Microdosing

This report summarises what Blossom’s database shows about microdosing, the practice of taking repeated, sub-perceptual doses of a psychedelic for everyday benefits. It is worth being clear what kind of page this is. It is not a condition page and not a treatment. It is about a hugely popular practice and a striking mismatch: between how reliably people say it helps them and how unreliably it performs when properly tested. That mismatch is the whole subject.

A note before the evidence

This page is a research summary, not medical advice, and nothing here is a recommendation to take psychedelics, including at low doses. It is worth saying plainly that the evidence below is, on the whole, discouraging about microdosing’s specific claims, and that the long-term safety of the practice is genuinely unknown. Read it as an honest accounting, not an endorsement.

Why microdosing is the field’s best natural experiment

Most psychedelic research has a blinding problem: a full dose is unmistakable, so participants know what they took, and expectation contaminates everything. Microdosing is the rare exception. Because a microdose is sub-perceptual by design, people genuinely may not know whether they have taken the drug or a placebo, which makes it one of the few psychedelic interventions that can be blinded properly. That is a gift for science, and the field has used it. The result is an unusually clean body of placebo-controlled evidence, and that evidence is not kind to the practice.

The most elegant study did not even need a lab. In a self-blinding citizen-science project, 191 people randomised their own capsules at home[1]eLife (2021), self-blinding citizen science (n=191): microdose improvements were matched by placebo; anecdotal benefits explained by the placebo effect: some got the real microdose, some a placebo, and none knew which. Everyone improved on measures of wellbeing and mood, but the placebo group improved just as much, and the small differences that did appear were explained by people working out which they had taken. The authors’ conclusion was blunt: the anecdotal benefits of microdosing can be explained by the placebo effect.

The controlled trials, one by one, come up null

The formal trials tell the same story across every claim. For attention, the flagship is a phase 2A randomised controlled trial of repeated low-dose LSD for adult ADHD, in which LSD did not beat placebo, and the placebo group improved slightly more[2]JAMA Psychiatry (2025), LSD-for-ADHD phase 2A RCT (n=53): LSD not more efficacious than placebo (AISRS -7.1 vs -8.9; placebo improved slightly more). For cognition and mood, two double-blind psilocybin trials found no reliable effect once results were corrected[3]Neuropharmacology (2026), two psilocybin double-blind placebo-controlled trials: no reliable benefit over placebo; initial signals vanished after correction, and a careful mushroom study found acute effects only in people who guessed correctly that they had the active dose[4]Transl Psychiatry (2022), psilocybin-mushroom DBPC (n=34): acute effects only in participants who correctly guessed their condition; expectation underlies the benefits. For pain, an LSD trial found no analgesia across the sample[5]Br J Pain (2025), LSD 15 µg microdose pain RCT: no analgesic effect in the whole sample (only a marginal post-hoc subgroup effect after the first dose). For migraine, a psilocybin trial saw roughly equal improvement in every arm, active drug and active placebo alike[6]Headache (2025), psilocybin vs active-placebo migraine RCT (n=18): no between-group difference (p=0.102); ~50% reductions in all arms.

The one positive controlled result is worth stating precisely, because it is so narrow. A mega-analysis pooling three blinded trials found that active microdosing raised a single creativity measure, the ratio of original to total responses, with no effect on any other creative or cognitive score[7]Neuropharmacology (2025), creativity mega-analysis (N=171, 3 DBPC trials): only the originality/fluency ratio rose; no effect on other divergent or convergent thinking. That is the strongest survivor in the entire controlled literature: one sub-measure of divergent thinking. It is a real finding, and it is a long way from "microdosing makes you more creative", let alone "microdosing improves your life".

Why the anecdotes and the trials disagree

The obvious objection is that millions of people swear by microdosing, and the surveys agree with them. A naturalistic study of nearly a thousand psilocybin microdosers found small-to-medium improvements in mood and mental health[8]Scientific Reports (2022), naturalistic cohort: psilocybin microdosers (n=953) showed small-to-medium mood/mental-health gains vs controls (n=180) over ~30 days (uncontrolled). So why believe the trials over lived experience? Because the trials are built to answer exactly the question the surveys cannot: is it the drug, or the expectation? Survey and open-label studies cannot separate the two; the people in them know they are microdosing, chose to do it, and expect it to work. When that expectation is held constant by a placebo, the drug’s advantage largely vanishes.

There is a more pointed piece of evidence still. A naturalistic cognition study testing people on and off their microdosing days found no link to better or worse performance in any cognitive domain[9]J Psychol Clin Psychiatry (2023), naturalistic cognition study (n=17): microdosing days were not linked to better or worse performance on any cognitive domain, which suggests that whatever microdosing does, it acts on how people feel rather than on measurable cognition. None of this means people are lying or imagining things. The placebo effect is real and powerful, and feeling better because you expect to is still feeling better. But it is not the same as the drug doing the work, and the distinction matters enormously for anyone deciding whether to take a controlled substance repeatedly on the strength of a pharmacological claim.

The two things the null trials do not settle

Intellectual honesty cuts both ways, and there are two real caveats. The first is that absence of a strong effect is not proof of no effect. The controlled trials are still relatively few, often small, mostly in healthy people, and possibly using doses too low to do anything, a point made fairly by a rapid review that judged a firm placebo verdict premature[10]J Psychopharmacol (2024), "Is microdosing a placebo?" rapid review (19 controlled studies): a pure-placebo verdict may be premature; it is not yet possible to determine whether microdosing is a placebo. A genuine but modest effect, or one confined to a particular population, has not been ruled out, only left unsupported.

The second caveat is the one that should worry enthusiasts most, and it is about safety rather than efficacy. Microdosing is, by definition, a chronic practice, often sustained for months or years, yet almost no one has studied what that does to the body. Short-term, a systematic review found side effects to be mild and short-lived[11]Neuropharmacology (2025), microdosing side-effects systematic review (31 studies): typically dose-dependent, mild and short-lived (raised BP, anxiety, mild cognitive impairment). But these drugs stimulate the 5-HT2B receptor implicated in heart-valve disease, and modelling suggests chronic dosing carries a real if unquantified cardiac risk that no study has been designed to measure[12]J Psychopharmacol (2023), modelling: classic psychedelics bind 5-HT2B with potency equal to or above 5-HT2A; valvular heart disease is a potential risk of chronic microdosing, but no purpose-designed study exists. So the practice sits in an uncomfortable place: its headline benefits look largely like placebo, while its main long-term risk is simply unknown.

Reading this honestly

So how should you read the microdosing story? As a humbling one. It is the part of psychedelic science where the experiments are cleanest and the popular claims fare worst. When microdosing is tested the way medicines are tested, double-blind and placebo-controlled, the benefits to mood, focus, cognition and wellbeing very largely fail to separate from placebo, and the small residue tracks expectation rather than chemistry. That does not make the experience fake; the placebo effect is genuine and people really do feel better. But it does mean the specific claim, that a tiny dose of a psychedelic pharmacologically sharpens or lifts you, is not supported by the best evidence, with a single narrow exception in creative thinking. The two honest reservations, that better trials might still find a modest or niche effect, and that the long-term safety of chronic dosing is genuinely unstudied, should temper both the dismissers and the believers. The most useful thing this literature offers an honest reader is a clean demonstration of how powerful expectation is, and a reminder that "it works for me" and "the drug works" are different claims, only one of which the evidence so far supports.