In the spring of 1968, experimental psychologist Donald Lewis and his colleagues published a study about memory that was well before its time. The researchers first trained rats to fear a particular sound. A day later, the animals heard the same sound followed immediately by an electroconvulsive shock to the head. After getting shocked, the animals forgot their fear of the sound. The old memory was gone.
The study was remarkable for its focus on memory retrieval, rather than memory formation. See, research up until then had suggested that a memory is only unstable (and thus vulnerable to change) in the minutes or hours after it’s first created. During this period, called ‘consolidation’, the memory moves into the brain’s long-term storage and, it was thought, into a stable and fixed molecular state. “For decades people had thought that once a memory is wired in the brain it stays there forever,” says Karim Nader, a neuroscientist at McGill University in Montreal. But Lewis’s study showed that wasn’t true: When a rat recalled a stored memory, the memory somehow became unstable again, making it vulnerable to erasure.
Lewis’s study contained a revolutionary idea, but it didn’t revolutionize the memory field. It was published in a top journal, Science, and followed up immediately by another group of researchers. But as that group reported the following year, also in Science, they couldn’t replicate Lewis’s findings. Over the next few decades, a couple of other studies came out suggesting that memories become unstable during recollection, but the idea never made it into the textbooks. “Part of the field believed it, but the other part of the field just didn’t believe it,” Nader says. “And the ones who didn’t believe it were the dominant ones.”
Nader gets credit for rekindling the idea in the late 1990s, while working as a postdoc in Joe LeDoux’s lab at New York University. Nader showed that, in rats, old memories can be erased by infusing a drug into the animal’s brain as it recalls the memory. Because the drug blocked protein synthesis, this experiment was evidence that memories go through a ‘reconsolidation’ process after being recalled, and that this process requires protein synthesis (just like the initial consolidation does).
Unlike Lewis’s study, Nader’s, published in Nature in 2000, did rock the field, not least because of its clinical implications. The results opened up the possibility that the frightening memories that haunt people with post-traumatic stress disorder could be erased — even long after they are formed.
A study out today in Nature Neuroscience takes this research a big step further by demonstrating not only that memory reconsolidation happens in people, but that it can be blocked with electroconvulsive therapy, or ECT.
Marijn Kroes and his colleagues at Radboud University Nijmegen, in the Netherlands, tested the memories of 39 people with severe depression who were already undergoing ECT. This rare treatment has an interesting cultural history. “People have the One Flew Over the Cuckoo’s Nest idea of it,” Kroes says, referring to the 1975 film in which Jack Nicholson’s character is forced — wide awake, whimpering and writhing in pain — to undergo ECT at a mental institution. “Luckily it’s nothing like that.”
Today ECT is used as a last resort for people with severe depression who do not respond to antidepressants, psychotherapy, or other treatments. It’s done in a hospital room, after the patient receives muscle relaxants and general anesthesia. A brief electric current passes through the brain, inducing a seizure. For reasons no one yet knows, it usually works: ECT had an 86 percent remission rate for people with major depression, according to one study.
Kroes wanted to find out whether ECT could erase memories during active recall, as it had for the rats in Lewis’s experiment nearly 50 years ago.
All of the participants first saw two narrated slide shows, each describing a different traumatic story with 11 pictures. In one of them, for example, two sisters leave their house to visit their brother at a nearby bar. As they walk past an alley, one of the sisters is kidnapped by a man and held at knife point:
The other story is similar in structure, but shows a boy getting hit by a car and then in surgery having his feet reattached to his legs. In the words of the paper, these are “high-arousing stories with negative valence.” I’ll say.
A week after the participants watched the slide shows, the researchers showed them the first slide from one of the stories as a memory trigger. A few minutes later, some participants were given ECT. After the treatment, they were given a memory test about the details of both stories.
Participants who did not undergo ECT got about half of the questions about the triggered memory correct, compared with 40 percent of questions about the other story (which they had not seen for a week).
In stark contrast, participants who received ECT seemed to have no memory of the story they had been reminded of — they scored a 25 percent on a four-answer multiple choice test, the same as guessing at random. The same participants showed significantly better recall — 35 percent — of the second, non-triggered story. ECT, in other words, selectively erased the memories that were being actively recalled, just as it had for Lewis’s rats.
“It’s a very elegant paper, compelling data, and it’s a difficult study to do,” says Daniela Schiller, a neuroscientist at Mount Sinai School of Medicine in New York, who was not involved in the work. “It’s very impressive they managed to do that, and that they even tried.”
Schiller’s experiments have also bolstered the reconsolidation hypothesis. She has shown, for example, that if people recall a fearful memory and then go through ‘extinction learning’ — meaning that they’re shown the fearful stimulus over and over again without any pain — they can erase the emotional sting of the memory. Other groups have shown something similar by giving people propranolol, a beta-blocker, immediately after recalling a memory.
The new study adds ECT to the list. There are still a lot of questions. For example, it’s not clear how ECT is disrupting reconsolidation. Or if it’s doing it at all: The effect could be partly or wholly due to anesthesia, though the researchers say this is unlikely. Most importantly, no one knows whether the procedure would work with old, real memories, as opposed to those artificially created in the lab.
Kroes and his colleagues are planning a clinical trial in which they will use ECT to lessen traumatic memories in people with PTSD. “Just to be clear: it’s a long way from being an actual clinical application,” he says. “A lot of experimental, fundamental science is often very difficult to translate into the real world.”
I asked all of these scientists a question that I’m sure they get every week: When are we going to be able to erase whatever memories we want, like in Eternal Sunshine of the Spotless Mind?
Kroes was quick to say, and rightly so, that ECT should only be used as a serious medical treatment. “You have to have some kind of disease state,” he says.
But as far as the technology goes, and what it could do, everybody told me that there’s no reason to think that we couldn’t play out a Spotless Mind scenario in the not-too-distant future.
“I wouldn’t have said so a few years, but there’s just more and more evidence, with different types of memories, different types of manipulations, and different species,” Schiller says.
Right or wrong, there’s certainly demand for it. Nader remembers the big public response to his 2000 Nature paper. “The day after it was published, a number of women emailed Joe [LeDoux] and asked, ‘Can you get rid of the memories of my ex-husband?’”
But would he be OK with that kind of application, should the technology advance as experts expect it will? ”For me,” Nader says, “I don’t think that would be the end of the world.”