Despite bold predictions from several tech companies about the future of neural interfaces, the science of brain expansion is still in its infancy. What do scientists think about all this hype that comes from Silicon Valley? Mikhail Lebedev, a neuroscientist working on neurocomputer interfaces at Duke University, recently received a $ 100,000 prize from Frontiers for a collection of work on brain expansion written over the past four years.
This prize should help him and his colleagues Ion Opris (a neuroscientist at the University of Miami) and Manuel Casanova (a physician at the University of South Carolina) put together an international conference on the topic next year. Singularity spoke with Lebedev to find out his views on the development of this area.
Is the hype we're seeing about neural implants and brain augmentation justified?
In the next 10 years, we will see the emergence of various kinds of realistic prostheses and many technologies for the rehabilitation of stroke and spinal cord injury survivors. The way it is described in these high-profile articles - like the fact that a person learns to type with the power of thought and receive several million electrodes implanted in the brain - all this will be, but in 20 years.
I could be wrong because new technologies are advancing rapidly. If 10 years ago it was normal to insert a half-millimeter electrode into the brain, now they are already nanoscale. Of course, decoding brain activity will remain a challenge for a long time.
Do we know enough about cognitive processes to work with them?
We have a basic understanding. We know that some areas of the brain are more “cognitive” (associated with learning and cognition functions) than others. Therefore, if you want to extract more additional information from the brain, you will have to place electrodes inside or over these areas. But we have a very mediocre idea of thought, so I don’t think that in the next 10 years we will learn to decode freely floating thoughts.
What does this mean for people's hopes of using brain expansion to communicate with AI? Is this realistic in the short or medium term?
Promotional video:
I think this is more than real, but the first success comes from augmented reality when you use your normal senses, which are quite good, to interact with AI. Let's call it exomozg. Thus, direct communication is really a good idea, but it is still limited by the number of channels for such pairing. The main problem is that we don't really understand the brain code, so we don't know how to make this interface efficient.
But my memory is limited, so augmented reality glasses would be extremely useful, as if the AI accompanies me as I move through the environment. It's not hard to imagine that the computer and the brain work together. So the brain gives examples, the computer learns, and the brain takes advantage of the computational power of the external device.
What kind of brain augmentation can you implement?
Take any brain function and you can try to supplement it. Among sensory functions, new senses can be added to the brain. For example, you can add a feeling of electromagnetic fields that we usually don't feel, and this will be a new feeling. You can place these new sensors around the perimeter of your head for panoramic vision. Of course, I would first experiment on animals.
You can also try to stimulate certain parts of the brain, but at the moment the vast majority of work shows that you can only suppress certain stages of data processing, and not improve. However, such suppression can be beneficial if applied. Imagine, for example, that a person solves certain problems, and the computer knows the correct answer - therefore, it sends a suppressive impulse to certain areas of the brain and inclines it towards a certain solution.
What are the main uses of a brain supplement?
There are two main branches. The first is non-invasive devices that are very easy to implement, and they seem to work. The only problem is that the quality of the signals they provide is limited. If you look at electroencephalogram (EEG) systems, they are represented by the activity of a huge number of neurons, and the most powerful EEGs are recorded during sleep. Thus, all actions associated with, say, precise motor functions become very small, and you cannot detect them on the EEG. In addition, EEGs suffer from all sorts of artifacts.
Of course, EEG devices aren't the only ones using non-invasive techniques. Functional near infrared spectroscopy (fNIR) is also a very good non-invasive method. It allows you to highlight certain types of activity, only it works very slowly.
The potential of invasive approaches has not yet been fully realized. We now have the ability to read, say, 100 neurons. In the future, when we read millions of neurons, we will be able to think about any decoding methods. The main obstacle today is that invasive surgery requires implanting a device in the brain.
What about pharmacological approaches to enhancing brain capacity?
Pharmacology isn't exactly my strong point, but drug formulators do amazing things. They can design molecules for specific tasks that can work for one receptor in the brain but not others, or one area of the brain but not others. In principle, all these methods can be improved and will solve specific problems.
It is even possible to modify brain cells genetically, as in optogenetics, which makes cells sensitive to light. This has not been fully implemented because there are many possibilities. Cells can be sensitive to magnetic fields, stretching, even mechanical movement, which is unusual for neurons. Or cells from another organism could be implanted into the brain. Any science fiction idea today seems quite feasible.
What are the potential disadvantages and cons of a brain supplement?
I am optimistic, so I see mostly pluses. We want to improve, we want to become less primitive people. The main drawback of all this will probably be the same as when taking drugs. Let's imagine a person who implants a device in the pleasure center in the brain and just constantly gets high. You are unlikely to like this, but it will attract many.
When interfering with the brain's motivation and pleasure systems, this can become a problem. Immediately, the military appears, who are taking control of their soldiers. In addition, any neurocomputer interface can act as a lie detector. You will notice things that you usually do not want to notice, you want to leave them in the ownership of other people.
Is there a risk that access to these technologies will be uneven?
I don’t worry about that, because surely rich people will be the first to gain access to systems for expanding brain functions. Yes, they will be expensive, clumsy, and poor performance. But as technology develops, they will get cheaper and everyone will have access. So there shouldn't be such a problem in a capitalist society.
ILYA KHEL