Maybe you have comforted a crying child by kissing her scraped knee and seen her tears turn to a smile. Perhaps you've stumbled half-asleep to the medicine cabinet, taken a pill for the headache that woke you, felt better — and the next morning discovered your relief came from a calcium tablet.

Or maybe you took your arthritic knee to a hospital where you had a completely fake procedure in which surgeons made incisions but didn't actually remove the cartilage whose deterioration causes osteoarthritis — after which you had less pain and began walking better than you had in years.

OK, you probably never experienced the last one. But scores of patients did. In a groundbreaking study, patients with osteoarthritis of the knee merely thought they had received arthroscopic surgery and their knee pain diminished as much as in patients who actually received the $5,000 procedure.

It is tempting to say that "mere thought" or "mere belief" caused these patients to feel and function better, just as the child's trust in her mother made her knee feel better and our belief in little white pills made the calcium tablet relieve a headache.

But doctors and scientists have learned there is nothing "mere" about how thoughts, beliefs and the power of the mind affect the body.

"What we believe and expect can significantly influence the outcome of a disease, how much pain we feel, even whether Parkinson's symptoms diminish," says neuroscientist Mario Beauregard of the University of Montreal, whose new book "Brain Wars" examines the basis for the placebo response in the brain.

The first rigorous experiment to look for the mechanism underlying the placebo response was reported in 1978; Dr. Jon Levine and colleagues at the University of California, San Francisco, found that patients given a placebo and naloxone, a drug that blocks the brain's natural morphine-like compounds, experienced more pain than patients given only a placebo.

"This study was the first to show that placebos relieve pain by activating the brain's natural painkillers," says Beauregard. "But it did more: It brought the placebo response in from the fringes of medical science."

It also paved the way for a flood of placebo studies. Among the many findings is that people expect more from expensive remedies than from cheap ones. For example, in a 2008 study, patients who thought they were getting a $2.50 tablet felt less pain from mild electrical shocks on their wrists than those told the pill cost 10 cents.

Expectations may seem like will-o'-the-wisps, but they have a real, physical signature in the brain. When we expect something good to happen, our brains release dopamine. In Parkinson's disease, production of dopamine in structures called the substantiae nigrae declines. Scientists led by neurologist Jon Stoessl of the University of British Columbia gave a small group of Parkinson's patients injections of a saline solution but told them it was medication. Sure enough, the patients' brains began producing more dopamine, and their movements became better controlled and less shaky — improvements equal to those in patients who received an actual Parkinson's drug. In Parkinson's disease, "the placebo effect is real, it's huge, and it's got a physiological basis," Stoessl told the journal Nature.

Natural painkillers and dopamine activation in the brain explain many placebo responses, but not all.

Pain and pooches

There is a third way to invoke a placebo response, and it has to do with some famous dogs. About a century ago, Russian physiologist Ivan Pavlov rang a bell over and over when giving his dogs food. Eventually, he rang the bell but did not feed them; the dogs salivated anyway. Might an association be formed between a signal and pain?

In one experiment demonstrating this phenomenon, scientists showed 40 volunteers two male faces on a computer screen. When the volunteers looked at one face, they got a mild burn on their forearm; when they looked at the other, they got a more painful burn. The volunteers became as conditioned as Pavlov's dogs.

In the next round when they saw the high-pain face and felt a burn, they rated it as more painful than when they saw the low-pain face and felt a burn — even though the applied heat was identical the second time around. The perception of pain, says Ted Kaptchuk of Harvard Medical School, who helped lead the 2012 study, depends on "what the nonconscious mind antici­pates despite any conscious thoughts."

For all the progress scientists have made in understanding the placebo response, mysteries remain. No one knows whether certain personality types, ages, genders or nationalities are more susceptible to the placebo response, though Beauregard notes that placebos don't work as well in skeptical people as they do in trusting souls.

Today, more doctors are applying the placebo effect by recommending relaxation therapy, meditation and other mind-based practices. Some of the benefit from seeing a doctor and receiving a treatment comes from our belief in that doctor and our expectation that the treatment will work. Will we have the same expectations of an online visit, explaining our symptoms impersonally over the Internet? If caring from a human being — remember Mom making the skinned knee feel better — triggers the placebo effect, can a computer?

"The healer is a crucial component of the placebo response," says Beauregard. "It would be a shame if we lost that" as technology advances.