The most honest criticism of the affirmations genre is also the most commonly misread: it’s just placebo. The criticism assumes that placebo means fake — a real effect producing an illusory result, or an illusory result producing a real feeling. Neither description is accurate.
This page is about what the placebo effect actually is, why the research on it is more interesting than either side of the debate tends to acknowledge, and how honestly naming it changes the case for the practices in this cluster.
What the placebo effect actually is
The familiar version of placebo is the sugar pill given to a patient who believes it’s real medicine and improves because of the belief. That version is real, but incomplete, because it implies that deception is the load-bearing variable — that the effect would evaporate if the patient knew.
It mostly doesn’t.
Ted Kaptchuk and colleagues at Harvard published a trial in 2010 in which patients with irritable bowel syndrome were randomized to open-label placebo or a no-treatment waitlist. Open-label means the participants were told directly: these are placebo pills. The intake form said so. The intake conversation said so. They were told placebos could still work through mind-body processes and were asked to take the pills twice daily.
Patients in the open-label group improved significantly more than the no-treatment group on both primary symptom measures and global improvement scales. Twice as many reported adequate relief.
The finding doesn’t claim everyone responded, or that the effect was large enough to replace active treatment for serious conditions. What it establishes is that deception is not necessary. When you remove deception and the response persists, expectancy becomes the most parsimonious remaining explanation — an anticipatory state the nervous system genuinely prepares for, independent of what the patient was told. Other explanations (conditioning, ritual, reporting effects) are possible; the study doesn’t rule them out. But they all require the person to be oriented toward the treatment in some way, which is expectancy by another name.
What expectancy does physiologically
Expectancy is not metaphorical. Neuroimaging studies by Tor Wager and colleagues have documented endogenous opioid release during placebo analgesia — the brain releases its own pain-reducing compounds when it expects relief, whether or not active medication is present. The subjective experience corresponds to a real neurochemical event, not a misremembered one.
The mechanism extends beyond pain. Neuroscientist Fabrizio Benedetti and colleagues have documented expectancy-driven changes in motor performance, respiratory function, and hormone release — each with measurable physiological correlates. The nervous system is a prediction machine; preparing for an expected state is part of how it operates. When the preparation is in a beneficial direction, the preparation itself is the benefit.
Psychologist Irving Kirsch formalized this as response expectancy theory in the 1980s: expectancies about the subjective outcomes of events can directly produce those outcomes by shaping experience. The expectation of calm, practiced consistently, shifts the probability of experiencing calm. Not universally, not magically — but reliably enough to measure and reliably enough to matter.
The nocebo: the same mechanism in reverse
If expectancy can produce beneficial changes, the same pathways can run toward harm. Nocebo is the expectation of harm that produces real harm.
Someone who is told that a treatment causes headaches is meaningfully more likely to report headaches, sometimes through the same physiological routes that make placebo work. Someone in a clinical trial assigned to the nocebo arm — who experiences side effects from an inert compound — is not confused or fabricating. The mechanism is real.
For sleep, this has more relevance than most articles in this space acknowledge. The story you carry to bed about whether tonight will go well is already setting a direction. “I never sleep well before a presentation” is a prediction, and predictions shape what happens. The sleep-effort loop — trying hard to fall asleep while monitoring whether it’s working — is a nocebo loop: the monitoring generates arousal, the arousal confirms the prediction, the prediction intensifies the monitoring. The mechanism is expectancy, not character.
This is part of why the hypnagogic state matters as a delivery window. It is the interval just before the prediction-machine shuts down for the night. What you’ve loaded into it shapes the direction of the drift.
How much of the affirmations cluster runs on expectancy?
A substantial portion. This is the honest answer, and it is worth keeping separate from the ways people typically misconstrue it.
When someone practices sleep affirmations for three weeks and notices a shift, several things are probably running in parallel. The expectancy of rest develops across consistent sessions. The language in the affirmations is processed at varying depth during sleep onset. The consistent practice forms a bedtime routine that itself signals the nervous system toward the wind-down transition. The repeated statements shape what the subconscious mind reaches for by default — which is a slower mechanism than expectancy but a real one.
Expectancy is not the whole story. But it is present, and it is working, and it is legitimate.
The most honest treatment of subliminal audio makes the same concession: the most plausible explanation for why some people who use subliminal audio report changes is expectancy — and expectancy, post-Kaptchuk, is its own worthwhile mechanism rather than a polite way of saying something didn’t work. Naming it is not a retreat. It is the accurate account.
What expectancy can’t reach
Expectancy operates most strongly where subjective experience has room to move: pain, anxiety, fatigue, mood, sleep onset. These are domains where the expected experience and the actual experience are partly constituted by the same thing.
Expectancy is much weaker for disease processes measured objectively regardless of self-report. This is why the expansive claims in the power of the subconscious mind tradition — including Coué’s original promise that his formula could heal organic illness — collapsed under scrutiny in the 1920s. The placebo effect did not resolve tuberculosis or produce documented tumor regression. The ceiling is real, and honestly stating it is the only way to keep the rest of the cluster’s claims credible.
Why “just placebo” is the wrong frame
Henry Beecher’s 1955 paper, “The Powerful Placebo,” is the founding document of the modern understanding that placebo responses in clinical trials are real and significant. His argument was that the mind’s anticipatory processing produces genuine physiological outcomes, and that this was consequential for how clinical research should be designed. The word “just” had not yet attached itself to placebo; it arrived later as dismissive shorthand.
The open-label finding does something useful to that shorthand. It removes the deception from placebo. Once deception is no longer in the mechanism, the objection loses its sharpest edge. If someone practicing nightly autosuggestion is partly working through expectancy, and expectancy is a documented physiological pathway that operates without deception, then what exactly is the dismissal doing?
The result of a practice is either real or it isn’t. If the result is real, the pathway that produced it is valid — whether or not that pathway matches a preferred theory of mechanism. This is not a methodological loophole. It is how medicine works for many interventions we don’t dismiss.
The smallest version of this practice
The practical implication is not to cultivate naive optimism or to pretend the ceiling doesn’t exist. It is to take the direction of your expectations seriously as a variable that you can actually move.
Before sleep is the interval that matters most, because it is where the nocebo and placebo compete for the same window. The question is not whether you believe perfectly, but what you are leaning toward when the night begins.
The autosuggestion tradition understood this intuitively. Coué’s comparative framing — every day, in every way, I’m getting better and better — was calibrated to stay inside the plausible range. A sentence the mind can almost believe is one expectancy can actually act on. A sentence the mind argues with generates arousal instead of preparation. The almost-believable version of the claim is not a failure of ambition; it is the version the mechanism can work with.
Murmora is built around the same logic. The sparse-whisper format — a brief line at the edge of sleep, in a guide voice chosen for the practice, with deliberate silence between each — is the delivery pattern most likely to land where expectancy operates. Dense continuous audio keeps the critical filter engaged; sparse audio arrives in the intervals where resistance is lower. The personalization matters for the same reason: a specific expectancy about a real goal you hold prepares the nervous system differently than a generic statement neither you nor anyone could argue about. That specificity is the mechanism, and it is worth choosing carefully.