Neuroscientists Say They've Found an Entirely New Form of Neural Communication(sciencealert.com)
sciencealert.com
Neuroscientists Say They've Found an Entirely New Form of Neural Communication
https://www.sciencealert.com/neuroscientists-say-they-ve-found-an-entirely-new-form-of-neural-communication
47 comments
the abstract was mentioned in this post https://news.ycombinator.com/user?id=SubiculumCode
> ...low‐frequency periodic activity (<1 Hz)... propagate with speeds around 0.1 m s−1
so thats a wavelength of ... at least 10cm! not much information density here... this is probably the reason nobody is interested much in these effects. Even if they can chaotically influence the state of the system, it won't contain much useful information...
> ...low‐frequency periodic activity (<1 Hz)... propagate with speeds around 0.1 m s−1
so thats a wavelength of ... at least 10cm! not much information density here... this is probably the reason nobody is interested much in these effects. Even if they can chaotically influence the state of the system, it won't contain much useful information...
One theory is that the waves themselves don’t carry much information, but set the stage for other kinds of transmission or computation.
In the rodent hippocampus, for example, theta oscillations (3-7ish Hz) don’t tell you much about the animals’ spatial position. The spiking rate of place cells contains some information about where the animal is, but it’s fairly coarse. However, combining spiking and the phase of the slow oscillation (e.g., cell is firing at the trough vs peak) provides much more precise information about where the rat is.
A related idea is that oscillations help route information by alternatingly facilitating and suppressing different areas. At one phase of the oscillation, information might preferentially flow from visual areas to higher-level areas. At another, the visual input might be suppressed while memory structures get “their turn to talk.”
In the rodent hippocampus, for example, theta oscillations (3-7ish Hz) don’t tell you much about the animals’ spatial position. The spiking rate of place cells contains some information about where the animal is, but it’s fairly coarse. However, combining spiking and the phase of the slow oscillation (e.g., cell is firing at the trough vs peak) provides much more precise information about where the rat is.
A related idea is that oscillations help route information by alternatingly facilitating and suppressing different areas. At one phase of the oscillation, information might preferentially flow from visual areas to higher-level areas. At another, the visual input might be suppressed while memory structures get “their turn to talk.”
note that my observation is not that the information contained is not "zero usefull information" but "not much usefull informmation". I certainly think a local clock would be usefull, and contains very little information.
But any large amount of information density is simply not supported by these observations due to simple wave aspects. A low information signal (for example highly predictable clock, or some other low frequency, high brain volume signal say perhaps a breathing rate not actually the breathing frequency) can still be important.
I was only replying to the information processing comment
But any large amount of information density is simply not supported by these observations due to simple wave aspects. A low information signal (for example highly predictable clock, or some other low frequency, high brain volume signal say perhaps a breathing rate not actually the breathing frequency) can still be important.
I was only replying to the information processing comment
?
I…don't see the argument here. I was just trying to provide some actual examples from neuro.
I…don't see the argument here. I was just trying to provide some actual examples from neuro.
Okay but you have to keep in mind that there are multiple waves, and what you might use them for. Think of how a sonar works. Sonars must associate and compare the timing of events, just like neurons. They use as many different frequencies for that as possible (used to be 4-8 frequencies but the state of the art is presumably much more)
Let's say I have brainwaves at 0.5Hz, 4Hz, 8Hz and 20Hz wave. And let's assume they are all non-harmonic. If you have an event, and you can quickly inspect the position and phase of all these waves, you now have an extremely accurate and long-period reference for the time the event occurred (think of the individual waves as the numbers on a clock).
So the accuracy would be a decent multiple of the highest frequency, so let's say 20-50ms resolution. And the repeat is the multiple of all the wavelengths, so this would give neurons a trivial way to understand or plan events up to 640 seconds away from the current moment, as well as an easy way to refer to the exact moment a prediction will come true, and how different events relate to eachother in time.
It's brilliant.
Let's say I have brainwaves at 0.5Hz, 4Hz, 8Hz and 20Hz wave. And let's assume they are all non-harmonic. If you have an event, and you can quickly inspect the position and phase of all these waves, you now have an extremely accurate and long-period reference for the time the event occurred (think of the individual waves as the numbers on a clock).
So the accuracy would be a decent multiple of the highest frequency, so let's say 20-50ms resolution. And the repeat is the multiple of all the wavelengths, so this would give neurons a trivial way to understand or plan events up to 640 seconds away from the current moment, as well as an easy way to refer to the exact moment a prediction will come true, and how different events relate to eachother in time.
It's brilliant.
The point is that the field strengths is too low or the distance to great for ephaptic coupling to occur.
> it is unknown the degree to which it contributes to standard computations and processing within different neural systems
And that is not even the topic of the article:
"the present study shows that the in-vitro slow periodic activity, mediated by the NMDA spikes, can propagate non-synaptically by a mechanism consistent with ephaptic coupling. This study implies that the effect of ephaptic coupling could play an important role in the propagation of neural activity under normal physiological conditions as well as pathological situations."
Important: "slow periodic activity"
Also important: it's in vitro: https://en.wikipedia.org/wiki/In_vitro
"In vitro (meaning: in the glass) studies are performed with microorganisms, cells, or biological molecules outside their normal biological context. Colloquially called "test-tube experiments"..."
What's new: "Results support the hypothesis that endogenous electric fields, previously thought to be too small to trigger neural activity, play a significant role in the self-propagation of slow periodic activity in the hippocampus."
In short, they claim to have proved that under some special conditions the propagation can happen even by what was believed to be too small fields.
But as it's the first paper about that, it's reasonable for the observers to wait for other scientists to repeat the experiment and review the details of it... before making any conclusions.
And that is not even the topic of the article:
"the present study shows that the in-vitro slow periodic activity, mediated by the NMDA spikes, can propagate non-synaptically by a mechanism consistent with ephaptic coupling. This study implies that the effect of ephaptic coupling could play an important role in the propagation of neural activity under normal physiological conditions as well as pathological situations."
Important: "slow periodic activity"
Also important: it's in vitro: https://en.wikipedia.org/wiki/In_vitro
"In vitro (meaning: in the glass) studies are performed with microorganisms, cells, or biological molecules outside their normal biological context. Colloquially called "test-tube experiments"..."
What's new: "Results support the hypothesis that endogenous electric fields, previously thought to be too small to trigger neural activity, play a significant role in the self-propagation of slow periodic activity in the hippocampus."
In short, they claim to have proved that under some special conditions the propagation can happen even by what was believed to be too small fields.
But as it's the first paper about that, it's reasonable for the observers to wait for other scientists to repeat the experiment and review the details of it... before making any conclusions.
I remember reading an article a while back and it went something like this. They had a digital circuit which implemented some function (might have been a filter, but no 100% sure). They let engineers design the circuit to perform the function. Then they let a computer iteratively (ML?) place the components to achieve that same function. It got so good that it was able to use less gates but some how worked just as well.
They were really confused and finally realized that unbeknownst to them the computer had taken advantage of the fact a moving charge creates a magnetic field and vice versa. This fact was exploited by the computer (inadvertently, lol).
They were really confused and finally realized that unbeknownst to them the computer had taken advantage of the fact a moving charge creates a magnetic field and vice versa. This fact was exploited by the computer (inadvertently, lol).
I think this is what you're looking for: https://www.damninteresting.com/on-the-origin-of-circuits/
"Dr. Thompson peered inside his perfect offspring to gain insight into its methods, but what he found inside was baffling.
The plucky chip was utilizing only thirty-seven of its one hundred logic gates, and most of them were arranged in a curious collection of feedback loops.
Five individual logic cells were functionally disconnected from the rest— with no pathways that would allow them to influence the output— yet when the researcher disabled any one of them the chip lost its ability to discriminate the tones.
Furthermore, the final program did not work reliably when it was loaded onto other FPGAs of the same type."
"Dr. Thompson peered inside his perfect offspring to gain insight into its methods, but what he found inside was baffling.
The plucky chip was utilizing only thirty-seven of its one hundred logic gates, and most of them were arranged in a curious collection of feedback loops.
Five individual logic cells were functionally disconnected from the rest— with no pathways that would allow them to influence the output— yet when the researcher disabled any one of them the chip lost its ability to discriminate the tones.
Furthermore, the final program did not work reliably when it was loaded onto other FPGAs of the same type."
Yes! This is the exact article I was talking about. I spent quite a bit of time yesterday trying to find it but had no luck.
Thank you so much
:)
Probably “An evolved circuit, intrinsic in silicon, entwined with physics” [1]. They let a genetic algorithm come up with a better FPGA circuit, but it no longer worked when non-contributing parts of the circuit were pruned.
[1] http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.50....
[1] http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.50....
And it didn't work in another FPGA chip of the same model.
You mean something like this? https://en.wikipedia.org/wiki/Evolved_antenna
>The discovery offers some radical new insights about the way neurons might be talking to one another, via a mysterious process unrelated to conventionally understood mechanisms, such as synaptic transmission, axonal transport, and gap junction connections.
One of the most humbling books I have ever read is Wetware by Dennis Bray. There is an unbelievable richness when it comes to computation and storage/transport of information in biological organisms and cells that are simply unparalleled.
I feel this is especially relevant given that we are (again) living in times where people talk a lot about the 'intelligence' or power of contemporary learning machines.
One of the most humbling books I have ever read is Wetware by Dennis Bray. There is an unbelievable richness when it comes to computation and storage/transport of information in biological organisms and cells that are simply unparalleled.
I feel this is especially relevant given that we are (again) living in times where people talk a lot about the 'intelligence' or power of contemporary learning machines.
I know it's early stuff, and I am not an expert on this by any means, but I wonder if any of this would be something applicable to a biological means for "backpropagation"?
Just to clarify: As far as I am aware, there hasn't been any biological equivalent for neural learning akin to what is called "backpropagation" as used by common implementations of artificial neural networks in order to learn and generalize a task or function.
That there has only been a "forward signal" from neuron to neuron observed, with no way for error signals to move "backwards" adjusting the "connection weights" of biological neurons (note: this is not how biological neurons work at all; that much I do know - I am just simplifying a bit here).
Could this new communication form potentially be used for such error propagation for learning? I know the article seemed to make the idea that it only happens in "slow waves" and during "sleep"; but I wonder if it isn't actually happening all the time...?
Well - now I am just speculating on a subject I have little to no knowledge in, so I will stop here.
Just to clarify: As far as I am aware, there hasn't been any biological equivalent for neural learning akin to what is called "backpropagation" as used by common implementations of artificial neural networks in order to learn and generalize a task or function.
That there has only been a "forward signal" from neuron to neuron observed, with no way for error signals to move "backwards" adjusting the "connection weights" of biological neurons (note: this is not how biological neurons work at all; that much I do know - I am just simplifying a bit here).
Could this new communication form potentially be used for such error propagation for learning? I know the article seemed to make the idea that it only happens in "slow waves" and during "sleep"; but I wonder if it isn't actually happening all the time...?
Well - now I am just speculating on a subject I have little to no knowledge in, so I will stop here.
> Scientists think they've identified a previously unknown form of neural communication that self-propagates across brain tissue, and can leap wirelessly from neurons in one section of brain tissue to another – even if they've been surgically severed.
Now it sounds curious why it can't leap wirelessly from one brain to another one this way.
Now it sounds curious why it can't leap wirelessly from one brain to another one this way.
I've already tested this 6 years ago with a friend.
We had this idea that it could be possible by putting our skulls close together. We've tried a variety of techniques from thinking about a specific number from 1-10, one of us visualizing the answer as hard as he could, the other one waiting passively for the answer to 'occur to him'. After making sth like 100 tries and writing down our answers - the results were negative, same as for pure chance.
Then we've realized that maybe abstract information cannot be sent this way so we agreed on just two messages - very happy vs really sad, where we would imagine extreme emotions. Obtained the same negative results.
Being only slightly less skeptic than is popularly considered reasonable (or more skeptic - skeptic about skepticism itself too) I'd say your conclusion is not strictly valid. Perhaps you were doing it wrong. Like people who can't swim are doing it wrong until they get it. Or perhaps ice skating can make a better example. Imagine you have never ever used you hand and don't remember how to - can you expect it to start moving the way you want just by imagining the motion (you probably are aware of the difference between inner experiences taking place when imagining your hand moving and when actually moving it already) and waiting it to occur?
There is a Ukrainian psychotechnologist Oleg Bakhtiyarov[1] who has written (in a fairly scientific manner) a book titled "active consciousness" (only available in Russian unfortunately as well as a number of his lectures on YouTube) on how to develop mind functions most of the people don't know they have. I (as well as many other people, mostly those practicing freediving) have successfully mastered some of the techniques he teaches and can this way guarantee his job is far from being pure science fiction, I have not came far enough to develop telepathic skills though (yet I believe those interested in trying should probably go this way).
[1]https://en.wikipedia.org/wiki/Oleg_Bakhtiyarov
There is a Ukrainian psychotechnologist Oleg Bakhtiyarov[1] who has written (in a fairly scientific manner) a book titled "active consciousness" (only available in Russian unfortunately as well as a number of his lectures on YouTube) on how to develop mind functions most of the people don't know they have. I (as well as many other people, mostly those practicing freediving) have successfully mastered some of the techniques he teaches and can this way guarantee his job is far from being pure science fiction, I have not came far enough to develop telepathic skills though (yet I believe those interested in trying should probably go this way).
[1]https://en.wikipedia.org/wiki/Oleg_Bakhtiyarov
You're point is valid, yet I still find it very unlikely.
At some point after numerous tries you might even start fooling yourself (and others). I'll give the guy a chance though.
Whatever mechanism that can send electrical signals wirelessly doesn't necessarily map to big concepts like numbers, words, or emotions. Whatever the effect, if it exists, is not necessarily observable just by examining consciousness. It might be observable in other ways, like electrodes.
I think that the effect works over very small distances, also there are two skulls in the way.
So, you're saying we need to slice open a couple skulls and put them side to side to have "thought" transfer? :)
That sounds like a good movie plot.
Well.... I have no idea! But I think that there's a big difference between thought transfer and creating physical interactions between brains or between a brain and a machine (for example I think it's well known that if you zap brains with big pulses of magnetism then there are observed impacts) or between a machine or another brain and a part of a brain. I think that actually it would be more like transferring a skin irritation rather than a thought. But lets go with the movie plot...
Also, maybe there are more profound ways of doing the interfacing! If we really want to get Dr. Evil about it, could you use transplant technology to conjoin people and then integrate their brains to invade and manipulate their minds? I wonder what the psychology of a mind made from two minds would be like (ok, I'm guessing that the torture , shock, fear and anguish of the process would mean that the resulting singular personality would be singularly crazy, but leave that aside...) could you make a "greater" or usefully different mind through this process?
Hell - why stop with two?! Why not run a program where you deliberately create seed minds by exposing children to different kinds of experiences and then integrating them to create novel kinds of thinkers!
To earn full Crazy Nazi badges you could then go on to bring animal minds into the equation - what happens if you add a chimp or a wolf into a shared brain structure?
Ok - just to be clear, anyone who does this should be executed, but I think that the idea could make a pretty decent Netflix mini series :)
Also, maybe there are more profound ways of doing the interfacing! If we really want to get Dr. Evil about it, could you use transplant technology to conjoin people and then integrate their brains to invade and manipulate their minds? I wonder what the psychology of a mind made from two minds would be like (ok, I'm guessing that the torture , shock, fear and anguish of the process would mean that the resulting singular personality would be singularly crazy, but leave that aside...) could you make a "greater" or usefully different mind through this process?
Hell - why stop with two?! Why not run a program where you deliberately create seed minds by exposing children to different kinds of experiences and then integrating them to create novel kinds of thinkers!
To earn full Crazy Nazi badges you could then go on to bring animal minds into the equation - what happens if you add a chimp or a wolf into a shared brain structure?
Ok - just to be clear, anyone who does this should be executed, but I think that the idea could make a pretty decent Netflix mini series :)
ah, you may enjoy "A Fire Upon the Deep" and "A Deepness in the Sky" by Vernor Vinge :)
I've read both! A long time ago for aFUtD but Deepness just last year. I'd forgotten the group mind Tines beings in aFUtD but just looked them up on wikipedia. I've recently been re-reading Iain M. Banks and the minds in his culture series are referred to as having group minds.
>Now it sounds curious why it can't leap wirelessly from one brain to another one this way.
What do you think you're doing when you write?
(I'm only half-joking)
What do you think you're doing when you write?
(I'm only half-joking)
I'm surprised to see no mention of the two mechanisms that jump out at me as candidate explanations:
a) standard electrical induction
b) standard conductance via the extracellular matrix (https://en.wikipedia.org/wiki/Extracellular_matrix)
W/r/t b), they do claim that they sliced the hippocampus and then observed activity induced across the preserved gap, but it's not clear (from my light read) that the tissues weren't still touching such that the axons/dendrites were gapped, but the ECM was not; if this were the case, then I'd expect signals to propagate across severed neuronal connections via standard conductance in the ECM.
More please!
W/r/t b), they do claim that they sliced the hippocampus and then observed activity induced across the preserved gap, but it's not clear (from my light read) that the tissues weren't still touching such that the axons/dendrites were gapped, but the ECM was not; if this were the case, then I'd expect signals to propagate across severed neuronal connections via standard conductance in the ECM.
More please!
This is extremely interesting and basically boils down to the following neurological observation:
Anything (molecules, electric charge, magnetic fields etc) that is originally created by brain activity and itself detectable by other parts of the brain - what ever those are (mostly neurons, axons or dendrites I suppose) was factored in during the evolution of the brain and is henceforth utilized as a signal. That's why the brain is so incredibly complex and the classical neural network model doesn't cut it.
In other terms if X is systematically caused by A and is then affecting some B then that X will at some point serve as a signal factored into the neural processing and is no longer just noise! This happens evolutionarily over millions of years as well as dynamically in context of neural plasticity to some extent.
The same concept is true for anything in the body - which is why the separation of the body into modules / organs is problematic. But the effect is much more potent in the brain due to the high density of signalling and processing.
Anything (molecules, electric charge, magnetic fields etc) that is originally created by brain activity and itself detectable by other parts of the brain - what ever those are (mostly neurons, axons or dendrites I suppose) was factored in during the evolution of the brain and is henceforth utilized as a signal. That's why the brain is so incredibly complex and the classical neural network model doesn't cut it.
In other terms if X is systematically caused by A and is then affecting some B then that X will at some point serve as a signal factored into the neural processing and is no longer just noise! This happens evolutionarily over millions of years as well as dynamically in context of neural plasticity to some extent.
The same concept is true for anything in the body - which is why the separation of the body into modules / organs is problematic. But the effect is much more potent in the brain due to the high density of signalling and processing.
It almost sounds as if there is neuron like behavior that emerges due to the information available in the raw electrical phenomena of the brain. It could be analogous to a machine learning circuit where the substrate itself responds neuron like to the overall electrical loading of the board.
Reminds me of Arc proteins
https://www.newsweek.com/breakthrough-memory-formation-virus...
Is this different from retinal waves?
https://en.wikipedia.org/wiki/Retinal_waves
A bit different.
The retinal waves consist of action potentials, which the canonical way that neurons communicate with each other. To do this, the cells form a specialized structure called a synapse: the presynaptic cell releases some neurotransmitter into it, which receptors (sensors) on the its post-synaptic partner detect. Here, the idea is that neurons can communicate, albeit less efficiently, without actually forming a synapse.
As an analogy, synaptic transmission is like plugging your phone into a speaker: there’s a direct, explicit connection. This coupling is more like when a speaker picks up some noise from the cell phone’s modem even though they’re not connected.
The retinal waves consist of action potentials, which the canonical way that neurons communicate with each other. To do this, the cells form a specialized structure called a synapse: the presynaptic cell releases some neurotransmitter into it, which receptors (sensors) on the its post-synaptic partner detect. Here, the idea is that neurons can communicate, albeit less efficiently, without actually forming a synapse.
As an analogy, synaptic transmission is like plugging your phone into a speaker: there’s a direct, explicit connection. This coupling is more like when a speaker picks up some noise from the cell phone’s modem even though they’re not connected.
It reminds me a little of this impossible reactionless space drive.
Yes, and now we also know that they are "very excited about it", because neural networks also work wireless.
I mean, the brain tissue is full of charged particles and molecules, there are membrane potentials which trigger molecule transports at synaptic gaps. Now they've discovered that there are electrostatic fields inside the brain? Duh? Imagine what happens when they find out there are also magnetic fields!
I mean, the brain tissue is full of charged particles and molecules, there are membrane potentials which trigger molecule transports at synaptic gaps. Now they've discovered that there are electrostatic fields inside the brain? Duh? Imagine what happens when they find out there are also magnetic fields!
I would guess they have checked for that.
I would also assume they have ruled out other things like neurotransmitters from the reaction time etc
I would also assume they have ruled out other things like neurotransmitters from the reaction time etc
I would like to add that I don't blame either these researchers or the ones with that drive.
The most honest expression of the scientific method is to try as hard as you can to disprove your own theory.
Sometimes it's just really really hard not to fool yourself, and statistics could be called the science of how to not screw yourself scientifically.
The most honest expression of the scientific method is to try as hard as you can to disprove your own theory.
Sometimes it's just really really hard not to fool yourself, and statistics could be called the science of how to not screw yourself scientifically.
did they in fact physically separate the tissue in 2 halves? or was it a localized cut? even if they separated in 2 halves, how did they measure the electrical activity? with conductors? or optically? even if optically, did they assume light could not cross from one sensor to the other or did they prevent light from the first optical electric field sensor from entering the second optical electric field sensor, say by using different wavelengths and filter for each? even if they excluded optical leaking, you still need to exclude the possibility that there is simply an ambient electric field in the lab whose correlation at 2 points we are measuring. So you need a control setup without brain tissue to compare the correlations.
Otherwise we might be investigating the equivalent of: dot of light moving at twice the speed of light! ... by reflecting a laser beam of a fast rotating mirror on a distant screen... yes the dot can move faster than light, but the dot is not a physical signal, the dots at different times are separate signals with a common cause, and no signal was speeding beyond c...
at some point you realize that there may be more important questions to invest time and effort in...
Otherwise we might be investigating the equivalent of: dot of light moving at twice the speed of light! ... by reflecting a laser beam of a fast rotating mirror on a distant screen... yes the dot can move faster than light, but the dot is not a physical signal, the dots at different times are separate signals with a common cause, and no signal was speeding beyond c...
at some point you realize that there may be more important questions to invest time and effort in...
Might be Dunning–Kruger speaking, but I'd be surprised if that was not the case, and I am surprised that this seems to be a new area of exploration.
I mean, think about it: we're talking about a network of electromagnetically unshielded electrical devices, swimming in a sea of conductors, that is believed to have evolved its functionality over time. Since evolution is a random optimization process, why wouldn't it exploit electromagnetic properties of the environment, if they're reliable enough? It's not like evolution understands the concept of a wire, or chemical signal.
(Also reminds me of that story of an FPGA design generated by genetic algorithm, which ended up exploiting the electrical properties of the particular chip/board onto which it was uploaded.)
I mean, think about it: we're talking about a network of electromagnetically unshielded electrical devices, swimming in a sea of conductors, that is believed to have evolved its functionality over time. Since evolution is a random optimization process, why wouldn't it exploit electromagnetic properties of the environment, if they're reliable enough? It's not like evolution understands the concept of a wire, or chemical signal.
(Also reminds me of that story of an FPGA design generated by genetic algorithm, which ended up exploiting the electrical properties of the particular chip/board onto which it was uploaded.)
I know of a Max Planck Institute researcher who has written up a theory on how dendrites are really antennas and how neurons really communicate wirelessly. He is waiting for his Nobel price any day now...
This will soon be cited by quacks as “evidence“ for their overpriced uselesss therapies.
Bioresonance, Homeopathy, Reiki, ...? „Brain cells communicate wirelessly, and my method integrates with that communication!“
Bioresonance, Homeopathy, Reiki, ...? „Brain cells communicate wirelessly, and my method integrates with that communication!“
Telepathy of a sort, at least over microscopic distances!
Anyway, I wonder if the fields can be externally applied for therapeutic purposes (inducing sleep etc)
Anyway, I wonder if the fields can be externally applied for therapeutic purposes (inducing sleep etc)
This is actually what I work on.
The umbrella term you’re looking for is transcranial electrical stimulation, or tES. It is also called tACS (AC for alternating current), tDCS (for direct current), or tRNS (for random noise), depending on what type of electricity is being used. In most cases, conductive electrodes are placed on the head and a weak current is passed between them. This is thought to generate an electric field that polarizes the neurons between them and interacts with the cells’ on-going electrical activity.
The idea itself is very old: someone in 46 AD apparently cured headaches by applying electric fish to their foreheads. It’s been rediscovered a few times since then, and there’s been a huge boom since about 2000, with people trying it for everything under the sun in both healthy people and a wide variety of patients, with a lot of hype.
Despite that, it’s still unclear how it works, or indeed if it works at all. The electric field that reaches your brain is rather weak—-the skin and skull shunt away a lot of the current. A lot of the human experiments are not the greatest (small sample size, strange methodological problems) and there has been some concern that many of the effects either aren’t real or are due to confounds (e.g., placebo effects, or changes in attention/motivation due to the tingling sensation it causes on the scalp), and there’s been a lot of “counter-hype” about how the whole thing is bunk.
I am fairly certain it does something, though it needs to be applied (and evaluated) carefully. Over the last few years, I—-and some awesome collaborators—-have been recording neural activity from monkeys receving tES. We find neural effects at multiple levels ranging from single cells to long-range functional connectivity, along with effects on the animals’ behavior. We’ve got a few papers here: http://packlab.mcgill.ca/publications.html (look for my name; the lab does a lot of different things). There’s a lot more to be done though.....
The umbrella term you’re looking for is transcranial electrical stimulation, or tES. It is also called tACS (AC for alternating current), tDCS (for direct current), or tRNS (for random noise), depending on what type of electricity is being used. In most cases, conductive electrodes are placed on the head and a weak current is passed between them. This is thought to generate an electric field that polarizes the neurons between them and interacts with the cells’ on-going electrical activity.
The idea itself is very old: someone in 46 AD apparently cured headaches by applying electric fish to their foreheads. It’s been rediscovered a few times since then, and there’s been a huge boom since about 2000, with people trying it for everything under the sun in both healthy people and a wide variety of patients, with a lot of hype.
Despite that, it’s still unclear how it works, or indeed if it works at all. The electric field that reaches your brain is rather weak—-the skin and skull shunt away a lot of the current. A lot of the human experiments are not the greatest (small sample size, strange methodological problems) and there has been some concern that many of the effects either aren’t real or are due to confounds (e.g., placebo effects, or changes in attention/motivation due to the tingling sensation it causes on the scalp), and there’s been a lot of “counter-hype” about how the whole thing is bunk.
I am fairly certain it does something, though it needs to be applied (and evaluated) carefully. Over the last few years, I—-and some awesome collaborators—-have been recording neural activity from monkeys receving tES. We find neural effects at multiple levels ranging from single cells to long-range functional connectivity, along with effects on the animals’ behavior. We’ve got a few papers here: http://packlab.mcgill.ca/publications.html (look for my name; the lab does a lot of different things). There’s a lot more to be done though.....
How does transcranial magnetic stimulation differ from applying electrical current directly?
The general idea is similar, but the details are quite different.
As you might remember from physics, moving a conductor relative to a magnetic field induces an electric current in the conductor. In a generator, the magnet is often fixed while the conductor is spun around by a water, wind, or steam-driven turbine. Neurons are conductive but tricky to move, so we move the magnetic field instead. TMS uses strong electromagnets to generate magnetic pulses. The changing field induces currents in the neurons.
Magnetic fields are not attenuated by the skull or scalp, so the resulting field is much, much stronger. In practice, TMS often produces an initial burst of activity, followed by a longer period of suppression. This can be used to temporarily and reversibly inactivate a tiny chunk of the brain, which is great for research. It may also be a way to “reset” neural circuits that have gotten stuck in some weird state (this is one hypothesis about depression, and TMS does seem to help with it).
I think of tES as “nudging” the neurons’ own activity towards some state that you want, while TMS is more like a sharp rap. It’s not obvious if one is always better than the other. TMS devices are pretty large (lots of capacitors) and fixed, while you could imagine tES being more suitable for use outside a clinic.
As you might remember from physics, moving a conductor relative to a magnetic field induces an electric current in the conductor. In a generator, the magnet is often fixed while the conductor is spun around by a water, wind, or steam-driven turbine. Neurons are conductive but tricky to move, so we move the magnetic field instead. TMS uses strong electromagnets to generate magnetic pulses. The changing field induces currents in the neurons.
Magnetic fields are not attenuated by the skull or scalp, so the resulting field is much, much stronger. In practice, TMS often produces an initial burst of activity, followed by a longer period of suppression. This can be used to temporarily and reversibly inactivate a tiny chunk of the brain, which is great for research. It may also be a way to “reset” neural circuits that have gotten stuck in some weird state (this is one hypothesis about depression, and TMS does seem to help with it).
I think of tES as “nudging” the neurons’ own activity towards some state that you want, while TMS is more like a sharp rap. It’s not obvious if one is always better than the other. TMS devices are pretty large (lots of capacitors) and fixed, while you could imagine tES being more suitable for use outside a clinic.
Thank you for taking the time to answer my question!
Literally what they pay me for :-)
Anything else you want to know?
Anything else you want to know?
It seems as though the authors are advocating for the role of extracellular electric fields to enable spiking. This is known as ephaptic coupling, which is a fairly well studied phenomenon, but it is unknown the degree to which it contributes to standard computations and processing within different neural systems.
https://en.wikipedia.org/wiki/Ephaptic_coupling