Individual expectations of rewards may explain why some people feel better after receiving fake drug treatments—a phenomenon known as "the placebo effect."

A new study using different brain imaging techniques linked the intensity of an individual's placebo effect to the amount of dopamine (a neurotransmitter involved in the pleasure and reward pathway) released in a midbrain region called the nucleus accumbens. Researchers at the University of Michigan at Ann Arbor specifically demonstrated that those who were more responsive to phony pills were also more likely to expect to win big in a gambling game.

"If you have the capacity to respond to reward, then you have the placebo effect," says neuroscientist and radiologist Jon-Kar Zubieta, senior author of the new study published this week in Neuron.

Zubieta and colleagues initially worked with 14 healthy volunteers who were told they would receive painful injections of a saline (benign saltwater) solution in their jawbones. That shot would be followed by a second one, which subjects were led to believe would either be a painkiller or a placebo. In actuality, everyone received a placebo. Participants continually self-reported (using a numerical scale) the degree of pain, and researchers monitored their brains with positron emission tomography (PET) scans.

The scans showed that half of the participants who believed they were receiving painkillers reported feeling significantly less pain than did other volunteers. Their dopamine levels were noticeably higher (than that of the others) from the moment they were told they were receiving authentic painkillers.

On a different day (so the original participants would think they were involved in two distinct studies), researchers had the 14 subjects, along with 16 new volunteers, play a game of chance; they scanned participants' brains via functional magnetic resonance imaging (fMRI) while they played. Participants were told they could win or lose a certain amount of money each round; they would then push a button to determine the real take. Several of the participants showed a flurry of activity involving dopamine release in the nucleus accumbens while awaiting the outcome, indicating that they were expecting a reward.

The people whose nucleus accumbens lit up during the game also reported greater relief from the sham painkillers. "What surprised me the most was the strong link between this element of reward processing and the fact that you can predict the placebo response," Zubieta says. "The placebo effect is a resiliency mechanism in the brain. … You don't [really] need the medication, you simply need to be convinced that something is going to work."

Tor Wager, a psychologist at Columbia University, says this work could be helpful in developing new therapies that manage the expectations of patients. "This study opens up a new avenue of studies for brain-based interventions," he says, "such as targeting the dopamine system" to increase the placebo effect.