|Editorial| tDCS and the Hype Cycle
Not long back, Dr. György Buzsáki of New York University (NYU) presented some research that, on the face skull of it (forgive the pun), put into question a whole host of exclamations made over the years by transcranial electrical stimulation (tES) research.
Let’s be clear from the outset. The study that got caught in the headlines was a side-note on a presentation about Rhythms in Mammalian Brains. This side note was that Buzsáki’s cadaver studies (grizzly, I know!) showed that up to 90% of the current was redirected by the skull. Buzsáki’s main bag is cortical oscillations and rhythms of the brain. The main purpose of the study was to uncover if sinusoidally modulated tES (alternating current, see figure 1) that mimic slow cortical oscillations and repetitive high-frequency Gaussian stimulus trains can entrain neuronal spiking.
It’s worth noting that these studies didn’t use transcranial direct current stimulation (tDCS).
It turns out you need a lot of juice to entrain neuronal spiking extracortically, much more than the standard 1-2mA. Buzsáki suggests >4mA is necessary, which can be dangerous to humans.
Unsurprisingly this has brought the mechanism of action for tES into question (though some suggest it hasn’t at all). If such a small amount of stimulation actually gets through to the brain, what is the mechanism for action? This is a hotly debated topic and the answer is elusive (I’ll leave it at that for now).
I will dedicate further posts to outline previous and current theories for the mechanism of action for tDCS (a form of tES, as each form of stimulation needs to be tackled independently).
As interesting as this mechanistic work is, I feel the point was massively missed. Rather than focusing on the lack of theory behind our mad search for answers on how tES works, the focus has been on the shunting effect.This, in turn, has snowballed into a messy debate over the legitimacy of previous tES research.
To highlight the overall reaction to this post, I’ve pulled out some quotes from some interested parties on twitter. Many delivered tweets that suggest tES was in trouble:
Alejandro de le Vega wrote, “I guess the positive flipside of the tDCS cadaver study is all those DIY kits aren’t doing much harm… or well much of anything really”.
Vaughan Bell wrote, “Cadaver study casts doubts on whether tDCS can actually stimulate the brain at all”.
Other’s wrote that this is nothing new:
Rachel Nowak wrote, “Didn’t we know this? Cadaver study casts doubts on tDCS, most current redirected by skin on skull”.
Some argued against the theoretical rationale:
JM Spinocchia wrote, “Hypothesised mechanism of tDCS is not neural stimulation but potentiation”.
It’s worth noting that tDCS is a common theme among tweets. However, as previously stated, tDCS wasn’t part of Buzsáki’s study. There seems to be a laziness in the social domain to group all forms of tES as the same thing. If some negative results (and possible positive results too) transfer from one form of tES to another this will likely lead to a very inaccurate picture as to the true efficacy of each independent tES device.
You might be thinking, “this is just twitter, who cares?” Okay, so let’s have a look at one of the sources of information that caused this wave of enthusiasm, the sciencemag article.
First things first, this article has done a fantastic job at grabbing a lot of attention both within and outside the tES field. With a snappy title sat under a grizzly picture, followed by some quality writing; this makes for some convincing journalism. Journalism being the key word! This site may have Science in its title but this piece of work is purely editorial. At least, in my opinion, it is. There are several reasons why I say this. The research that the author discusses was presented at a conference prior to the peer-review process. What makes up the bulk of this article is a collection of different opinions from within the discipline (albeit their opinions may be grounded in research) laid out in a back and forth format. What it isn’t, is a collection of facts and findings.
Which is fine if this article is taken at that level. What seems to be a popular move in Science at the moment, is to readily accept new information that discredits a field (which may well be in response to an awareness of positive publication bias) without the same scepticism as we do with positive results.
Assume for a moment that the results were the opposite: That the cadaver study found that tES did all the things that the discipline claims (realistically this fictitious scenario is problematic because dead brains… are well, dead; all the claims come from the living but play along anyway). In this scenario; would we be so quick to accept this information? Maybe ten years ago we would, when the revitalised novelty of tDCS/tES was steadily increasing in popularity.
We’ve been riding the tDCS wave for a while now and according to the Gartner Hype Cycle, it may well be overdue a pretty drastic fall in popularity.
Below are the Gartner and Heathers Hype Cycles. As you can see the Heathers version is just the Gartner version adapted for comedic effect. These cycles show the response pattern of a new technology, which also works nicely in the context of a new idea. Allow me to walk you through the process, using technology as an example. (1) The new technology emerges. (2) The expectations for this technology are overinflated. (3) This reality becomes apparent with a crash in enthusiasm for said technology. (4) The expectations for the technology become more realistic. (5) Uses for the technology become more refined.
A great example is Speech Interpretation and Recognition Interface (or SIRI, as most of us know it). When it first came to our attention (trigger) the outlandish idea was that we’d never have to type again (peak of inflated expectations). We soon realised that SIRI wasn’t all it was cracked up to be (trough); one word is mistaken for several similar sounding words and vice-versa. Developers get better at working within the limitations and users have some degree of faith restored in the technology (slope of enlightenment). Finally, the developers become expert at working within these limits and users become comfortable using the device (plateau of productivity).
The Heathers version of the Hype Cycle describes the span of the scientific research hype, rather than the Gartner Hype Cycle, which is talked about more in a marketing sense. The one discernable difference (other than the use of language) is the last section: The Unfundable Zone. Finding out all the limitations and refining a technology is really the end point for research, whereas, in a marketing sense, the bugs have been squashed, the limitations are known and what is left is arguably the most lucrative position.
I’m going to go out on a limb and hazard a guess at the location of tES in the Heathers Hype Cycle (because it’s more fun than the Gartner version) and say it’s approaching the peak of Mount Bullshit. The lovely graphic tells us what’s on the other side; a period of uncertainty and unpopularity for research that uses tES. As we travel down The Slough of Despond there is an intrinsic risk that the technology or idea may be cast aside. When an idea in science is cast aside, there is the possibility that we throw out not only the bad stuff but the good stuff too. Everything that was learnt heading towards and beyond Mount Bullshit has been marred bad science by association.
So if tDCS deserves to be thrown off the cliff of Mount Bullshit, maybe we should be a little more cautious when it comes to accepting any result, as opposed to having our sceptics eye keenly watching the positive results and just haphazardly accepting the negative ones.
I’m Not Defending tES
This hasn’t been a defence of the tES literature, nor has it been an attempt to tear down the cadaver research. The research coming out of Dr. György Buzsáki’s lab has great potential. Any research that attempts to understand the mechanism of action behind various forms of tES is doing bloody good work and I look forward to reading the published article.
I’ve purposefully not gone into the arguments for and against this research. This has been done in detail elsewhere. If you want to have a look at some of the back and forth, check out Neurocritic’s blog and the sciencemag post (make sure to check both comments sections, that’s where the action happens).
One thing I will mention is that I don’t think it’s particularly helpful to group all forms of tES together as if they are the same thing. Without a doubt, tDCS has taken an indirect hit via the popularity of the sciencemag expose. tDCS wasn’t used in the experiment, yet a large volume of comments have associated tDCS directly with the results of this research.
A mechanistic model of coordinated rhythmic movement as a model for assessing the efficacy of tDCS.
Neurocritic,. (2016). The Neurocritic: What We Think We Know and Don’t Know About tDCS.Neurocritic.blogspot.co.uk. Retrieved 22 June 2016, from http://neurocritic.blogspot.co.uk/2016/04/what-we-think-we-know-and-dont-know.html
Quintana, D. & Heathers, J. (2016). 9: What happens if your research is wrong?. Everything Hertz. Retrieved 22 June 2016, from https://soundcloud.com/everything-hertz/9-what-happens-if-your-research-is-wrong
Underwood, E. (2016). Cadaver study casts doubts on how zapping brain may boost mood, relieve pain. Science. http://dx.doi.org/10.1126/science.aaf9938
Voroslakos, M., Olivia, A., Brinyiczki, K., Zombori, T., Ivanyi, B., Buszaki, G., & Berenyi, A. (2015). Targeted transcranial electrical stimulation protocols: Spatially restricted intracerebral effects via improved stimulation and recording techniques. Presentation, Neuroscience 2015.