Herbal medicines and nootropics
18 items where marketing usually runs ahead of the data
Experimental material. Read before using anything from here.
Nothing on this page is a medical recommendation, prescription or treatment plan. It is an organized translation of protocols circulating in research communities and, where it exists, of what published trials tested. The two are marked differently — and they are not equivalent.
Most of the compounds here have no FDA approval for human use. Several are sold labeled "research use only", which means they have not gone through purity, sterility or dosage controls for human consumption. A community-reported dose is not a validated dose: it is what someone reported having done.
Talk to a licensed health professional before considering any of these compounds. If you already use one and feel anything unexpected, seek care — do not wait for the next routine test.
Summary
Ashwagandha, rhodiola, bacopa, ginkgo, berberine, curcumin, tongkat ali and ten more. Risk generally low, evidence generally weak and marketing generally strong — the table separates the three, item by item. It gives the dose range used in studies, the strength of the evidence and the caveat for each, including those with a toxicity signal or a serious interaction.
How to read this page
Over-the-counter herbal products and nootropics occupy an awkward place: risk generally low, evidence very uneven and marketing always strong. The tables separate the three.
The RCT/meta column gives the number of randomized trials and meta-analyses indexed in PubMed, checked item by item on September 4, 2026. It is a number, not an opinion — and in some cases it contradicts the product's fame in both directions.
Adaptogens and cognition
| Feature | Usual range in studies | For what | RCT/meta | What the number says |
|---|---|---|---|---|
| Ashwagandha (KSM-66) | 300–600 mg/day of the standardized extract | Stress, cortisol, sleep, testosterone | 67 | Real base for stress and sleep. Nordic hepatotoxicity alerts: avoid in liver disease and thyroid disease |
| Rhodiola rosea | 200–600 mg/day (3% rosavins / 1% salidroside) | Mental fatigue, stress | 70 | Decent volume, small studies and extract standardization that varies widely between brands |
| Bacopa monnieri | 300 mg/day (50% bacosides), for 8–12 weeks | Memory and learning | 14 | Modest effect and only after weeks. Gastrointestinal discomfort is common |
| Panax ginseng | 200–400 mg/day of the standardized extract | Fatigue, cognition | 400 | Far more studied than assumed. The problem is heterogeneity, not scarcity. Interacts with anticoagulants and antidiabetics |
| Lion's Mane (Hericium erinaceus) | 500–3,000 mg/day | Cognition, nerve growth factor | 2 | Two. The base is almost entirely preclinical — the fame has no human backing |
| Ginkgo biloba | 120–240 mg/day (EGb 761) | Cerebral circulation, memory | 88 | Here volume does not save it: the large dementia-prevention trials were negative. Increases bleeding risk |
| Piracetam | 1.2–4.8 g/day | Classic racetam nootropic | 371 | High volume, but concentrated in dementia, stroke and cognitive decline — not in healthy people. In Brazil it is a prescription drug |
| Phosphatidylserine | 100–300 mg/day | Memory, post-exercise cortisol | 19 | The old evidence used a bovine source; current products are soy-derived, with less robust data |
Metabolic and hepatic
| Feature | Usual range in studies | For what | RCT/meta | What the number says |
|---|---|---|---|---|
| Berberine | 500 mg, 2–3×/day, before meals | Blood glucose, lipid profile | 177 | Solid base, with an effect comparable to that of drugs in some trials. Inhibits CYP3A4 and is a serious interactor — do not combine with a statin or immunosuppressant without guidance |
| Curcumin with piperine | 500–1,000 mg/day of curcuminoids | Inflammation, joint pain | 44 (osteoarthritis only) | Good base for osteoarthritis. The piperine that increases absorption also increases that of other drugs |
| Silymarin (milk thistle) | 200–400 mg/day | Liver protection | 33 | Reasonable volume, inconsistent results in hepatitis. Sold as detox, which is not a clinical outcome |
| Bergamot (Citrus bergamia) | 500–1,000 mg/day of the extract | Cholesterol | 24 | Twenty-four trials with a lipid outcome. The broad query returns 53, but part of it is essential oil in aromatherapy, which has nothing to do with cholesterol. Studies mostly Italian and small |
| Cinnamon (extract) | 1–6 g/day | Blood glucose | 23 | The problem is not a lack of trials, it is inconsistent results among them. Cassia cinnamon contains coumarin, hepatotoxic at high doses; Ceylon has far less |
| Corosolic acid (banaba) | 10–50 mg/day | Blood glucose | 8 | Eight trials, all small. It is the lowest number in this table |
Testosterone, libido and the rest
| Feature | Usual range in studies | For what | RCT/meta | What the number says |
|---|---|---|---|---|
| Tongkat Ali (Eurycoma longifolia) | 200–400 mg/day of the standardized extract | Testosterone, libido, stress | 8 | Little, but it exists. The effect on testosterone shows up mainly in those who started from a low level |
| Fadogia agrestis | not established in humans | Testosterone | 0 | Zero randomized trials and no human study. I checked the three articles PubMed tags as human: one is a rat study, one is a market survey in Ghana and the third is in vitro antimalarial screening. Signal of testicular and renal toxicity in rodents |
| Shilajit | 250–500 mg/day | Testosterone, energy, fulvic acid | 7 | Minimal base. Risk of heavy-metal contamination in unpurified product — demand a lab report |
| Yohimbine HCl | 5–10 mg before training or fasted | Fat loss, erectile dysfunction | 278 | The volume misleads: most use yohimbine as a pharmacological probe in psychiatric research, not as a fat-loss agent. Anxiety, tachycardia and blood pressure spikes. Do not use with MAOIs or with hypertension |
What changes when you look at the number
- Lion's Mane has 2 trials. It is the item with the widest gap between fame and backing on this page.
- Ginkgo has 88 and even so is not recommended to prevent dementia — because the large trials came out negative. Study volume is not synonymous with favorable evidence.
- Yohimbine and piracetam have hundreds of trials, but in contexts that are not those on the label. A high count requires checking what the studies are about.
- Fadogia agrestis has zero. It was popularized by podcast, not by evidence.
References
- Counts of randomized trials and meta-analyses obtained from PubMed on September 4, 2026, with the filter Randomized Controlled Trial[Publication Type] OR Meta-Analysis[Publication Type], item by item.
- Fadogia agrestis — the three articles indexed as human: Ogunro & Yakubu 2022 (Wistar rats), van Andel et al. 2012 (Ghanaian herbal market), Sanon et al. 2003 (in vitro antimalarial screening).