The latest additions to the placebo effect family might be the rudest. First there was placebo, which uses your body's own tools to make you feel better after you try a treatment you imagine will help you. Then there was nocebo, placebo's evil twin: it makes you feel worse only because you think you will. Now researchers have discovered that placebo and nocebo effects can be triggered subliminally, which is like finding out that the good and evil twins have both been living in your basement without you knowing it.

Usually, placebo and nocebo look like cases of our own expectations manipulating us. Someone swallows his favorite headache remedy or visits a doctor, and his body, expecting to feel better, ramps up production of its own pain-relief molecules. Someone else steps onboard a plane and begins to feel nauseous, simply because her body has learned that airplanes mean queasiness. If we were more ignorant of our circumstances, the effects wouldn't be there.

But there seem to be some cues we can take in subliminally, without noticing them. So researchers led by Karin Jensen at Harvard Medical School wondered whether visual signals that are too brief to reach our consciousness—but perhaps not too brief for certain areas of our brains to snag as they pass—can trigger placebo and nocebo effects too.

For their visual signals, the team chose photos of male faces. "We know from previous studies that faces can be detected and processed very quickly in the brain," Jensen says. Their models came from a set of photos created for use in psychology experiments.

The researchers carried out two experiments, the first of which was a classic test of placebo and nocebo. Subjects were shown pictures of two expressionless male faces over and over. Each time they saw face A, they felt a painfully hot sensation on the forearm. Face B was paired with heat that was milder, but still uncomfortable. (The A and B models alternated between different subjects—just in case one man's face really was more painful to look at.) During the conditioning part of the experiment, subjects saw each face 25 times. This taught them to expect higher pain with face A and lower pain with face B.

Then came a second series where subjects saw the same two faces as before, with a few new ones mixed in as controls. With each face they saw, subjects rated the pain they felt from the heat instrument on a 100-point scale. The twist was that in this part of the experiment, the heat level was exactly the same every time. But subjects consistently reported high pain for face A and low pain for face B. When they saw a new face, subjects reported an intermediate level of pain (which corresponded to what they were actually feeling).

This first experiment showed the researchers that pairing faces with painful heat stimuli could create both a placebo effect (when subjects rated moderate heat as less painful because they saw face B) and a nocebo effect (when subjects found moderate heat more painful, thanks to face A). So they moved on to the second experiment. In this round, the visual signals would be "nonconscious,"or subliminal.

A new group of subjects went through the same conditioning sequence as before. Then they were given a testing sequence using face A, face B, and the new (control) faces, all paired with the same moderate heat on the arm. But the faces in this sequence flashed on the screen for just 12 milliseconds, compared to 100 milliseconds in the earlier experiment.

12 milliseconds is fast. Too fast, in fact, for subjects to consciously process the faces zipping by. They reported that they couldn't tell who was who (and a separate experiment confirmed that people can't recognize faces shown this quickly).

But, as the researchers report this week in PNAS, the pain scores still matched the faces subjects said they couldn't see. Face A got significantly higher pain scores than face B, with the control faces scoring in the middle—and don't forget that, once again, subjects were actually feeling the same degree of heat every time.

Even though the pictures flashed too briefly to enter conscious awareness, they seem to have snuck in through the brain's back door. These visual cues made subjects experience more or less pain than they should have, even though they had no idea what they'd seen.

There were only 20 subjects in each experiment; it would take further studies to show how consistent or how powerful the subliminal placebo and nocebo effects are. But the fact that they found an effect at all is exciting news to the researchers. "To the best of my knowledge, there has not been an experiment [previously] where placebo/nocebo effects have been activated by nonconscious cues," Jensen says.

The common assumption, Jensen says, is that placebo and nocebo rely on the signals we're paying attention to (pills, needles, drug commercials) and the results we expect (relief, discomfort, alarming side effects). But this study "proves that we don't need to be aware of the cue to elicit a conditioned response," Jensen says.

Don't expect to start seeing mysterious images flashing at you in the doctor's office. The subjects in Jensen's study had to be trained to associate photos of faces with high or low pain. And even if there were another kind of image that automatically produced a placebo effect in a wide audience (teddy bears? puppies?), our brains might not be able to recognize it as quickly as a human face.

But the idea that placebo and nocebo effects can be triggered by cues patients don't even notice could be important for healthcare, Jensen says. Certain conditions such as asthma, depression, and irritable bowel syndrome are known to respond well to placebos. Maybe doctors' offices and hospitals in the future will tailor everything patients see—from the posters on the wall to the instruments on the counter to the fish swimming in the lobby aquarium—to encourage placebo and avoid nocebo. Or maybe we'll be able to use the same tricks at home to keep ourselves feeling our best. Let's kick those weird placebo relatives out of the basement and put them to work.

Karin B. Jensen, Ted J. Kaptchuk, Irving Kirsch, Jacqueline Raicek, Kara M. Lindstrom, Chantal Berna, Randy L. Gollub, Martin Ingvar, & and Jian Kong (2012). Nonconscious activation of placebo and nocebo pain responses PNAS : 10.1073/pnas.1202056109

Image: freya.gefn/Flickr