Written By: Joe Taylor

Source: Sara Cheshire “Brain Pictures” 13 June 2008.

Science fiction movies across time have frequently shown robotic limbs and prosthetics that could easily function as substitutes for our organic appendages and in some cases even perform much better than them as well. Just science fiction? Well maybe not. There have been advances in neuroscience technology that have allowed this fantasy to possibly become a reality in the near future. With the development of Brain-Computer Interfaces or BCIs this dream gets closer with each passing day. Bidirectional Brain Computer Interfaces advance the traditional goal of controlling prosthetic devices to combining neural decoding and encoding within a single neuroprosthetic device. This expands the range of applications for such a device to include things such as inducing Hebbian Plasticity for rehabilitation, reanimating paralyzed limbs, and even enhancing memory. The company that is currently making the biggest strides toward this area of industry and research is Elon Musk’s neuroscience startup Neuralink Corp., which is developing a next-generation brain-computer interface. This paper will cover the science behind this technology and the current advances being made in this field of research.

Source: Ed Grabianowski “How Brain-computer Interfaces Work” 2 November 2007.
Source: Ed Grabianowski “How Brain-computer Interfaces Work” 2 November 2007.
Source: Ed Grabianowski “How Brain-computer Interfaces Work” 2 November 2007.

To begin, there needs to be a clear understanding of what a BCI is. A Brain-Computer Interface (BCI) is a system that facilitates a direct communication pathway between an enhanced or wired brain and an external device (Shein 2017). BCIs are nothing new, however. The earliest simple brain machine was the Cochlear Implant invented in 1961. Deep Brain Stimulation is also being used to treat like Parkinson’s and epilepsy. What a cochlear implant and deep brain stimulation have in common, is that they are both low-dimensional brain machine interfaces, meaning there are not many wires or channels going into the brain. The techniques researchers use to study the brain signals include EEG (Electroencephalogram) and fNIRS (functional near-infrared spectroscopy). Techniques like the fMRI and MEG (Magnetoencephalography) are too clumsy for reasonable BCI applications. This work used to be targeted towards heling disabled patients, but it is now spreading into more mainstream applications where the brain provides an extra channel of input from the user to the computer. (Shein, 2017). Brain Gate is a research organization that has been working on developing and testing BCI devices to restore communication, mobility, and the overall independence of people who suffer from spinal cord injury, stroke, and ALS. In 2013, President Obama announced a $100 million in funding for the Brain Research through Advancing Innovative Neurotechnologies (BRAIN) initiative to help neuroscientists conduct research that could lead to new treatments for autism and mood disorders (Shein, 2017). Federal agencies like DARPA responded to this by forming the Neural Engineering System Design (NESD) to enhance research capabilities in neurotechnology. This interest by federal and commercial entities sparked a growth in the global BCI market, which was valued at $723.64 million in 2014 in 2014 and has grown on a compound annual growth rate of 10% to $1.36 billion in 2019. It still projected to reach $3.85 billion in 2027. BCI technologies face a number of obstacles before they can become more widespread. The existing systems require a lot of setup and require invasive surgical implantation or non-invasive headset technology that is difficult to calibrate, which prevents practicality for everyday use. Also, there is major concern for the privacy of the device’s users. Many people feel uncomfortable with the idea that a BCI system could tap into their inner most thoughts, because that may be accessed by the anyone with access to the system. A final obstacle is the issue of whether the brain will accept or reject the introduction of foreign objects such as wiring and also how to create devices that will last a lifetime so that further surgical procedures won’t be needed to repair or replace devices. BCI devices also give off heat. If the devices are not encased correctly, they could severely damage the brain (Shein, 2017).

The next part of my paper focuses on the evolution of BCI technologies. BCI has made enormous strides in the past two decades. After an energized start there was a surprising lull in the field until the 1990s. With the introduction of multi-electrode recordings and fast cheap computers, the field saw a rebirth (Rao, 2019). Building on the advances in BCI technology, researchers began to explore bi-directional BCIs (BBCIs) which combine decoding and encoding in a single neuroprosthetic device system. The problem with controlling a prosthetic hand is that it involves using recorded neural responses, stimulating somatosensory neurons to provide proprioceptive feedback, and ensuring that stimulation artifacts do not corrupt the recorded signals. Several artifact reduction methods have been proposed, but the main solution that this paper is going to focus on is combining decoding with encoding (Rao, 2019). One of the first studies to combine encoding and decoding was O’Doherty et al. who showed that stimulation of somatosensory cortex (encoding) could be used to instruct a rhesus monkey which of two targets to move a cursor to. The cursor was controlled using a BCI based on linear decoding to predict the X-coordinate and Y-coordinate of the cursor. A later study by the same group demonstrated true closed loop control. Monkeys used a BCI based on primary motor cortex recordings and Kalman-filter-based decoding to explore virtual objects on a screen with artificial tactile properties. The monkeys were rewarded if they found the object with particular artificial tactile properties. During brain-controlled exploration of an object, tactile information was delivered (encoded) to the somatosensory cortex via intracortical stimulation. Information was encoded as a high-frequency biphasic pulse train presented in packets at a lower frequency. Because stimulation artifacts masked neural activity after each pulse, an interleaved scheme of 50 ms recording (decoding) and 50 ms stimulation (encoding) was used. The monkeys were able to select the desired target within a second or less (Rao, 2019). BBCIs are a much deeper dive into the word of BCIs. The terminology used in further explanation of BBCIs may be too confusing for the average reader, so this is as far as this paper will go.

As the popularity of BCIs grows, it is only natural that this technology would gain greater commercial attention form the likes of billionaire entrepreneurs like Elon Musk and Bryan Johnson. Elon Musk, the CEO of Tesla and SpaceX is backing another company called Neuralink, which is a BCI venture to create devices that can be implanted into human brains to eventually improve memory and interface with computer systems (Shein, 2017). In 2019, Musk unveiled a significant advance toward a therapeutic device. They put together a submission to the U.S Food and Drug Administration to start testing technology in humans. The company is currently focusing on patients with severe neurological conditions but hopes to safe enough to one day be an elective procedure like Lasik. One of the most important tests for the company is to show that they can monitor brain activity and then decode it. At an event in San Francisco in 2019, “Neuralink described a tiny probe with nearly 3,100 electrodes laid out across about 100 flexible wires, or threads, each individually inserted into rat brains by a custom-made surgical robot. The device can monitor the activity of upward of 1,000 neurons at a time, according to the company” (Hernandez 2019). A BCI that uses flexible threads as electrodes, reduces the risk of damage to the soft tissues of the brain (Mint, 2020). Although the idea of this kind of of technology is exciting, neuroscientists cautioned that it is still too early to tell how quickly these devices could be safely used in patients.

Source: Sara Cheshire “Brain Pictures” 13 June 2008.

Brain-Computer Interface Technology could easily be considered the final frontier of science because once we can fully understand our own minds, we can gain a sophisticated understanding of what exactly makes us humans. There is still a long road ahead for BCI technology, but with all the organizations and individuals investing into the field and all the neuroscientists hard at work in the field, the future is looking bright.

References:

Brain tech is coming of age, but will it make you smarter? (2020, Jan 23). Mint Retrieved from https://search-proquest-com.proxy2.library.illinois.edu/docview/2343973262?accountid=14553

Hernández, D., & Mack, H. (2019, Jul 17). Elon musk’s neuralink shows off advances to brain-computer interface; company putting together submission to the FDA to start testing the technology in humans. Wall Street Journal (Online) Retrieved from https://search-proquest-com.proxy2.library.illinois.edu/docview/2258495718?accountid=14553

Rao, R. P. (2019). Towards neural co-processors for the brain: combining decoding and encoding in brain–computer interfaces. Current Opinion in Neurobiology, 55, 142–151. https://doi-org.proxy2.library.illinois.edu/10.1016/j.conb.2019.03.008

Shein, E. (2017). Overcoming Disabilities: Brain-computer interfaces hold the promise of fully featured replacements for body parts that don’t work or are missing. Communications of the ACM, 60(11), 17–19. https://doi-org.proxy2.library.illinois.edu/10.1145/313783