By Paloma Salmeron-O'Brien
Have you ever wondered what it would be like to connect your brain to a computer? A lot of sci-fi settings have played with the idea of integrating the mind with external tech: Pacific Rim and The Matrix are two well-known examples. But outside of fiction, the idea of connecting a human brain to a computer poses many additional questions; What is the reality of this technology? What applications does it have? Is it even attainable? A review posted in Frontiers in Systems Neuroscience aims to shed some light on this.
BCI technology, or brain-computer interface, directly connects neurological activity to computers and “other external devices”. The overarching intent of BCI is to offload simple mental tasks, broaden one’s connection to external stimuli, and strengthen one’s own capacities. BCI can allow for singular (brain to machine) and bidirectional (two-way) communication between the brain and a device, all without lifting a finger. In the case of bidirectional communication, this can take the form of one brain receiving feedback from an external device or, theoretically, the connection of multiple brains. Recent studies have found that BCI can be used to fix motor impairments by reinvigorating synaptic plasticity, or the brain’s ability to reconfigure itself and adapt to new information. When you learn, your neurons form new connections, or synapses, to store that newly absorbed information. This process of restructuring the brain's connections can also be used to restore cognitive and motor functions previously lost due to faulty connections. Besides restoration, BCI can also be used to augment cognitive and motor functions beyond our biological limitations.
This technology has a wide range of applications from medical to recreational. According to the Brain/Neural Computer Interaction Horizon 2020 project, BCI has 6 major potential applications, including Restoration, Replacement, and Improvement. These are the arguably more important and attainable applications of the six and primarily deal with the use of BCI as a medical treatment. Examples include the restoration of motor faculties of individuals suffering from locked-in syndrome, a condition in which patients are conscious but their body is completely paralyzed. BCI can also act as a neuroprosthesis to support motor skills, making it easier for the brain to signal the body how to move. The Research category extends from these medical applications, using BCI to monitor brain activity in real-time. Past medical uses, Enhancement deals with using BCI as a middleman to make it easier for people to understand and use different technology. Related to this, Supplementation explores pairing BCI and VR. This menagerie of possibilities has led to many different proposed projects involving BCI, such as lie detection, ultra-immersive video games, and robot piloting. As a result of the technology’s far-reaching implications, BCI research has risen exponentially since its conception.
Though some of its potential applications may seem a little out there, BCI isn’t as new of a concept as one might expect. Precursors to BCI have been in the works since 1973 when J.J. Vidal recorded stimulated electrical activity in the brain using a non-invasive technique invented in 1929. From its primitive beginnings, BCI technology has diversified into many different forms.
Different BCI systems are distinguished by the way they interact with the brain. Passive BCI deals with involuntary or ambient brain activity (such as fatigue). Active BCI, on the other hand, interprets voluntary cognitive activity. There is also Reactive BCI, which records activity generated in response to an external stimulus. BCI is further categorized by the physical computer-brain interface used. Non-invasive BCI does not require the insertion of intracortical sensors, unlike invasive BCI. However, though invasive methods require more drastic alterations made to users, the benefits of using invasive BCI include better signal-noise ratios and cleaner localization of brain activity. BCI typically interacts with the peripheral nervous system as its window into the brain. Modern advancements in neurotechnology have enabled BCI to accurately decode neurological signaling and deliver signals to target areas of the brain in return. Together, these functions can be used to encourage neuroplasticity.
Now, as innovative, widely-applicable, and revolutionary as BCI appears to be, any new technology comes with associated risks. Where BCI is concerned, these risks typically originate from our still-developing understanding of the brain. Invasive BCI, for instance, has been found to be more successful than non-invasive methods, specifically in the case of medical applications such as restoring mobility. However, the surgery to implant this technology is associated with some serious risks due to the sensitivity of the nervous system. As such, this technology isn’t recommended for “neurologically intact people”. However, in more extreme cases such as patients suffering from locked-in syndrome, the real possibility of significantly increasing their quality of life outweighs the risks of the operation. Furthermore, while the surgery at present carries risk, these associated issues may be temporary. Studies are being conducted into the effects of implantation, and one such study found no ill effects following one year since surgery. In the case of non-invasive BCI, one downside is that it is difficult to maintain an acceptable signal-to-noise ratio. This is due to the dynamic nature of brain activity as well as differences between the brains of individual users. To mitigate this variation, BCI technology needs to be calibrated to each specific user, a process that is often found to be frustrating and tedious. Researchers are attempting to mediate this calibration requirement by training BCI systems on data from multiple people. Using commonalities between people means the system requires less unique information from each individual user, reducing training time. Though there are hurdles currently blocking the full implementation of BCI, researchers are already finding possible solutions to mitigate these issues.
BCI also provokes ethical concerns that must be addressed. As previously discussed, one of the biggest and currently most attainable outcomes of BCI integration is the treatment of patients with locked-in syndrome and other similar ailments. However, getting consent from these patients to use BCI technology on them is challenging. Concern with BCI is also attributed to the nervous system's inherent sensitivity and our still limited understanding of the brain. Far more testing is required before BCI can be implemented as anything more than a treatment for extreme medical conditions. There may be adverse long-term effects associated with BCI that we have yet to discover. Another obvious concern is the potential for this technology to be used maliciously. While almost any new invention can be used harmfully, the direct exploitation of the mind as a possible consequence of BCI misuse cannot be taken lightly. The ethically sound implementation of BCI would at least require extensive testing, efforts to educate the public on the technology’s pros and cons, and the gradual introduction of the technology so as to ‘acclimate’ the general public. Furthermore, certain protocols and safeguards would have to be implemented in order to protect the health, well-being, privacy, and safety of BCI users. Technology as personalized as BCI requires a lot of factors to be considered for its success. Despite this, the tech shows strong promise for a variety of applications.
Author’s note: Like with anything, I see good and bad here. For instance, would I like to control three extra pairs of limbs to effectively become a cyborg homo-arachnid? Absolutely. Do I want Elon Musk streaming unskippable Spotify ads directly to my cranium? Definitely not.
Works Cited
Saha, S., Mamun, K. A., Ahmed, K., Mostafa, R., Naik, G. R., Darvishi, S., Khandoker, A. H., & Baumert, M. (2021). Progress in Brain Computer Interface: Challenges and Opportunities. Frontiers in Systems Neuroscience, 15, 4. https://doi.org/10.3389/FNSYS.2021.578875/BIBTEX