IMPACT COMMUNICATION
AND INNOVATION
AND INNOVATION
Elon Musk is not the first to propose creating a company or project for brain implants, but it is a fact that the publicity and money involved in his ventures make a difference when it comes to believing whether something will actually move forward or not. But beyond the potential buzz and the investment to be secured, there are also technical (and ethical) characteristics that must be considered when creating and popularizing an invasive device like the one proposed by Neuralink.
With that in mind, Rick Liebling from The Adjacent Possible newsletter recommended the article “A Brief Tour Through the Wild West of Neural Interfaces” byShelly Fan for SingularityHubas a starting point for a conversation on the subject. In it, Fan discusses an article published in the journal Nature, one of the top references for scientific papers, which points to the fact that electrodes are increasingly falling into disuse in favor of other types of neural interfaces capable of using stimulants other than electricity, such as light, chemical substances, or even two different sources at once.
According to the research, while we are often familiar with cybernetic implants, there are other possibilities that also involve genetic engineering, though these have so far only been tested in animals—precisely for that reason. In any case, Fan argues that this type of proposal should not be reserved only for the laboratory, but should reach the shelves of the nearest store—which may take a good few years, but is not an impossible hypothesis.
In fact, electrodes are one of the most advanced and established brain interface technologies in the scientific community, and one reason for this is that our neurons communicate primarily using electrical “spikes,” which are, however, supported by chemical reactions. As Fan writes, “The logic is sound: an implanted electrode can ‘listen’ to and record the conversation between neurons, and increasingly powerful software can help decode this signal.”
In this sense, when we finally manage to decode this communication, we can theoretically also create data to be transferred from a device to our brain. “When our internal code is broken—as in the case of Alzheimer’s, depression, or epilepsy—external interventions can organize the nervous system’s information and restore its function,” Fan explains. “The problem,” argue MIT professors Dr. James Frank, Marc-Joseph Antonini, and Polina Anikeeva, “is that we will eventually need to go beyond classic implantable electrodes because they are not specific enough. In fact, they do not target the types and groups of neurons needed to reproduce a thought or an action. The result? Undesirable and unpredictable side effects.”
On the other hand, that is not quite how our brain works. Patterns of neural activity occur from different perspectives: individual neurons, neural networks, brain waves... and all at different times as well. “It’s as if we were trying to play a specific piano piece by reading a score: you have to hit the right notes and hold them with the right force and for a certain amount of time, or the neural ‘music’ will sound like something else entirely,” Fan explains.
Therefore, the future of neural implants, according to the authors of the article published in Nature, is a product capable of integrating multiple levels of brain activity. The easiest way to do this is by integrating multiple functions—electricity, light, chemistry, magnetism—into a single implant. In the case of the MiNDS (Miniaturized Neural Drug Delivery System), it is possible to record deep regions of the non-human primate brain and also stimulate them. However, for long-term implants, they need to be soft and biocompatible with the brain to reduce scarring issues.
Finally, Fan also comments on the real possibility of connecting the brain to a computer. For this, the device must be truly intelligent and enclosed in its own system. “Instead of stimulating neurons based on a pre-programmed protocol, these implants measure brain activity and learn to interfere only when they detect anomalies. In this way, the implants are less likely to distort the brain’s normal function—that is, they won’t mess with a person’s sense of self and self-control,” the author describes.
As for the use of light, optogenetics has proven to be a field of research that has revolutionized neuroscience. In this case, the method involves genetically inserting light-sensitive proteins, called opsins, into neurons. When activated by light—which is usually blue, but practically the entire spectrum can be used—the protein forms a tunnel through which ions can pass. This is how neurons naturally function, Fan explains, so light can act as an artificial trigger to activate or inhibit specific neurons. Thanks to light, scientists have been able to “create” false memories in mice or erase unwanted ones as well.
And since such a technique uses genetic engineering methods, one of the biggest targets of interest for its use is blindness, especially because eyes are easy to access and are naturally capable of processing light. “Preliminary studies have already found that natural light, through optogenetic proteins within the retina, can help mice ‘see’ and react to images similarly to normal mice, even though the proteins are less sensitive than the eyes’ original biological sensors,” the article describes.
Because of this, researchers are trying to find a way to create more sophisticated solutions, for example with small, wireless devices that can be inserted under the scalp. Instead of using batteries, these devices can harvest energy using externally oscillating magnetic fields. Furthermore, these devices could also be digitally controlled to modulate their intensity and frequency for light activation and neural activity.
Another study, Fan describes, also used an existing optogenetics technique, but in a transformative way. Compared to blue light, near-infrared wavelengths penetrate the skull and brain tissue easily without scattering, which allows deeper regions of the brain to be modulated. A team of engineers at Berkeley, for example, created nanoparticles that would serve as lamps, converting this type of wavelength into traditional optogenetic light—which is blue-green—inside the brain. With this, the team was able to activate neurons inside a mouse’s brain by using this type of light from outside the skull, thereby targeting neurons related to depression so they could release dopamine.
Regarding the chemical methodology, Fan comments that the idea is not to replicate old ways of releasing substances into the brain, as was previously used to release dopamine in the brains of Parkinson’s patients. In fact, in this case, the main ways of using chemical substances in the brain are with the help of drugs specifically designed to activate certain neurons or a combination with optogenetics to control neural circuits.
The first case, Fan comments, is even similar to optogenetics, but instead of using light-sensitive proteins, scientists use proteins specifically created to open channels that will only fit with drugs designed for that purpose. The technology, known as DREADD (Designer Receptors Exclusively Activated by Designer Drugs), allows scientists to refine their control over neural networks. However, these devices also need genetically altered neurons to function, so there are no laboratory tests yet—at least, not for now. In any case, the idea is that this type of solution will reach the market as a kind of pill that you take and are able to have specific control over your neural circuits.
As for the second form, a recently published study combined drugs with optogenetics inside an “optofluidic” implant. That is, it is a soft implant with a kind of replaceable plug that can be refilled for chronic chemical stimulation—dopamine, then, is one of the favorite substances, as it can be used for both depression and Parkinson’s, in addition to being able to reduce damage to neural tissues. “The implant can be controlled remotely with a smartphone, is minimally invasive, and uses very little energy. For now, the implant is used in animals to stimulate neural network functions, but the team is interested in starting to test it in humans,” Fan details.
Finally, electricity re-emerges as a neural stimulant methodology following the success of the Walk Again project, which used brain implants like the Utah Array to help people with paralysis regain movement. The microelectrodes, in this case, use microfabrication techniques to create these silicon-based electrodes. They are the inspiration for a new generation of implants such as the Neuropixel, which is capable of providing a thousand recording areas on its 10mm surface.
With this, scientists have also begun to explore soft, biocompatible electrodes that can increase an implant’s shelf life. Since the brain and hard electrodes are not compatible in their mechanical properties, these types of electrodes cause an intense scarring reaction around the implant, which can make it useless at some point. “The media darling, Neuralink, for example, uses jellyfish-like soft electrodes that capture brain activity from much smaller groups of neurons and, in theory, would be able to reduce the brain’s rejection response. Another idea, the Neural Lace, adopts a sandwich-like structure that allows for flexibility according to its host and integration with brain tissues. In mice, the tissue was able to consistently stimulate the brain for eight months after the implant,” Fan writes.
Another example the author highlights is Neurogrid, which uses biocompatible material at the electrode tip to reduce noise. According to Fan, a recent study found that these types of materials are able to carry ions more easily and transport them between the brain and the electrodes.
Finally, one of the conclusions evaluated by Fan is that the future of neural implants is, at the very least, bimodal. “The transition of this new generation of electrical implants into clinics could provide more precise alternatives than what we have today. But when combined with intelligent algorithms that allow for closed-loop stimulation, and the reach of chemical substances and the precision of optogenetics, we will finally be able to play with our brains just as we play musical instruments.”
