The potential to restore vision with Neuralink’s “Blindsight” neural interface technology

A biocompatible, safe, and reliable, effective human brain-to-computer neural interface has long been a subject of science fiction. Recently, however, Neuralink (a neurotechnology company founded by Elon Musk) has been developing innovative ways to interface the human brain with computers. Neuralink's neural interface technology enables bidirectional communication between external devices and the brain with unprecedented levels of precision. This is because of its array of 96 flexible and small electrode threads that contain 32 electrodes on each thread, for a total of 3072 electrodes Figure 1.1 In addition, Neuralink also developed a novel robotic insertion approach to allow for micron-level precision of probe insertion into the brain to aid in targeting specific regions while avoiding cerebral vasculature.1 This approach allowed a spiking yield of up to 85.5% to be achieved in freely moving rats.1 The nervous system communicates through action potentials or "spikes" and a high spiking yield is essential to closely replicate neural activity for better integration and performance.Figure 1: Prototype of Neuralink's packaged sensor device. A – Integrated circuit which can process 256 data channels, B – Polymer threads on substrate made of parylene-c, C – Enclosure made of titanium (top removed), D – USB-C connector. Note that this diagram is based on a Neuralink prototype from 2019, and Blindsight will likely vary from this design 1 The many potential promising applications of Neuralink's interface include: Assistive technology, enhanced learning, neurofeedback to improve mental health and for brain monitoring research. Neuralink's first product telepathy reportedly allowed a quadriplegic person to move pieces and play chess using only his thoughts and the brain implant to control a computer.2 Neuralink's next planned product "Blindsight" also reportedly has had early success in trials in nonhuman primates.3 This paper explores the potential and challenges for safe and effective vision restoration through neural interfaces and the possible ethical considerations of this potentially transformative technology. Blindness is a debilitating condition affecting millions worldwide and can severely limit independence and decrease quality of life. Current tactile and auditory assistive technology can improve independence and navigation but have significant limitations. For example, a cane only detects ground-based objects and obstacles.4 Newer visual restoration technologies (e.g., cortical and retina implants) and virtual and augmented reality technologies have emerged as potential options with modest success to date.5-8 Neural interfaces have the potential to restore vision in cases where the eye is unable to capture and/or transmit visual information, by directly interfacing with neural tissue. Recent advances in wireless data, power transmission, and electrode design have led to immense advances in this area in recent years.9 Around the world, multiple sites are currently testing cortical prosthetics to restore vision. The most common design of a cortical prosthetic features a camera to capture visual information, which is then fed to a video processing unit, converted to a sequence of commands, and then transferred to the implant to deliver patterned stimulation.9 In one study, multiple arrays in monkeys studied for 3 years showed a decreased signal quality, decreased electrode impedences, and impaired performance on visual tasks over time.10 Ongoing research and innovation in this area will be required to overcome these challenges to make neural interfaces a viable solution for vision restoration. The long-term effects of neural implants are not yet fully understood, and close monitoring of any adverse effects will be essential. Access to having a neural implant may be limited by availability or cost, which may further exacerbate preexisting health-care disparities. Particularly as much of the world's blindness is in developing countries, where there is a lack of ophthalmic services.11-13 Finally, considerations should also be made on the potential implications of use when not medically indicated, such as for enhancing vision beyond human capabilities. New and unique ethical questions created by brain interfaces will have to be addressed including risk–benefit and cost–benefit analysis. Neural implants could also potentially generate significant amounts of sensitive and private data surrounding an individual's visual experience and neural activity. These data must be closely safeguarded to protect privacy, prevent misuse, and potential exploitation or misuse. Human brain to computer neural interface may be moving from the realm of science fiction to the world of science fact. Careful review of the safety, efficacy, reliability, and durability of such devices bring hope to the visually impaired. Novel ethical questions however will arise including the possibility of two-way rather than one-way (brain-to-computer) interface and how that might impact our notions of reality and virtual or augmented reality. Author contributions E. W. – Conceptualization, writing, J. O. – Conceptualization, writing, A. G. L. – Review, intellectual support. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.