The Placebo Effect: How Belief Changes Health

The placebo effect is a real improvement caused by expectation, not an active treatment. Learn how it works in the brain, where it helps, and what it cannot do.

Category: Neuroscience · Created: 2026-10-04 · Updated: 2026-10-04 · 4 min read

The placebo effect is a measurable improvement in symptoms that follows a treatment with no active ingredient, such as a sugar pill or a saline injection. It is caused by the person's expectations and the context of care rather than by the treatment's chemistry. The effect is real, but it is narrower than popular accounts suggest. It mainly changes symptoms that the brain helps to generate or regulate, such as pain, nausea, and fatigue. It does not shrink tumours or cure infections.

What is a placebo?

The word is Latin for "I shall please". In medical research a placebo is the control in a clinical trial: one group receives the real treatment, another receives an inert lookalike, and comparing the groups shows what the drug does beyond natural recovery and expectation. In a double-blind trial neither the participants nor the researchers know who is receiving which, so that neither can bias the outcome. Placebos matter because people often improve for reasons that have nothing to do with the treatment.

In 1955 the anaesthetist Henry Beecher published "The Powerful Placebo," claiming that about a third of patients responded to placebos. Later analyses showed that his figure mixed in other effects, including natural recovery and regression to the mean: people tend to seek treatment when their symptoms are at their worst, so they would often improve anyway. A 2010 Cochrane review by Asbjørn Hróbjartsson and Peter Gøtzsche found little effect of placebo on objective outcomes, but some effect on subjective ones such as pain and nausea. Telling real effects from coincidence is the problem described in correlation and causation, and it is why trials rely on hypothesis testing.

How does the placebo effect work in the brain?

Expecting relief activates real neural pathways. In a classic 1978 experiment by Jon Levine and colleagues, dental patients who got relief from a placebo lost it when they were given naloxone, a drug that blocks opioid receptors, which suggested that the relief depended on the body's own opioids, the endorphins. Brain scans show reduced activity in pain-processing regions and activation of pathways that damp pain signals on their way up the spinal cord. In Parkinson's disease, a placebo can trigger the release of dopamine in the brain, as shown in a 2001 study using imaging.

Learning plays a part as well. If a pill, a clinic, or a ritual has been followed by relief many times, the association alone can produce a response, a process like classical conditioning as in Pavlov's dogs. The chemical messengers involved are the same neurotransmitters that operate at every synapse.

What makes a placebo stronger?

Price and branding: in a 2008 experiment by Rebecca Waber and colleagues, a placebo painkiller described as costing $2.50 per pill relieved pain more than the same pill described as costing 10 cents. The price acts as an expectation, a kind of anchoring.
Dramatic procedures: injections tend to work better than pills, and large pills better than small ones. In a 2002 trial of arthroscopic surgery for knee osteoarthritis, patients who received only a sham operation with small incisions did as well as those who had the full procedure.
The relationship with the clinician: in a 2008 trial on irritable bowel syndrome led by Ted Kaptchuk, patients received sham acupuncture, and those treated by a warm, attentive practitioner improved more than those who got a brief, businesslike visit.

Open-label placebos and the nocebo effect

Remarkably, a placebo can sometimes work even when the patient knows it is a placebo. In a small 2010 trial on irritable bowel syndrome, patients given honestly labelled placebo pills reported more relief than patients given no treatment. The results are promising but come from small studies, and researchers are still testing them.

The opposite also exists. The nocebo effect occurs when negative expectations cause real symptoms. In many drug trials a sizeable share of the people taking the placebo report side effects such as headaches and fatigue, and patients who are warned about a side effect are more likely to report it. This is one reason why doctors think carefully about how they describe risks.

What the placebo effect cannot do

Placebos do not treat underlying disease. There is no good evidence that they shrink tumours, clear infections, or change objective measures such as blood tests. That is why placebo-controlled trials remain the standard for testing treatments, and why an experience of getting better after a remedy is weak evidence that the remedy worked, a point central to the scientific method.

The effect has a useful lesson, though. Part of what we call treatment is the care that surrounds it: attention, explanation, ritual, and trust. A treatment that works can work even better when delivered well, and a good clinician makes use of expectation honestly, without deception, to help the biology do its work.