It is nothing new for Musk and his Tesla and SpaceX companies to subvert people’s three views time and time again and push technology and engineering to the extreme. At the end of August, he showed the new generation of brain-computer interface equipment from Neuralink, which he founded, and their latest progress in brain-computer interface technology through a live broadcast on the whole network.
In this article, let’s take a look at what a brain-computer interface is, what are its main application areas, and whether it is a gimmick for financing or a driving force for the progress of human civilization.
What is a brain-computer interface
Brain-computer interface, also known as brain-computer interface BCI, or brain-machine interface BMI, refers to the construction of interfaces and interconnection paths between the brain and external machines, and the direct exchange of information.
The brain-computer interface is not a completely new concept that has only been put forward in recent years. In fact, as early as the 1970s, scientists have begun to study the specific relationship between neurons and functions such as movement and perception. In many novels and movies, the brain-computer interface technology has reached an extraordinary height, such as the remote control of Avatar, and the switching between virtual and reality in The Matrix, etc., which also greatly magnified people’s brain Expectation and curiosity of computer interface technology.
Brain-computer interface workflow
In order to clarify the nature of the brain-computer interface, we must start with the working methods of neurons. As we all know, neurons are the most basic unit that makes up the brain and nervous system. In the human brain, there are about 100 billion neurons. These neurons combine to form a complex neural network.
Specifically, neurons are connected to each other through synapses, and the main function of synapses is to release and transmit chemical substances called neurotransmitters. When a nerve impulse occurs, these neurotransmitters are released from one end of the synapse, and then transmit and act on the next neuron. In this process, electrical signals are generated and appear in the form of electrical pulses. All the experiences and sensations of our human body can be attributed to the conduction of neuronal electrical signals.
Therefore, the essence of the brain-computer interface is to perceive and detect the electrical activity of neurons in the brain through certain means, and translate and convert it into the control of external machines. Or vice versa, it converts external events into electrical impulse signals and transmits them back to the brain to intervene in the electrical activities that affect neurons, so that the subject can get corresponding feelings and experiences.
Another important theoretical basis of the brain-computer interface is that people find that there are partitions in the cerebral cortex, and different areas are responsible for different functions. For example, the frontal lobe is responsible for high-level cognitive functions, such as learning, language, decision-making, abstract thinking and emotion; the parietal lobe is responsible for the processing and integration of somatosensory, spatial and visual information; the temporal lobe is responsible for hearing, smell, and long-term memory; the occipital lobe Responsible for vision and so on.
Due to the existence of brain partitions, it is possible to study the electrical activities of different areas of the brain and their corresponding specific behaviors or feelings. This largely makes the perception and control of neurons more precise. For example, if you want to control the robotic hand to grab objects through your consciousness, you may need to intervene in the electrical activity of neurons in the parietal lobe of the brain.
In fact, research on brain-computer interfaces has been around for decades. The most representative example is a series of studies from the University of Pittsburgh. In 2008, they implanted 15 to 25 control units in the monkey’s cerebral cortex and connected them to a manipulator placed next to the monkey’s shoulder. At the same time, the monkey’s own hands were stuffed into two tubes and could not move. After a period of training, the monkey controlled the manipulator through the brain-computer interface, and successfully grabbed the marshmallow on the needle in front of him and fed it into his mouth. This research also hit the academic circle at the time, and related papers were also published in Nature in 2008. The full text of this paper has been uploaded to Knowledge Planet “Lao Shi Tanxin-Advanced Edition”, welcome to scan the code at the end of the article to enter the planet to view.
From this example, we can see that the brain-computer interface has unlimited imagination in the medical field. For example, for those patients who unfortunately suffer from spinal or brain damage and thus lose their mobility, the brain-computer interface can be used to control the manipulator, or even the mechanical exoskeleton, to restore mobility to a certain extent. Similarly, the brain-computer interface can also be used to repair other nerve injuries, such as enhancing the patient’s ability in language, vision, and hearing. Musk mentioned in his speech that every one of us will be more or less threatened by neurological diseases. In addition to trauma, there are also the decline in perception and memory that comes with age, as well as contemporary people. Common depression, anxiety, insomnia and many other emotional problems. What Musk and his Neuralink company want to solve is to help people repair brain nerves through brain-computer interfaces, so as to solve the various problems mentioned above.
There are two main ways to realize the brain-computer interface: non-invasive and invasive. Their criterion is the method of obtaining neuron electrical signals. The non-invasive brain-computer interface is to directly collect scalp EEG, which is similar to doing EEG. But because the scalp and neurons are separated by one hundred and eight thousand miles, the accuracy of the EEG signals collected in this way is very low, and the discrimination is not enough. This kind of non-intrusive brain-computer interface is usually only used in some scenarios that do not require high precision and complexity, such as simple game control, or judgment on simple problems, and so on.
The invasive brain-computer interface, as the name suggests, is to directly implant microelectrodes, probes and other devices into the cerebral cortex through surgery, and directly interact with brain neurons. Just like the monkey experiment at the University of Pittsburgh introduced above. The main advantage of the invasive brain-computer interface is that it has little interference and can accurately obtain neuron electrical signals in a certain area, and it can also accurately stimulate this area.
In fact, people have a long history of using this invasive technique to treat neurological diseases. For example, in order to treat Parkinson’s disease in the middle and late stages, people will use a method called deep brain stimulation (DBS). To put it simply, this method is to implant electrodes into the patient’s skull and exert an effect on the nervous system of the human brain by means of electrical impulses. This treatment method has been used internationally for more than 30 years. It has indeed shown good application prospects in the treatment of Parkinson’s disease, epilepsy, and some neurological and mental diseases.
However, the main problem of invasive brain-computer interfaces is invasiveness. The entire implantation process requires a craniotomy, and it is conceivable that this in itself will bring greater potential risks. In addition, rejection of implanted devices, brain damage, and many other problems are also hindering the development of invasive brain-computer interface technology.
Neuralink’s main technological innovations
Then we return to Musk and his Neuralink company. Their main goal is to optimize the intrusive brain-computer interface to make it more stable and cheaper, so that it can be widely used. To this end, Musk and Neuralink also published an in-depth paper:
(The full text of this paper has also been uploaded to Knowledge Planet “Lao Shi Tanxin-Advanced Edition”, please scan the code at the end of the article to enter the planet to view)
In summary, so far, the main technological innovations made by Musk and Neuralink are as follows.
First of all, there is a new type of flexible material for implanted electrodes. For invasive brain-computer interfaces, it is definitely hoped that the implanted electrodes or probes are as small as possible, so that on the one hand, more electrodes can be implanted, and on the other hand, the damage to the brain will be less. So the filament-like material of Neuralink is only 5 microns in width, which is 10 times thinner than hair. After such a flexible material is implanted in the brain, it can follow the brain for small movements, which greatly reduces the damage to the brain tissue and nervous system.
In each filament, there are 32 electrodes, and multiple strands of such filaments can form a larger electrode array. For example, in the device named Link v0.9 shown this time, it contains 1024 electrodes. This number of electrodes is an order of magnitude higher than traditional invasive brain-computer interface research.
Neuralink’s second technological innovation is a surgical robot used to implant electrodes. It is conceivable that in order to operate the electrode wire 10 times thinner than the hair on the surface of the brain, it is unrealistic to operate it manually. Therefore, the use of robots for electrode implantation is almost an inevitable choice. The surgical robot developed by Neuralink, like an ultra-high-precision sewing machine, can implant electrode wires into specific areas of the brain one by one, and automatically detect and avoid blood vessels in the brain.
The third major achievement, which I am more concerned about, is the brain-computer interface chip N1 developed by Neuralink and the brain-computer interface device Link V0.9 demonstrated this time. Let me talk about the N1 chip. Its size is 4 mm by 5 mm, and it can process 1024 electrode channels at the same time. The N1 chip is connected to the electrode wire, and is responsible for collecting electrical signals from neurons, then performing signal amplification, analog-to-digital conversion, and finally output.
In the version shown last year, this device was placed behind the implanter’s ear. In the version shown this time, all the components are integrated and directly implanted into the skull, so that no equipment and connections can be seen from the outside.
Specifically, the size of this circular device is 23 mm and the thickness is 8 mm. Its main body contains the wire base, N1 chip, battery, and other data transmission and charging modules. It is worth noting that all the data transmission and charging process of Link V0.9 are done wirelessly, which avoids changing the battery or opening a USB port on the head for charging and data transmission like the first product they showed.
One of the key links of Musk’s demonstration was to demonstrate some basic functions of Neuralink brain-computer interface devices through the three little pigs. For example, the actual walking posture of the piglet can be estimated through the sampled neural signal. To be honest, compared with the aforementioned video of a monkey controlling a robotic hand, it is much less shocking. Some people also mentioned that such a function can be achieved through non-invasive brain electricity or electromyography.
In addition, Musk also demonstrated the process of writing neural signals through electrode wires. This is more of a feasibility demonstration, which is to prove that electrical stimulation can be written through the wire electrode in engineering and affect the response of surrounding neurons. But this does not mean that the electrical stimulation of these writes is meaningful, or what kind of impact they will cause, let alone how to precisely control these writes. This is also the main reason why many people in academia and industry question or disdain this conference. In the final analysis, there are not many actual “black technology” debuts at this conference.
Brain-computer interface technology: financing gimmick or civilization promoter
Lao Shi believes that from the beginning, we may not have too much expectations for these far immature technologies, but this does not mean denying any small progress. Imagine if we take the time dimension from the press conference on August 28, 2020, and stretch the time axis forward by one year, ten years, one hundred years, or one thousand years. From the perspective of God, we can Seeing that all current civilizations are extremely imaginary to us before.
But if you travel to decades ago, tell the people at the time that they will be inseparable from mobile phones and computers in the future, and they can use these Electronic products to solve almost any problems in life and work, and they can talk to anyone you know at any time. If people talk on the phone, they will most likely say that you are crazy. If you travel to hundreds of years ago, tell the people at that time that they can go to any corner of the planet within a day; or there are various panacea that can extend people’s life span by decades , They will definitely say you are crazy.
So, if someone tells you now, people can transfer data directly to the brain like downloading files, so as to easily learn various languages and skills, or adjust their emotions and feelings by controlling certain parts of the brain, or By repairing the neural network of the brain, it can cure Parkinson’s, Alzheimer’s, depression, anxiety and other neurological and mental diseases-would you say they are crazy?
What we need to know is that the advancement of technology and civilization has never been achieved overnight. Each step forward may be very trivial, but decades, hundreds, or even thousands of years of such progress can trigger revolutions in civilization and technology. Students who have seen the three bodies should know that low-dimensional creatures cannot visually imagine higher-dimensional worlds, but we can continue to explore and quantify those higher-dimensional unknowns through the advancement of technology.
In fact, the most important question is, when the future really comes, when all this happens, do you choose the blue pill, wake up from the dream after swallowing it, continue to believe what you want to believe, or choose Red pill, to explore the truth behind this surface?
Concluding remarks
In fact, Musk’s current net worth has reached 100 billion U.S. dollars, has surpassed Buffett, and became the fifth richest man in the world. At least in terms of Neuralink’s current scale, Musk really does not need to seek external financing. Only one-thousandth of the assets is enough to maintain the normal development of Neuralink.
Musk also said at the beginning of this conference that he does not need financing, and the main significance of this conference is to attract all kinds of talents to join Neuralink.
Perhaps in our lifetime, we will not be able to see the real input of data to the brain through brain-computer interface technology, and thus make our lives more efficient and beautiful. However, every step towards this goal, no matter how small, is worthy of recognition.
The Links: FZ1000R25KF1 6DI30A-060
0 Comments for “Brain-computer interface technology: financing gimmick or civilization promoter”