The Science and Snake Oil of Neurostimulation
What if learning a skill only required donning a cap that used electric currents to tickle your brain in just the right way?
This is the question posed in an explanatory video released in tandem with a new report which claims to have used transcranial direct current stimulation (tDCS) to improve the acquisition of piloting skills in test subjects. It certainly is an enticing idea, one which could put an end to spending hours buried in textbooks or wasting away in corporate training seminars—but could it ever really be that easy?
TDCS is a form of neurostimulation in which a constant, low current is delivered to a certain area of the brain via electrodes attached to the scalp. While tDCS is a relatively novel form of neurostiumaltion (as opposed to other methods such as Transcranial Magnetic Stimulation), using direct currents to stimulate parts of the brain for therapeutic purposes dates back to at least the early 19th century, when the Italian physicist Giovanni Aldini used a direct current to improve the mood of a patient suffering from “melancholy madness.”
The therapeutic use of stimulating the brain with direct currents has gone in and out of fashion in the 200-odd years since Aldini’s experiments, although in the last 15 years there's been a resurgence of interest in using tDCS for everything from cognitive enhancement to treatments for neurodegenerative diseases, thanks to technical improvements that have made jolting the brain with electricity safe for human use.
The most recent tDCS experiments were carried out by the California-based HRL Laboratories, a research unit owned by Boeing and General Motors. The results coming out of HRL, published in the February issue of Frontiers in Human Neuroscience, claim to have used tDCS activate the brain areas critical for piloting as observed in six expert commercial and military pilots in a number of test subjects as they used a flight simulator, thereby decreasing the variability in performance among those learning to pilot an aircraft.
According to a press release from HRL, “While previous research has demonstrated that tDCS can both help patients more quickly recover from a stroke and boost a healthy person's creativity, study is one of the first to show that tDCS is effective in accelerating practical learning.”
The HRL team was led by Matthew Phillips, a neurocognitive systems researcher, who had 32 test subjects come in for four consecutive days of flight training in a simulator. After receiving basic instructions on how to operate the flight simulator, the test subjects donned a cap outfitted with electrodes which would be stimulating the prefrontal and motor cortex areas of their brains.
In a double blind experiment, some of the test subjects received neurostimulation patterned after the brain activity of expert pilots throughout the duration of their flights, while others received sham stimulation (their cap would stimulate their brain for the first minute of the flight simulator, and then it would shut off, unbeknownst to the subject).
According to Phillips, the objective of HRL’s experiments was to determine whether tDCS was effective in modulating the acquisition of a skill in a group of people. Basically, if you’re handed a bunch of newbies in a particular field and after a training session these new recruits are all performing at a similar level, that speaks to the efficacy of that training regimen. What Phillips and his team wanted to see was whether tDCS could assist in this process by decreasing the variance within a group trying to acquire a certain skill (in this case, learning to pilot and land an aircraft).
At the conclusion of their experiments, Phillips and his colleagues found that there was in fact a decreased variability in training among the test subjects who received stimulation to their prefrontal cortex—those who received stimulation showed more consistency in their performance than those who did not.
Given the team’s aims in the experiment, this is an encouraging result, but it is not the same as using tDCS to improve a subject’s piloting skills, as headlines from the HRL press release (“Learn to Fly a Plane from Expert-Pilot Brainwave Patterns”) and other media reports (“Scientists Discover How to Upload Knowledge to Your Brain” or even Gizmodo’s skeptical take) might have you believe. The subjects always performed poorly on the tasks in the experiment (as measured in the amount of g-force at landing) and didn’t show significant improvement over the course of the experiment—the only thing that changed is that neurostimulation allowed them to fail with more consistency.
“It's the kind of thing that just plagues the field and makes tDCS look so silly"
“There are many ways of measuring and characterizing the behavioral performance in this experiment,” Phillips told me. “We wanted to see if there was any benefit of applying tDCS to a population that is being trained in a very complex skill such as piloting an aircraft. It’s not necessarily about lowering the g-force at landing—which wasn’t significantly different between the stim and sham groups— but that behavioral performance is more consistent throughout the day, which we speculated might lead to further effects down the line.”
HRL’s results are encouraging for the application of tDCS to improving some aspects of working memory, but whether or not tDCS will ever prove to be useful for improving your piloting skills (or anything else, for that matter) is the subject of much debate. Although there have been some encouraging initial results in the use of tDCS for treating depression, its efficacy in other areas, such as memory improvement, skill enhancement, or the treatment of neurodegenerative diseases like Alzheimer’s, is much more hotly contested.
While the researchers at HRL see an emerging consensus from researchers that tDCS is an effective for bolstering working memory, others remain more skeptical about the benefits that can be derived from this type of neurostimulation. Chief among the skeptics is Jared Horvath, a Post-Doctoral researcher at the Melbourne Graduate School of Education, who published a meta-analysis in 2015 examining research that claimed tDCS modulated everything from working memory to language production. After compiling this exhaustive analysis however, Horvath was surprised to find that tDCS did not appear to have any significant effect on any cognitive measure whatsoever.
“Working memory is one of the more impressive areas in terms of getting results from tDCS,” Horvath told me. “But when you do a meta-analysis and bring all the working memory data together, you find that tDCS has no significant effect on working memory. It doesn't mean that tDCS is not improving some people's working memory under some conditions, but it’s just so crazy variable that we can't predict when, where or how. And if I can't tell you exactly what's going to happen with some sort of certitude, then the results become meaningless.”
The problem, as Horvath sees it, is that researchers working with tDCS were trying to run before they learned to walk. It seemed that every new publication dealing with this particular brand of neurostimulation announced some new, and often contradictory, miracle application for tDCS: Appreciate abstract art! Detect Threats! Facilitate insight! Better…tongue twisters?!
Yet despite these apparent applications of tDCS, the scientists were generally unable to say why or how they were coming about. And if the goal of the science is eventual clinical application, then the inability to explain the phenomenon means it amounts to little more than an interesting spectacle.
“Whatever you're looking for, apparently tDCS helps,” said Horvath. “But if you can't explain it, then it's really just an aberration. It’s cute, but it’s an anecdote.”
As Horvath pointed out, when you’re using small enough sample sizes, it’s pretty easy to make experiments ‘demonstrate’ anything you want them to. Furthermore, due to the singular nature of each individual’s brains, replicating studies where subjects were performing incredibly complicated tasks (like flying an airplane, for instance) was next to impossible without a detailed understanding of what was physically happening in the brain while it was being stimulated—information the researchers didn’t, and for the most part still don’t, have.
Without being able to explain precisely what was happening to subjects receiving tDCS or replicate successful tDCS experiments, this new technology was of very little use to anybody. It certainly wasn’t getting closer to clinical applications and perhaps even worse, the proliferation of miracle tDCS applications without explanations meant that neuroscientists and the public at large were having a hard time taking tDCS research very seriously.
Yet in the aftermath of Horvath’s paper, he said he has noticed a widespread return to the basics among researchers looking at tDCS and its effects on the human brain. Rather than having small groups of test subjects perform complicated tasks (where it would be difficult to describe the brain states of the subjects even without trying to factor in the tDCS effect), researchers are increasingly beginning to look at how tDCS affects the brains on much simpler tasks in order to get a basic understanding of the brain mechanisms at work. Horvath cited his own research, which involves subjects being stimulated with tDCS and then proceeding to push a button hundreds of times, as an example of this neo-simplicity within tDCS experiments.
While it might make for less glamorous headlines, this type of work is necessary to get the data that will turn tDCS from a futuristic fiction into a science with manifold clinical applications.
“Now I think most of the field is predicated on going backwards to rigorous foundational research,” said Horvath. “We’ve had our fun, but now it’s time to get back down to work.”
Although Horvath acknowledges there is some merit to the latest HRL research, he cites it as an example of the type of research that generates good headlines without contributing much helpful data to the field.
“I'm going to be honest: I have no idea what the point ? It’s not a bad study, it’s just really difficult to reconcile with everything else that’s going on in the field right now.”
One of the important aspects of the recent HRL study was their use of electroencephalography (EEG) and functional near infrared spectroscopy (fNIRS) to monitor the brains of the subjects while they were piloting the aircraft. Phillips and his team are still in the process of parsing through this data and they hope that it will ultimately reveal the underlying brain states which would explain the decrease in skill acquisition variability they saw in their piloting experiments.
In the meantime, Phillips is in agreement with Horvath that the brain mechanisms at work in tDCS are complicated, but as he also points out, the HRL study is just a first step in trying to figure out what’s going on. Until that point, it’s unlikely that tDCS will see much application outside of laboratory experiments.
“Pinning down the brain states associated with the baseline of the task is a huge challenge,” said Phillips. “Piloting an aircraft can take years to learn, and we had four days. It’s a simple paradigm to assess some of the aspects that go into longer term learning process. It’s a complex picture, but this is just one of the initial first steps to peer into the brain.”