When we become cyborgs
The present and future of neural interfaces
Alexander Kaplan, Post-science
The desire to simplify communication between a person and a computer led to the creation of neural interfaces. Brain–computer interfaces register the electrical activity of the brain and provide a direct connection between it and the device. Now neurointerfaces are used primarily in medicine, helping patients to learn to walk and communicate again. However, the possibilities of technology are not limited to this.
How neural interfaces work
Neurointerface technology is called the brain–computer interface, because it is the computer that registers, processes the electrical reactions of the brain and transforms these signals into commands for executive systems. Brain cells transmit messages to each other electrically, and such echoes of the life of neurons can be registered on the skin surface of the head, despite the fact that they are extremely weak, on the order of a millionth of a volt.
The main element in neurointerface technology is a module for recognizing a person's intentions or images arising in his head from the electrical activity of the brain: a person has just presented some action of his own, for example, a hand movement, wants to do something, and signals about this can already be registered and transmitted via an electronic circuit to an executive device that will perform the planned action. It turns out that with the help of neurointerface technology, a person can perform a planned action without the help of nerves and muscles, a direct command of the brain to the executive device. One of the difficulties of this technology is that not every intention can be recognized by the electrical activity of the brain. For example, no one has yet been able to decipher from the electrical activity of the brain whether a person is thinking about a tangerine or a steam locomotive, about the sea or mountains, and so on. As a rule, it is possible to decipher intentions or images associated with the movement of body parts with a good probability. And then, most importantly, this deciphered intention can be used as a mental command for external actuators, for example, for a toy car with a motor: imagine the left hand is a command to turn right, the left – to the left, and so on.
Of course, the decoding procedure begins with learning algorithms "by examples": segments of the electrical activity of the brain are taken, against which a person presented this or that image. If the algorithms learn to distinguish them, then they can be forced to "observe" the current electrical activity and report to the system as soon as they find a fragment that looks like an example with the appropriate way. This will be a sign for the formation of the first team and so on for each of the tested images or intentions.
This type of interfaces, as you can see, is based on focusing a person's attention to his inner intentions and images. You can get 4-6 mental commands in it. Another type of neural interfaces is based on focusing attention to external objects, for example, to the letters of the alphabet winking on the monitor screen. Here it turned out that if a person is interested in a particular letter, then the reaction in the electrical activity of the brain to the illumination of this letter will differ from reactions to other letters. Thus, it will give out which letter at the moment the person wants to type in the text. So, letter by letter, the user will be able to type text without making any muscle effort, without touching the keyboard. These are the so-called communication neurointerfaces.
Where neurointerface technology is used
Most of all, neurointerfaces are now used in rehabilitation medicine to help paralyzed patients. The first type of interfaces is for controlling simulators, when a mechanical exoskeleton controlled by the patient's mental intentions begins to develop a paralyzed arm. The second type of neurointerfaces is useful when patients lose their speech: in this case, they will be able to type texts with their mental efforts alone.
A program for entering text using a neurointerface // giphy.com
Helmet or chip: what neural interfaces look like
Implanting a chip into the brain, which is a package of electrodes, contacts with nerve cells, is very dangerous. Foreign bodies are rejected by the body. The chip won't live in the brain for a long time, a couple of years at most, and a healthy person doesn't need it. Over the past 20 years of the development of neurointerface systems, such electrode complexes have been implanted in the brain of only 15-20 patients. This was done for medical reasons, in particular to enable paralyzed patients to operate the manipulator.
There can be no request from a healthy person to implant such devices in the brain: after all, for this you will need to do a neurosurgical operation. We feel quite comfortable without these electrodes. But despite this, tens or even hundreds of thousands of people are walking around with implanted electrode complexes – though for a completely different reason. For example, to prevent epilepsy attacks. The fact is that, as a rule, initially an epileptic seizure occurs in deep structures of the brain and then spreads throughout the brain – a person loses consciousness, convulses. If this happens once a month, you can somehow put up with it. But if this happens several times a day, it's not life anymore. It turned out that it was enough to pass a weak electric current through the part of the brain in which an epileptic seizure was born in order to prevent this attack. For this procedure, an electrode is implanted into the patient's brain, with the help of which early signs of convulsive activity are detected, and immediately electrical impulses from a miniature stimulator previously implanted under the scalp are fed through this electrode into the epileptic focus. Patients do not even notice at what point they could have had an attack, because it is immediately stopped.
So the technology of implanted electrode complexes itself is widespread, but not at all in order to expand a person's memory, ergonomically adjust a computer to him or so that he can quickly type texts with the power of thought.
The device, which reads the electrical activity of the brain from the skin surface of the head, consists of three parts. The first part is an electrode complex, which is a number of small metal discs that are applied directly to the skin surface of the head to register the electrical potential. There are many such electrodes that can be placed – in my laboratory, for example, up to 130 such electrodes can be fixed on the head, but only 8 electrodes are enough for a communication neurointerface so that a person can type letters with mental effort. The electrode complex is pressed against the skin surface of the head with a special rubber cap.
The second part is the device to which the wires from the electrodes go and which should significantly amplify the signals of brain activity, since they are very weak, one millionth of a volt – such a weak signal the computer simply cannot accept in order to process it.
The third part of this system is computer technology: signals are received at the input, and from potential fluctuations all this turns into numbers that describe the amplitudes of these potentials. Next comes the digital signal processing: its various quantitative characteristics are calculated, its similarity with other signals is determined, and so on – the latter can be found out if we compile catalogs of signals that we registered in advance, when, for example, a person imagined movement with his right or left hand. At the output, we get a command that can be applied to external actuators.
The process of learning the game using a neurointerface // youtube.com
Neural interfaces of the future
Of course, the most attractive idea seems to be the development of neural interfaces for healthy people, in particular for typing texts with one mental effort, without movements and voice. Such neural interfaces, as already mentioned, have been developed, but they work very slowly – 10 characters per minute. Meanwhile, we even type the test with two fingers from the keyboard at a speed of more than 90 characters per minute. This is optimal for patients, because they get tired quickly, and they have no other option, and for a healthy person it is too slow.
In addition to typing, neural interfaces can help a person manage information flows. For example, often the user's computer screen is filled with several windows, among which only one may be necessary at the moment. If an encephalogram could be used to determine which part of the screen is more interesting to a person at the moment, then it would be possible to control the content and parameters of the screen with this encephalogram in such a way as to make a person's work as comfortable as possible. Such interfaces will help a person to automatically adjust information display devices, putting them in the most optimal mode for human perception.
There are, of course, special situations: for example, for a pilot operating an airplane, you need to see the entire space of the monitor reflecting the flight parameters equally well. But we are talking about the simple work of a person on a computer, when some routine accounting or scientific information is processed, websites or books are viewed, but, as a rule, attention is paid only to part of the screen. In this case, pre-selected markers in the EEG by means of neurointerfaces, as it were, are led across the screen with a flashlight, highlighting for a person exactly the information that he needs at the moment.
You can go further and try to find such specific changes in the EEG, by which it will be possible to determine a person's more subtle intentions – for example, flipping through the pages of the Internet, finding information in it that is still vaguely expressed in words, but already in demand in the depth of thought, and so on. In this way, it would be possible to realize the dream of any computer user: for this computer to anticipate a person's actions before he had time to think about them. Now this approach is being developed by scientific teams of many laboratories around the world, and there are already first results, but implementation in specific gadgets is still far away.
In the near future, it can be assumed that compact highly dynamic systems will be developed to optimize the ergonomic relationship of a person with a computer. The next question is whether it will be possible to create neural interfaces with which it will be possible to control not buttons and motors, but directly the memory cells of the processor. If it succeeds, it will be a real revolution in neurointerface technologies and in general in the relationship between man and machine: silicon processors will directly begin to serve the information and analytical activity of the brain. At the same time, a person will not turn into a cyborg, he will still control external computing facilities, but the channel of this communication will become so dynamic that these external processors can become like the third hemisphere of the brain. This also applies to memory expansion: it would be possible to store a hundred phone numbers in a hundred memory cells – it looks on the verge of fiction, but still theoretically possible.
Is cyborgization possible?
A completely different area of neurointerface technologies is the management of artificial organs of one's own body: an artificial kidney, heart, and so on, which are already being tested and are about to enter real medicine. Apparently, here, too, neurointerfaces will prove useful for adjusting the modes of operation of artificial organs to the current activity of the body and to the plans of future action: for example, if a person intends to dramatically accelerate movement or climb stairs, then it will first be necessary to strengthen the work of the heart. These intentions will be "heard" by neurointerfaces and transmitted to the appropriate artificial heart mechanisms.
Of course, at first glance, fantasies about transplanting a human brain into a cybernetic body look much better here, and then there is no need to construct individual organs. But these are just fantasies that, in my opinion, will never come true. After all, the brain as an organ lives only because other organs provide it not only with oxygen and glucose, as many people think, but also with a complex cocktail of all kinds of substances – from amino acids and vitamins to hormones and a lot of other tissue regulators. In addition, a huge number of nerve fibers from all sensory organs and from the receptors of internal organs approach the brain. To disconnect the brain from all this is like pulling the processor out of its usual slot in a computer.
Nevertheless, there is no escape from biological aging: if someone thinks to reproduce the fantasy of the writer Belyaev "Professor Dowell's Head" in reality, this does not mean that the human brain will live as long as the cybernetic platform supporting it with nutrient solutions will work.
Theoretically, it would be possible to create an artificial brain, for example, on silicon elements, which to some extent would reproduce the structures of the natural brain and would live almost forever. However, now it seems unrealistic to rewrite the contents of the human brain on a silicon carrier. After all, in order to rewrite, for example, some picture from a USB flash drive to a computer, you need to know the data formats, the codes in which the picture is recorded, the meanings of code symbols, and so on. If we don't know anything about it, we will rewrite some numbers from the computer, and they won't turn into a picture in the end - they will remain numbers. No technological breakthrough will help here, because even theoretically it is impossible to know the communication codes of nerve cells, which are different in each pair and change every second due to the arrival of new information.
Therefore, connecting to 86 billion nerve cells of the human brain, and even decoding everything and collecting human thoughts from it, is an unrealistic task. However, every year scientists open up a wider path in studying the subtle mechanisms of the brain, in finding out the causes of its diseases and ultimately in knowing who we are.
About the author:
Alexander Kaplan – Doctor of Biological Sciences, Psychophysiologist, Professor of the Department of Human and Animal Physiology, Head of the Laboratory of Neurophysiology and Neurointerfaces at the Faculty of Biology of Lomonosov Moscow State University.
Portal "Eternal youth" http://vechnayamolodost.ru