by Jo Marchant
Proponents say this phenomenon has the potential to slash drug doses for transplant patients like Karl-Heinz, as well as those suffering from allergies, autoimmune disorders and even cancer. But it’s far from mainstream medicine, and most immunologists barely acknowledge that it even exists.
—
IMAGINE TAKING a plump, yellow lemon from your fruit bowl. Its skin is smooth to touch, glossy and dotted with pores. Now put the lemon on a plate and cut it into quarters. Juice drips down the knife blade onto your fingers and the smell hits you: sharp and sour. You pick up one of the segments and notice how its flesh glistens, the light shining off hundreds of tiny liquid compartments, each full to bursting. Then you bite into it, sucking the rush of acid juice onto your tongue.
Did your mouth pucker as you read that paragraph? Did your salivary glands tingle into action, preparing your tongue for the imminent attack of acid? If so, you must have eaten a lemon before, and you have learned the appropriate physiological response. But here’s the crucial point. You no longer have to physically eat a lemon to experience these changes. Your body triggers them automatically in response to the sight, the smell—or just the thought—of a lemon, well before you actually taste the juice.
This form of learning, in which a mental cue drives a physical response, is called conditioning. It was famously discovered by a Russian physiologist named Ivan Pavlov in the 1890s. Pavlov was studying how dogs started salivating when he brought them food. Then he noticed that they started to salivate as soon as he entered the room, whether he was carrying food or not. The dogs had learned to associate his presence with being fed. After a while, they responded to him just as they did to their meat.
Pavlov showed that he could train the dogs to associate any stimulus—an electric shock, say, or a light or bell—with dinner. Once the association was learned, that signal on its own was enough to make the dogs drool. It’s a beautiful example of how the body doesn’t just react blindly to physical events and changes—lemon juice hitting our tongue, for example. It uses psychological cues to stay one step ahead.
Such anticipatory responses prepare us for important biological events such as eating or sex. Your tummy rumbles when you perceive signs—the clock, perhaps, or news headlines on the radio—that tell you it’s time for lunch. You get excited by the smell of a lover’s perfume, or the sound of their voice. (Psychologists have conditioned volunteers to become sexually aroused by neutral images from guns to penny jars, simply by pairing them with erotic film clips.) The memory of a song your mother used to sing to you at bedtime slows your heart rate and calms you down.
Other conditioned responses have evolved to protect us, preparing us to flee danger, or encouraging us to avoid it. If someone is bitten by a dog in childhood, the sight of a dog later in life may be enough to send their heart racing in fear (this is the basis of many phobias). If we eat a food that gives us a stomach upset, the mere thought or smell of that food may be enough to make us feel sick again. In some cases, even a particular place that we associate with sickness can trigger symptoms. This is why many people having chemotherapy get sick as soon as they arrive at the hospital, before their treatment session even starts.
This much is fairly well-known. Pavlov’s work on those salivating dogs is world famous. What’s less familiar to most scientists, let alone the rest of us, is that conditioning can also trigger placebo responses. If we swallow a pill that contains an active drug, we learn to associate that pill with a particular physiological change. Later, if we receive a look-alike placebo, we can experience the same change. It’s an automatic response in the body that happens regardless of whether we know the pill is fake. But it is triggered via conscious psychological cues—such effects don’t occur if we’re given a placebo while sedated, for example, or without knowing that we’ve taken it.
Placebo responses based on physiological conditioning often occur in addition to responses based on conscious expectation. For example, Benedetti told me that across his trials, the percentage of volunteers who respond to a placebo painkiller is extremely variable, anything from 0–100% depending on the circumstances. But if he first gives them a series of identical-looking injections containing an active drug, the proportion who subsequently respond to the placebo soars to a reliable 95–100%. “You can bet that virtually all patients will respond,” he says—even if they know that the final injection isn’t real.2
Could such responses be useful in medicine? We heard in chapter one how North Carolina pediatrician Adrian Sandler tested the hormone secretin as a treatment for autism, and found that it was no more effective than placebo. Yet he was struck by how dramatically the children in both groups improved, and he was unable to leave that revelation behind. Any drug that helped as much as the placebo had in his study would be leapt on as a potent treatment. Yet because this remedy involved the mind rather than a pharmaceutical, it was being ignored. In his spare time, Sandler started reading up on placebos, and wondered how he might be able to use them—without deceiving his patients.
The most prevalent diagnosis among the children he saw every day was attention deficit hyperactivity disorder (ADHD). As the name suggests, these kids were inattentive, hyperactive, impulsive. They were constantly talking and fidgeting, they were unable to wait their turn, and they found it impossible to focus in school. Medication helped them to control their symptoms but still caused problems, from irritable outbursts when the drug wore off in the evening to weight loss and stunted growth. “It becomes a balancing act in the clinic,” he says, “trying to find [a dose] that is giving sufficient benefit without giving excessive side effects.”3
Sandler wondered whether a placebo might help these children to manage their symptoms on a lower dose of drug. He decided to give his placebos honestly, as part of a regime that would hopefully harness the power of both expectation and conditioning. Seventy ADHD patients, aged six to twelve, completed his two-month trial.
These children were split randomly into three groups. One group underwent a conditioning regime. For one month they received their normal medication, but also swallowed a distinctive green-and-white capsule alongside their drug—they knew this was inert, but Sandler hoped that they would learn to associate it with the physiological response to their active medication. For the second month they received half their usual drug dose, as well as the placebo capsule.
Sandler compared these patients against two control groups, neither of which received any conditioning. One group received their full dose of medication for the first month and a half dose for the second month—just like the conditioning group. The last group got a full dose all the way through.
Sandler published his results in 2010. As expected, in the half-dose control group, the children’s symptoms got significantly worse in the second month of the trial. But the conditioned group remained stable, doing just as well as the full-dose patients. In fact, there were hints that kids in this group did even better, suffering fewer side effects than those on the full dose of the drug.4
It is the first and only trial in which honest placebos have been given to children. Sandler says that parents and kids alike embraced the idea, and that more than half of them wanted to keep taking the placebo once the study was over. “It is the best medicine I’ve had,” one child told him afterwards. “I think it tricked the brain into thinking it would work.” Sandler’s study is small and preliminary, but combined with Benedetti’s findings, it hints that doctors could use simple conditioning procedures to boost the effectiveness of placebos, without any deception required.
For me, that’s an exciting finding. By using expectation and conditioning together, ethical placebos could potentially help to reduce drug doses for millions of patients around the world, in conditions from pain and depression to Parkinson’s and ADHD.
There’s something else about conditioned responses, however, that opens up an entirely new landscape of possibility. These learned, unconscious associations aren’t limited to the subjective
symptoms—like the distractibility of those with ADHD—that are shaped by conventional placebo effects. They can also influence the immune system, providing a route by which the mind can become a weapon in the body’s fight against disease. The mind, in other words, can do much more than help us to feel and perform better. Through conditioning, it might make the difference between life and death.
Scientists denied this was possible until just a few decades ago. Then they were forced to rethink their ideas by two accidental discoveries and a brave teenager named Marette.
—
IN 1975, a psychologist named Bob Ader at the University of Rochester in New York was investigating the phenomenon of taste aversion, in which we feel nauseated by a food that has made us sick in the past. He wanted to know how long such learned associations last, so he took a group of rats and fed them several doses of water sweetened with saccharin. This would usually be a treat, but in this experiment he paired the water with injections that made the animals feel sick. Later, Ader gave the rats the sweetened water on its own. Just as he expected, they associated the sugary taste with feeling ill and refused to drink it.
So Ader force-fed it to them using an eyedropper, to see how long it would take for them to forget the negative association. The experiment should have been fairly routine, but what actually happened to the rats seemed like black magic. All Ader gave them during this stage of the experiment was sweetened water, containing no drug whatsoever. But they didn’t stop feeling sick. Instead, one by one, they died.5
To work out what had killed them, Ader looked more carefully at the chemical he had used to make the rats feel sick in the first place. It was a drug called cytoxan, which, as well as causing stomach pain, suppresses the immune system. The dose Ader had used in his experiment was too low to be fatal, so he came to a radical conclusion. When he conditioned the rats, they didn’t just learn to feel sick. The extra “doses” of sweetened water also suppressed their immune systems to the point where they succumbed to fatal infections. It was a stunning finding, suggesting that conditioning reaches far beyond known responses such as salivation, heart rate and blood flow. Our immune systems are vulnerable too.
At the time, this was seen as tantamount to pseudoscience by the immunology establishment. “The immune system and nervous system were considered to be completely independent systems,” says Manfred Schedlowski, a medical psychologist at the University of Essen in Germany.6 “Immunologists thought [Ader’s finding] was crazy.” Biologists were convinced that the immune system worked alone, responding to foreign invaders or damage without any help from the brain. Ader died in 2011, but according to his daughter, Deborah, he attributed his insight to the fact that as a psychologist rather than an immunologist, he hadn’t been schooled in this dogma. “I just didn’t know any better,” he would say. “I didn’t know that the immune system wasn’t supposed to be connected to the brain.”7
So although Ader’s finding was striking, it wasn’t accepted at first. His main problem was that back in the 1970s, he couldn’t explain how conditioning of the immune system could possibly work. He was up against generations of immunologists who were convinced that the brain and the immune system don’t communicate. They weren’t about to change their minds without direct proof of a physical link between the two.
A few years later, they got that proof. David Felten, a neuroscientist working at the Indiana University School of Medicine, was using a powerful microscope to look at body tissues from dissected mice, in order to track where different nerves traveled in the body. In particular, he was interested in the network of the autonomic nervous system, which controls body functions such as heart rate, blood pressure and digestion. Our nerves are divided into the central nervous system, which comprises the brain and spinal cord, and the peripheral nervous system, which runs throughout the body. The peripheral nervous system is in turn divided into two branches. One, the somatic nervous system, deals with conscious messages—it carries our instructions to the muscles so that we can move around and relays sensations such as warmth and pain back to the brain. The second, the autonomic nervous system, controls those physiological systems not usually thought to be under conscious control.
When Felten followed the different branches of the autonomic nervous system, he saw it connecting up with the animals’ blood vessels, just as he expected. But then he saw something that seemed totally wrong—nerves running right into the heart of immune organs such as the spleen and thymus (where the body’s white blood cells develop and are stored). As he later told a reporter for PBS: “We saw nerve fibers all over the place, sitting right smack in the middle of some of these cells of the immune system.”8
He checked and rechecked his results, making sure that his tissue slices were identified correctly. “I was almost afraid to say anything. I was worried that we had missed something, and we’d look like a bunch of dufuses.” But there was no escaping the fact that nerves were directly connecting with cells of the immune system. It was incontrovertible evidence of a hard-wired connection between the immune system and the brain.
Felten recalls that when he first published his results in 1981,9 he was laughed at. But he received encouragement from Jonas Salk, the great U.S. virologist who developed the vaccine that eradicated polio in the 1950s. Felten was so touched by Salk’s words that he learned them by heart: “This research area could turn out to be one of the truly great areas of biology in medicine,” said Salk. “You’ll meet some opposition. Continue to swim upstream.”10
Felten started to collaborate with Ader, as well as Ader’s colleague Nicholas Cohen, and shortly afterwards he moved to join them at the University of Rochester. These three researchers are now broadly credited with founding a field of research known as psychoneuroimmunology. They championed the idea that in protecting us from illness, the brain and the immune system work together.
Felten’s group went on to discover a complex web of connections. As well as hard-wired nerve connections, they found receptors for neurotransmitters—messenger molecules produced by the brain—on the surface of immune cells, as well as new neurotransmitters that could talk to those cells. And they found that the lines of communication went in both directions. Psychological factors such as stress can trigger the release of neurotransmitters that influence immune responses, while chemicals released by the immune system can in turn influence the brain, for example triggering the drowsiness, fever and depressive symptoms that confine us to bed when we are ill.
Meanwhile Ader continued working on conditioned immune responses. The idea of Pavlovian conditioning had soaked into popular culture, but it was usually portrayed as a dubious means for authorities to exert mind control over the masses. In Aldous Huxley’s novel Brave New World (1932), toddlers destined for factory work are conditioned to avoid books and flowers using shrill noises and mild electric shocks, while in Anthony Burgess’s A Clockwork Orange (1962), the protagonist is fed a drug to make him nauseous then forced to watch footage of violent acts. Ader wanted to know if conditioning could instead be harnessed to fight disease.
—
MARETTE FLIES was a cheerful high-school student from Minneapolis, Minnesota. She had a mop of dark, curly hair and a pale, moon-shaped face, and she loved playing the trumpet.
Then, in 1983, when she was 11, she was diagnosed with a life-threatening condition called lupus erythematosus. It’s an autoimmune disease, in which the immune system mistakenly attacks the body’s own cells. Some autoimmune conditions target a specific organ or cell type: rheumatoid arthritis eats away at the joints, for example, while diabetes kills the cells in the pancreas that make insulin. But with lupus, the immune system wages war on the whole body—the joints, skin, and in severe cases, the heart, kidney, lungs and brain.
Marette was initially treated with steroids to suppress her rampant immune system. She hated taking them—they made her face look like she had “swallowed a blimp,” she complained,11 and her hair fell out. She’d wake up in the morning wi
th hair covering her pillow. Then she’d eat breakfast, and more hair would fall in her food.
Despite the drug treatment, Marette’s condition deteriorated rapidly over the next two years. At first she could still play the trumpet (against her doctors’ advice) but then she developed kidney damage, seizures, high blood pressure and bouts of pneumonia. Her immune system also destroyed a vital clotting agent in her blood, causing episodes of severe bleeding. Her condition was so serious that her doctors were considering giving her a hysterectomy, because they feared that when she started to menstruate, she might die from blood loss. Then, in September 1985, her heart began to fail.
With Marette’s life in imminent danger, her doctors decided that they had no choice but to put her on a much more powerful immunosuppressant drug. It was cytoxan, the same drug that Ader had used in his experiments on rats. Its use in humans at the time was experimental, and it is highly toxic. The long list of side effects includes vomiting, stomachaches, severe bruising, bleeding, and kidney and liver damage, as well as life-threatening infections and cancer. Cytoxan was Marette’s only chance of surviving her lupus, but it was almost as dangerous as the condition itself.
Karen Olness, a pediatrician now at Case Western Reserve University in Ohio, was one of the doctors caring for Marette at the time, using biofeedback and hypnosis to help the teenager cope with the stress and pain of her condition. She had become fond of Marette and was struggling to resign herself to the fact that her patient was not likely to survive this latest crisis. Then Marette’s mother, who was a psychologist, showed Olness a copy of one of Ader’s papers, which had been published in 1982.12