In the 2004 movie Eternal Sunshine of the Spotless Mind Jim Carrey has his memory erased to forget a painful relationship. Scientists are a long way off performing this type of procedure in people, but studies suggest that it is, indeed, possible.
Credit: Focus Features
JASON NICOLL IS TORMENTED by the past. In 1994, when he was 21 years old, he was deployed with the Australian Defence Force to serve six months in Rwanda. This was at the height of the genocide.
"I saw dead bodies, people who'd stood on land mines, people who'd been shot or hacked up with machetes. Once, I carried a young girl, about eight or nine years old, who'd been shot in the chest.
"The round had torn most of her spine out. There was a massive hole, and the way she was looking at me … I put her on the table for the medics to start working on her. It wasn't long afterwards that she died."
Fifteen years on, Nicoll still has flashbacks and nightmares. Certain smells - a rotting banana in a child's school bags, for example - bring terrible memories flooding back, sending him dry-retching into the yard of his house in Forrest Beach, Queensland.
"There was always something rotting around the place - whether it was fruit or meat or humans," he says of Rwanda. It wasn't until 2006, when Nicoll suffered a breakdown, that he was finally diagnosed with post-traumatic stress disorder (PTSD), and given a medical discharge from the Australian Army.
SINCE THEN, HE'S COMPLETED a two-month course of PTSD treatment, had anger management therapy and seen a variety of psychiatrists, psychologists and counsellors.
He's learned strategies to help him cope. But nothing has been able to take away the horror associated with those memories. And he's only one of more than a million Australian sufferers, according to a 2007 Australian Bureau of Statistics report
About one third of people with PTSD don't really benefit from existing treatments, says Mark Creamer, director of the Australian Centre for Posttraumatic Mental Health at the University of Melbourne.
"That is a concern," he says. "We need to think about what alternatives might be available." The good news is that teams of scientists around the world are doing just that.
A recent flurry of research is suggesting new ways to get rid of the fear associated with certain memories, and even to delete those memories - permanently. Not only is this work throwing light on PTSD, it's helping to answer fundamental questions about how we form memories, how we store them and how we access them. This new understanding is leading to developments that would have been unthinkable a decade ago.
In London, neuroscientists have read people's memories by scanning their brains. A team in the U.S. is building an artificial human memory centre in their lab. Another group thinks we'll soon be able to turbo-charge our memory by consciously altering electrical activity in our brains. "It's an exciting time for researchers in the field," says Elizabeth Loftus, a leading expert on memory at the University of California at Irvine.
THIS WORK MAY BE EXCITING, but it is equally as controversial. If murderers are traumatised by memories of their actions, should they be allowed to have their guilty consciences wiped clean?
If scientists can peer into brains to see memories, should suspected criminals be scanned, to check if they've been at a crime scene?
If, like the character played by Jim Carrey in the 2004 film Eternal Sunshine of the Spotless Mind, you'd like to have unpleasant memories of a past love affair wiped - well, why not?
But then, what if you accidentally lose pleasant memories too? And what about survivors of war crimes: do they have a duty to remember their suffering, so that the perpetrators might one day be punished?
"The main ethical problems might be that we change who we are - that we make ourselves believe untrue things, or that we make decisions we would not have agreed were right if we knew or felt the relevant facts," says Anders Sandberg, a neuroethicist at the University of Oxford's Future of Humanity Institute in Britain.
Loftus worries about the extent to which people should have 'freedom of memory'. "Should they have freedom to dampen their personal plight, or do they have an obligation to society to remember? Another concern is whether people should be obligated to take the drug.
"Consider soldiers returning from the horrors of war, who are at high risk of developing PTSD. Taking the drug might help them, and also reduce the costs of treating the illness, should it develop. We vaccinate people to prevent disease - so why not this?"
Of course, it isn't only former soldiers who suffer from PTSD. About one third of people who experience a traumatic event, such as a physical attack or a nasty car accident, go on to develop the disorder. To understand why, scientists are investigating how memories are created and stored.
Your sense organs continuously send signals to a part of the brain called the thalamus. The thalamus directs this information to an almond-shaped region called the amygdala. The amygdala then performs a quick rough-and-ready emotional assessment. If it recognises something as potentially threatening - such as a sudden loud bang - it triggers the body's stress response.
Meanwhile, signals from the thalamus have also been sent to the prefrontal cortex, a region of the brain responsible for higher mental functions but with slower responses. If the prefrontal cortex ultimately decides the threat isn't serious - perhaps it recognises that the bang is just a car backfiring - it tells the amygdala to calm down.
But if the danger is real - if a soldier sees a colleague get shot, for example - the rush of stress chemicals yells at the hippocampus, one of the brain's main memory centres, to make sure it remembers that a loud bang could mean deadly gunfire.
In this way, you're guaranteed to remember associations that are vital for your survival. After that, your brain is primed to recognise any sudden loud noise as dangerous. Over a few months, the mere hint of something similar would be enough to send your stress levels soaring.
Over time, repeated exposure to loud noises that turn out to be merely cars backfiring or fireworks will over-ride the old memory, in a process called 'extinction'. In some people, for reasons poorly understood, extinction doesn't happen - and they develop PTSD.
"In the worst cases, it's almost as though the memory has been locked in an unchangeable form," says Creamer. "And not only the memory of what actually happened, but of their reaction - so the person responds in exactly the same way during these re-experiencing moments."
'Exposure therapy' is the standard frontline treatment for PTSD, and the related anxiety problem phobia. By repeatedly exposing the patient to the triggers that normally make them afraid, but doing it in a safe location, such as a lab, the idea is that they'll gradually learn to lose their fears.
While this technique works very well for about a third of PTSD patients, for others - including Nicoll - it provides no relief. So some researchers are turning to drug therapy instead to see if it might help.
In 2001, Roger Pitman, at Harvard Medical School in Boston, asked trauma patients arriving at the emergency room of the Massachusetts General Hospital if they were willing to take part in his study.
If they agreed to take part, they were either given a placebo, or a drug called propanolol - commonly used to reduce blood pressure. As soon as possible after treatment for their injuries, Pitman recorded each patient's description of their trauma.
Three months later, he hooked the same people up to various biological sensors, played to them the recordings of their interviews of their traumatic incident, and asked them to picture what had happened to them. The results were interesting.
While nearly half of the dummy-pill group showed excessive stress, none of the patients who'd taken the propanolol did. The memories of the patients in the latter group hadn't been wiped - but they didn't experience the same fear when recalling the memory.
For the first time, it suggested a way doctors might be able to prevent PTSD developing by using a drug. Pitman has since moved on to study the effects of morphine and the U.S. abortion drug mifepristone, or RU486, in rodents, as a potential treatment for people who have already developed PTSD.
Drugs such as propanolol bind to the same receptors in cells as adrenalin: the suspicion is that they might stop the memory being labelled by the amygdala as 'emotionally charged' - and so preventing it being so firmly imprinted in the brain.
And it seems the effects can be long lasting, if not permanent. In 2007, a paper in the Journal of Psychiatric Research concluded that 10 years on, victims of rape or car crashes who'd been given propanolol not long after the incident showed much less stress than other victims when recalling their trauma.
A team at the University of Amsterdam, in the Netherlands, recently found that volunteers who had been trained to be afraid of a picture of a spider would have their fear erased if they took propanolol. The fear would remain erased even after another period of training designed to reinstate it.
Why? Reliving fearful memories triggers another burst of adrenalin, which reinforces that memory. Lead researcher Merel Kindt thinks that by using the propanolol to stop adrenalin binding to cells, they stopped the unpleasant memory from being reinforced - and broke the fear association permanently.
The Dutch work makes use of a phenomenon known to memory researchers for years - after a memory is jogged, there is a short period of time when it's open to being modified, or 'reconsolidated'. "It seems we have this window of opportunity," explains Marie Monfils at the University of Texas in Austin.
"When you retrieve a memory, it becomes susceptible to disruption - but only for a certain period of time." Monfils wondered if it might be possible to use standard extinction therapy within this time window to effectively treat a phobia or PTSD, without using drugs.
The team taught rats to associate a musical tone with a small electrical shock. Pretty soon, just the sound alone made the rats freeze in fear. When the team then played the tone over and over without the shocks - in classic exposure therapy - the rats showed less and less fear. But the effect didn't last.
A month later, the rats were just as scared of the tone, as though the exposure therapy had never happened. The breakthrough came next. After training another group of rats, Monfils and her team sounded the tone once. They then waited for an hour, during which the animals' brains retrieved the unpleasant memory.
Then the researchers played the tone over and over and over. They found that hour was crucial - the rats' fear of the tone disappeared, and it appeared to have gone for good.
Neuroscientists Daniela Schiller and Elizabeth Phelps at New York University recently tried a very similar approach with people. They trained people to associate the sight of a blue dot with an electric shock.
And they found that exposure therapy during the 'reconsolidation window' permanently stopped the fear when they were later shown the dot.
Monfils thinks that standard exposure therapy creates a non-fearful memory associated with the trigger, which then exists alongside the original fearful memory.
But extinction therapy during the critical reconsolidation time period actually overwrites the original memory, effectively erasing it. "This technique seems to be rather specific - allowing control over what is recalled and manipulated," Schiller says.
OTHER RESEARCHERS, MEANWHILE, are trying different approaches to easing bad memories. Some are focussing on cortisol, for example - a hormone involved in the stress response, which seems to be present at a much lower level than normal in the brains of patients with PTSD.
These studies don't reveal much about how memory deletion might actually happen in the brain. But in March 2009 came a bombshell. A paper published in the U.S. journal Science described a way to identify the specific brain cells that were storing a fearful memory, and to wipe them - and the memory - out.
This leads the researchers to think that the kind of memory mapping and deletion that was fiction in Eternal Sunshine of the Spotless Mind could one day become fact.
"Funnily enough, I have never seen this movie," admits Sheena Josselyn, co-researcher at the Hospital for Sick Children in Toronto, Canada.
"Although I have heard of it, and Jim Carrey grew up only a few short miles from my office! Technologically, we're a long way off performing this type of procedure in people," she stresses. "But our studies suggest that it is, indeed, possible."
Josselyn and her husband Paul Frankland, who works at the same hospital, collaborated to focus on a protein called CREB (cyclic adenosine monophosphate response element-binding protein - to give it its full unwieldy name).
Earlier work by the pair, as well as by other groups, has shown that this protein is crucially important in the formation of memories - not just in people, but also in sea slugs, fruit flies and mice. Too little CREB stops memories being formed, while increasing CREB in the brain seems to actually boost an animal's capacity to remember.
Josselyn and Frankland wondered if the brain cells in the amygdala that make CREB might in fact store fearful memories. Using electrical shocks, they trained mice to become afraid of a musical tone.
Then they infected the animals with a virus that targeted only brain cells expressing CREB, and the virus made the cells susceptible to a diphtheria toxin. When the pair injected this toxin, only these cells - which turned out to be 10 to 20% of neurons in the amygdala - died.
Afterwards, the rats completely forgot to be afraid of the tone. This memory was gone. Yet others didn't seem to be damaged. "We examined a number of other memories and found they were intact," says Josselyn.
"Of course, we did not exhaustively look at all memories, and it is certainly possible that we did impair others," she adds.
While there's more work to be done, this research might lead to new ways of destroying fearful memories in people - though probably by interfering with the CREB production, rather than destroying brain cells.
However, Mark Creamer, for one, is sceptical that only target memories could be erased. "I think the idea that you could get rid of a single memory without affecting others is still fantasy land, quite honestly.
But another point is, even if you could do it, I don't think people would want to. For most people with PTSD, these have been very important and powerful experiences, and they don't want to forget. They want to be able to remember them without being incapacitated by them."
Would memory researchers themselves use erasure techniques? "I would do it if traumatised by witnessing some horrific event to minimise chances of PTSD," says Loftus. "But I work with other scientists, who say, 'no - I want to hang onto my memories - good and bad'."
Despite all the terrible things he has witnessed, Jason Nicoll says he doesn't want to forget. "I want to remember," he says, "because it keeps me in a place where I know the world is not a safe place. I'd like to keep the memories - just not suffer from them."
If painful memories can be erased - at least in theory - what about others? At the University of Cambridge, in Britain, Barry Everitt and his team have been getting rats hooked on cocaine in a bid to find out.
They've discovered that by giving the rats a chemical that interferes with a brain receptor for glutamate - a compound known to be involved in learning and recall - they can wipe an animal's drug-related memories. It's well known that when human addicts recall memories to do with drug taking, they're more likely to relapse. So the team hopes that this kind of treatment might help addicts stay clean.
But what about other troubling memories - such as embarrassing ones or guilty ones? "It's a very interesting question," agrees Josselyn. Because it would be unethical to attempt extreme memory manipulation experiments in people, researchers use mice. And, while mice certainly show fear, embarrassment and guilt might be uniquely human emotions, she says.
Discovering how to tamper with these sorts of memories is likely to lie further in the future. In the meantime, Josselyn and Frankland's work is giving memory researchers plenty of food for thought. And not only because of the implications for PTSD.
In fact, the significance of the CREB study goes far beyond understanding how to erase fearful memories. For the first time, scientists have been able to point to specific brain cells and say, "there's a memory".
"This is important because neuroscientists have been searching for the physical trace of memory for decades," Josselyn explains. And the pair aren't the only ones now laying the brain's age-old memory secrets bare.
About 500 km southeast of Josselyn's lab, researchers led by neuroscientist Todd Sacktor at the State University of New York, in Brooklyn, have found a brain molecule that seems to be crucial for all sorts of memories to be formed.
An experience - such as meeting someone or seeing a film - creates a particular pattern of activation of brain cells. The theory is that by developing thicker or more efficient lines of communication between this network of cells, the brain retains a memory. But how it does this has long been a mystery.
Ten years ago, two leading brain scientists from Harvard University identified 117 molecules that seemed to have some part to play in the process. However, no one knew how important each of these molecules might be, or exactly how they might help memory formation.
Sacktor's lab went to work to answer the question. Eventually, they hit pay dirt. They found that at the precise moment that a cell was stimulated by a neighbour, molecules of a chemical called PKMzeta became active in that cell.
In other words, it looked as though PKMzeta might be vital for a memory to be formed. Further work in the lab, and by other teams around the world, has now confirmed this.
PKMzeta works by doubling a type of receptor for the chemical messenger glutamate at synapses - the point of connection between two nerve cells.
One study in Israel even found that a single dose of a drug that interferes with PKMzeta can make rats forget their dislike of a taste that had made them sick months earlier. Over at the University of Alabama in Birmingham, USA, another team is also picking apart the process of long-term memory.
The proteins that form connections between brain cells - and which seem to be essential for a memory to be retained - are constantly destroyed and replaced. So how do brain cells 'remember' which neighbours they're linked to?
Work by Courtney Miller and David Sweatt suggests the answer lies at least in part in a process that adds methyl 'caps' onto DNA. This process alters gene expression, and so protein production, in the cell.
The pair found rapid changes in methylation patterns of cells in the hippocampus for an hour after animals were trained using a mild electric shock. A day later, these patterns had returned to normal.
However, methyl shuffling was now happening up in the cortex. After seven days, this cortical shuffling stabilised, and stayed constant.
Miller and Sweatt think they've watched short-term memories forming in the hippocampus slowly turning into long-term memories in the cortex, but it's not clear yet how altered protein production could be involved in storing memories.
"While it's too early to say if DNA methylation can explain how memories are preserved, our findings do support the idea that it is playing an important role," Miller says.
This work was based on rats, so it's hard to know how long memories might hang around in the hippocampus before moving to the cortex in people. But Christine Smith and Larry Squire at the Memory Research Lab at the University of California in San Diego, think they know the answer.
In January 2009, they reported in the Journal of Neuroscience a study of 15 volunteers who had their brains scanned while they answered questions about news events that had taken place over the past 30 years.
As memories approached 12 years old, activity in the volunteers' hippocampus and amygdala gradually decreased, then levelled off. But in the cortex, the opposite happened.
In the cortex, activity stayed pretty constant for recent incidents, but increased with the age of the news event.
This explains why patients with damage to their hippocampus have trouble forming new memories and also why they can't remember events in the years immediately before their amnesia, while their memory of the distant past remains intact.
So what might this flood of new insights into memory mean for people with short-term memory problems - or for healthy people who want help not to forget, but to remember?
IN A LAB AT THE University of Southern California in Los Angeles, Theodore Berger is leading work on the first brain prosthesis. The artificial hippocampus consists of a silicon chip that communicates with the brain via an array of electrodes.
So far it's been trialled only on animals. But the team hopes that in the future, it could be tested on people with damage to their own hippocampus. When it comes to healthy people, though, they'll be more likely to turn to drugs to boost memory.
Some drugs - such as Aricept, manufactured by Japanene company Eisai - were developed to treat Alzheimer's disease, but it seems that they might also boost memory in normal people.
Meanwhile, scientists at the University of California in Irvine are working on a compound known as CX717, which boosts the brain chemical glutamate - the same chemical targeted in the rat cocaine work.
Volunteer pill-poppers who had their memory tested while they stayed awake during the night showed much better powers of recall than those given dummy pills.
In the future, Todd Sacktor thinks that stimulating brain cells to increase their production of PKMzeta might form the basis for a very powerful and targeted memory aid.
But there could also be another alternative - one based on software on a PDA or an iPhone.
Enhancing your memory might sound like only a good thing, but some researchers are as concerned about the ethics of this as they are about erasing memory. Should students sitting exams be able to use the techniques?
If they were banned, how could anybody actually enforce this? Will only richer people have access to the drugs and devices? Anders Sandberg isn't too worried about access.
If the drugs really work, and it's believed they are unjustly distributed, the government could introduce subsidies for them, he says. "A much harder problem is whether we should give them to kids.
"I think informed adults have a right to decide for themselves if they want to enhance. But whether parents should be allowed to decide for their kids is harder to decide," he says.
As far as current memory research and ethical concerns go, however, one particular recent breakthrough is creating at least as much debate as memory enhancement, if not more.
In March 2009, Eleanor Maguire, a neuroscientist at University College London, and her colleague Demis Hassabis, revealed that they'd been able to read a person's memory by scanning their brains.
Volunteers put on a virtual reality headset then made their way to and from four rooms in a virtual building. Meanwhile, the scientists used fMRI (functional magnetic resonance imaging) to monitor activity in their hippocampuses.
Software analysed the data, looking for patterns in brain activity across the volunteers. And the pair found that certain cells in the hippocampus encode a person's location - and they could use this memory data to actually predict where in the building a volunteer was standing.
This work shows for the first time that memories of our environments are encoded in the brain in a predictable way. And it has neuroscientists wondering about what might be possible with brain scanners in the near future.
Might police be able to scan a suspect's brain to see if they have been to a crime scene, for instance? Hassabis says that he thinks that could be achievable in as little as 10 years.
Neuroethicist Anders Sandberg argues that, as a society, we now need to try to figure out the 'proper' way of using some of these techniques, particularly memory enhancers and deleters - and this will take time.
"Ethical principles are not going to be as important as the actual social norms, habits and culture that will surround them," he says. "But this requires experimentation, trial-and-error experiences and a lot of negotiation. We cannot plan this ahead, because we just don't know the full implications right now."
In the meantime, PTSD sufferers like Jason Nicoll are in desperate need of effective new treatments. "If there was something out there that might make my life better, I'd try it. I'd try anything," he says.
Emma Young is a science journalist and author in Sydney. She has been a reporter for The Guardian and a columnist for The Sydney Morning Herald. Emma Young won 2010 Health Journalist of the Year for this report
