7 ms·

BrainNet: A Multi-Person Brain-To-Brain Interface for Direct Collaboration

120 points by legionof7 8y ago

seiferteric 8y ago
This is awesome, but am I wrong in thinking that EEG + TMS is a pretty low bandwidth signal? Not sure how useful this will ever be for "cooperative problem solving by humans using a "social network" of connected brains" as the article mentions.
- MrLeap 8y ago
  I'd be interested to know an estimated bitrate. Texting is a very low bandwidth signal that has a lot of utility. If it's comparable to what a person can do with two thumbs I can imagine some really awesome things I could create with this tech. I wish I could justify purchasing the hardware, or had an audience. Haha.
- jdpigeon 8y ago
  Yes, very low bandwidth. Active BCIs based on EEG are extremely unreliable and/or exhausting to use
- darkmighty 8y ago
  I'm also worried about direct interference between EEG and TMS (and EMI is quite common) -- I'd prefer they let the user perform the action physically instead of being electrical-in, electrical-out. The researchers come from an accredited institution, so it probably checks out, but seems like an obvious oversight.
carapace 8y ago
Okay, this is crazy cool, but you don't need hardware to do this! Now that you know this is possible, go read about Charles Tart's mutual hypnosis experiment: https://s3.amazonaws.com/cttart/articles/april2013articles/Psychedelic+Experiences+Associated+with+a+Novel+Hypnosis+Procedure+Mutual+Hypnosis.pdf https://s3.amazonaws.com/cttart/articles/april2013articles/P... That fact that they were able to nail down this kind of communication using the hardware and setup that they did is an existence proof that this sort of communication and synchronization are possible (and not that difficult to establish), however, we are already equipped with sophisticated sensory and signalling systems, more than capable of supporting this sort of thing without extra hardware.
- Gys 8y ago
  One benefit of using hardware is not all persons have to be in the same location: "The Senders' decisions are transmitted via the Internet to the brain of a third subject, the "Receiver," who cannot see the game screen."
  - carapace 8y ago
    That's a good point, and it immediately makes me wonder what kind of e.g. camera and monitor might be enough to enable remote synchronization between brains. There are so many options. You could transmit signals carrying things like pulse and breath rate as well as visual and audio. My sense of urgency comes from my desire that folks reading this on HN realize that the "magic sauce" is in the brains not the comm channel. I've been trying to get traction on this idea for years: if you want a mind-machine interface you keep the hardware simple and let the most advanced processor do the tough part of the job. (In case it's not crystal clear, the most advance processor in ANY technological system is the one between your ears, eh?) Metaphorically, the hardware should be seen as one of those cheap, dumb modems that offload most of the work to the CPU on the motherboard. With a little bit of hypnosis and the crudest of GSR (galvanic skin response) sensors you can make a brain-machine interface at home that would blow your mind (not literally.) Anticipating the next question: "If it's so easy why isn't everyone doing it?" Well, smart-alec, it's because it requires a major adjustment of one's self-image, that most people can't handle or even imagine, so (for example) it could never become a mass-market consumer product. We have had the technology for decades. The limiting factor is belief. That's why this "BrainNet" is exciting: not because they used such a crude interconnect, but because they did it in a way that opens a wedge for the idea that this stuff is possible at all. As soon as you grasp that the doors fling open to all sorts of fantastic possibilities. (And pitfalls: you and your brain are not the same thing and will occasionally find yourselves working at cross-purposes. At times like that you'll probably NOT want your brain to be directly wired into your "smart" home and have all your passwords, eh? Like I said, this requires a major adjustment of one's self-image.)
- codeulike 8y ago
  I read the paper. Its not exactly scientific.
baldeagle 8y ago
For an example of this in science fiction, check out the Nexus series by Ramez Naam.
- minkzilla 8y ago
  I've had that book sitting on my shelf for a couple years now. I read the first couple of chapters, but the writing seemed poor and I couldn't really get into it. Is it worth giving it another go?
  - baldeagle 8y ago
    yes, very much so if you want an action movie / philosophy story about linking minds like boxes on the internet. I think it is some of the most forward leaning scifi I've read in a while, but it does have a strong political bend to it.
aussieguy1234 8y ago
You will be assimilated
moonbug 8y ago
"speech"?
- oelmekki 8y ago
  I don't know if it was intended to be tongue in the cheek from you, but this is stereotypical dropbox-like HN comment. "- We found a way to enable simple telepathy. "- Not impressed. I can already communicate by speaking." If it was indeed intended, thanks a lot, it made my day :)
  - red75prime 8y ago
    I am afraid this simple telepathy is of the same type of impressiveness as SHRDLU (AI program) or acoustic levitation. It is impressive, but underlying approach seems to be hard to scale in certain directions (bandwidth, model world complexity, payload mass). So, no fast progress should be expected.
Gys 8y ago
So I wonder if some people are more compatible, better readable, connectable, solvers etc then others. Then I can imagine an industry around people that can rent out their brain remotely to help solve problems. Kind of an AirBnB for brains ;-)
- ben_w 8y ago
  “Mechanical Turk: you don’t have to be a cyborg to work here, but it helps”?
  - RosanaAnaDana 8y ago
    Rebranded: Maniacal Turk
- v_lisivka 8y ago
  It's better to just connect to computer and use it as additional long therm memory and quick problem solver. It's very helpful when you have problem to hard to your own brain to solve, then think about or do something else, then have ready to use solution in the short-term memory - you just know it. However, intermediate steps from problem to solution are omitted, of course, because they are performed by separate thing. It's good for quick solving of problems, but bad for learning. It's like using of calculator for math: calculator shows you answer quickly, but you learn nothing. Such interface will allow to effectively use persons with average IQ for problem solving, like calculators allows to do math easily for just anybody, but it will not generate geniuses. However, such direct interaction with computer or an other person can lead to schizophrenia, so it must be clearly marked: "here are my own memories, here are memories implanted by the computer", which is not an easy thing to do. Sometimes we cannot separate even our own dreams from our own memories. If such system will be used in wrong way, you will need years to recover, to forget badly implanted memories.
  - taneq 8y ago
    > Such interface will allow to effectively use persons with average IQ for problem solving, like calculators allows to do math easily for just anybody, but it will not generate geniuses. Motorbikes don't generate world class sprinters either but they sure get you from A to B faster than walking.
r41nbowdash 8y ago
>8-channel Cyton system that's the interesting bit, because they got the results on a portable, consumer-grade device
logronoide 8y ago
Can’t wait to have one of this for our weekly code reviews.
ArtWomb 8y ago
DARPA's Next-gen Non-surgical Neuro-tech (N3) program wants brain interfaces for drone pilots https://spectrum.ieee.org/the-human-os/biomedical/bionics/darpa-wants-brain-interfaces-for-able-bodied-warfighters https://spectrum.ieee.org/the-human-os/biomedical/bionics/da... But I think one interesting peacetime application focuses on brain training and deep meditation. One barrier to achieving trancelike states of flow and concentration is overcoming the constant involuntary barrage of "noise". Having a feedback loop may assist in getting there. And in geriatric medicine, keeping the mind sharp is seen to improve longevity and quality of life. Designing custom games to improve neuroplasticity in the aging brain could allow for independence well into late retirement.
sgentle 8y ago
Lightly edited extract from the paper, for anyone curious about the method: BrainNet relies on two well-known technologies: Electroencephalography (EEG) for non-invasively recording brain signals from scalp and transcranial magnetic stimulation (TMS) for non-invasively stimulating the visual cortex. The Senders convey their decisions of "rotate" or "do not rotate" by controlling a horizontally moving cursor using steady-state visually-evoked potentials (SSVEPs): to convey a "rotate" decision, Senders focused their attention on a "Yes" LED light flashing at 17 Hz placed on one side of their computer screen; to convey a "do not rotate" decision, they focused on the “No” LED light flashing at 15 Hz placed on the other side. The direction of movement of the cursor was determined by comparing the EEG power at 17 Hz versus 15 Hz, with a higher power at 17 Hz over that at 15 Hz moving the cursor towards the side near the “Yes” light, and vice versa for the "No" light. A "rotate" ("do not rotate") decision was made when the cursor hit the "YES" ("NO") side of the screen. In trials where the cursor did not reach either side of the screen due to trial time elapsing, the decision closest to the last location of the cursor was chosen as the subject’s decision. The decisions of the two Senders were sent to the Receiver’s computer through a TCP/IP network and were further translated into single pulses of transcranial magentic stimulation (TMS) delivered to the occipital cortex of the Receiver. The intensity of the stimulation was set above or below the threshold at which the Receiver will perceive a flash of light known as a phosphene: a “Yes” response was translated to an intensity above the threshold, and “No” was translated to an intensity below the threshold. The Receiver made his/her decision based on whether a phosphene was perceived and this decision was conveyed to the game by the Receiver using the same SSVEP procedure used by both Senders.
hour_glass 8y ago
The TMS product they are using to stimulate the receivers brain is pretty interesting. https://www.magstim.com/products/ https://www.magstim.com/products/ I'm pretty skeptical that you could use magnetism to transfer any sort of useful information into someone's brain. I'm guessing it just gives a sensation, or maybe some vague visual effect since it is used on the occipital lobe in this experiment.
- amelius 8y ago
  Question: assume you have a closed hull around some domain D, and record the electric and magnetic fields E and H at the entire hull; at a later moment you prescribe E and H at the hull; will this setup regenerate the original values of E and H inside the domain D?
  - ckemere 8y ago
    To maybe short-circuit your idea, magneto-encephalography (MEG) uses magnetic field sensors to sense brain activity. Coherent currents from ~100K neurons produce femto-tesla signals (about 1 billionth of the earth's magnetic field). This currently requires liquid helium cooled superconductors (though there is some work trying on nanofabricated room-temperature ultraprecise field sensors). To induce effects, the TMS magnets produce reasonable fractions of a Tesla.
- ckemere 8y ago
  TMS relies on Faraday's Law - a large transient magnetic field induces electrical currents in the brain. When these currents are large enough, they perturb the activity of neurons nearby. (Neurons communicate based on pulses of dynamics in the electrical potential across their cell membranes.) There are also second order effects related to the alignment of the neural structure to the induced field lines, etc. TMS of motor cortex can induce finger twitches and TMS of visual cortex can induce phosphenes or other visual disturbances. Single TMS pulses have effects that wear off very rapidly, but repetitive TMS may be able to induce therapeutic plasticity (it has been reported to potentially induce seizures - following the normal law that nothing medically useful has no side effects).
allenleein 8y ago
I'm developing a theory about brain's network called: Ghosts in the Shell. I'm trying to decentralizing my brain to a network of brains. Ghost in the Shell (Japanese: 攻殻機動隊シェルのゴースト) is a 1989 Japanese media franchise originally published as a seinen manga series of the same name written and illustrated by Masamune Shirow. In that post-cyberpunk iteration of a possible future, computer technology has advanced to the point that many members of the public possess cyberbrains, technology that allows them to interface their biological brain with various networks. The level of cyberization varies from simple minimal interfaces to almost complete replacement of the brain with cybernetic parts, in cases of severe trauma. This can also be combined with various levels of prostheses, with a fully prosthetic body enabling a person to become a cyborg. We are still way far from that kind of individual advanced tech in 2018, though we are slowly becoming the cyborg with a supercomputer in our pocket. With the sheer amount of information available in the world today, we have become overwhelmed. Our mind is in a constant racing state. Inspired by Ghost in the Shell, I believe the future of human thinking is to decentralize ourselves to a thinking network like "GHOSTS in the Shell". I think we should try to turn our brain into composable and functional brains network by starting to view everything as a function and iterate it like a machine, together. We need to elevate our mind in this digital age. Not only as human, but also as human + machine. (https://github.com/allenleein/brains/wiki https://github.com/allenleein/brains/wiki)
- legionof7 8y ago
  This is extremely interesting and it really ties in with my work on decentralized networks. I'm sure there are lots of intersections with AI and biotech also.
  - allenleein 8y ago
    Thanks! Yes, my ultimate goal is to build a network of brains with this kind of gadget from MIT Media Lab: (https://www.media.mit.edu/projects/alterego/overview/ https://www.media.mit.edu/projects/alterego/overview/) AlterEgo is a non-invasive, wearable, peripheral neural interface that allows humans to converse in natural language with machines, artificial intelligence assistants, services, and other people without any voice—without opening their mouth, and without externally observable movements. --- Love to connect! Here is my email: allenleein@gmail.com

120 points by legionof7 8y ago

This is awesome, but am I wrong in thinking that EEG + TMS is a pretty low bandwidth signal? Not sure how useful this will ever be for "cooperative problem solving by humans using a "social network" of connected brains" as the article mentions.

MrLeap 8y ago
I'd be interested to know an estimated bitrate. Texting is a very low bandwidth signal that has a lot of utility. If it's comparable to what a person can do with two thumbs I can imagine some really awesome things I could create with this tech. I wish I could justify purchasing the hardware, or had an audience. Haha.
jdpigeon 8y ago
Yes, very low bandwidth. Active BCIs based on EEG are extremely unreliable and/or exhausting to use
darkmighty 8y ago
I'm also worried about direct interference between EEG and TMS (and EMI is quite common) -- I'd prefer they let the user perform the action physically instead of being electrical-in, electrical-out. The researchers come from an accredited institution, so it probably checks out, but seems like an obvious oversight.

carapace 8y ago

Okay, this is crazy cool, but you don't need hardware to do this! Now that you know this is possible, go read about Charles Tart's mutual hypnosis experiment: https://s3.amazonaws.com/cttart/articles/april2013articles/Psychedelic+Experiences+Associated+with+a+Novel+Hypnosis+Procedure+Mutual+Hypnosis.pdf https://s3.amazonaws.com/cttart/articles/april2013articles/P... That fact that they were able to nail down this kind of communication using the hardware and setup that they did is an existence proof that this sort of communication and synchronization are possible (and not that difficult to establish), however, we are already equipped with sophisticated sensory and signalling systems, more than capable of supporting this sort of thing without extra hardware.

Gys 8y ago
One benefit of using hardware is not all persons have to be in the same location: "The Senders' decisions are transmitted via the Internet to the brain of a third subject, the "Receiver," who cannot see the game screen."
- carapace 8y ago
  That's a good point, and it immediately makes me wonder what kind of e.g. camera and monitor might be enough to enable remote synchronization between brains. There are so many options. You could transmit signals carrying things like pulse and breath rate as well as visual and audio. My sense of urgency comes from my desire that folks reading this on HN realize that the "magic sauce" is in the brains not the comm channel. I've been trying to get traction on this idea for years: if you want a mind-machine interface you keep the hardware simple and let the most advanced processor do the tough part of the job. (In case it's not crystal clear, the most advance processor in ANY technological system is the one between your ears, eh?) Metaphorically, the hardware should be seen as one of those cheap, dumb modems that offload most of the work to the CPU on the motherboard. With a little bit of hypnosis and the crudest of GSR (galvanic skin response) sensors you can make a brain-machine interface at home that would blow your mind (not literally.) Anticipating the next question: "If it's so easy why isn't everyone doing it?" Well, smart-alec, it's because it requires a major adjustment of one's self-image, that most people can't handle or even imagine, so (for example) it could never become a mass-market consumer product. We have had the technology for decades. The limiting factor is belief. That's why this "BrainNet" is exciting: not because they used such a crude interconnect, but because they did it in a way that opens a wedge for the idea that this stuff is possible at all. As soon as you grasp that the doors fling open to all sorts of fantastic possibilities. (And pitfalls: you and your brain are not the same thing and will occasionally find yourselves working at cross-purposes. At times like that you'll probably NOT want your brain to be directly wired into your "smart" home and have all your passwords, eh? Like I said, this requires a major adjustment of one's self-image.)
codeulike 8y ago
I read the paper. Its not exactly scientific.

baldeagle 8y ago

For an example of this in science fiction, check out the Nexus series by Ramez Naam.

minkzilla 8y ago
I've had that book sitting on my shelf for a couple years now. I read the first couple of chapters, but the writing seemed poor and I couldn't really get into it. Is it worth giving it another go?
- baldeagle 8y ago
  yes, very much so if you want an action movie / philosophy story about linking minds like boxes on the internet. I think it is some of the most forward leaning scifi I've read in a while, but it does have a strong political bend to it.

aussieguy1234 8y ago

You will be assimilated

moonbug 8y ago

"speech"?

oelmekki 8y ago
I don't know if it was intended to be tongue in the cheek from you, but this is stereotypical dropbox-like HN comment. "- We found a way to enable simple telepathy. "- Not impressed. I can already communicate by speaking." If it was indeed intended, thanks a lot, it made my day :)
- red75prime 8y ago
  I am afraid this simple telepathy is of the same type of impressiveness as SHRDLU (AI program) or acoustic levitation. It is impressive, but underlying approach seems to be hard to scale in certain directions (bandwidth, model world complexity, payload mass). So, no fast progress should be expected.

Gys 8y ago

So I wonder if some people are more compatible, better readable, connectable, solvers etc then others. Then I can imagine an industry around people that can rent out their brain remotely to help solve problems. Kind of an AirBnB for brains ;-)

ben_w 8y ago
“Mechanical Turk: you don’t have to be a cyborg to work here, but it helps”?
- RosanaAnaDana 8y ago
  Rebranded: Maniacal Turk
v_lisivka 8y ago
It's better to just connect to computer and use it as additional long therm memory and quick problem solver. It's very helpful when you have problem to hard to your own brain to solve, then think about or do something else, then have ready to use solution in the short-term memory - you just know it. However, intermediate steps from problem to solution are omitted, of course, because they are performed by separate thing. It's good for quick solving of problems, but bad for learning. It's like using of calculator for math: calculator shows you answer quickly, but you learn nothing. Such interface will allow to effectively use persons with average IQ for problem solving, like calculators allows to do math easily for just anybody, but it will not generate geniuses. However, such direct interaction with computer or an other person can lead to schizophrenia, so it must be clearly marked: "here are my own memories, here are memories implanted by the computer", which is not an easy thing to do. Sometimes we cannot separate even our own dreams from our own memories. If such system will be used in wrong way, you will need years to recover, to forget badly implanted memories.
- taneq 8y ago
  > Such interface will allow to effectively use persons with average IQ for problem solving, like calculators allows to do math easily for just anybody, but it will not generate geniuses. Motorbikes don't generate world class sprinters either but they sure get you from A to B faster than walking.

r41nbowdash 8y ago

>8-channel Cyton system that's the interesting bit, because they got the results on a portable, consumer-grade device

logronoide 8y ago

Can’t wait to have one of this for our weekly code reviews.

ArtWomb 8y ago

DARPA's Next-gen Non-surgical Neuro-tech (N3) program wants brain interfaces for drone pilots https://spectrum.ieee.org/the-human-os/biomedical/bionics/darpa-wants-brain-interfaces-for-able-bodied-warfighters https://spectrum.ieee.org/the-human-os/biomedical/bionics/da... But I think one interesting peacetime application focuses on brain training and deep meditation. One barrier to achieving trancelike states of flow and concentration is overcoming the constant involuntary barrage of "noise". Having a feedback loop may assist in getting there. And in geriatric medicine, keeping the mind sharp is seen to improve longevity and quality of life. Designing custom games to improve neuroplasticity in the aging brain could allow for independence well into late retirement.

sgentle 8y ago

Lightly edited extract from the paper, for anyone curious about the method: BrainNet relies on two well-known technologies: Electroencephalography (EEG) for non-invasively recording brain signals from scalp and transcranial magnetic stimulation (TMS) for non-invasively stimulating the visual cortex. The Senders convey their decisions of "rotate" or "do not rotate" by controlling a horizontally moving cursor using steady-state visually-evoked potentials (SSVEPs): to convey a "rotate" decision, Senders focused their attention on a "Yes" LED light flashing at 17 Hz placed on one side of their computer screen; to convey a "do not rotate" decision, they focused on the “No” LED light flashing at 15 Hz placed on the other side. The direction of movement of the cursor was determined by comparing the EEG power at 17 Hz versus 15 Hz, with a higher power at 17 Hz over that at 15 Hz moving the cursor towards the side near the “Yes” light, and vice versa for the "No" light. A "rotate" ("do not rotate") decision was made when the cursor hit the "YES" ("NO") side of the screen. In trials where the cursor did not reach either side of the screen due to trial time elapsing, the decision closest to the last location of the cursor was chosen as the subject’s decision. The decisions of the two Senders were sent to the Receiver’s computer through a TCP/IP network and were further translated into single pulses of transcranial magentic stimulation (TMS) delivered to the occipital cortex of the Receiver. The intensity of the stimulation was set above or below the threshold at which the Receiver will perceive a flash of light known as a phosphene: a “Yes” response was translated to an intensity above the threshold, and “No” was translated to an intensity below the threshold. The Receiver made his/her decision based on whether a phosphene was perceived and this decision was conveyed to the game by the Receiver using the same SSVEP procedure used by both Senders.

hour_glass 8y ago

The TMS product they are using to stimulate the receivers brain is pretty interesting. https://www.magstim.com/products/ https://www.magstim.com/products/ I'm pretty skeptical that you could use magnetism to transfer any sort of useful information into someone's brain. I'm guessing it just gives a sensation, or maybe some vague visual effect since it is used on the occipital lobe in this experiment.

amelius 8y ago
Question: assume you have a closed hull around some domain D, and record the electric and magnetic fields E and H at the entire hull; at a later moment you prescribe E and H at the hull; will this setup regenerate the original values of E and H inside the domain D?
- ckemere 8y ago
  To maybe short-circuit your idea, magneto-encephalography (MEG) uses magnetic field sensors to sense brain activity. Coherent currents from ~100K neurons produce femto-tesla signals (about 1 billionth of the earth's magnetic field). This currently requires liquid helium cooled superconductors (though there is some work trying on nanofabricated room-temperature ultraprecise field sensors). To induce effects, the TMS magnets produce reasonable fractions of a Tesla.
ckemere 8y ago
TMS relies on Faraday's Law - a large transient magnetic field induces electrical currents in the brain. When these currents are large enough, they perturb the activity of neurons nearby. (Neurons communicate based on pulses of dynamics in the electrical potential across their cell membranes.) There are also second order effects related to the alignment of the neural structure to the induced field lines, etc. TMS of motor cortex can induce finger twitches and TMS of visual cortex can induce phosphenes or other visual disturbances. Single TMS pulses have effects that wear off very rapidly, but repetitive TMS may be able to induce therapeutic plasticity (it has been reported to potentially induce seizures - following the normal law that nothing medically useful has no side effects).

allenleein 8y ago

I'm developing a theory about brain's network called: Ghosts in the Shell. I'm trying to decentralizing my brain to a network of brains. Ghost in the Shell (Japanese: 攻殻機動隊シェルのゴースト) is a 1989 Japanese media franchise originally published as a seinen manga series of the same name written and illustrated by Masamune Shirow. In that post-cyberpunk iteration of a possible future, computer technology has advanced to the point that many members of the public possess cyberbrains, technology that allows them to interface their biological brain with various networks. The level of cyberization varies from simple minimal interfaces to almost complete replacement of the brain with cybernetic parts, in cases of severe trauma. This can also be combined with various levels of prostheses, with a fully prosthetic body enabling a person to become a cyborg. We are still way far from that kind of individual advanced tech in 2018, though we are slowly becoming the cyborg with a supercomputer in our pocket. With the sheer amount of information available in the world today, we have become overwhelmed. Our mind is in a constant racing state. Inspired by Ghost in the Shell, I believe the future of human thinking is to decentralize ourselves to a thinking network like "GHOSTS in the Shell". I think we should try to turn our brain into composable and functional brains network by starting to view everything as a function and iterate it like a machine, together. We need to elevate our mind in this digital age. Not only as human, but also as human + machine. (https://github.com/allenleein/brains/wiki https://github.com/allenleein/brains/wiki)

legionof7 8y ago
This is extremely interesting and it really ties in with my work on decentralized networks. I'm sure there are lots of intersections with AI and biotech also.
- allenleein 8y ago
  Thanks! Yes, my ultimate goal is to build a network of brains with this kind of gadget from MIT Media Lab: (https://www.media.mit.edu/projects/alterego/overview/ https://www.media.mit.edu/projects/alterego/overview/) AlterEgo is a non-invasive, wearable, peripheral neural interface that allows humans to converse in natural language with machines, artificial intelligence assistants, services, and other people without any voice—without opening their mouth, and without externally observable movements. --- Love to connect! Here is my email: allenleein@gmail.com