Expectancy Effects in Microdosing: Belief as Predictor

Microdosing Placebo & Expectancy Effects

81 microdosers tracked

In 81 people who microdosed for four weeks, how much benefit they expected before the first dose predicted how much their well-being improved — a result the study's own authors called suggestive of a significant placebo response.

Kaertner et al., Scientific Reports, 2021

Belief first

baseline expectancy, recorded before any dosing, predicted subsequent well-being gains

No difference

active and placebo groups improved equally in the Szigeti 2021 self-blinding trial of 191 participants

AEB

activated expectancy bias: correctly guessing your dose amplifies belief-driven benefit

Kaertner et al., Scientific Reports, 2021

View as image ↗

The most surprising finding in microdosing research isn't about psilocybin. It's about the person taking it. In a 2021 study of 81 people who microdosed psychedelics, Kaertner and colleagues tracked mental health outcomes over four weeks.^[3] Well-being improved. Anxiety fell. Depression scores declined. But when the researchers asked what predicted the size of each person's improvement, one of the strongest answers was a number participants wrote down before they took a single dose: their expectancy score, meaning how much they believed, going in, that microdosing would help them.

This is the expectancy effect in microdosing, and it sits at the center of the field's hardest methodological problem. You cannot understand the placebo debate in microdosing research without understanding why expectancy is so difficult to separate from pharmacology, why psychedelics are especially vulnerable to it, and what researchers have learned trying to design around it. This article covers the full evidence chain: from the observational studies that could not separate belief from drug effect, to the formal framework that explains exactly how belief contaminates even the most rigorous trials, to the 2024 review arguing that the field has swung too far toward a placebo verdict. For a broader overview of the field, see the complete guide to psilocybin microdosing.

Why the First Wave of Microdosing Studies Couldn't Separate Belief from Drug Effect

The early evidence for microdosing benefits came from observational research: no control conditions, no blinding, just tracking people who were already microdosing and recording what happened.

Polito and Stevenson's 2019 study is the most-cited example.^[1] Over six weeks, 98 self-selected microdosers provided daily ratings of psychological functioning. Reported functioning rose on dosing days, with limited evidence of carry-over to the days between doses. Comparing before and after the study, the researchers found reductions in reported depression, stress and distractibility — alongside an increase in reported neuroticism, a result rarely mentioned when this study is cited as evidence that microdosing works.

Polito and Stevenson were explicit about what their design could not tell them: whether psilocybin caused those changes. The participants chose to microdose. Many came from online communities where positive experiences are shared constantly. There was no comparison group, and nobody was taking a look-alike capsule.

The same paper contains a second study that is far more relevant to expectancy than the first, and far less often quoted. Polito and Stevenson surveyed 263 naïve and experienced microdosers about what they expected microdosing to do. Expectations were uniformly large and wide-ranging — considerably larger than the outcomes the 98 actual microdosers reported. More revealing still, the effects participants believed were most likely to change were unrelated to the pattern of effects that actually showed up in the data.

Expectancy Bias, Measured

Belief Ran Ahead of the Data — and Pointed the Wrong Way

Polito and Stevenson ran two studies side by side: one tracking what microdosers reported, another measuring what people expected microdosing to do. The gap between them is the clearest single demonstration of expectancy bias in the microdosing literature.

What people believed microdosing would do

n = 263Naïve and experienced microdosers rating expected effects

• Large benefits
• Wide-ranging benefits across many domains
• Change concentrated in the domains people cared most about

Expectations were uniformly high — held by naïve and experienced participants alike.

mismatch

What microdosers actually reported

n = 98Daily ratings across a six-week microdosing period

• Higher psychological functioning on dosing days only
• Little evidence of carry-over to following days
• Lower depression, stress and distractibility — but neuroticism rose

The domains people most expected to change were unrelated to the domains that actually shifted.

Why this matters: if reported benefits were simply expectations being fulfilled, the effects people expected most should be the effects they reported most. They were not. Expectancy shapes the microdosing experience, but it does not straightforwardly dictate it.

ExpectationObservation

Source: Polito & Stevenson, PLoS ONE, 2019 (Study One n = 98; Study Two n = 263)

That mismatch cuts in an interesting direction. It confirms that expectancy in this population is enormous, which is the core worry. But it also shows that reported outcomes were not simply expectations echoed back — if they were, the biggest expected effects should have been the biggest reported ones, and they were not. Expectancy shapes the microdosing experience. It does not appear to dictate it in a simple one-to-one way.

A 2020 systematic review by Ona and Bouso pulled together this first generation of evidence and found the same structural pattern across studies.^[2] Self-reported benefits were common. Placebo effects could not be ruled out in any study lacking blinded controls.

This was not a failure of Polito and Stevenson's study design. For a first-pass investigation into a barely-studied phenomenon, observational work is the appropriate starting point. The problem comes when these results are treated as evidence that psilocybin produces the benefits, rather than evidence that microdosers who expect benefits tend to report benefits. Those are meaningfully different claims, and the design of these early studies cannot tell them apart.

The Kaertner 2021 Study: Measuring Expectancy Effects in Microdosing Directly

To measure the expectancy effect directly, researchers need to do something most microdosing studies skip: quantify what participants believe before the study starts, then compare that number to their outcomes.

Kaertner and colleagues did exactly this in 2021.^[3] Using a prospective web-based design, they recruited people who were already planning a weekly microdosing regimen and surveyed them before, during and after a core four-week test period. At baseline, before any dosing, participants rated how much improvement they anticipated. Eighty-one participants completed the primary endpoint.

At four weeks, the cohort reported increased psychological well-being and emotional stability, and reductions in state anxiety and depressive symptoms, along with gains in resilience, social connectedness, agreeableness, nature relatedness and aspects of psychological flexibility. Then came the finding the paper is known for: baseline positive expectancy scores predicted subsequent improvements in well-being. The authors read this as "suggestive of a significant placebo response" and closed by cautioning against zealous inferences about microdosing's therapeutic value.

Key Study

Kaertner et al. (2021) · Scientific Reports

Population

81 adults already planning a weekly psychedelic microdosing regimen, surveyed online before, during and after a 4-week period

Intervention

Self-directed microdosing — substance, dose and schedule chosen by participants, not controlled by researchers

Comparator

None. There was no placebo arm and no non-microdosing control group.

Finding

Well-being, emotional stability, anxiety and depressive symptoms all improved at 4 weeks. Positive expectancy measured at baseline predicted the subsequent improvement in well-being.

narrow-scope A prospective uncontrolled design can show that belief predicts outcome. It cannot show what the drug did, because nothing was held constant to compare against.

Kaertner et al., Scientific Reports, 2021 (PMID 33479342)

Here is where this study is most often misreported, including in otherwise careful summaries. It is frequently said that Kaertner found expectancy predicted outcomes "better than the drug did," or better than whether participants received psilocybin or placebo. That comparison never happened. There was no placebo. There was no controlled dose. Every participant took what they chose to take, and knew it.

Claim Reach Check

What the Kaertner 2021 Design Can and Cannot Establish

Kaertner and colleagues followed 81 people through a self-directed microdosing regimen with no placebo group and no controlled dose. That design supports a claim about prediction and cannot support a claim about comparison — a distinction frequently lost when this study is summarised.

Within reach of the design

✓

Well-being, anxiety and depressive symptoms improved over four weeks

Measured directly in all 81 completers at the primary endpoint.

✓

Baseline expectancy predicted who improved most in well-being

Expectancy was recorded before the first dose, so it cannot have been caused by the outcome.

✓

The improvements are partly a placebo response

The authors describe the expectancy-outcome link as "suggestive of a significant placebo response" — an inference, not a measured contrast.

no comparator

Beyond reach of the design

×

Expectancy predicted outcomes better than psilocybin did

There was no placebo arm and no controlled dose. "Drug versus belief" was never a comparison this study could run.

×

Psilocybin contributed nothing to the improvement

With no comparator, the drug's contribution is unmeasured — not measured at zero.

Source: Kaertner et al., Scientific Reports, 2021 (n = 81, prospective, uncontrolled)

The distinction matters because the two claims license very different conclusions. "Belief predicts who improves" is well-supported and genuinely important — it means any study that tracked outcomes without measuring baseline expectancy was blind to one of the strongest predictors in its own dataset, and the vast majority of early studies did not measure expectancy at all. "Belief beats the drug" is a causal comparison that requires a comparator this study did not have.

Why Psychedelics Are More Vulnerable to Expectancy Than Most Drugs

Expectancy effects appear throughout medicine. Patients who believe a treatment will work show better outcomes than patients who don't, even when the treatment is pharmacologically inert. This happens with pain medications, antidepressants, and blood pressure drugs. Expectancy is not a quirk of psychedelics. But several features of psychedelics make expectancy particularly powerful in microdosing research.

The first is perceptibility. Even a sub-perceptual dose, one small enough that the person does not feel high, involves a psychoactive compound that can produce subtle shifts in sensation, thought, or emotional tone. Many people can detect that something has happened. When you can tell you have taken a drug, your beliefs about what that drug does become active. This "break-blind" problem is manageable in trials for drugs like statins or antihistamines, which produce no detectable signal at therapeutic doses. For psychedelics, it is a structural challenge that follows every study design.

The second is cultural saturation. Microdosing exists within a thick layer of enthusiastic advocacy. Online communities with hundreds of thousands of members share almost exclusively positive experiences. Popular books frame the practice as transformative. Podcasts circulate accounts of improved creativity, mood, and focus. People who seek out microdosing research have typically moved through this environment first. By the time they enroll in a study, many have absorbed months of positive priming. Their expectancy scores are not neutral starting points, as the 263-person expectation survey in Polito and Stevenson's paper demonstrated directly.^[1]

The third is self-selection in research recruitment. When researchers advertise a microdosing study, the participants who respond are disproportionately people with high enthusiasm for psychedelics and strong prior beliefs about their benefits. The Ona and Bouso 2020 systematic review identified this explicitly as a structural source of bias in the field.^[2] A sample composed entirely of eager self-selected believers is structurally different from a random population sample. The expectancy distribution is skewed from the start, before the first capsule is dispensed.

These three factors compound each other. The enthusiastic, self-selected, culturally-primed microdosing research participant is precisely the hardest population in which to isolate pharmacological drug effects from belief-driven ones.

The Szigeti 2021 Trial: Both Groups Improved, Which Means What

The most ambitious attempt to control for these problems came from Szigeti and colleagues in 2021, in a citizen science study completed by 191 participants.^[4] At the time, it was the largest placebo-controlled trial conducted with any psychedelic. The design was elegant: participants followed online instructions to build placebo control into their own microdosing routine at home, preparing capsules — some containing psychedelic material, some inert — under a randomization and QR-code system that kept them blinded to condition until data collection ended.

The primary results were striking in their symmetry. All psychological outcomes improved significantly from baseline over the four-week dosing period in the microdose group. The placebo group also improved. When the researchers ran formal between-group comparisons, no significant differences were observed.

Key Study

Szigeti et al. (2021) · eLife

Population

191 people already microdosing, self-blinding at home via an online citizen-science protocol

Intervention

4 weeks of self-administered microdosing under participant-implemented randomization

Comparator

Visually identical inert placebo capsules, allocated by QR-code randomization

Finding

All psychological outcomes improved from baseline in the microdose group — and in the placebo group too. No significant between-group differences. Small acute and post-acute differences that did appear were explained by participants breaking blind.

null A null between-group result in a design with imperfect blinding means the drug effect could not be separated from expectancy — not that the drug effect is zero.

Szigeti et al., eLife, 2021 (PMID 33648632)

There was one exception. Small but statistically significant differences appeared on acute scales — emotional state, drug intensity, mood, energy and creativity measured in the hours after dosing — and on one post-acute scale, anxiety. At first read this looks like evidence of a real drug effect.

The interpretation is more complicated. Participants did not maintain perfect blinding. A substantial proportion correctly guessed which condition they were in, likely because they detected subtle physiological or perceptual cues. When the researchers examined who reported those benefits, the reports clustered in participants who had broken blind. The apparent drug effect tracked the break-blind rate rather than the allocation.

That result did not stand alone for long. In 2022, Cavanna and colleagues ran a double-blind, placebo-controlled trial in 34 people newly starting to microdose with Psilocybe cubensis, using a controlled 0.5 g dose of dried mushrooms.^[5] Acute subjective effects were significantly more intense on the active dose than placebo — but only among participants who correctly identified their experimental condition. Measures of well-being, creativity and cognitive function showed no benefit; the few changes detected pointed slightly toward cognitive impairment.

For a detailed account of the self-blinding design, its findings, and what they do and do not establish, see Is Microdosing Just Placebo? The Self-Blinding Studies and The Imperial College Microdosing Placebo Trial: What It Found.

Activated Expectancy Bias: The Formal Name for What Was Happening

The pattern Szigeti and colleagues observed, where break-blind status predicted acute reported benefit, was clear enough that they returned to it formally in a 2023 paper introducing the concept of Activated Expectancy Bias (AEB).^[6]

The framework begins with a basic observation. In a well-blinded trial, participants in both arms guess their condition at roughly chance rates, because there is no systematic cue. In a trial with a perceptible psychoactive substance, participants in the active arm have a detection advantage, so their correct-guess rate exceeds chance. Guessing is no longer balanced across arms.

That asymmetry does something specific. Correct identification activates a person's pre-existing expectations. They now believe, on the basis of felt evidence, that they took the real drug. If they arrived expecting psilocybin to improve their mood and creativity, those expectations are now switched on — and switched on more often in the active arm than the placebo arm. What they report afterwards reflects pharmacological activity combined with the psychological effect of believing they received something that works. In self-reported outcome data, those two components cannot be separated.

Mechanism of a False Positive

How a real drug effect turns into a measurement artifact

The trap in microdosing research is that the pharmacology itself is what breaks the blind. Each step below is individually well-evidenced; together they explain how a trial can report a benefit that the drug did not cause.

1. Step 1:

   The dose is perceptible

   Pharmacology

   Psilocybin is psychoactive even at microdose levels. Cavanna 2022 confirmed the pharmacology was active — EEG theta power fell on the real dose.

2. Step 2:

   The participant guesses correctly

   Pharmacology

   A faint shift in energy or visual texture is enough. In both Szigeti 2021 and Cavanna 2022, stronger acute effects appeared only among participants who correctly identified their condition.

3. Step 3:

   Expectation switches on — unequally

   Expectancy

   Those who believe they got the real thing now expect to improve. Because correct guessing is more common in the active arm, expectancy is no longer balanced across groups.

4. Step 4:

   Self-reported mood genuinely rises

   Expectancy

   Expectancy is a real mechanism, not an imaginary one. Kaertner 2021 found baseline expectancy predicted subsequent improvement in well-being.

5. Step 5:

   The trial records a drug effect

   Measurement artifact

   The difference between arms gets attributed to psilocybin. Szigeti 2023 named this activated expectancy bias and showed it can inflate effect estimates and manufacture false positives.

The correction

Szigeti’s 2023 Correct Guess Rate Curve estimates what a perfectly blinded trial would have shown using data from an imperfectly blinded one. Re-applied to the original self-blinding dataset, the residual microdose–placebo differences proved susceptible to the bias. Hence the authors’ distinction between a trial that merely has a placebo group and a trial in which participants are genuinely blind.

Source: Szigeti et al., eLife, 10, 2021; Cavanna et al., Translational Psychiatry, 12(1):307, 2022; Kaertner et al., Scientific Reports, 11(1):1941, 2021; Szigeti et al., Scientific Reports, 13(1):12107, 2023.

Szigeti and colleagues developed a statistical tool, the Correct Guess Rate Curve (CGRC), to measure and partially correct for this bias. Applied to the 2021 dataset, the small acute differences between active and placebo shrank substantially.

The implication is counterintuitive. Acute effects, measured in the hours after dosing when the substance is most detectable, are the data points most contaminated by AEB. A surge in mood or creativity scores immediately after dosing is precisely the kind of result most likely to reflect activated expectancy rather than clean pharmacological activity. Researchers and readers often treat acute effects as the most direct evidence that a drug is doing something real. For perceptible psychoactive substances, they may be among the least reliable evidence available.

For a technical breakdown of how this plays out in the original self-blinding dataset, see The Imperial College Microdosing Self-Blinding Study: What It Actually Proved.

Does Expectancy Research Mean Microdosing Does Not Work?

The evidence on expectancy points in a clear direction: belief drives a substantial portion of the reported benefit in uncontrolled microdosing studies. But there is a precise thing this evidence establishes, and a different thing it does not.

What it establishes: the studies conducted to date cannot demonstrate that psilocybin causes the reported benefits. Polito and Stevenson 2019 found benefits in conditions where expectancy was entirely uncontrolled, and separately showed how large expectations run in this population.^[1] Kaertner 2021 showed baseline belief predicted who improved most.^[3] Szigeti 2021 found both arms improving equally in the most rigorous design available.^[4] Cavanna 2022 found acute effects confined to those who broke blind.^[5] Szigeti 2023 showed how that mechanism manufactures false positives.^[6]

What it does not establish: that psilocybin is pharmacologically inert at sub-perceptual doses. A null between-group result is not proof that the drug does nothing. It means the design could not detect a drug effect above the noise created by expectancy and blind-breaking.

That argument was made forcefully in 2024, when Polito and Liknaitzky published a rapid review of every controlled microdosing study with a placebo comparator they could identify — 19 in total, risk-of-bias assessed with the Cochrane tool.^[7] Their reading was that across this literature, low doses of LSD and psilocybin do produce changes relative to placebo in neurobiology, physiology, subjective experience, affect and cognition. They then set out eight reasons why concluding "microdosing is mostly placebo" is premature and possibly wrong: there have been only a small number of controlled studies; those studies had small samples; there is evidence of dose-dependent effects; only a narrow range of doses has been tested; the doses used may have been too low; only non-clinical populations have been studied; the studies remain susceptible to selection bias; and the measured impact of expectancy is itself small.

Fact Check

Has microdosing been shown to be nothing but placebo?

The claim

“Placebo-controlled trials have proven that microdosing is just a placebo effect.”

Headline coverage of the Szigeti 2021 self-blinding study, which reported no significant differences between microdose and placebo groups.

What research shows

Overstated

Szigeti 2021 found no between-group difference on primary psychological outcomes, and the acute differences that appeared were explained by blind-breaking. Cavanna 2022 independently reproduced that pattern. Both are genuine null results for benefit.

Polito and Liknaitzky's 2024 review of 19 placebo-controlled studies concluded the opposite verdict is also unproven: the trials are few, small, restricted to non-clinical populations and narrow dose ranges, and show some dose-dependent effects. Their stated conclusion is that it is not yet possible to determine whether microdosing is a placebo.

Szigeti et al., eLife, 2021; Cavanna et al., Translational Psychiatry, 2022; Polito & Liknaitzky, Journal of Psychopharmacology, 2024

Evidence Review

1. Naturalistic surveys & daily-diary studiescompleted

   Large, consistent positive self-reports — with expectancy and self-selection entirely uncontrolled (Polito & Stevenson 2019; reviewed in Ona & Bouso 2020).

2. Small placebo-controlled trialscompleted

   Szigeti 2021 (n=191) and Cavanna 2022 (n=34) found no benefit beyond placebo, with the differences that did appear traceable to blind-breaking.

3. Adequately powered RCT in a clinical population with verified blind integritymissing

   No trial has yet combined a clinical sample, a range of controlled doses, adequate power, and formal measurement of blinding success.

4. Systematic synthesis of controlled trialscurrent

   Polito & Liknaitzky 2024 synthesised 19 placebo-controlled studies and concluded the placebo question cannot yet be resolved — precisely because the trial above does not exist.

Expectancy effects in microdosing are large, measurable and systematically under-controlled. Whether psilocybin adds benefit on top of them is genuinely unresolved in both directions: the positive evidence is confounded, and the null evidence comes from trials too small, too short and too narrowly dosed to rule a real effect out.

Polito & Stevenson 2019; Ona & Bouso 2020; Szigeti 2021; Cavanna 2022; Polito & Liknaitzky 2024

Whether psilocybin at sub-perceptual doses changes brain activity independently of belief is investigated through methods that do not rely on self-report — and Cavanna's EEG findings suggest it does something measurable. Whether any such activity translates to meaningful lived benefit is covered in depth in Can Doses Too Small to Feel Still Change the Brain?.

The honest summary: expectancy effects are large, measurable, and systematically under-controlled in microdosing research. The people most likely to report benefiting are enthusiastic believers with high baseline expectancy, which is exactly who enrolled in the early observational studies that generated much of the field's initial evidence. That observation does not prove microdosing is a placebo. It means the question has not yet been answered with sufficient methodological rigor to know.

Safety

Moderate

Expectancy findings are not a treatment recommendation in either direction

Concern

Microdosing is frequently taken up as a self-directed substitute for treatment of depression or anxiety, often on the strength of exactly the uncontrolled, expectancy-saturated reports described in this article. Polito and Stevenson 2019 also recorded an increase in reported neuroticism over their six-week study, and Cavanna 2022 found small changes in the direction of cognitive impairment.

What the research says

No controlled trial has yet tested microdosing in a clinical population, so there is no evidence base supporting it as a treatment for any diagnosed condition — and none establishing its safety over sustained use.

Particularly relevant for: Anyone considering microdosing in place of, or alongside, treatment for a mental health condition

What to do

Discuss any change to mental health treatment with a qualified clinician. This article summarises research and is not medical advice.

Polito & Stevenson, PLoS ONE, 2019; Cavanna et al., Translational Psychiatry, 2022; Polito & Liknaitzky, Journal of Psychopharmacology, 2024

For readers interested in the dosing schedules behind this research — Polito and Stevenson's participants chose their own regimens, and the daily rating scale they completed was adapted from Fadiman's protocol^[1] — see The Fadiman Protocol: The Original Microdosing Schedule. For the specific domain where expectancy contamination appears most pronounced, see Microdosing and Mood Improvement: Clinical and Survey Data.

The Bottom Line

Expectancy effects are the central methodological challenge in microdosing research. A 2021 study of 81 microdosers found that baseline expectancy scores predicted four-week improvements in well-being, a result its authors called suggestive of a significant placebo response — though with no placebo arm, that study could not weigh belief against drug. In the best-controlled trials available, a self-blinding study of 191 participants and a double-blind study of 34, both groups improved equally, and the only apparent between-group differences tracked with participants breaking blind. Szigeti et al. 2023 formalized this as Activated Expectancy Bias. A 2024 review of 19 placebo-controlled studies then argued the reverse conclusion is equally unproven, since the controlled trials remain few, small, and narrowly dosed. Whether psilocybin produces psychological benefit above expectancy remains genuinely unresolved — not because the research is weak, but because the methodological problem is structural: psychoactive substances that can be detected contaminate any blinded comparison.

Frequently Asked Questions

What is the expectancy effect in microdosing?

An expectancy effect occurs when a person's belief that a treatment will work produces measurable improvement independent of any pharmacological activity. In Kaertner et al. 2021, baseline expectancy scores, meaning how much participants expected microdosing to help before taking anything, predicted subsequent improvements in well-being across a four-week period in 81 microdosers. The authors described this as suggestive of a significant placebo response. Because that study had no placebo group, it shows that belief predicts who improves — not that belief outperforms the drug.

How much of microdosing is placebo?

The most rigorous evidence available does not resolve this question, which is the accurate answer rather than a hedge. The Szigeti 2021 self-blinding study of 191 participants found both active and placebo groups improved equally over four weeks, with no significant between-group differences on primary psychological outcomes, and Cavanna et al. 2022 reproduced that pattern in a 34-person double-blind trial. The Ona and Bouso 2020 systematic review concluded that placebo effects could not be ruled out in any study lacking blinded controls. But the Polito and Liknaitzky 2024 rapid review of 19 placebo-controlled trials concluded it is not yet possible to determine whether microdosing benefits are pharmacological or placebo-driven, arguing that the controlled studies so far are too few, too small, and too narrowly dosed to settle it.

Does belief affect microdosing outcomes?

Yes, measurably. Kaertner et al. 2021 found that baseline expectancy scores, measured before any dosing, predicted subsequent well-being improvements in 81 microdosers. In the Szigeti 2021 trial and again in Cavanna et al. 2022, participants who correctly guessed they had taken the active substance reported greater acute effects than those who could not tell, suggesting that the belief of having taken the real drug, not just the drug itself, shaped reported experience. Notably, Polito and Stevenson's 2019 survey of 263 people found expectations were far larger than reported outcomes and pointed at different domains, so belief shapes the experience without simply dictating it.

What is activated expectancy bias in drug trials?

Activated Expectancy Bias (AEB) is a concept formalized by Szigeti et al. in 2023 to describe what happens in psychoactive drug trials when participants correctly identify which condition they are in. In a well-blinded trial, guessing your condition is roughly chance-level, since there is no reliable cue. In a psychedelic trial, the active substance produces detectable sensations, raising the correct-guess rate in the active arm above chance. When those participants realize they took the real drug, their pre-existing positive beliefs become activated and amplify their reported outcomes, producing a between-group difference that the drug did not cause. Szigeti's team developed the Correct Guess Rate Curve (CGRC) as a statistical tool to measure and partially correct for this inflation.

Can positive thinking explain microdosing benefits?

Expectancy, a specific and measurable form of belief about treatment outcome, explains a substantial portion of self-reported microdosing benefit in uncontrolled studies. But positive thinking does not fully explain those benefits, for three reasons. First, expectancy effects are neurobiological, not merely attitudinal: belief engages real physiological pathways. Second, Polito and Stevenson 2019 found the effects people most expected to change were unrelated to the effects actually reported, which is not what a pure expectation-fulfilment account predicts. Third, the controlled evidence leaves the question of pharmacological benefit above expectancy open rather than answered — Polito and Liknaitzky's 2024 review found dose-dependent effects across 19 placebo-controlled studies and concluded the placebo verdict is premature.

Sources & References

1. 1RTHM-01113·Polito, Vince et al. (2019). “A systematic study of microdosing psychedelics..” PloS one.Study breakdown →PubMed →↩
2. 2RTHM-01271·Ona, Genís et al. (2020). “Potential safety, benefits, and influence of the placebo effect in microdosing psychedelic drugs: A systematic review..” Neuroscience and biobehavioral reviews.Study breakdown →PubMed →↩
3. 3RTHM-01371·Kaertner, L S et al. (2021). “Positive expectations predict improved mental-health outcomes linked to psychedelic microdosing..” Scientific reports.Study breakdown →PubMed →↩
4. 4RTHM-01441·Szigeti, Balázs et al. (2021). “Self-blinding citizen science to explore psychedelic microdosing..” eLife.Study breakdown →PubMed →↩
5. 5RTHM-00748·Cavanna, Federico et al. (2022). “Microdosing with psilocybin mushrooms: a double-blind placebo-controlled study..” Translational psychiatry.Study breakdown →PubMed →↩
6. 6RTHM-01797·Szigeti, Balázs et al. (2023). “The difference between 'placebo group' and 'placebo control': a case study in psychedelic microdosing..” Scientific reports.Study breakdown →PubMed →↩
7. 7RTHM-00900·Polito, Vince et al. (2024). “Is microdosing a placebo? A rapid review of low-dose LSD and psilocybin research..” Journal of psychopharmacology (Oxford.Study breakdown →PubMed →↩