Nootropics: The Four With Real RCTs, The Twenty That Don't

The nootropic aisle is where supplement marketing goes to lose its mind. Twelve-ingredient proprietary blends. Molecules from Soviet-era pharmacology with almost no modern human data. Powders sold on Reddit threads with the confidence of published trials. The category, as it exists in 2026, is mostly novelty over evidence.

The honest version of this list is short. Four molecules have serious human randomized-controlled-trial support for cognitive enhancement in healthy adults. A fifth has partial support in specific populations. Everything else — and it is a long list of everything elses — is running on tradition, animal studies, or the dopamine hit of trying something new.

This piece names the four. Then it names the twenty most-marketed nootropics that do not clear the bar, and what the evidence actually says about each. If you buy nootropics, you should be able to place any given ingredient in one of the two buckets.

What a nootropic actually is

The word was coined by Romanian chemist Corneliu Giurgea in 1972. Giurgea gave the category five criteria that would rule out most of what gets sold under the name today:

1. Enhances learning and memory.
2. Enhances resistance of learned behaviors to conditions that would disrupt them.
3. Protects the brain against physical or chemical injuries.
4. Improves cortical/subcortical control mechanisms.
5. Very low toxicity, few side effects, absence of typical stimulant pharmacology.

Modern usage has stripped this down to "anything you take to think better." That is fine for casual conversation. It is not fine when a bottle is charging you sixty dollars and citing Giurgea on the label.

The four with real RCT support

These four have multiple human trials in healthy or near-healthy adults, published in indexed journals, showing meaningful cognitive effects at defined doses.

1. Caffeine plus L-theanine

This is the boring one nobody wants to hear first. Caffeine alone improves reaction time, sustained attention, and vigilance in a large body of trials, at doses of 40 to 200 mg. The tradeoff is jitter, elevated heart rate, and disrupted sleep if used late. L-theanine, an amino acid from green tea, blunts the sympathetic side effects without blunting the cognitive effect. Multiple RCTs of the combination (Owen 2008, Giesbrecht 2010, Kelly 2008) show attention and task-switching improvements above caffeine alone.

The reproducible protocol is caffeine 100 mg with L-theanine 200 mg, morning use only, ideally after 90 minutes awake to let endogenous cortisol clear. This gets its own dedicated post in the Vyvata blog for readers who want the deeper cut.

2. Creatine monohydrate

Creatine has been the most-studied strength supplement for two decades. The cognition angle is newer and quieter, and it is real. Rae 2003 showed working memory and processing speed improvements in vegetarians (who have lower baseline creatine) at 5 g per day. Benton 2011 replicated in omnivores. McMorris 2007 showed protection against sleep-deprivation-induced cognitive decline at 20 g per day for a week — meaningful for shift workers, new parents, or anyone with disrupted sleep.

The reproducible protocol is 5 g per day, monohydrate form, consistent daily use. Effects on cognition are modest at baseline and become clearer under fatigue. Creatine also gets its own Vyvata write-up covering both the muscle and brain angles.

3. Bacopa monnieri

This is the herbal one that clears the bar. Roodenrys 2002 ran a double-blind placebo-controlled trial in adults at 300 mg per day of standardized bacopa extract (55 percent bacosides) for 12 weeks and found significant improvements in verbal learning and memory consolidation. Stough 2001 showed similar effects at similar doses. Multiple subsequent trials, including in older adults and in students, have replicated the memory-formation signal.

The critical caveat. Bacopa is slow. Meaningful effects take 8 to 12 weeks to emerge. This is not an acute focus tool. The trials that show nothing are almost always the short-duration trials. If you buy bacopa and stop at four weeks, you will conclude it does not work, and you will be wrong.

The reproducible protocol is 300 mg per day of extract standardized to 55 percent bacosides (Bacognize and CDRI-08 are the two most-studied standardized forms), taken with food to reduce GI upset, for at least 12 weeks before evaluating.

4. Modafinil

Modafinil is a prescription wakefulness-promoting agent approved for narcolepsy, shift-work sleep disorder, and obstructive sleep apnea. It is not a supplement. It is on this list because the off-label "study drug" use is common enough that leaving it out would be dishonest, and because the human RCT data on cognition in healthy adults is stronger than for any nootropic in the supplement category.

The Battleday 2015 systematic review in European Neuropsychopharmacology pooled 24 trials and found robust improvements in attention, executive function, and learning, particularly on complex tasks and under sleep restriction. Effects are largest in sleep-deprived subjects. Tolerance and dependence appear to be lower than for traditional stimulants, but the drug is not risk-free — headache, nausea, insomnia, and rare severe skin reactions are documented.

The honest note is this: modafinil requires a prescription for good reason. It is not something to source from grey markets. Any nootropic conversation that presents it as an over-the-counter option is misleading you.

Honorable mention: high-dose omega-3 in specific populations

EPA-heavy omega-3 at 1 to 2 g per day has moderate evidence for mood in depression and for cognitive function in older adults with mild cognitive impairment. Effects in healthy young adults are smaller and less consistent. This does not clear Giurgea's bar for a general-purpose nootropic, but it is worth flagging as the closest fifth entry.

The twenty (or so) that do not clear the bar

This is the longer list. Not every ingredient here is useless. Some have interesting mechanisms, some have small effect sizes in small trials, and some may earn a promotion when better data lands. But none of them belong in the same conversation as the four above.

The racetam family

• Piracetam. The original nootropic, synthesized by Giurgea. Most human trials are older than 1990, methodologically limited, and studied in cognitively impaired populations rather than healthy adults. The healthy-adult signal is thin.
• Aniracetam. Fat-soluble racetam. Human data is almost entirely in dementia or age-related cognitive decline. Healthy-adult evidence is close to nonexistent.
• Oxiracetam, pramiracetam, phenylpiracetam. Each has a small niche literature, mostly in Eastern European clinical populations. Phenylpiracetam has a mild stimulant component that shows up in athletic doping tests. Not enough healthy-adult RCT data to recommend.

The Russian pharmacology set

• Noopept. Sold as a nootropic in the US supplement grey market. Human data is almost entirely from Russian clinical trials in cognitively impaired populations. Not FDA-approved. Not well-characterized in healthy adults.
• Semax and Selank. Peptide nootropics with essentially no Western clinical trial infrastructure. Sold as intranasal sprays. Buy-at-your-own-risk category.
• Dihexa. A synthetic peptide with impressive animal data on synaptic plasticity and no meaningful human evidence. The bottles on the market are experimental research chemicals, not supplements.

The choline family

• Alpha-GPC. Some small trials in athletes on power output and growth-hormone response. Cognition data in healthy adults is thin and mixed. Not a general-purpose nootropic; possible niche in acute physical performance.
• CDP-choline (citicoline). Better characterized than alpha-GPC. Some data on attention in adolescents and on recovery from ischemic stroke. Healthy-adult cognitive enhancement data is modest.
• Uridine. Combined with fish oil and choline in some stacks based on preclinical rodent work by Wurtman. Human data is limited.

Phospholipids and their neighbors

• Phosphatidylserine. The one in this bucket closest to a modest recommendation. Some trials show small improvements in memory and processing speed in older adults. Effect sizes are small. Not in the same tier as bacopa.

The mushroom pushed hardest

• Lion's mane (Hericium erinaceus). The Mori 2009 study in mild cognitive impairment at 3 g per day showed some improvement over 16 weeks. The evidence in healthy adults is much thinner. Retail-dose mushroom-coffee servings are usually well below the trial dose. Interesting mechanism, weak retail evidence.

The old herbal names

• Ginkgo biloba. Mostly null in healthy adult cognition trials. Solomon 2002 in JAMA found no cognitive benefit at 40 mg three times daily in older adults over six weeks. Some smaller signals in circulation and tinnitus. Not a nootropic in the modern sense.
• Panax ginseng. Modest mental-fatigue signal covered in the adaptogen category. Not a memory enhancer.

The chemistry-adjacent oddballs

• Methylene blue at low doses. Interesting mitochondrial mechanism. Some rodent data on memory. Human RCT data on cognition is thin. High doses are pharmacologically active and interact with SSRIs. Not a beginner nootropic; not a recommended nootropic at any level currently.
• Nicotine (via lozenge or gum). Has real acute cognitive effects — attention, working memory. Trades against addiction potential and cardiovascular load. Excluded from most nootropic conversations for reasonable ethical reasons.

The stack ingredients

• Tyrosine. Some evidence for cognitive preservation under acute stress and sleep deprivation. Baseline healthy-adult effects are small.
• Rhodiola. Genuine mental-fatigue evidence, covered in the adaptogen category. Better classified as an adaptogen than a nootropic.
• Ashwagandha. Same note — an adaptogen with modest downstream cognitive effects via stress reduction, not a direct memory enhancer.

The proprietary blend problem

The single biggest issue with the retail nootropic category is proprietary blends. A bottle labeled "Cognitive Complex, 1,200 mg" that lists ten ingredients without individual doses is doing three dishonest things at once.

• It hides underdosing. Bacopa needs 300 mg per day of standardized extract to show effect. A ten-ingredient blend at 1,200 mg total cannot contain 300 mg of bacopa and 300 mg of anything else. Something in that blend is at a trace dose.
• It hides stimulants. Caffeine, synephrine, and other stimulants often hide inside a proprietary complex. That is why the label lists the total blend weight rather than the per-ingredient amounts.
• It prevents comparison. You cannot compare Product A's bacopa dose to Product B's if neither one lists it. Which is the point.

The rule is simple. If a nootropic product will not disclose the milligram amount of every ingredient, do not buy it. The good brands disclose. The bad ones hide behind proprietary blends and marketing language.

The novelty trap

The nootropic culture, especially on Reddit and in adjacent forums, has a chronic bias toward novelty. Whatever is newest gets the loudest voice. The four molecules at the top of this post are old news. Caffeine, creatine, bacopa, and modafinil have been on the honest list for at least fifteen years. That is not because innovation stopped. It is because most of the new molecules do not clear the bar.

The tell is that a person who chases the latest nootropic every three months rarely reports lasting benefit from any of them. Chasing novelty produces a very consistent 30-day placebo response and no long-term change. Sticking with a boring stack that has evidence behind it produces small, real, cumulative effects that add up over years.

A starter nootropic protocol

1. Foundation first. Sleep 7 to 9 hours. Adequate protein. Aerobic and resistance training. Nootropics on a broken foundation do very little.
2. Add caffeine plus L-theanine. 100 mg caffeine, 200 mg L-theanine, morning only. Run for 30 days and note the delta on focused work.
3. Add creatine. 5 g per day of monohydrate. Effects on cognition are subtle unless you are also under sleep deprivation, where they become obvious.
4. Optional: add bacopa. 300 mg per day of 55 percent bacosides extract. Commit to 12 weeks minimum before evaluating.
5. Do not stack six things at once. Add one thing at a time. Wait long enough to see whether it did anything. Drop what does not work.

The honest summary: the nootropic category is 90 percent hype and 10 percent quiet, boring, well-supported basics. Caffeine plus L-theanine, creatine, bacopa, and a prescription for modafinil in specific circumstances are the entire evidence base for cognitive enhancement in healthy adults. Everything else is a bet with worse odds. The beginner move is to master the boring four before spending a dime on anything else.