Can electric currents make us smarter and cure mental illness? Elizabeth Finkel reports.
“It was the best of times, it was the worst of times, it was the age of wisdom, it was the age of foolishness, it was the epoch of belief, it was the epoch of incredulity …”
– Charles Dickens, A Tale of Two Cities.
Dickens could have been describing the latest methods to tweak our brains. Forget cognitive behavioural therapy or drugs: 21st century therapists are wielding electrical currents like soldering irons, claiming to tweak the circuitry of our grey matter without nicking the skull.
Not so long ago, the idea of zapping the brain was horrifying. Jack Nicholson set the tone when he played RP McMurphy in the 1975 movieOne Flew Over the Cuckoo’s Nest which was set in a lunatic asylum.
Wild and wolfish but not insane, McMurphy’s unruliness is “treated”. Held down by white-uniformed orderlies, the electrodes attached to his head render him a jerking, writhing mass.
These days, people are clamouring to zap their brains. You can even buy DIY kits online at Foc.us, advertised by young men with chic red headsets attached to their forehead. The ads say “Unleash your inner game” and “Take charge”. Powered by a nine-volt battery, the kits deliver what is known as transcranial direct current stimulation or tDCS. They are being snapped up by video gamers and others who want to amp up their concentration or zap themselves out of depression.
Do they work? If so, what are they doing – and can they cause harm? No one knows for certain. The devices have bolted half-baked out of the lab. Researchers seem ambivalent about possible harms, and some question if the piddly two-milliamp current does anything at all. Some users singe their scalp, but otherwise there’s no evidence of ill effects. But if the devices do affect the brain, then overzealous users could be damaging themselves.
These battery-powered kits have escaped the lab, but they have a cousin that is far too big and costly to do so. It delivers transcranial magnetic stimulation or TMS. TMS is attracting less buzz but the evidence for its effects are far more solid – the US Federal Drug Administration approved the technique for depression in 2008, and other countries including Australia, Israel, Brazil and Canada have followed suit. Large clinical trials are planned to test the technique on those suffering from schizophrenia, autism, obsessive-compulsive disorder, auditory hallucinations, stroke and memory loss.
There is a sense that a paradigm shift is afoot. At one level it seems unbelievable. The rationale behind the technology is at odds with our traditional understanding of the brain.
Like African drum talkers the team seems to have found a way to speak the brain’s own language.
Most discussions about the brain begin with how little we know about this circuit of 86 billion cells. Europe’s Human Brain Project aims to map those circuits one cell at a time, or even one connection at a time, in an endeavour termed connectomics. All this speaks to the oceans of work required to map a human brain from the bottom up.
Yet psychiatrist Paul Fitzgerald, a TMS pioneer based at Monash Alfred Psychiatry Research Centre in Melbourne, tells me of a case in which he treated a person for autism with TMS. A few days later the person blogged that he no longer had trouble making eye contact. Is it really possible to rewire a brain circuit so easily?
Let’s take a tour of this brave new world of mind-tweaking.
You could begin with the electric fish prescribed for headaches in ancient Rome. But a slightly more scientific starting place would be the city of Bologna in the mid-18th century. Luigi Galvani discovered that the mystical force of electricity was responsible for the movements of living flesh, having found that a spark of current made a frog’s muscle twitch.
In 1874, American physician Robert Bartholow showed that the brain was the source of the muscle-twitching signals. In an experiment so brutal it beggars belief he administered electrical currents to the exposed brain of a woman called Mary Rafferty. Her muscles twitched. He increased the current until she became distressed, convulsed and went into a coma. She died 72 hours later.
By 1937 Italian physicians were using electro-convulsive therapy to treat the manic symptoms of schizophrenia. Their logic was partly based on the observation that epilepsy and schizophrenia were rarely diagnosed in the same patient, suggesting that convulsions might protect against schizophrenia. The asylums were overflowing and doctors were ready to try desperate treatments – some included injecting inmates with high doses of insulin to induce a coma.
Shock therapy turned out to be helpful at reducing manic symptoms, although it also induced memory loss. But after these convulsions, peoples’ mood generally improved. Shock therapy was soon being touted as a panacea for a number of mental illnesses. Its abuse, indelibly portrayed by Hollywood, stigmatised the technique. These days, shock therapy is carefully used on consenting adults under anaesthesia, and is a mainstay for treating severe drug-resistant depression, although memory loss continues to be a consequence.
TMS and tDCS escaped the stigma of shock therapy because they began as a way of diagnosing patients, not of treating them. For instance in the late 1970s, researchers wanting to monitor patients with multiple sclerosis applied direct currents, like that in tDCS, across the skull to measure how fast the impulses generated in the brain were travelling to the muscles. But the skull has high electrical resistance; it is hard to get the current through the bone and doing so can be painful.
So researchers opted to try magnetic stimulation. A pulsating magnetic field easily crosses the skull and induces an electric field at right angles. So by focusing a magnetic beam below the surface of the skull, an electric current could be generated in the brain in a more precise and painless way. Anthony Barker and colleagues first put this into practice at the University of Sheffield in 1985. Meanwhile, new techniques such as PET scans or functional MRI (fMRI) allowed researchers to peer into a living human brain. When people performed different activities, these machines showed different parts of their brain lighting up.
These techniques pinpointed a difference in the brains of people with depression. The oversized wrinkly outer layer of the brain known as the cerebral cortex is where we analyse sensory inputs, control muscle movements and think. And within this vast 1,500-square-centimetre processor are two small spots with an oversized name – the dorsolateral prefrontal cortex. Lying just behind the temples and only a couple of centimetres in diameter, they are the command centres that control executive functions such as drive and decision making.
These command centres integrate inputs from other parts of the brain and an important input comes from deep within, from an area not accessible to TMS – the amygdala. The almond-shaped structure attaches emotional weight to thoughts. In people with depression the amygdala seems hyperactive while the left dorsolateral prefrontal cortex tends to be underactive. That may explain why depressed people ruminate and lose their drive. Could stimulating the left dorsolateral prefrontal cortex boost the command centre and restore the balance?
CREDIT: ANTHONY CALVERT
In 1996, the Lancet reported what looked like a promising result. After receiving daily TMS directed to the left dorsolateral prefrontal cortex for five days, 11 of 17 patients with drug-resistant depression showed significant relief for more than two weeks.
TMS achieved full clinical respectability in 2008 when US authorities approved the technique for use in patients with depression. An international trial in which all the researchers used the same technique was decisive. It showed that about 50% of severely depressed patients who had failed to respond to drugs responded to TMS. In Australia, Paul Fitzgerald’s group in Melbourne and a team led by psychiatrist Colleen Loo at The Black Dog Institute at the University of New South Wales were part of that trial. “I was very sceptical,” Fitzgerald says of when he first tested the technique in 2000, “but I was pleasantly surprised”. He recalls the case of 57-year-old Jan Steele, plagued by depression for most of her life. She found it hard to tolerate the treatment which can cause headaches, yet a month or so later she called him to announce that her depression had gone.
Fitzgerald’s group has since refined the way TMS is delivered. Jerome Maller – a neuroscientist and self-described neuro-geek who likes to build robots in his spare time – has been behind much of the improvement. Ten years ago, pinpointing the position of the dorsolateral prefrontal cortex involved navigating your way there with a rough set of instructions.
The typical procedure involved an operator standing behind a seated patent and sliding a magnetic coil across the patient’s head. For starters, the operator needed to focus the beam on to the top of the cortex, but that depth varies from person to person.
Finding the right depth somewhat resembles the way sonar is used on a boat to get a sounding of the sea floor. Sonar operators hear a ping when they strike bottom – TMS operators rely on a muscle twitch. They slowly move the magnetic beam over the motor cortex, a finger-sized area above the ears which carries a map of every muscle in the body. Tweaking a particular spot will jerk a particular muscle.
A thumb jerk makes for a convenient signal, so the operator moves the coil across the motor cortex to zero in on the part that corresponds to the thumb. The minimum magnetic power needed for a thumb jerk is termed the motor-evoked potential. It sets the strength the operator will use to stimulate other parts of that individual’s brain.
So how does the operator find the tiny dorsolateral prefrontal cortex in a depressed patient? After marking the position on the skull that corresponds to the thumb jerk, the convention was to move the beam forward five or six centimetres to land on the target. Hardly a precise technique. Maller refined it using MRI to locate the exact position.
Maller and Fitzgerald have also been calculating what kind of magnetic pulses work best to tweak the brain. So far, the best results have comes from using a theta rhythm – bursts of 10-hertz activity followed by rests. It’s the rhythm that emanates from an actively learning brain and, if applied to isolated neurons in a dish, it will trigger them to form lasting connections. On the other hand, a slow monotonous one-hertz rhythm seems good for toning down brain activity – it’s used for the right-hand dorsolateral prefrontal cortex which is slightly overactive in depression. Like African drum talkers, the team seems to have found a way to speak the brain’s own language. And patients have responded.
Maller’s iPhone bears a picture of a figure eight-shaped chocolate cake – a replica of a TMS coil baked by a grateful patient. She’d been depressed since her teenage years but during her session something changed. “So this is what it feels like not to be depressed, I’d forgotten,” she told Maller. Her treatment lasted only 15 minutes. Years later she is still depression-free. Her case is “not uncommon”, says Maller, although the length and number of treatments differs for each patient.
So what does TMS actually do to the brain? A neuron is like a clever wire that solders itself to others to form a circuit. The strength of the solder can grow weaker or stronger, largely depending on how often the circuit is used – neurons that fire together wire together. Neurons are also sensitive to the charge around them; they fire when their internal charge starts to neutralise or “depolarise”. The TMS-induced current probably depolarises groups of neurons, triggering them to fire together. That strengthens circuits in the command centre, helping it override the emotional signals coming from the hyperactive amygdala.
Perhaps a depressed brain is like a stream with an eddy in it and TMS sends a surge of activity that clears the eddy. It’s a nice metaphor and for the moment, as good as anything in terms of an explanation.
The truth is people do not understand how TMS works. And that does not discredit the technique, says Loo. She points out that people do not understand how lithium – a mainstay of treatment for bipolar disorder – works either. But there’s little doubt that TMS can treat depression, thanks to the well-controlled clinical trials.
A do-it-yourself movement has sprung up around battery-charged kits, with practitioners eager to experiment on themselves.
CREDIT: ANTHONY CALVERT
Alas, the situation with tDCS couldn’t be more different.
A TMS machine costs tens of thousands of dollars and can only be used in a clinic by a carefully trained operator, but you can pick up a nine-volt tDCS kit online for a couple of hundred dollars, or make your own for less. And there are no laws against zapping yourself with one.
The technique escaped into the do-it-yourself world after a 2010 study funded by the US defence department. It showed that tDCS doubled performance scores on DARWARS Ambush! – a training game to identify concealed threats and used by recruits headed for Iraq. The result was published by Vincent Clark and colleagues at the Mind Research Network in New Mexico in the journal Neuroimage. Clark had first carried out fMRI to locate the areas that were most active as the subjects learnt the task, which he believes explained the phenomenal result. “I think that’s key,” says Clark. His finding was replicated by Brian Falcone and Raja Parasuraman at George Mason University in Virginia – but because Clark’s name appeared on the publication, it was technically not an “independent” study. A similar study by Andy McKinley and colleagues at the Airforce Research lab in Ohio used tDCS to train airmen to recognise fuzzy objects in radar images, and found an improvement of only 25%.
She imagines a future in which depressed patients would be sent home
with the headsets
There’s been no shortage of reports on the effects of tDCS. There are claims it can enhance the learning of mathematical tasks and language, improve working memory, the ability to concentrate on boring tasks, reduce chronic pain and improve the recovery of motor function in people who’ve had a stroke. No wonder do-it-yourself enthusiasts are on to it!
No one expects these home-based experimenters to make a sensible contribution to science. Some researchers aren’t even too concerned about the possibility of harm among this group. People have ended up with skin burns – the electrodes heat the flesh which is why damp sponges are used. But “the brain is very good at wiping out any effects of NIBS [non-invasive brain stimulation], hence the difﬁculty of getting any effects at all, let alone ones that last. We do have to be mindful of the possible dangers of using higher currents and longer, repeated stimulation”, wrote Vincent Walsh, a neuroscientist at University College London, in the journal Brain Stimulation in 2013.
But Walsh is more concerned about what his own colleagues are doing in the labs than in the personal experiments of gamers. “The bullshit claims about therapies are masking what might be some useful science. I’m really embarrassed that an area of my research has come to this,” Walsh wrote in an email to me.
While his own papers have shown that tDCS improved numerical skills, he has since become extremely circumspect about some of the claims and is concerned by how they have been picked up by the media – and the public. There are, for instance, rumours that parents are trying the technique on their children.
So what’s going on here? Again, no one really knows how tDCS works. One proposed mechanism is that by generating a current through the watery tissues of the brain it changes the propensity of neurons to fire. Strangely, the technique is supposed to produce opposite effects depending on whether brain tissue lies under the positive terminal (anode) or negative terminal (cathode). Tissue lying close to the anode is supposed to grow more excitable, while tissue lying beneath the negative terminal grows less responsive. Many experienced neuroscientists find this idea mystifying!
They also wonder how a nine-volt battery can get an electric current reliably flowing past the resistance of the skull and into the brain. By contrast, a TMS pulse generates a momentary current that is thousands of times greater than that of tDCS. And it clearly has an effect – as the thumb jerk shows.
CREDIT: ANTHONY CALVERT
Evidence that tDCS can send current flowing into the brain did come from a report from Michael Nitsche and Water Paulus at the University of Gottingen in 2000. They showed that after tDCS was applied they could measure a change in the general excitability of the brain – specifically the motor-evoked potential, a measure of how much electrical power is required to induce a muscle jerk. Voila: tDCS could change the human brain. Their finding opened the floodgates to testing its effects. “Given the technique was easy, cheap and safe, of course people ran with the ball,” explains Neil O’Connell, a clinical researcher at Brunel University, London. The problem is “there’s been an awful lot of exploratory studies but no one replicates them with rigour”.
The use of tDCS to relieve chronic pain, for instance, has been disappointing. Some early studies from Harvard suggested the technique helped sufferers. But last year O’Connell participated in a thorough Cochrane review of all the published papers on the effects of tDCS in people suffering fibromyalgia, back pain or pain after a stroke. There was no statistically significant effect. O’Connell and colleague Benedict Wand from the University of Notre Dame in Western Australia said so in a report in theBritish Medical Journal this April. It was titled: “Transcranial direct current brain stimulation for chronic pain: Not recommended; early promise is fading fast as trial methods improve.”
This finding is no surprise to Jared Horvath. Five years ago, he began researching the effects of tDCS at the Berenson-Allen Center for Non-invasive Brain Stimulation at Harvard Medical School. Two years in, he could find no significant effects. So after relocating to the University of Melbourne to do a PhD, he changed tack. He gave up on his own experiments and spent the next two years analysing more than 1,000 published papers which claimed that tDCS enhanced the mental performance of healthy subjects. About 400 of the papers looked convincing – they included sham controls (people who thought they were being treated but weren’t receiving a current) and the statistics appeared solid.
But the true test of science is replication by an independent lab. And that was a big challenge – different labs were all using different protocols. For instance some reported memory and language improvements based on a single session; others employed multiple sessions but all with different treatment schedules. Only 59 treatments were similar enough to allow him to compare the results – all were single session stimulation. The bottom line: Horvath could find no replicable result. His study generated a huge controversy when published online in Brain Stimulation this January.
In another paper published in January 2014 in Frontiers in System Neuroscience, he argued why it would be hard to get replicable results. It is difficult to deliver the technique in a reliable way. The thickness of a person’s hair or the sweatiness of their skin can change the amount of current delivered. Whatever else a person is doing with their brain at the time of tDCS – say bouncing a ball or doing mental maths – can also change the apparent degree of brain stimulation received, as measured by the muscle twitching test.
And it turns out it’s not always so easy to make sure that the sham control is properly done. Ideally, neither the subject nor the operator should know whether the stimulation is fake or not. But that’s not always the case, agrees O’Connell who outlined the problem in a paper in PLoS ONE.
But the coup de grace was the muscle twitching test or motor-evoked potential. These evoked muscle twitches had given tDCS credibility by showing that the technique changed the excitability of the brain motor cortex. Horvath found these measures fluctuated wildly – by 10-fold within a single experiment. Overall, since 2000 the reported measures of how much tDCS affects the muscles have been declining. “That’s a compelling finding,” says O’Connell. “Motor-evoked potential is the only solid measure we’ve got.”
Walsh backs up Horvath’s analysis. “The evidence for therapeutic efficacy is about as strong as the evidence for crystals replacing vaccination. No one will face the fact that not a single major cognitive claim has replicated across labs. No one seems surprised that most of the claims come from a handful of labs. The bad science needs to be outed in the literature, not the newspaper. That’s where I’m putting my efforts.”
So what does this mean? Maybe the field of tDCS is a victim of what’s known as the decline effect. This is often seen in drug studies. Typically, an initial set of publications report a significant effect which vanishes under the weight of larger studies. A well-known example includes the relative benefits of second-generation antidepressants.
Colleen Loo is convinced that tDCS will not go the way of the decline effect – at least not for the treatment of depression. For Loo, the controversy invokes a case of déjà vu. She witnessed the same contradictions with early trials of TMS. “Studies of TMS always had positive and negative results – you expect that.”
How much difference can the charge from a nine-volt battery make to our brains? Experts are divided.
Loo is also not surprised by Horvath’s finding that single sessions of tDCS had little effect on memory or other cognitive enhancements. But she is confident of her own findings on the use of the technique for depressed patients. In her trials, patients receive 20 sessions of anodal stimulation directed at the dorsolateral prefrontal cortex over a month. She uses particular patterns of stimulation. “Just like TMS, the way we stimulate is important.” She is also confident that her patients can’t tell the difference between fake and real treatment – and nor can the operators. So far her trials, and those of colleagues in Brazil, are showing moderate improvements to patients with depression.
She believes these trials will help refine and improve the technique so that it follows the trajectory of TMS. “My own hunch having developed both is that they are very similar in efficacy.”
But what excites Loo about tDCS is its low cost and portability. She imagines a future in which depressed patients would be trained to use it in the clinic and then sent home with the headsets. Fitzgerald imagines a similar treatment for people in the early stages of dementia.
Clark, the researcher who found the stunning effects of tDCS on war-training games, believes that with more funding researchers will replicate each other’s findings and the pay-off will be huge. “We’ve been focused on pharmaceuticals for so long. Applying direct energy to brain areas could be a huge revolution in medicine. ”