What is the nocebo effect?

What is the nocebo effect?

The nocebo effect is what happens when expecting harm produces harm: new symptoms, or more intense symptoms, that follow from the belief that something will hurt or make you ill. The word comes from Latin, 'I shall harm', coined as the counterpart of placebo, 'I shall please'.

The two effects are twins. In the placebo effect, an inert pill, a confident clinician or a ritual of care produces real relief because the brain expects relief. In the nocebo effect, a warning, a frightening word or a bad past experience produces real pain because the brain expects pain. Neither requires deception, and neither requires the person to be gullible. They happen to physicians, to skeptics and to people who know exactly what is going on.

This is the point that gets lost. A placebo response is often described as 'nothing happening', and a nocebo response as 'imagining it'. Brain imaging studies reviewed by Benedetti and colleagues show otherwise: expectation changes activity in the circuits that produce pain, and in placebo analgesia it recruits the brain's own opioid system. The pain a person feels after a nocebo is produced by the same machinery as any other pain. It is real pain with an unusual trigger.

What are some examples of the nocebo effect?

The clearest examples come from medicine itself, because medicine is full of warnings. Colloca and Barsky's review in the New England Journal of Medicine collects many of them.

• Side effects from the leaflet. People taking an inert pill in a clinical trial report side effects, and the ones they report tend to match the ones they were warned about. A share of the side effects attributed to active drugs appears in the placebo arm as well.
• 'This will hurt.' Patients told a needle will sting or burn report more pain than patients given neutral or reassuring wording for the same procedure.
• The imaging report. Words such as 'degeneration', 'bulging' and 'wear and tear' describe findings that are common in people with no pain at all, yet they land as a verdict. Many clinicians now argue that the report itself can become a nocebo.
• Watching someone else. Observing another person react badly to a treatment increases the chance that you will too. Expectation can be caught socially.
• Prior experience. If a movement hurt last time, the brain predicts that it will hurt this time, and the prediction shapes what is felt before the movement is complete.

Notice that every example involves information: a word, an image, a memory. None involves new damage. That is the signature of the nocebo effect, and it is also what makes it treatable. Information can be changed.

How does expectation change pain in the brain?

The short answer is that pain is built partly from prediction, and expectation is the brain's prediction about what the body is about to feel. A useful way to hold this is to think of expectation as a forecast. Most forecasts only describe the weather. In the body, the forecast helps make it.

Prediction and precision

On the predictive-processing account of the brain, which is a framework with growing but incomplete evidence rather than settled fact, the brain continually guesses what signals it will receive and compares the guess with what arrives. The gap is prediction error. How much weight the brain gives its own guess versus the incoming signal is called precision. A strong, confident expectation of pain is a high-precision prediction, and the brain leans on it even when the body sends mild or ambiguous signals. Our guide to predictive coding walks through this in detail.

The unpleasantness is built separately

Pain has at least two dimensions: where and how intense it is, and how unpleasant and threatening it feels. Howard Fields's lab produced direct evidence that these can come apart. In rodents, lesions of the anterior cingulate cortex removed the learned aversion to a painful stimulus while leaving the basic pain reflexes intact. The animals still registered the stimulus, but it stopped teaching them to avoid the place where it happened. The anterior cingulate is also one of the regions most consistently changed by expectation in human imaging studies, along with the insula and prefrontal cortex.

This matters for nocebo because expectation acts most strongly on the second dimension. Atlas and Wager's review of expectancy and pain describes how cues that predict pain, even cues that were neutral a few minutes earlier, come to increase reported pain and the brain's response to it. A tone, a color, a room or a word can become a pain-predictive cue, and from then on it changes the forecast.

There is also chemistry. Research suggests nocebo hyperalgesia involves anticipatory anxiety and the signaling molecule cholecystokinin, while placebo analgesia recruits the brain's own opioids. Blocking one pathway can blunt one effect without touching the other. The two effects are twins, but they are not the same event run backwards.

Why does this matter in chronic pain?

Because in chronic pain the nocebo is not a single warning before a single injection. It is a daily nocebo, assembled from many sources over years, and each source makes the next flare more likely.

• A scan report full of words that sound like damage.
• A clinician who said 'you have the spine of an eighty-year-old' or 'be careful with that back'.
• A knee described as 'bone on bone'.
• Family who flinch when you bend.
• The memory of the last time this movement hurt, which is also a forecast of the next time.

Each of these is information about the body, and the brain updates its predictions accordingly. Over time, the expectation of harm stops being a guess and becomes a stubborn prior: a default assumption that this body is fragile and this movement is dangerous. Priors that have been confirmed many times are hard to shift, and chronic pain supplies confirmation daily. Movement hurts, which proves it was dangerous, which sharpens the forecast for next time.

This loop is what the fear-avoidance literature describes from the outside. The appraisal that a sensation means damage is what the Pain Catastrophizing Scale measures. The avoidance that follows is kinesiophobia. Seen through the lens of expectation, both are consequences of a forecast that has become too confident to be corrected by ordinary experience.

None of this means the tissue is irrelevant. Arthritis, nerve injury and disease are real inputs, and the brain's forecast is often built on a real event. The claim is narrower: once the forecast is established, it contributes to the pain in its own right, and reducing that contribution is a legitimate target even when the original cause cannot be undone.

How does expectation shape the sense of self in pain?

Expectation does not stop at the body part. The brain uses the same predictive machinery to build its model of the whole body and, on current theories of self, the sense of being a particular person living in it. What you repeatedly expect of your body becomes part of who you take yourself to be.

A person who expects to be fragile learns to move as a fragile person, plan as a fragile person and introduce themselves as one. 'I have a bad back' becomes a fact about the self rather than a report about a body part. Research on the brain's resting networks, which our guide on whether the brain feels pain touches on, suggests that self-referential thought and rumination share circuitry, which is one reason a story about a broken body is so sticky. It is rehearsed every time the mind wanders.

This is the thread that runs through the whole therapeutic neuroscience series: our sense of self is constructed, and chronic pain builds a different self. The nocebo effect is one of the builders. Each confirmed expectation of harm adds a brick to a self organized around threat and guarding.

The same logic cuts the other way. A self that was learned can be relearned. When a person repeatedly experiences a feared movement going well, the prediction about that movement updates, and over many repetitions so does the prediction about the person who performs it. Nobody should promise that this happens quickly or for everyone. The point is that the fragile self is a model, and models can be revised when the evidence changes.

Can you use expectation on purpose?

To a degree, yes, and researchers have been trying for two decades. The evidence is strongest for three approaches, and none of them is 'think positive'.

Education that changes what a sensation means

If pain is partly forecast, then changing the forecast changes the pain. Pain neuroscience education tries to do this by giving people an accurate account of how pain is produced, so that a twinge on bending reads as a sensitive system rather than a tearing disc. The effect sizes are modest on their own and larger when education is paired with movement; we review the evidence in does pain neuroscience education work?.

Careful language from clinicians

Because words trigger nocebo effects, words can also prevent them. Describing a procedure in neutral terms, translating an imaging report into what the findings actually mean for function, and avoiding phrases such as 'wear and tear' are low-cost changes with a plausible mechanism. Colloca and Barsky treat clinician communication as one of the main levers on both placebo and nocebo responses.

Open-label placebo

The strangest finding in this field is that placebos may work when people know they are placebos. In a randomized trial of 80 people with irritable bowel syndrome, Kaptchuk and colleagues told participants openly that the pills were inert, explained that placebo responses are real, and found greater global improvement over three weeks than in a no-treatment group. The trial was small, short and unblinded by design, the condition was IBS rather than musculoskeletal pain, and replications have been mixed. It is best read as proof that expectation can be engaged without deception, not as a treatment anyone should rely on.

What does this mean if you have chronic pain?

First, that your pain is real, whatever has contributed to it. Second, that some of what contributes may be information you were given, and information can be revisited.

If you are the person in pain

• Ask what the words on your imaging report mean for what you can do, rather than how they sound. Many findings are common in pain-free people of the same age.
• Notice the forecasts. 'This is going to hurt' before a movement is a prediction, and predictions can be tested in small, safe doses.
• Track function alongside pain. A chronic pain scale tells you intensity; what you can do this week compared with last week tells you whether the forecast is changing.
• Be wary of anyone who tells you the fix is attitude. Expectation changes through experience, with support, over weeks.

If you are a clinician

Your words are treatment. The same findings can be delivered as a verdict or as context, and the evidence on nocebo suggests the delivery changes the outcome. Explaining what a sensation means before a patient has to interpret it alone is one of the cheapest interventions in pain medicine.

How Karuna uses this

Karuna's 12-week program is built on therapeutic neuroscience and treats expectation as a target. Weekly one-on-one sessions with a pain coach teach how pain is produced, so that sensations can be reinterpreted. The VR embodiment exercises are designed to create safe prediction errors around movement: the brain sees the body moving comfortably, repeatedly, and the forecast of harm is given evidence against it. A clinician reviews every case before it starts and provides oversight throughout. Many people improve substantially and others do not; the measured outcomes, and the limits of that data, are on how it works, and the free consultation exists to say whether the program fits.

Expectation-based approaches assume that serious disease has been ruled out. See a doctor promptly for new weakness or numbness, loss of bowel or bladder control, unexplained weight loss, fever with pain, pain after a major injury, or chest pain. If you are having thoughts of harming yourself, call or text 988 in the US.