our quality page.

*These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.

Which supplements for cognitive function did we compare? Eight labels

These are real supplements sold for brain health or thinking. We chose them to cover what Google's results for "best supplements for cognitive function" pointed to on November 9, 2026: Neuriva, which the results' related questions name; single bottles of ingredients the results' summary names (omega-3s, creatine and B vitamins); and two brain formulas sold through major retailers (Focus Factor and Natrol Cognium), plus a magnesium L-threonate sold for memory. We read each label on the brand's own website that day. Life Extension, Thorne, Nordic Naturals and Nature Made sell theirs on their own sites, and each was in stock. Focus Factor, Natrol and Neuriva show the label on their own sites and sell through other stores. For single- and two-ingredient products we list the printed amount; for formulas we list the ingredients by name. The last column says what research, if any, the product page points to, in the brand's own terms. Product names are trademarks of their owners and are used here only to identify the products.

Product	What it is	Caffeine per serving	Cognitive ingredients on the label	Blends on the label	Research the product page points to
Avantera Elevate (our pick; we make it)	Multi-ingredient formula	95 mg, from green tea leaf	Caffeine and L-theanine, bacopa standardized to 50% bacosides, CDP-choline, rhodiola and lion's mane, plus turmeric, ginger and black pepper extract	None; all nine doses listed	The decoder on this page names a thinking test for each of its six focus and memory ingredients
Focus Factor Original	Multivitamin with a brain blend	None listed (the brand says it has none)	Bacopa (as Bacopin), phosphatidylserine, DMAE, DHA and huperzine A, among others in a blend, plus a full multivitamin	One named blend, printed with one total weight	A placebo-controlled study of the formula; names a word-recall measure
Natrol Cognium	Single ingredient	None listed (the brand says stimulant free)	Silk protein hydrolysate (Cognium), 100 mg a tablet; the label says 1 tablet morning and evening	None	Published studies; names a memory recall score
Neuriva Original	Two ingredients	None listed (the page says decaffeinated)	Coffee fruit extract, 100 mg, and phosphatidylserine, 100 mg	None	Neurofactor, which the page says was tested for its effect on BDNF, a brain protein; no thinking test named
Life Extension Neuro-Mag Magnesium L-Threonate	Single ingredient	None listed	Magnesium L-threonate (as Magtein), 2,000 mg, giving 144 mg of magnesium in 3 capsules	None	Magtein; a placebo-controlled trial in 44 adults aged 50 to 70; no thinking test named
Thorne Creatine	Single ingredient	None listed	Creatine monohydrate, 5 g a scoop	None	None named on the page
Nordic Naturals Ultimate Omega	Fish oil	None listed	Omega-3s: 650 mg of EPA and 450 mg of DHA in 2 soft gels	None	None named on the page
Nature Made Super B-Complex with C	B vitamins	None listed	Eight B vitamins, plus vitamin C	None	None named on the page

Of the seven other products, two name a thinking measure on the product page (a word-recall measure for Focus Factor and a memory recall score for Natrol Cognium), two name a study or an ingredient brand without naming a thinking test (Neuro-Mag and Neuriva), and three name no research. Avantera Elevate is the only one of the eight that contains caffeine.

The decoder works on any of these. Where a page names a measure, look it up in the table above; where it names a study but no test, the study itself will list the tests.

Which part of cognitive function do you want support for?

Start with the part of thinking you care about most, then look at which ingredients were tested on it.

If you want support for	Ingredients tested on it	What to look for on the label
Alertness and reaction speed on the day	Caffeine, alone or with L-theanine	The caffeine amount printed, so you can count it with your coffee
Staying on task and switching between tasks	Caffeine with L-theanine; CDP-choline for attention errors	Both amounts of the pair printed; citicoline (CDP-choline) by name
Learning and recalling new information	Bacopa, as a daily habit over weeks; CDP-choline for episodic memory	Bacopa standardized to bacosides, with the percentage printed
Brain health over the long run	Diet, sleep and exercise first; some people use omega-3s or B vitamins	Our guide to the best brain supplements covers long-term brain health

If you already drink coffee, count your caffeine before you add a supplement that contains it. The FDA cites 400 mg of caffeine a day as an amount not generally associated with negative effects for most adults. If you're pregnant or nursing, take medication, or are sensitive to caffeine, ask your doctor before starting any supplement for cognitive function.

For memory and focus together, see the best supplements for memory and focus. For the research on memory ingredients alone, see our memory supplement guide, and for focus, the best nootropic ingredients for focus.

Do supplements for cognitive function work?

Specific ingredients have moved specific test scores in placebo-controlled trials of healthy adults, and that's the evidence to judge a supplement on. Health groups are cautious about the category: the Global Council on Brain Health, convened by AARP, reviewed brain-health supplements and said it could not endorse any ingredient, product or formulation designed for brain health, and that for most people the best way to get nutrients for brain health is a healthy diet. We agree with the facts behind that. Sleep, food, exercise and time with other people come first, and no supplement replaces them.

The trials in the decoder show the other half: caffeine with L-theanine improved attention switching within two hours (Camfield 2014), bacopa improved word-list learning after 12 weeks (Stough 2001), and citicoline cut attention errors over 28 days (McGlade 2012). The same table shows trials where nothing moved. If you try a supplement, judge it against the part of thinking it was tested on, over the time frame it was tested.

How can you check our pick yourself?

Every label fact we state about Elevate on this page is printed or published somewhere you can open.

Claim	Where it's stated	How to check
Six focus and memory ingredients, each with a named test on record: caffeine, L-theanine, bacopa, CDP-choline, rhodiola, lion's mane	The decoder on this page, and the Sources list	Open each study's PubMed link and read the tests it reports
200 mg of CDP-choline and 300 mg of bacopa (standardized to 50% bacosides), plus 95 mg of caffeine with 200 mg of L-theanine	Supplement Facts panel on the product page	Read the CDP choline, bacopa, caffeine and L-theanine lines
Nine ingredients, every dose listed, no proprietary blends	Supplement Facts panel	Count the lines and read each amount
Every batch third-party tested	Product page, and certificates of analysis on the quality page	Open a certificate and match the product
$49.95 for a 28-day supply, 30-day money-back guarantee on your first order	Product page buy box	Read the price and the guarantee line

Frequently asked questions

What is the best supplement for cognitive function?

Our pick for cognitive function you can check test by test is Avantera Elevate, which we make. Each of its six focus and memory ingredients has been tested in healthy adults on a named thinking test, and all nine doses are on the label. Other products fit other needs: Focus Factor Original is a multivitamin with a brain blend, Natrol Cognium and Life Extension Neuro-Mag are single ingredients sold for memory, and fish oil, creatine and B vitamins are single ingredients that search results for this question often name.

What tests measure cognitive function in supplement studies?

Studies use short timed tasks, each aimed at one part of thinking. The Stroop test, rapid visual information processing and the continuous performance test measure attention. Simple and choice reaction time measure processing speed. Digit span backward measures working memory. Word-list learning and paired associates measure learning and recall, and the Bond-Lader scale measures how alert people say they feel. A result on one test doesn't carry over to the others.

Do supplements for cognitive function work?

Some ingredients have moved specific test scores in placebo-controlled trials of healthy adults: caffeine with L-theanine on attention switching within two hours, and bacopa on word-list learning after 12 weeks, for example. Other trials of the same ingredients found no difference. Health groups such as the Global Council on Brain Health put diet, sleep and exercise first, and so do we.

How long do supplements for cognitive function take to work?

It depends on the test. Caffeine and L-theanine results were measured within an hour or two of a single dose. Bacopa's learning and recall results came after about 12 weeks of daily use and weren't seen at 5 weeks. Judge a supplement over the time frame its ingredients were tested on.

Is there a supplement for cognitive function without caffeine?

Yes. Of the eight products we compared, Focus Factor Original, Natrol Cognium, Neuriva Original, Life Extension Neuro-Mag, Thorne Creatine, Nordic Naturals Ultimate Omega and Nature Made Super B-Complex with C list no caffeine. Avantera Elevate contains 95 mg of caffeine, about a cup of coffee, so it isn't the one to choose if you want none.

What does "clinically studied" mean on a brain supplement label?

Usually that an ingredient was tested in a trial, not the finished product. To see what the claim means, ask three questions: which test was used, after one dose or after weeks, and who was studied. Then look for the trials of the same ingredient that found nothing.

Sources

1. Baba Y, Inagaki S, Nakagawa S, et al. Effects of l-theanine on cognitive function in middle-aged and older subjects: a randomized placebo-controlled study. Journal of Medicinal Food, 2021. PubMed
2. Benson S, Downey LA, Stough C, et al. An acute, double-blind, placebo-controlled cross-over study of 320 mg and 640 mg doses of Bacopa monnieri (CDRI 08) on multitasking stress reactivity and mood. Phytotherapy Research, 2014. PubMed
3. Sathyanarayanan V, Thomas T, Einöther SJ, et al. Brahmi for the better? New findings challenging cognition and anti-anxiety effects of Brahmi (Bacopa monniera) in healthy adults. Psychopharmacology, 2013. PubMed
4. Koozehchian MS, Newton AT, Mabrey G, et al. Dose-response effects of short-term Rhodiola rosea (Golden Root Extract) supplementation on anaerobic exercise performance and cognitive function in resistance-trained athletes. Nutrients, 2025. PubMed
5. Noreen EE, Buckley JG, Lewis SL, et al. The effects of an acute dose of Rhodiola rosea on endurance exercise performance. Journal of Strength and Conditioning Research, 2013. PubMed
6. Docherty S, Doughty FL, Smith EF. The acute and chronic effects of lion's mane mushroom supplementation on cognitive function, stress and mood in young adults: a double-blind, parallel groups, pilot study. Nutrients, 2023. PubMed
7. Grozier CD, Alves VA, Killen LG, et al. Four weeks of Hericium erinaceus supplementation does not impact markers of metabolic flexibility or cognition. International Journal of Exercise Science, 2022. PubMed
8. Smit HJ, Rogers PJ. Effects of low doses of caffeine on cognitive performance, mood and thirst in low and higher caffeine consumers. Psychopharmacology, 2000. PubMed
9. Haskell CF, Kennedy DO, Milne AL, et al. The effects of L-theanine, caffeine and their combination on cognition and mood. Biological Psychology, 2008. PubMed
10. Mátyus RO, Szikora Z, Bodó D, et al. Promising, but not completely conclusive: the effect of l-theanine on cognitive performance based on the systematic review and meta-analysis of randomized placebo-controlled clinical trials. Journal of Clinical Medicine, 2025. PubMed
11. Dassanayake TL, Wijesundara D, Kahathuduwa CN, et al. Dose-response effect of L-theanine on psychomotor speed, sustained attention, and inhibitory control: a double-blind, placebo-controlled, crossover study. Nutritional Neuroscience, 2023. PubMed
12. Stough C, Downey LA, Lloyd J, et al. Examining the nootropic effects of a special extract of Bacopa monniera on human cognitive functioning: 90 day double-blind placebo-controlled randomized trial. Phytotherapy Research, 2008. PubMed
13. Owen GN, Parnell H, De Bruin EA, Rycroft JA. The combined effects of L-theanine and caffeine on cognitive performance and mood. Nutritional Neuroscience, 2008. PubMed
14. Giesbrecht T, Rycroft JA, Rowson MJ, De Bruin EA. The combination of L-theanine and caffeine improves cognitive performance and increases subjective alertness. Nutritional Neuroscience, 2010. PubMed
15. Einöther SJ, Martens VE, Rycroft JA, et al. L-theanine and caffeine improve task switching but not intersensory attention or subjective alertness. Appetite, 2010. PubMed
16. Camfield DA, Stough C, Farrimond J, et al. Acute effects of tea constituents L-theanine, caffeine, and epigallocatechin gallate on cognitive function and mood: a systematic review and meta-analysis. Nutrition Reviews, 2014. PubMed
17. McGlade E, Locatelli A, Hardy J, et al. Improved attentional performance following citicoline administration in healthy adult women. Food and Nutrition Sciences, 2012;3(6):769-773. Publisher (open access)
18. Haskell CF, Kennedy DO, Wesnes KA, et al. Cognitive and mood improvements of caffeine in habitual consumers and habitual non-consumers of caffeine. Psychopharmacology, 2005. PubMed
19. Gerolymos C, Saddier E, Boyer L, et al. Cognitive and affective effects of L-theanine: a systematic review and meta-analysis of 31 randomized trials. Molecular Psychiatry, 2026. PubMed
20. Jówko E, Sadowski J, Długołęcka B, et al. Effects of Rhodiola rosea supplementation on mental performance, physical capacity, and oxidative stress biomarkers in healthy men. Journal of Sport and Health Science, 2018. PubMed
21. Kongkeaw C, Dilokthornsakul P, Thanarangsarit P, et al. Meta-analysis of randomized controlled trials on cognitive effects of Bacopa monnieri extract. Journal of Ethnopharmacology, 2014. PubMed
22. Stough C, Lloyd J, Clarke J, et al. The chronic effects of an extract of Bacopa monniera (Brahmi) on cognitive function in healthy human subjects. Psychopharmacology, 2001. PubMed
23. Morgan A, Stevens J. Does Bacopa monnieri improve memory performance in older persons? Results of a randomized, placebo-controlled, double-blind trial. Journal of Alternative and Complementary Medicine, 2010. PubMed
24. Nakazaki E, Mah E, Sanoshy K, et al. Citicoline and memory function in healthy older adults: a randomized, double-blind, placebo-controlled clinical trial. Journal of Nutrition, 2021. PubMed
25. Kumar N, Abichandani LG, Thawani V, et al. Efficacy of standardized extract of Bacopa monnieri (Bacognize) on cognitive functions of medical students: a six-week, randomized placebo-controlled trial. Evidence-Based Complementary and Alternative Medicine, 2016. PubMed
26. Global Council on Brain Health (AARP). Brain health supplements. AARP
27. U.S. Food and Drug Administration. Spilling the beans: how much caffeine is too much? FDA