# Placebo effect: how belief changes your brain

- Category: **Science > Psychology & Neuroscience**
- Publisher: **Psyll Magazine** - [https://psyll.com](https://psyll.com)
- Author: **Laura J. Grays** - [https://psyll.com/laura](https://psyll.com/laura)
- Original article: [https://psyll.com/articles/science/psychology-neuroscience/placebo-effect-how-belief-changes-your-brain](https://psyll.com/articles/science/psychology-neuroscience/placebo-effect-how-belief-changes-your-brain)

---

![Main image](https://psyll.com/assets/image/placebo-effect-how-belief-changes-your-brain.jpg)

**The placebo effect is real neuroscience, not imagination. Discover how belief triggers brain chemicals that genuinely reduce pain and improve health symptoms.**

---

The placebo effect, often summed up in the phrase "mind over matter," is one of the most studied and most misunderstood phenomena in clinical medicine. It is not a trick of the imagination, and it is not "nothing happening." It describes something stranger and more interesting: a measurable physiological change that occurs because a person believes a treatment will help them, even when that treatment has no active pharmacological ingredient at all.

![Discover the neuroscience of belief and how the brain creates real medicine.](https://psyll.com/assets/image/discover-the-neuroscience-of-belief-and-how-the-brain.png)

I want to be precise about what that means, because the phrase "it's just a placebo" has done a disservice to decades of careful neuroscience. When researchers say placebo effects are real, they mean real in the way a blood test is real - detectable changes in brain activity, neurotransmitter release, and bodily function that show up on a scanner or a hormone panel, not just on a patient's mood.

As Tor Wager and Lauren Atlas put it in their influential *Nature Reviews Neuroscience* paper, placebo effects are best understood as brain-body responses to context information that promote health and wellbeing. That single sentence reframes the whole topic. The brain is not being fooled. It is doing exactly what it evolved to do - using context, history, and expectation to prepare the body for what's coming next.

## How the brain turns belief into biology

Understanding the placebo effect means moving past the old, lazy idea that it's "just psychological." The mechanisms are wired into our neural architecture, shaping how we feel pain, how we respond to stress, and how several of our internal systems regulate themselves.

![The placebo effect represents measurable neurobiological changes, not imagination.](https://psyll.com/assets/image/the-placebo-effect-represents-measurable-neurobiological.png)

The effect is mediated by an interplay of cognitive and emotional processes - expectation, learning, memory, and motivation - that recruit multiple brain systems and several distinct neurochemical pathways. This is part of why placebo responses show up so consistently in clinical trials across nearly every field of medicine, from gastroenterology to neurology to oncology.

### The endogenous opioid system

The best-studied mechanism behind the placebo effect is its interaction with the brain's own opioid system. This system governs pain modulation, and placebo analgesia - pain relief produced by an inert treatment - is the textbook example of it in action.

When someone expects pain relief, that expectation can trigger the release of endorphins, the body's natural painkillers. These endorphins bind to opioid receptors in the brain, reducing pain perception in a way that looks remarkably similar to how an actual analgesic drug works. Neuroimaging studies have repeatedly shown reduced activity in pain-processing regions such as the anterior cingulate cortex, the periaqueductal gray, and the amygdala during placebo analgesia, which is fairly strong evidence that something biological - not merely attitudinal - is happening.

### The dopamine system's role

Pain isn't the only domain where this shows up. The dopamine system plays an outsized role too, and nowhere is that clearer than in Parkinson's disease research.

In one of the field's foundational studies, researchers using PET imaging found that patients with Parkinson's disease released substantial amounts of dopamine in the striatum after receiving a saline injection they believed was active medication. A later study went further, showing that the *strength* of a patient's expectation directly predicted how much dopamine was released, particularly in the ventral striatum - the brain's reward circuitry. Belief, in other words, wasn't incidental to the response. It was driving it.

![The brain's endogenous opioid system releases natural painkillers to reduce pain.](https://psyll.com/assets/image/the-brains-endogenous-opioid-system-releases-natural.png)

This dopamine surge can mimic the therapeutic effect of real dopaminergic drugs, producing measurable improvements in motor performance. It also implicates structures like the striatum and the ventromedial prefrontal cortex (vmPFC) in a broader story about motivation, reward, and anticipation - one that extends well past Parkinson's disease into how we respond to any treatment we expect to work.

### Lesser-known players: endocannabinoids and cholecystokinin

The opioid and dopamine systems get most of the attention, understandably, but they aren't the whole story. The endocannabinoid system, which helps regulate pain, mood, and appetite, has also been identified as a contributor to placebo analgesia in certain contexts - particularly when opioid blockers are used experimentally and the analgesic effect persists anyway, pointing researchers toward a parallel pathway.

![Placebos increase dopamine activity, mimicking real drugs to improve motor skills.](https://psyll.com/assets/image/placebos-increase-dopamine-activity-mimicking-real-drugs-to.png)

The neuropeptide cholecystokinin has been linked to placebo's *failure* modes as well - it appears to work against opioid-mediated placebo analgesia, and is more closely tied to the nocebo response, which I'll get to shortly. Taken together, this tells us the placebo effect isn't a single switch. It's a network of overlapping biochemical systems, several of which can be engaged or disengaged depending on context.

## The brain regions doing the work

Neuroimaging has mapped out, with increasing precision, which brain regions activate or quiet down during a placebo response. A few deserve specific mention.

* **Prefrontal cortex, anterior insula, and nucleus accumbens.** These regions show increased activation during placebo treatment, especially when a patient holds genuinely positive expectations about the outcome. The prefrontal cortex in particular is central to cognitive control, expectation formation, and decision-making - it's doing a lot of the "believing" work.
* **Posterior insula.** In placebo analgesia, treatments aimed at reducing pain correspond with *less* activity here. This region is heavily involved in constructing the subjective, felt experience of pain, so a drop in its activity tracks closely with a drop in reported pain.
* **Basal ganglia.** Beyond its well-known role in motor control, this subcortical structure links sensory information to action and shows activity changes under placebo conditions, tying into the motor benefits seen in conditions like Parkinson's.
* **Ventromedial prefrontal cortex, insula, amygdala, and hypothalamus.** This cluster underlies placebo effects on pain, autonomic responses, and even some immune signaling. The vmPFC is considered a hub for processing relationships among concepts, which gives it a natural role in forming and sustaining expectations over time.

![Complex networks of cortical and subcortical regions orchestrate the placebo response.](https://psyll.com/assets/image/complex-networks-of-cortical-and-subcortical-regions.png)

A team led by Greg Scherrer and Chong Chen at the University of North Carolina School of Medicine, working with collaborators at Stanford, the Howard Hughes Medical Institute, and the Allen Institute for Brain Science, published findings in *Nature* identifying a specific circuit connecting the rostral anterior cingulate cortex to the pontine nucleus and on to the cerebellum. It's worth being precise here: this discovery came from mouse models, not human patients. Mice were conditioned in a two-chamber apparatus, learning to associate moving into one chamber with relief from a hot floor, and the researchers tracked which neurons lit up as that expectation took hold.

The team found something genuinely unexpected: the pontine nucleus, a brainstem structure with no prior established role in pain processing, turned out to be studded with an unusually dense concentration of opioid receptors. When mice expected pain relief, signaling along this cortico-ponto-cerebellar pathway intensified, and artificially activating the pathway produced pain relief on its own, even without any placebo conditioning. As Scherrer put it, neurons in the cerebral cortex communicate with the pons and cerebellum to adjust pain thresholds based on expectations - something he described as both unexpected and exciting given decades of prior assumptions about pain circuitry. It's an animal study, so it isn't a direct map of the human brain, but it gives researchers a concrete, testable circuit to chase in a species where invasive techniques like optogenetics are possible, and the NIH has flagged the pontine nucleus as a promising new target for pain-relief therapies going forward.

## Learning shapes belief: classical conditioning and expectation

The placebo effect isn't purely an in-the-moment cognitive event. It is also deeply shaped by learning, and particularly by classical conditioning. If a patient repeatedly receives a distinctly colored pill alongside real pain relief, the color itself can eventually start triggering a pain-reducing response on its own - even once the "medication" is switched to an inert version.

![The brain does not distinguish imagined treatment from reality, releasing chemicals.](https://psyll.com/assets/image/the-brain-does-not-distinguish-imagined-treatment-from.png)

This is why the *context* surrounding a treatment matters as much as the treatment itself. Wager and Atlas were explicit about this point, writing that placebo interventions do not, by definition, have any direct therapeutic effects on the body, but that all treatments are delivered in a context including social and physical cues, verbal suggestions, and treatment history - a context the brain actively interprets, which can elicit expectations, memories, and emotions that influence health outcomes in both brain and body.

That context includes things clinicians don't always think to control for: a confident tone of voice, the setting of a hospital versus a kitchen table, the shape and color of a pill, even a patient's history with prior treatments. None of it is "real medicine" in the pharmacological sense, and all of it measurably shapes outcomes.

## What the numbers actually say

It's tempting to wave at "the placebo effect" as a vague, overarching force, but the data are more specific than that, and worth sitting with.

* A widely cited study in *Science Translational Medicine*, examining migraine pain medication, found that a placebo achieved roughly **50% of the effectiveness** of an active analgesic when participants were told the drug had genuine pain-relieving properties. That's not a footnote - that's half the effect size of an actual medication, generated entirely by expectation.
* Estimates suggest that somewhere around **one in three people** will experience a placebo response across various health conditions, which tells you this isn't a rare quirk affecting a suggestible minority. It's closer to a baseline feature of how human physiology responds to care.
* Placebos have measurable effects on nausea and vomiting, asthma symptoms, erectile dysfunction, blood pressure, physical performance metrics like endurance and strength, depression and anxiety symptoms, and specific phobias.
* In sport-science research, a 2024 systematic review update found a moderate-to-large pooled placebo effect on athletic performance (Cohen's *d* of roughly 0.67), with nutritional placebos - athletes told an inert substance was a performance-enhancing supplement - producing the strongest gains of any category studied.

![Approximately 1 in 3 individuals will experience measurable placebo responses.](https://psyll.com/assets/image/approximately-1-in-3-individuals-will-experience-measurable.png)

It's worth being clear-eyed about where this does and doesn't apply. Placebos tend to be most effective for symptoms that are heavily modulated by the brain - pain, stress-related insomnia, fatigue, and the subjective burden of nausea during cancer treatment, for instance. They do not shrink tumors, heal fractures faster, or alter objective markers of disease progression. That distinction matters enormously, and I'll come back to it.

![Placebo treatments for pain can be up to 50% as effective as active analgesics.](https://psyll.com/assets/image/placebo-treatments-for-pain-can-be-up-to-50-as-effective-as.png)

### Why this matters so much in clinical trials

Because placebo responses are reliable and often substantial, regulators require that new drugs demonstrate efficacy *beyond* placebo before approval. This isn't bureaucratic caution for its own sake - it's the only way to confirm that a benefit comes from a drug's actual pharmacology rather than from the ritual of treatment itself.

Interestingly, non-blinded trials - where patients or clinicians know who's receiving the active treatment - tend to show disproportionately large placebo responses. That gap is itself informative: it tells researchers how much of an observed benefit might be coming from expectation and bias rather than biology, which is exactly why blinding remains the backbone of trial design.

![Modern clinical trials must rigorously account for the brain's baseline healing power.](https://psyll.com/assets/image/modern-clinical-trials-must-rigorously-account-for-the.png)

A 2025 scoping review in *Health Psychology Review* synthesized dozens of meta-analyses and confirmed that individual factors (like age), clinical factors (like baseline symptom severity), and psychological factors (like expectation itself) all shape the size of the placebo response, which is part of why effect sizes vary so much from one trial to the next. There's also a quieter methodological debate worth knowing about: researchers have recently scrutinized whether trials should compare new drugs against an inert placebo or an "active placebo" - a substance that mimics a drug's side effects (like dry mouth) without its therapeutic action, to better preserve blinding. A 2025 meta-epidemiological analysis found the impact of this choice is genuinely uncertain and can shift estimated drug effects in either direction, which is one reason trial design in this area is still evolving rather than settled.

## A note on the famous wartime story

Many popular accounts of placebo history repeat a vivid anecdote about Henry Beecher, the Harvard anesthesiologist, running out of morphine while treating wounded soldiers and substituting saline injections that worked anyway. It's a great story. It also appears, on closer historical examination, to be at least partly embellished or apocryphal - medical historians have been unable to find it documented in Beecher's own published work, and some trace its popularization to later retellings rather than to Beecher's actual case records.

What *is* well documented is Beecher's 1955 paper "The Powerful Placebo," in which he analyzed data across multiple clinical studies and estimated that roughly a third of patients' improvement in various trials could be attributed to placebo response rather than active treatment. That paper, not the battlefield legend, is what actually launched modern placebo-controlled trial design, and it deserves the credit more than the story does.

## The etymology and the early history

The word "placebo" comes from the Latin for "I shall please." For centuries it described sham treatments given mainly to comfort patients rather than cure them. By the 18th century, physicians had begun formally noting the psychological effects inert treatments could produce. Dr. John Haygarth's 1799 experiments with rheumatism patients, using fake wooden "tractors" instead of the metallic ones popularized by the practitioner Elisha Perkins, are often cited as among the earliest documented placebo-controlled comparisons, showing that the *belief* in a device's healing power, not the device itself, was doing the work.

## Beyond positive thinking

It's worth saying plainly: contemporary placebo science is not a rebranding of "the power of positive thinking." It describes a genuine brain-body connection capable of producing physiological and cognitive shifts that, in specific symptom domains, can rival the effects of active drugs. As wellness author David R. Hamilton has written, to a large extent the brain doesn't distinguish real from imaginary, and this underpins some aspects of the placebo effect - when a person imagines something is happening, the brain treats it as real and releases the chemical substances that correspond to that belief.

That's a striking claim, and the research backs a version of it, though I'd add a clinician's caveat: this mechanism explains *symptom modulation*, not disease cure, and conflating the two is where a lot of well-meaning placebo enthusiasm goes wrong.

## What placebos cannot do

This is the part of the conversation that gets skipped too often, and I think it's the most important section in this entire piece. Placebos can make people feel genuinely better. They can reduce the *experience* of pain, fatigue, nausea, and low mood. What they cannot do is alter the underlying biological course of a disease.

![Placebos alter subjective symptom perception but do not cure underlying diseases.](https://psyll.com/assets/image/placebos-alter-subjective-symptom-perception-but-do-not-cure.png)

A placebo will not shrink a tumor. It will not heal a fracture faster. It will not clear an infection. It acts on symptom perception and subjective wellbeing - real and valuable targets in their own right - but not on the objective pathology driving those symptoms. Confusing the two is not just a scientific error; in clinical settings, it can be a dangerous one, since it risks delaying treatments that actually address the disease itself. If you're managing chronic pain in particular, it's worth reading about how structured approaches like CBT and mindfulness can [rewire how the brain processes pain](https://psyll.com/articles/science/psychology-neuroscience/how-to-rewire-your-brain-to-manage-chronic-pain) - a related but distinct mechanism from placebo response, and one with a more durable evidence base for long-term symptom management.

## The nocebo effect: the placebo's unwelcome twin

Expectation cuts both ways. The nocebo effect is what happens when *negative* expectations about a treatment, procedure, or condition produce genuinely adverse outcomes. Warn a patient extensively about a pill's possible side effects, even when the pill is inert, and a meaningful number will report experiencing those exact side effects.

This has real clinical consequences. Informed consent requires disclosing risks, but how those risks are framed and discussed appears to influence how often patients actually experience them - a genuinely difficult ethical tension that researchers and clinicians are still working through.

![Negative expectations can trigger the nocebo effect, creating real adverse symptoms.](https://psyll.com/assets/image/negative-expectations-can-trigger-the-nocebo-effect.png)

A 2025 study published in *eLife* sharpened this picture in a way that I think deserves wider attention. Using a within-subject design in healthy volunteers, researchers directly compared the magnitude and durability of placebo and nocebo effects on experimental pain, both immediately and at a one-week follow-up. The nocebo effects were consistently stronger than the placebo effects at both time points - a pattern the researchers describe as consistent with an evolutionarily sensible "better safe than sorry" strategy, where the nervous system weights threat-related expectations more heavily than reassuring ones. The same study found something clinicians may find uncomfortable: higher perceived competence in the person delivering a negative suggestion was associated with a *stronger* nocebo response, meaning that the very credibility that normally helps a treatment work can backfire when the message being delivered is a warning rather than a reassurance. That asymmetry is a big part of why some researchers now argue nocebo effects deserve at least as much clinical attention as placebo effects, not less.

## Open-label placebos: when honesty doesn't kill the effect

Perhaps the most genuinely surprising development in recent placebo research is the open-label placebo - a treatment where patients are told outright that the pill contains no active medication, and still experience benefit.

This finding mattered enough to overturn a long-standing assumption in the field, namely that deception was a *necessary* ingredient of the placebo response. A 2010 randomized trial in patients with irritable bowel syndrome found that those given openly labeled placebo pills reported significantly greater symptom improvement than a no-treatment control group. A larger 2021 follow-up trial at Beth Israel Deaconess Medical Center, involving 262 adults with IBS, compared open-label placebo against both a double-blind placebo arm and a no-pill control. Sixty-nine percent of participants who received the open-label placebo reported a clinically meaningful improvement in their symptoms, and that improvement was statistically indistinguishable from the double-blind group - meaning, remarkably, that not knowing whether you're getting a real treatment provided no extra benefit over knowing for certain you weren't.

![Patients experience benefits even when they know the treatment is completely inert.](https://psyll.com/assets/image/patients-experience-benefits-even-when-they-know-the.png)

As the trial's senior investigators, Ted Kaptchuk and Anthony Lembo, have noted, this finding has real implications for how the placebo effect might be harnessed ethically in clinical practice, since it suggests the ritual of treatment - the ceremony of taking a pill, the attention of a clinician, the explicit suggestion that something good is happening - can mobilize the body's own regulatory systems even without any pretense involved.

The pattern has since been tested more broadly, with mixed but informative results. A randomized trial in chronic low back pain found that adding three weeks of open-label placebo to usual care produced significantly greater pain reduction and disability improvement than usual care alone, and a separate brain-imaging trial using an open-label saline injection found similar benefits at one month, accompanied by measurable increases in connectivity between the prefrontal cortex and brainstem pain-regulation circuits - a neural fingerprint that looked a lot like the one seen with deceptive placebos. Open-label placebo trials in episodic migraine have shown improvements in migraine-related disability and quality of life, although results here have been more inconsistent than in IBS or back pain, and at least one trial found no reduction in headache frequency itself despite quality-of-life gains. That inconsistency is a useful reminder that open-label placebo is a promising tool, not a guaranteed one, and its effect size appears to vary meaningfully by condition.

## Frequently asked questions

**Is the placebo effect "fake" or "all in your head"?**
No, in the sense that matters clinically. The changes are measurable in brain imaging, neurotransmitter levels, and physiological markers. It is "in your head" only in the sense that the brain is the organ generating the response, which is also true of every other sensation you have ever had.

**Can placebos cure a disease?**
No. They can meaningfully reduce the subjective burden of symptoms like pain, fatigue, and nausea, but they do not alter the underlying pathology - tumors, fractures, infections, and similar objective disease processes are unaffected.

**Does the placebo effect work if you know you're taking a placebo?**
Often, yes. Open-label placebo trials in conditions like irritable bowel syndrome and chronic back pain have shown meaningful symptom improvement even when patients are told explicitly that the pill is inert.

**What is the opposite of the placebo effect?**
The nocebo effect - when negative expectations about a treatment produce genuine adverse symptoms, even when the treatment itself is harmless. Recent research suggests nocebo effects may actually be stronger and more durable than placebo effects in some contexts.

**Why do clinical trials need a placebo group?**
Because the placebo response is large and reliable enough that without a comparison group, researchers can't tell whether an observed improvement came from the drug or from the context and expectation surrounding treatment.

## Closing thoughts

The placebo effect is one of the clearest demonstrations we have that the brain and body are not separate systems running in parallel - they're one integrated system, constantly trading information. Expectation, learning, and context all leave biochemical fingerprints: endorphins released to dampen pain, dopamine surging to ease motor symptoms, neural circuits adjusting their thresholds based on what we believe is coming next.

None of that licenses sloppy thinking about what placebos can do. They are not a substitute for treating the disease itself, and pretending otherwise does patients a disservice. But understood correctly, the placebo effect is a genuine, scientifically grounded reminder that context, care, and expectation are not soft, secondary variables in medicine. They're active ingredients in their own right - measurable, mechanistic, and worth taking seriously in how we design and deliver care.

![The human brain possesses an incredible, intrinsic capacity to heal the body.](https://psyll.com/assets/image/the-human-brain-possesses-an-incredible-intrinsic-capacity.png)

## Key takeaways:

* The placebo effect is a **neurobiological phenomenon** in which belief in a treatment produces genuine physiological or psychological improvement, even when the treatment itself is pharmacologically inert.
* It involves *measurable* changes in brain activity, neurotransmitter release, and bodily function - not merely a shift in mood or imagination.
* The **endogenous opioid system** is a central mechanism: expecting pain relief can trigger endorphin release, reducing pain perception in a way similar to actual analgesic drugs.
* The **dopamine system** plays a major role in conditions like Parkinson's disease, where placebo injections have been shown to release dopamine in the striatum and improve motor performance.
* Key brain regions implicated include the *prefrontal cortex*, *anterior insula*, *posterior insula*, *basal ganglia*, and *ventromedial prefrontal cortex*; a 2024 mouse study in *Nature* also identified a specific circuit - from the cingulate cortex through the pons to the cerebellum - involved in placebo pain relief.
* In a notable *Science Translational Medicine* study, a placebo achieved roughly **50% of the effectiveness** of an active analgesic when participants believed the drug had genuine pain-relieving properties.
* Approximately **1 in 3 people** experience a placebo response across a range of health conditions, including nausea, asthma, anxiety, and physical performance.
* Placebos do **not** cure underlying disease - they cannot shrink tumors, heal fractures, or clear infections; their effects are limited to symptom perception and subjective wellbeing.
* The **nocebo effect** is the placebo's negative counterpart: negative expectations about a treatment can produce genuine adverse symptoms, even from an inert substance.
* **Open-label placebo** research shows that patients can still benefit even when explicitly told the treatment contains no active medication, challenging the long-held assumption that deception is required.

## Sources:

* Nature Reviews Neuroscience: The neuroscience of placebo effects - [https://www.nature.com/articles/nrn3976](https://www.nature.com/articles/nrn3976)
* Nature: Neural circuit basis of placebo pain relief (UNC/Scherrer lab study) - [https://www.unc.edu/posts/2024/07/31/brain-circuits-offer-placebo-effect-pain-relief/](https://www.unc.edu/posts/2024/07/31/brain-circuits-offer-placebo-effect-pain-relief/)
* PAIN journal: Open-label placebo vs double-blind placebo for irritable bowel syndrome - [https://journals.lww.com/pain/abstract/2021/09000/open_label_placebo_vs_double_blind_placebo_for.13.aspx](https://journals.lww.com/pain/abstract/2021/09000/open_label_placebo_vs_double_blind_placebo_for.13.aspx)
* Science: Expectation and dopamine release - mechanism of the placebo effect in Parkinson's disease - [https://www.science.org/doi/10.1126/science.1060937](https://www.science.org/doi/10.1126/science.1060937)
* National Institutes of Health (PMC): Clinical neuroscience and neurobiology of placebo and nocebo effects - [https://pmc.ncbi.nlm.nih.gov/articles/PMC11593399/](https://pmc.ncbi.nlm.nih.gov/articles/PMC11593399/)

## Author

- **Author**: Laura J. Grays
- **Job title**: Senior Clinical Biopsychologist
- **Author profile**:  [https://psyll.com/laura](https://psyll.com/laura)
- **About author**: Laura J. Grays has spent her career mapping the intricate biological bridges between mind and body. Transitioning from molecular neuroscience research to clinical psychosomatic medicine, she investigates how chronic stress, cognitive aging, and psychological resilience interact at the cellular level to shape long-term health outcomes. She provides deeply grounded, evidence-based insights into mental well-being and longevity, deliberately steering away from wellness trends and toward the underlying biological mechanisms that determine how we age, how we recover, and how we heal.

## **License**

This article is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0).

You are free to copy, redistribute, and share this article in any medium or format, provided that:

- Attribution is given to the original author.
- A visible link to the original article is included: https://psyll.com/articles/science/psychology-neuroscience/placebo-effect-how-belief-changes-your-brain
- Any modifications are clearly indicated.

License: [https://creativecommons.org/licenses/by/4.0/](https://creativecommons.org/licenses/by/4.0/)