Ashwagandha Complete Research Review
Affiliate disclosure: The Sourcing Standards section contains an affiliate link to a product I use myself. If you purchase through it, NeuroEdge Formula earns a commission at no extra cost to you. This guide also documents a liver-safety signal that argues against ashwagandha for some readers entirely — that section costs me money and it stays in regardless.
⚕️ Educational Information, Not Medical Advice
This guide is educational and is not medical advice. I’m a cognitive enhancement researcher, not a medical doctor. Cases of clinically apparent liver injury have been reported with ashwagandha — anyone with existing liver disease should avoid it, and everyone should stop immediately and seek medical advice if jaundice, dark urine or persistent itching develop. Ashwagandha can also raise thyroid hormone levels, is not recommended in pregnancy or breastfeeding, and may interact with thyroid medication, sedatives and immunosuppressants. Consult a qualified healthcare provider before starting, particularly if you take medication or have a pre-existing condition.
Ashwagandha: A Stress Adaptogen, Not a Nootropic
By Peter Benson, Cognitive Enhancement Researcher | 18+ Years Independent Research · Last reviewed & citations re-verified: July 2026
Ashwagandha gets marketed constantly as a cognitive enhancer, and that framing oversells what the evidence shows. Read the human trials closely and a clearer picture emerges: this is, first and foremost, a stress and cortisol-regulating adaptogen. The cognitive benefits some studies report — improved memory, faster processing — show up specifically in stressed populations, alongside reduced cortisol and better sleep, which strongly suggests the cognitive gain is downstream of lower stress rather than a direct nootropic effect in its own right.
That distinction matters for how you use it. Over 18+ years of testing compounds on myself, ashwagandha has earned a place in my rotation — but as a stress-management tool I reach for during genuinely demanding periods, not as something I expect to sharpen my thinking directly. This guide covers the mechanism, what the best human trials do and don’t support, a liver-safety signal that most coverage omits entirely, and how to source and dose it properly. For where it sits among the broader options, start with our Nootropics & Supplements guide.
Ashwagandha: Dosage & Timing
Doses used in the trials — a starting reference, not a prescription. Read the cautions before the FAQ first, especially the liver signal and thyroid interaction.
| Studied dose | 600mg/day of a standardised root extract — the landmark stress trial used 300mg twice daily. Cognitive and sleep trials have used 300mg once daily. |
| Form to look for | Root-only extract, standardised to ~5% withanolides by HPLC, “root extract” stated explicitly on the label — not whole-plant or leaf-inclusive. |
| When to take it | With food. Split doses (morning and evening) or a single daily dose both have trial support. Many take the larger share in the evening for its calming effect. |
| Time to effect | Not acute. Judge it at 8–12 weeks — the strongest trial results appeared at 60 days. Sleep often shifts first, within 1–3 weeks. |
| How to start | Begin at the low end and hold. Note the liver watch-window is roughly the first 12 weeks — see the cautions block below. |
What ashwagandha actually does
Ashwagandha (Withania somnifera) is classified as an adaptogen — a compound proposed to help the body regulate its stress response rather than acting on one specific pathway. Its active constituents, withanolides, appear to modulate activity along the hypothalamic-pituitary-adrenal axis, the system governing cortisol release under stress, and may also interact with GABAergic signalling, which would help explain the reported calming effects.
This mechanism predicts exactly the pattern seen in the human trials: measurable changes in stress hormones and subjective stress scores, with cognitive changes appearing as a secondary effect in the subset of trials that also measured cognition. It does not predict an acute, direct effect on attention or memory the way a stimulant or a cholinergic compound would — which is why ashwagandha behaves so differently from Bacopa or Lion’s Mane in practice.
The stress and cortisol evidence
The single most-cited human trial is a double-blind, randomised, placebo-controlled study of a high-concentration full-spectrum root extract, which enrolled 64 adults with a history of chronic stress and gave them either 300mg twice daily or placebo for 60 days. Serum cortisol fell by 27.9% in the treatment group against 7.9% on placebo, a difference reaching strong statistical significance, alongside substantial reductions across the General Health Questionnaire, the Depression Anxiety Stress Scale and the Perceived Stress Scale (Chandrasekhar et al., 2012).
Three limits are worth stating plainly, because this one trial carries an enormous amount of commercial weight. Sixty-four participants is small — and three were non-compliant, leaving 61 in the efficacy analysis. It was a single-centre study in Indian adults aged 18–54, so generalisation isn’t automatic. And 60 days tells you about short-term effects, not about a year of continuous use or what happens when you stop. A subsequent systematic review and meta-analysis (12 trials, 1,002 participants) pooled the ashwagandha literature more broadly and found statistically significant reductions in both stress and anxiety versus placebo — more reassuring than resting the whole case on a single study. Its authors were careful, though, to grade the overall certainty of that evidence as low, with substantial inconsistency between trials (Akhgarjand et al., 2022). So the direction of effect is well supported; the precise size of it isn’t yet nailed down.
It’s also worth being precise about extracts. Other trials using different proprietary preparations — different withanolide concentrations, different extraction methods — report broadly similar directional effects, but the specific figures above belong to one extract at one dose. Switching products doesn’t guarantee you replicate them. And there’s one honest wrinkle worth flagging: the biochemical signal — cortisol coming down — is the most consistently replicated finding across trials, whereas the felt side is a little softer. The NIH’s complementary-health centre, reviewing the same field, judges ashwagandha’s effect on anxiety specifically to still be unclear, even while accepting it may help stress and sleep (NCCIH). Both can be true at once — a fairly reliable cortisol effect, and genuine uncertainty about exactly how much better people feel.
The cognitive evidence — and its limits
A separate randomised, double-blind trial using a different (sustained-release) extract gave stressed but cognitively healthy adults (130 randomised, 125 completing) 300mg daily for 90 days, measuring cognition directly with the CANTAB battery rather than a self-report questionnaire. The treatment group showed significantly better recall memory and fewer errors on pattern-recall tasks than placebo — alongside, and this is the important part, a lower Perceived Stress Scale score, reduced cortisol, and better sleep quality measured at the same visit (Gopukumar et al., 2021).
That clustering — memory improving alongside stress, cortisol and sleep all improving together — is exactly the pattern you’d expect if the memory benefit is a downstream consequence of reduced stress rather than a direct cognitive mechanism. This doesn’t make the finding less real or less useful. It does mean you should expect ashwagandha to help your cognition specifically to the extent that stress is currently what’s holding your cognition back — not as a general-purpose memory enhancer for someone who isn’t particularly stressed to begin with. Consistent with that, the NIH’s complementary-health centre still rates the evidence for ashwagandha and cognitive function as insufficient (NCCIH). If you want compounds with direct cognitive evidence, our beginners guide ranks them.
The liver signal most coverage omits
I take ashwagandha and I’ve written about it favourably, which is exactly why this section needs to be here rather than buried in a disclaimer. Clinical trials of ashwagandha have not shown liver enzyme elevations. But the NIH’s LiverTox database records that cases of clinically apparent liver injury have been reported, typically appearing two to twelve weeks after starting, presenting with jaundice and itching, and grades ashwagandha a likely cause of such injury.
Proportion matters here in both directions. In the best-documented case series, the injury pattern was cholestatic or mixed with prolonged jaundice, but no patient developed liver failure and liver tests normalised over one to five months — most reported cases resolve fully once the supplement is stopped, and they are rare relative to how many millions of people take ashwagandha uneventfully. That said, LiverTox notes rarer severe outcomes, including fatal cases and instances requiring emergency liver transplantation, and advises avoiding it in anyone with significant pre-existing liver disease, where the margin for a bad outcome is thinner (NIH LiverTox).
Practically this means three things. If you have liver disease, don’t take this compound. If you’re healthy and taking it, know that the window of concern is roughly the first three months, and that jaundice, dark urine or persistent unexplained itching are stop-immediately-and-see-a-doctor signals rather than wait-and-see ones. And it’s a further argument for root-only extracts: the human safety data is almost entirely on root preparations, so leaf-inclusive or whole-plant products are the less-characterised, higher-uncertainty option. Our nootropic safety guide covers this in the wider context of supplement risk.
Root vs whole-plant extracts — why sourcing matters
Not all ashwagandha extracts come from the same part of the plant, and this isn’t a minor detail. The National Center for Complementary and Integrative Health notes that ashwagandha products vary — they typically contain root, leaf, or combined root-and-leaf extracts — and that clinical trials have used a range of preparations. That variation matters, because the human safety and efficacy evidence is concentrated overwhelmingly on root extracts: the landmark stress trial used a root-only extract, as do the standardised extracts behind most of the positive cognitive and sleep studies. Leaf and whole-plant material is far less well characterised in people, which makes a stated root extract the conservative, better-evidenced choice rather than a marketing nicety.
Practically, check the label for “root extract” specifically rather than assuming any product labelled ashwagandha is equivalent. The extract used in the landmark stress trial was explicitly root-only, standardised to at least 5% withanolides by HPLC, and produced without alcohol or synthetic solvents.
Evidence hierarchy: ashwagandha by outcome
Named Protocol
The NeuroEdge Stress-Resilience Protocol
Built around the trial evidence rather than a generic ashwagandha dose.
Core dose
300mg root extract standardised to 5% withanolides, twice daily — 600mg total. This matches the dose used in the strongest stress and cortisol trial.
Timeline
Minimum 8 weeks before judging it. The strongest trial results appeared at 60 days. Track subjective stress and sleep, not just how sharp you feel.
Watch window
Reported liver injury cases cluster between weeks 2 and 12. Jaundice, dark urine or persistent itching in that window means stop and seek medical advice.
Cycling
Trials used continuous dosing, but situational use around known stressful periods is a defensible approach — and limits cumulative exposure.
🧮 Worked example — an 8-week test that can actually detect what ashwagandha does
If the cognitive benefit is downstream of stress, then testing cognition alone will miss it — you have to watch the whole chain. Here’s how to run a self-test that matches the mechanism:
Track three things, not one. Before you start, log a weekly baseline for each of: sleep quality (a simple 1–10 each morning is fine), subjective stress (same scale, each evening), and a cognitive measure you can repeat (a timed recall or attention task). The point is to be able to see whether cognition moves with stress and sleep or independently of them.
Run 600mg/day for a full eight weeks. Judging at two or three weeks is the most common way people wrongly conclude it “doesn’t work” — the landmark trial’s strongest effects came at 60 days. Expect sleep to shift first, often within the first two to three weeks, before anything else moves.
Read the pattern, not the single number. The tell-tale result isn’t “my memory score went up.” It’s “my memory score went up in the same weeks my sleep and stress improved.” If cognition tracks the stress and sleep lines, you’ve found a stress-mediated benefit — which is real, and is exactly what to expect. If your stress and sleep didn’t move at all but you still feel sharper, be sceptical: with an unblinded n=1, that’s the profile of expectation rather than effect.
Decide on that basis. If stress wasn’t limiting you to begin with — good sleep, low baseline stress — and nothing moved, that’s not a failed protocol. It’s the honest answer that this particular compound wasn’t matched to your particular bottleneck.
Peter’s Testing Notes
First-person, n=1 — reported honestly
I initially tested ashwagandha expecting a nootropic-style effect and was, frankly, unimpressed — I wasn’t tracking stress or sleep closely at the time, only Creyos scores, and those barely moved over the first month. It wasn’t until I started logging subjective stress and sleep alongside the cognitive testing that the actual pattern emerged: my Creyos scores improved modestly, but only in weeks where sleep quality and self-rated stress had also improved. That’s when I recognised it wasn’t behaving like Bacopa or Lion’s Mane at all — it was behaving like a stress intervention with a cognitive side effect.
I now use a standardised root extract at 300mg twice daily, reserved for periods I know will be genuinely demanding — a heavy travel month, a major deadline — rather than continuously year-round. Across roughly six years of intermittent use the pattern has held: better sleep within two to three weeks, a subjective sense of quieter mental noise by week four to six, and only then, if at all, a modest improvement on cognitive testing. The one time I switched to a cheaper non-standardised extract to save money, I didn’t get the same effect in a comparable timeframe. I can’t rule out other factors in that month, but it was enough to convince me standardisation isn’t purely marketing.
One change I’ve made since reviewing the LiverTox data: I no longer treat this as a compound to run indefinitely without thought. Intermittent use was originally a habit born of cost and cycling instinct. I now think it’s also the more sensible approach given a hepatotoxicity signal that, while rare, is real enough for the NIH to have graded it. That’s a genuine revision to my own practice prompted by writing this article, and I’d rather say so than quietly update the protocol.
Sourcing standards
Three checks: root-only sourcing stated explicitly, a declared withanolide percentage (5% is the research-grade benchmark), and published third-party testing. Given both the efficacy and the safety arguments for root extracts specifically, this is a compound where reading the label properly genuinely matters.
Ashwagandha KSM-66 — Nootropics Depot
Root-only, withanolide content stated on the label, with published third-party testing — the three things that matter for both efficacy and the liver-safety argument above.
Deliberately no magnesium pairing recommendation here. Magnesium is a sensible stress-and-sleep compound in its own right — covered in its own guide — but the ashwagandha-plus-magnesium combination has never been tested as a combination, and recommending untested pairings is precisely what our stacking guide argues against.
Using it for what it’s actually for
Illustrative examples, not real individuals or testimonials. The scenarios below are composite situations built to show common usage patterns — they do not describe specific people and contain no invented outcomes presented as results. Individual responses vary; nothing here is a typical-results claim.
When stress is the actual bottleneck
Someone under sustained work pressure is the profile the trials best match. The realistic thing to expect first isn’t feeling smarter — it’s the background hum of stress easing and sleep steadying, with any focus improvement following as a consequence of not running depleted all day. This is ashwagandha used for what the evidence actually supports.
When it’s mistaken for a memory pill
Buying it on memory claims and expecting a study-drug effect is the classic mismatch. Where memory does seem to improve for a stressed person, the more likely explanation is that lower baseline anxiety reduces the forgetting that comes from being distracted and overloaded — not a direct memory mechanism. Expectations set here decide whether someone judges it fairly.
When a health condition means checking first
Because ashwagandha can raise thyroid hormone, a thyroid condition is a genuine reason to ask a doctor before starting rather than assuming it’s fine — and to consider rechecking thyroid levels a few months in. That’s the proportionate middle ground between avoiding every supplement out of fear and ignoring a real interaction.
When the extract quietly changes
Reordering without checking and ending up with a cheaper whole-plant product is an easy way to lose the effect you had — and the safety profile you wanted. “Ashwagandha” on a label doesn’t guarantee a consistent intervention. Reading for “root extract” and a stated withanolide percentage each time is the habit that prevents it.
Cautions & Interactions
Liver — the signal to take seriously. This is the least-appreciated risk and the most important. Clinical trials showed no enzyme elevations, but documented cases of clinically apparent liver injury have occurred 2–12 weeks after starting (jaundice, dark urine, itching). Anyone with existing liver disease should avoid ashwagandha entirely. If those symptoms appear, stop and seek medical advice promptly — this is a stop-now signal, not a wait-and-see one.
Thyroid. Ashwagandha can raise thyroid hormone levels. If you have a thyroid condition or take thyroid medication, this warrants medical supervision and periodic monitoring — not a reason to avoid it automatically, but a genuine interaction.
Autoimmune conditions and immunosuppressants. Because it may stimulate immune activity, caution applies with autoimmune disease or immunosuppressant medication.
Sedatives, blood sugar, pregnancy and surgery. It may add to the effect of sedatives, and may interact with diabetes and blood-pressure medication. It is not recommended in pregnancy or breastfeeding, and NCCIH advises against it before surgery. Because it may raise testosterone, those with hormone-sensitive prostate cancer should also avoid it.
This is not a complete list of interactions or precautions. For most healthy adults at standard doses of a root extract ashwagandha is well tolerated — but this is one supplement where a conversation with a pharmacist or doctor first is genuinely worthwhile rather than boilerplate, since interactions are precisely their expertise.
Key takeaways
| → | This is a stress and cortisol adaptogen first — cortisol fell 27.9% versus 7.9% on placebo in the landmark trial (Chandrasekhar et al., 2012). |
| → | Cognitive benefits appear downstream of stress reduction — memory improvements cluster with lower cortisol and better sleep, not as an isolated effect. |
| → | Rare but documented liver injury has been reported, typically 2–12 weeks after starting. Avoid entirely with liver disease; stop at jaundice or persistent itching (NIH LiverTox). |
| → | The human evidence is almost entirely on root extracts — check the label for “root extract” rather than assuming any ashwagandha product is equivalent. |
| → | Give it a minimum of eight weeks — the strongest trial results appeared at 60 days, and judging sooner risks a false negative. |
Frequently asked questions
Is ashwagandha actually a nootropic?
Not in the direct sense most people mean. The strongest and most replicated evidence is for stress and cortisol reduction — the landmark trial found serum cortisol fell 27.9% versus 7.9% on placebo over 60 days. Cognitive benefits have been demonstrated in at least one well-designed trial using the CANTAB battery, but they appeared alongside reduced stress, lower cortisol and improved sleep measured at the same visit, which suggests the memory effect is a downstream consequence rather than an independent cognitive mechanism. Expect it to help your thinking to the extent stress is what’s limiting it.
Can ashwagandha damage your liver?
Rarely, but it has been documented and it’s worth knowing about. Clinical trials have not shown liver enzyme elevations, but the NIH’s LiverTox database records cases of clinically apparent liver injury typically appearing two to twelve weeks after starting, presenting with jaundice and itching, and grades ashwagandha a likely cause. Most cases resolve fully after stopping, though LiverTox notes rarer severe outcomes including fatal cases and emergency liver transplantation — so people with pre-existing liver disease should avoid it entirely. If you develop jaundice, dark urine or persistent itching, stop and seek medical advice promptly.
What’s the difference between ashwagandha extracts?
The most important distinction is root-only versus whole-plant or leaf-inclusive. Most clinical safety and efficacy evidence comes from root extracts specifically, and leaf and whole-plant material is far less well characterised in humans. The second distinction is standardisation: research-grade extracts state withanolide content, typically at least 5% measured by HPLC. A product that names neither the plant part nor the withanolide percentage isn’t the intervention the trials tested.
How long before ashwagandha starts working?
Give it a minimum of eight weeks. The landmark stress trial measured its results at 60 days, and other positive trials ran 8 to 12 weeks. Judging it inside the first month risks a false negative on a compound whose own supporting evidence didn’t show its strongest effects until two months in. In my own use the sequence has been consistent: sleep improves first, within two to three weeks; a subjective sense of reduced mental noise follows around weeks four to six; measurable cognitive change, if it comes at all, comes last.
Who should not take ashwagandha?
Anyone with existing liver disease should avoid it, given the documented hepatotoxicity signal. It’s not recommended during pregnancy or breastfeeding. It can raise thyroid hormone levels, so it warrants medical supervision alongside a thyroid condition or thyroid medication. Because it may influence immune activity, caution applies with autoimmune conditions or immunosuppressant medication; it may have additive effects with sedatives; and NCCIH advises against it before surgery and for people with hormone-sensitive prostate cancer. For most healthy adults at standard doses of a root extract it’s well tolerated — but this is one supplement where a conversation with a pharmacist or doctor first is genuinely worthwhile rather than boilerplate.
Get “7 Days to a Sharper Brain” — Free
Peter Benson’s personal daily protocol, rebuilt from 18 years of testing
Seven evidence-based interventions, in the exact order that makes each one more effective — from sleep foundation to your complete assembled daily stack.
Join 2,000+ readers optimising their cognitive performance. Unsubscribe anytime.
Scientific references
1. Chandrasekhar, K., Kapoor, J., & Anishetty, S. (2012). A prospective, randomized double-blind, placebo-controlled study of safety and efficacy of a high-concentration full-spectrum extract of ashwagandha root in reducing stress and anxiety in adults. Indian Journal of Psychological Medicine, 34(3), 255–262. PMID: 23439798 · PMC3573577
2. Gopukumar, K., Thanawala, S., Somepalli, V., Rao, T. S. S., Thamatam, V. B., & Chauhan, S. (2021). Efficacy and safety of ashwagandha root extract on cognitive functions in healthy, stressed adults: A randomized, double-blind, placebo-controlled study. Evidence-Based Complementary and Alternative Medicine, 2021, 8254344. PMID: 34858513 · PMC8632422
3. National Institutes of Health. Ashwagandha. LiverTox: Clinical and Research Information on Drug-Induced Liver Injury. NCBI Bookshelf. NBK548536
4. Akhgarjand, C., Asoudeh, F., Bagheri, A., et al. (2022). Does ashwagandha supplementation have a beneficial effect on the management of anxiety and stress? A systematic review and meta-analysis of randomized controlled trials. Phytotherapy Research, 36(11), 4115–4124. PMID: 36017529 · DOI: 10.1002/ptr.7598
5. Björnsson, H. K., Björnsson, E. S., Avula, B., et al. (2020). Ashwagandha-induced liver injury: A case series from Iceland and the US Drug-Induced Liver Injury Network. Liver International, 40(4), 825–829. DOI: 10.1111/liv.14393
6. National Center for Complementary and Integrative Health. Ashwagandha: Usefulness and Safety. NCCIH
Peter Benson
Cognitive Enhancement Researcher | 18+ Years Independent Research
Peter has tested ashwagandha intermittently over six years, tracking stress, sleep and cognitive performance together rather than in isolation. This guide was re-verified against source and revised to keep its liver-safety section prominent rather than buried. He is not a clinician — this guide is educational, not medical advice.
Last reviewed: July 2026
