In the early 1960s, Kandel decided to conduct classic Pavlovian conditioning studies on aplysia, or sea slugs, which have relatively few neurons. More important, aplysia possess what Kandel has described as the “largest nerve cells in the animal kingdom. You can see them with your naked eye.” That made them easy to manipulate in a laboratory. Kandel removed neurons and placed them in a petri dish. By stimulating the neurons with an electrode, he was able to map the entire neural circuit required to cause a common reflex. (The reflex he chose forces the slug’s gills to retract when they are disturbed, in much the same way that a threatened porcupine will raise its quills.)
Scientists were already aware that making a memory requires chemical activity in the brain. But neurons are programmed by our DNA, and they rarely change. On the other hand, synapses, the small gaps between neurons, turn out to be highly mutable. Synaptic networks grow as we learn, often sprouting entirely new branches, based on the way that chemical messengers called neurotransmitters pass between neurons. “The growth and maintenance of new synaptic terminals makes memory persist,” Kandel wrote in his book In Search of Memory: The Emergence of a New Science of Mind (2006). “Thus, if you remember anything of this book, it will be because your brain is slightly different after you have finished reading it.”
Nader was thrilled by the idea that one could watch an organism form a memory. “I was not trained as a neuroscientist in memory or in consolidation,” he told me recently on the phone from McGill University, where he is now a professor of psychology. “Kandel talked about the physiology of the neuron on the most basic level, and I was amazed. But I didn’t understand why a thing like that—the complete chemical production required to form a memory—would happen just once. I looked at the data and thought, What makes us so certain that after our memories are formed, they are fixed forever?”
The prospect that a memory might be altered simply by being recalled was heretical; LeDoux urged Nader not to waste his time. But he was determined, and LeDoux didn’t interfere. Early in 1999, Nader and his colleagues devised an experiment in which they trained a group of rats to fear a tone. Conditioning, for rats and most species, including ours, is relatively straightforward: a researcher will pair a neutral stimulus like a tone or a color with something unpleasant, usually a shock. The results are quick and definitive; replay the tone, even without the shock, and the rat will freeze in place, crouching as low as it can. Its fur will stand on end, and its blood pressure will soar. The next time the rat (or human) hears the tone, the electrical circuitry in its brain responds as powerfully as if it were also experiencing the shock, and the synapses associated with that memory will grow stronger.
After teaching the rats to fear the tone, Nader waited 24 hours, to give their memories time to consolidate. Then he played the tone again and injected the antibiotic anisomycin into the rats’ lateral amygdala, the area that houses fearful emotions. Anisomycin has been shown to prevent neurons from producing the proteins necessary to store a memory. If memories are formed just once, Nader reasoned, the drug should have no effect. “The idea,” he said, “was that if a new set of proteins was required, then the drug should prevent the memory from being recalled.” That is exactly what happened. Rats that received the drug within four hours of recalling the memory forgot their fear. Two weeks later, when Nader again tested the rats, those with blocked memories responded as if they had never heard the tone. Rats in two control groups—one of which received no shot, the other of which received a placebo injection that did nothing to prevent synapses from making new proteins—remained terrified.
Nader’s data could not have been clearer, or more unsettling. He had demonstrated that the very act of remembering something makes it vulnerable to change. Like a text recalled from a computer’s hard drive, each memory was subject to editing. First you have to search the computer for the text and then bring it to the screen, at which point you can alter and save it. Whether the changes are slight or extensive, the new document is never quite the same as the original.
Many people in the field treated Nader’s findings with contempt. James L. McGaugh, of the University of California at Irvine’s Center for the Neurobiology of Learning and Memory, and one of the nation’s leading neuroscientists, argued, like most of his colleagues, that once long-term memories are established, they are there to stay. “Occasionally the seduction of simplicity embarrasses the field,” McGaugh and two colleagues wrote at the time. He compared work on reconsolidation like Nader’s to notoriously inaccurate research, begun in the 1960s but long since debunked, suggesting that it was possible to transfer intelligence from one animal to another through “memory molecules.” “We should be careful not to laugh in retrospect at such ideas,” McGaugh wrote, “if we remain attracted to other more contemporary simple explanations of the complex phenomena of learning and memory.”
Scientists around the world soon set out to repeat Nader’s study, and the results of experiments in dozens of species, from fruit flies to mice, supported his conclusions. The dogma of consolidation made no sense. It is one thing, of course, to erase a fear created in a laboratory and applied to rats, and another to do it with humans. Daniela Schiller was in Israel at the time, finishing her doctorate. Using an animal model, she had studied the relationship between emotion and neural circuitry in schizophrenia. When Schiller learned of Nader’s findings, she wondered if it would be possible to reactivate a traumatic memory in humans and then block the fears associated with it, much as Nader had done in rats. With her father’s advancing age never far from her mind, she became determined to find out.
Daniela Schiller is tall and trim, with steel-blue eyes and dark-blond hair. When she strides through her laboratory at Mount Sinai, Schiller—nearly always dressed in understated outfits designed by her sister, Yael, in Tel Aviv—carries herself more like a Middle European aristocrat than like a woman who grew up in a scruffy suburb of Tel Aviv. Schiller’s mother is Moroccan, and she says that her father, who suffers from emphysema, sounds like a sort of Polish Darth Vader. “You hear him before you see him,” she said. Schiller is the youngest of four children; her two older brothers and her sister stayed in Israel, and her parents still live in the house where she grew up. Science always appealed to Schiller. “I would mix sand from the backyard with all sorts of materials I found at home and turn it into weird solids and liquids,” she told me. The concoction “looked like a top-secret chemistry set, in my little mind, so I asked a neighbor to hide it in her backyard. After a few days she asked me to take it back. She was worried it might blow up or something.”
The winter Schiller started working at NYU, she noticed her boss, Joseph LeDoux, playing guitar at a Christmas party with Tyler Volk, a professor of biology. Schiller is a drummer, and she soon found a lab mate who played bass. The four formed the Amygdaloids, which, despite the gimmicky name, is far better than one might suspect of a band born in a brain lab. At NYU, Elizabeth Phelps asked Schiller to work on a study that might determine whether humans would respond the way rats did to Nader’s experiments. But the drug used for rats was far too toxic to use on people. Instead Schiller used propranolol, a common beta blocker that, because it latches on to receptors in a variety of proteins, has been shown to interfere with the formation of memories. She applied to the university for permission to carry out the experiment and waited for a response; she has not yet received one.
During a laboratory meeting, however, Schiller’s colleague Marie Monfils mentioned that after behavioral training, a group of rats in one of her experiments seemed to lose their fear. The finding was serendipitous; Monfils had originally been studying something else. But the comment provided Schiller with what she describes as her “eureka moment.” Until then memory reconsolidation had been blocked only by physical intervention, either drugs or electric shocks. If, as scientists have suggested, reconsolidation evolved so that memory could be augmented with new information, then behavior modification ought to have the same effect as a drug. “I
suddenly realized that we had never tested that theory,” Schiller told me. Monfils agreed to carry out a behavioral study of rats, and Schiller would do the same with humans.
The theory was borne out by both experiments. Schiller trained 65 people to fear a colored square by associating it with a shock. The next day, the sight of the square alone was enough to revive their fearful reactions. Then Schiller divided the subjects into three groups. By presenting the squares many more times with no shock, she attempted to teach them to overcome their fear. That is called extinction training. The results were dramatic: people who saw the squares within 10 minutes of having their memories revived forgot their fear completely. The others, who were not shown the squares again until hours later, remained frightened.
Schiller’s study, which was published in Nature in 2010, offered the first clear suggestion that it might be possible to provide long-term treatment for people who suffer from PTSD and other anxiety disorders without drugs. And the effect seemed to last; a year later, when the researchers tested the subjects again, the fear response still had not returned.
Schiller moved to Mount Sinai in 2010. Since then she has pursued three central goals in her research: tracing the neural mechanism, or signature, that causes memory to update in the human brain; determining whether drugs might work safely in humans; and establishing a protocol that therapists could use to treat patients. (Scientists have already found that behavioral interference during reconsolidation appears to alter glutamate receptors in the amygdala, which might explain how memories are rewritten during the treatment.)
On a particularly harsh winter morning in February, I joined Schiller and one of her postdocs, Dorothee Bentz, at the Mount Sinai School of Medicine’s Brain Imaging Core. Despite its impressive, Matrix-like name, the Core is a closet-sized room filled with computers and electrical machinery. The gauges and ominous-looking dials seem to belong on an old radio set. Bentz attached electrodes and sensors to my arms and to my right wrist, told me to take a deep breath, and then started ramping up the voltage. I watched the meter as the needle jumped.
“Do you feel that?” Schiller asked, somewhat remotely. “It’s twenty volts, a small charge.” I said no. She moved the lever to 30. Yes, but only barely, I told her. Finally, at 40 volts, I began to feel the shock. It was by no means a dangerous level; nonetheless, it was a sensation that few people would welcome. Schiller was planning to do to me what she had spent so much time doing to others: teach me to fear a meaningless symbol. Colored spheres began to float onto a computer screen in front of me, in no particularly discernible pattern: just a random, rapid-fire procession—purple, yellow, and blue. It didn’t take long to realize that nearly every time a blue sphere appeared a shock would follow; by the time I felt the voltage, my pulse and heart rate had already spiked in anticipation. The shock itself quickly became superfluous.
The day after learning to fear the spheres, Schiller’s subjects see them again many times—but without the accompanying shock. “If you present a negative memory over and over again, without anything bad happening, it is possible for most people to overcome the fear,” Schiller explained. Extinction training has for a long time been one of the principal treatments for many phobias and fears; psychiatrists refer to it as exposure therapy. The more you see something, the less it scares you, and the less it scares you, the more able you are to deal with it. There has always been a problem, though, in using extinction to treat people who have experienced profound trauma: the process leaves them with a pair of memories: blue sphere predicts shock; blue sphere doesn’t predict shock. Over time the two memories can compete for expression. That is a significant characteristic of anxiety disorder. People will be fine for months or years, but if they encounter a particularly stressful situation, the fear memory often overwhelms the calm memory.
Schiller’s study demonstrated that the competing memories can become one. “If we zap it at just the right time, there are no new memories,” she told me with a look of restrained satisfaction. “There is a different memory. You will still know what happened, and the information will be available to you. But the emotion will be gone.”
Schiller has applied for funding to continue the research. Deep budget cuts have made it harder to get money than ever before, though, and her initial three-year grant at Mount Sinai has nearly come to an end. I asked what would happen if she received no money. “I’m back on the street,” she said, shrugging. “But I believe we can find a way to make PTSD less terrible. From the research perspective, you really do get very optimistic. Of course, I am careful not to try and overhype it. Translating research into better human lives is never easy.”
Not long after my fear test, I took the train to Philadelphia to speak with Edna Foa, who is the director of the Center for the Treatment and Study of Anxiety, at the University of Pennsylvania Medical School. Foa is one of the nation’s leading experts on the psychopathology of anxiety disorders, and she has written widely on PTSD. We met in her office at the medical school, which looks onto the oddly serene urban landscape of Center City. I asked if she thought scientists would ever really be able to write the pain out of a patient’s mind.
“That is the critical question,” she replied, stressing that she is a clinician, not a neuroscientist. “This is the most exciting prospect I think I have ever seen for treating people with severe anxiety-based disease. It isn’t easy to banish demons caused by war, trauma, and rape.” Freud argued that repressed memories, blocked unconsciously, were like infections, capable of deepening and festering unless they were brought to the conscious mind and resolved. Many psychiatrists have taken the opposite approach. “There has always been a group that says we could reignite a trauma by asking people to deal with the memory,” Foa said. “In this thinking, keeping the memory suppressed was actually better. That was a strong belief in the early era of psychiatry: Put it behind you. Don’t deal with it. Go on with your life. The idea behind counseling was to soothe the patient, to find ways to make him as comfortable as possible.”
Only in the past decade have researchers determined that while the original memory may be inhibited, it doesn’t vanish. Foa said that the idea of rewriting memories, rather than destroying them, appealed to her. But she added that reconsolidation raises a paradox: in order to update our most painful memories, we have to revisit them. That is never easy to do. Foa described a patient who was raped more than a decade ago by her boyfriend and several of his friends. She suffered badly from PTSD, found it impossible to maintain relationships, and had recently entered therapy. “Instead of asking herself what actually happened, she would immediately say it was all her fault,” Foa said. “She always said the same thing: ‘I didn’t fight them. If I had, they would have stopped.’
“But she never dealt with it, and that is why she had PTSD,” Foa went on. “We asked her to tell the story of that New Year’s Eve and repeat it many times.” As people work through the story again and again, they learn to distinguish between remembering what happened in the past and actually being back there. For people with PTSD, this distinction is not easy to make. The next step was to bring those memories to the surface—and when, finally, the woman did that, she realized that her terror and her rape were not her fault.
I asked Foa if she had considered the ethical complexity involved in tampering with a person’s memory. “Of course,” she replied. “But you do have to look at the whole picture. We are talking about helping people who have been severely traumatized, and in many cases they are unable to function. Nobody is suggesting that we rewrite the memory of someone who had a bad date or a fight with his mother.”
In practice it may be hard to draw an ethical line that would satisfy patients, doctors, and the public. Few people would deny effective treatment to victims of severe brutality. But any treatment available to those who need it will almost certainly be available to others. “Memory erasure remains a possible but unproven hypothesis,” Joseph LeDoux has written, adding that editing memories “i
s definitely possible and has broad implications. We are nothing without our memories, but sometimes they also make us less than we could be . . . Although some ethicists argue that memory should not be tampered with, every special date and anniversary, every advertisement, every therapy session, every day in school is an effort to create or modify memory. Tampering with memory is a part of daily life. If we take a more realistic view of just how much we mess with memory, the dampening of memories that produce emotional responses in traumatized individuals might seem less malevolent.”
Reconsolidation has already been shown, in promising if limited research, to help treat drug addiction. Addicts are compelled by the same persistent emotional memories that drive other disorders. “The biggest problem for most addicts is how to deal with relapse,” Schiller told me. “Let’s say somebody is drug-free and then goes and hangs out with friends at a park. He might see a cue associated with his drug use, and that will induce a craving that will cause him to seek the drug.” Reconsolidation presents a chance to disrupt that process; you don’t lose the memory—you just lose the pleasant feeling it creates.
The idea is simple enough: you cannot be addicted to a desire that you don’t remember. Jonathan Lee, a behavioral neuroscientist now at the University of Birmingham, in England, has already put that notion to a test. He used Pavlovian conditioning to induce cravings in rats, by pairing light with a narcotic. The next time he showed the animals the light, they automatically reached for the drug. But, as was the case with Nader’s experiments, when Lee interrupted the process of reconsolidation, the association disappeared. Researchers in the U.S. and China have had similar success with human addicts. Once again, timing was critical: the effect worked only if extinction training took place within 10 minutes of retrieving the old memory. “If you block that association, you can erase the craving,” Schiller said. “This is the first time we have seen a treatment like that lead to a cessation of addiction.” Even six months later, the addicts showed no sign of relapse, suggesting, as with Schiller’s work, that when fearful memories are disturbed at the right moment, the fear may be gone for good.
The Best American Science and Nature Writing 2015 Page 32
