Inside the N1: How Neuralink Gives the Paralyzed a Second Life

Imagine controlling a computer, playing video games, and communicating with the world using only your thoughts. For Noland Arbaugh, a 30-year-old man paralyzed below his shoulders from a spinal cord injury, this is no longer science fiction. In January 2024, Arbaugh became the first person to receive Neuralink's revolutionary brain-computer interface implant, and his life changed forever. As of December 2025, 13 patients across four continents have received Neuralink implants, marking a pivotal moment in neurotechnology and demonstrating that restoring independence to people with severe paralysis is now a clinical reality.

This blog post provides a comprehensive technical and clinical overview of Neuralink's technology, how the implant actually works, the patients currently benefiting from it, the risks involved, and what this means for the future of neurotechnology and brain-computer interfaces in general.

What Is Neuralink and Why Does It Matter

Neuralink is a brain-computer interface company founded by Elon Musk in 2016 with the mission to help people with severe paralysis, neurological disorders, and other debilitating conditions regain autonomy over their lives. The company developed the N1 implant, colloquially known as "The Link," which is a fully implantable brain-computer interface that allows users to control external devices such as computers, smartphones, and robotic arms using only their thoughts.

The significance of Neuralink lies not just in the technology itself but in what it represents: a bridge between the human brain and digital devices. For someone with severe paralysis who cannot move their limbs, communicate verbally, or perform basic daily tasks, Neuralink offers a pathway to reclaiming independence and dignity that would otherwise be impossible. Unlike eye-tracking systems or other assistive technologies that require partial motor control, Neuralink reads neural signals directly from the motor cortex of the brain, interpreting the user's intention to move before any physical movement would occur.

The Technical Architecture: Understanding The Link

The N1 implant consists of three main components: the external casing and electronics, the flexible electrode threads, and the surgical robot used for implantation.

The Implant Itself

The N1 is a hermetically sealed disc measuring approximately 23 millimeters in diameter and 8 millimeters in thickness, roughly the size of a coin. It is surgically embedded flush with the skull and sits entirely beneath the scalp, making it cosmetically invisible. The device is powered wirelessly through inductive charging, which means users do not need to plug in any cables or deal with percutaneous wires that penetrate the skin. This eliminates one of the major infection risks associated with earlier brain-computer interface designs.

Inside the N1 housing are custom-designed application-specific integrated circuits (ASICs) that perform critical signal processing functions. These ASICs amplify the extremely weak electrical signals produced by individual neurons, digitize them, and perform initial preprocessing before transmitting the data wirelessly via a high-bandwidth Bluetooth connection to external devices such as a laptop or smartphone. The device includes its own battery that provides all-day battery life on a single charge.

The Flexible Electrode Threads

Where Neuralink truly innovates is in its approach to electrode design. Rather than using rigid electrode arrays, which have been the standard for decades in brain-computer interface research, Neuralink developed ultra-thin, highly flexible threads made from biocompatible polyimide polymer. These threads measure between 4 and 6 micrometers in width, making them substantially thinner than a human hair (which is approximately 70 micrometers).

Each N1 implant includes 64 of these threads, and each thread contains 16 individual electrodes, resulting in a total of 1,024 electrode channels per implant. This represents approximately 10 times more electrodes than the Utah Array, the gold-standard rigid electrode array that has been used in brain-computer interface research for decades. The high electrode density allows Neuralink to record from many more neurons simultaneously, capturing much more detailed information about neural activity and enabling more sophisticated decoding of brain signals.

The flexibility of these threads is critical for long-term functionality. Unlike rigid electrodes, flexible threads can move with the brain as it naturally shifts and changes over time. This reduces the mechanical mismatch between the implant and brain tissue, which in turn reduces inflammation, glial scarring, and the loss of electrode-to-neuron contact that has plagued earlier rigid electrode designs. The threads are capable of both recording neural activity and delivering electrical stimulation back to the brain, though current applications focus primarily on reading rather than writing neural signals.

The Surgical Robot

Implanting 64 threads with 1,024 electrodes into the brain with sufficient precision to target the motor cortex while avoiding blood vessels and critical brain structures is far beyond what a human neurosurgeon could accomplish manually. This is where Neuralink's R1 surgical robot enters the picture.

The R1 is a specialized robotic surgical system with an insertion head featuring a tungsten-rhenium needle measuring just 25 micrometers in diameter. This needle is many times thinner than a blood vessel and can be controlled with sub-millimeter precision. The robot is capable of inserting up to six electrode threads, totaling 192 electrodes, per minute. The system uses computer vision and advanced imaging to identify blood vessels and other critical structures, automatically adjusting insertion angles to avoid them. This level of precision and automation helps ensure that electrode placement is optimal and that the risk of tissue damage is minimized compared to manual surgical implantation.

How Does The N1 Actually Decode Thoughts Into Actions

The process of converting neural signals into computer commands involves several sophisticated steps that blend neuroscience, signal processing, and machine learning.

Step One: Neural Signal Recording

When you think about moving your arm, even though the command never reaches your paralyzed muscles, the motor cortex of your brain still generates electrical signals that would normally travel down the spinal cord to initiate movement. These signals represent populations of neurons firing in coordinated patterns that encode information about movement direction, speed, and magnitude.

The electrode threads pick up these electrical signals, which are measured in microvolts (millionths of a volt) and change on timescales of milliseconds. The signals are extremely noisy because electrodes pick up activity from many neurons simultaneously as well as background electrical noise from the brain and body. The N1's internal ASICs amplify these weak signals and digitize them, converting analog neural activity into digital data.

Step Two: Signal Processing and Feature Extraction

Raw neural data from 1,024 channels sampling at high frequencies produces enormous amounts of data that must be compressed and processed before it can be transmitted wirelessly. The ASICs and external software work together to identify meaningful features in the neural data. One common approach involves detecting "spikes," the brief, sharp voltage deflections that occur when individual neurons fire action potentials. By identifying spikes and associating them with specific neurons, the system creates a more interpretable representation of neural activity.

The processed data is then transmitted wirelessly to the external computer running the decoding software. Current Neuralink systems achieve this with latency below 200 milliseconds, which is fast enough to feel intuitive and natural for most applications like cursor control.

Step Three: Neural Decoding and Machine Learning

The key innovation that makes Neuralink practical for real-world use is the decoder algorithm, which is a machine learning model trained to map patterns of neural activity to intended actions like mouse movements or button clicks.

The training process, called "body mapping" by Neuralink engineers, involves asking the patient to think about moving specific body parts (such as moving the cursor left, right, up, or down) while recording the corresponding neural activity. These recordings become the training data for the machine learning model. Modern decoders often use algorithms such as recurrent neural networks or other neural network architectures that can learn the complex, nonlinear relationship between neural firing patterns and the user's intended action.

Critically, this decoding process is highly personalized. Every brain is different, every person's motor cortex is organized differently, and the specific neural signals that correspond to movement intention are unique to each individual. Neuralink's decoders are trained on each patient's neural data specifically, which allows the system to adapt to natural variations in neural organization and even to changes in neural activity patterns over time as the patient learns to use the device.

Step Four: Real-Time Control

Once the decoder is trained, the system operates in real time. The patient thinks about moving a cursor or performing an action, that neural activity is recorded, processed, and decoded within the 200-millisecond window, and the corresponding action is executed on the computer. The user receives visual feedback on the screen, which allows them to make small adjustments to their mental commands in a feedback loop, gradually refining their control.

Importantly, the user does not need to consciously think through the technical details of each step. With practice, the same neural areas that normally control movement learn to interact with the Neuralink system through natural feedback and adaptation, much as the motor system learns to play piano or throw a baseball.

The Patients and What They Have Achieved

As of December 2025, 13 Neuralink patients have received implants across four hospital sites and three continents. Their achievements demonstrate the real-world capabilities and potential of the technology.

Noland Arbaugh: The First Patient

Noland Arbaugh, a 30-year-old man, became the first person to receive the N1 implant in January 2024 at Barrow Neurological Institute in Phoenix, Arizona. Arbaugh suffered a spinal cord injury in 2016 after a diving accident, leaving him paralyzed below his shoulders with quadriplegia. Before Neuralink, Arbaugh could move only his head and shoulders and relied on others for nearly all daily tasks. He spent most of his time in bed and used assistive devices such as a mouth-held stick to interact with computers.

Within hours of the N1 implant surgery, neuron spike detection was confirmed, and within days, Arbaugh demonstrated the ability to move a computer cursor with his thoughts. Over the following months, he progressed to playing video games such as Chess.com and Civilization VI, streaming his gameplay online, and even setting records for cursor control speed using a brain-computer interface. He was able to set his own world record for BCI cursor control, demonstrating that Neuralink could achieve performance levels exceeding what previous brain-computer interface systems had accomplished. Arbaugh could browse the web, send emails, and engage in social media, all using his thoughts to control the mouse and keyboard. He moved back into his family home and reported a profound improvement in his quality of life and sense of independence.

Arbaugh's journey was not without challenges. Several electrode threads partially retracted from his brain tissue in the weeks following surgery, a phenomenon referred to as "thread pull-out." This led to a temporary decrease in the number of functional electrodes and a reduction in data transmission speed. However, Neuralink engineers updated the decoding algorithms to compensate for this change, and Arbaugh's performance remained strong. The company has since modified the implantation technique to reduce the likelihood of thread retraction in subsequent patients.

Alex Conley: The Second Patient

The second patient to receive the N1 implant was a man named Alex (full name Alex Conley), an automotive technician with a spinal cord injury similar to Arbaugh's. Alex received his implant in July 2024 at Barrow Neurological Institute. Notably, Alex's implantation procedure differed slightly from Arbaugh's, as Neuralink made modifications to the surgical technique and potentially the chip design based on lessons learned from the first patient.

Alex's recovery was smooth with no reported complications. He rapidly demonstrated proficiency with the device and began using it to play video games and learn computer-aided design (CAD) software. By August 2024, less than one month after surgery, Alex was already creating 3D designs using CAD tools controlled entirely through his neural interface, showcasing not just cursor control but advanced digital manipulation capabilities.

Brad Smith: The Third Patient, Speaking Through AI Voice

In early 2025, Neuralink announced that a patient named Brad Smith, who has advanced ALS (amyotrophic lateral sclerosis), had received the N1 implant. Brad's case is particularly significant because he represents a different patient population than the first two, who had spinal cord injuries. ALS is a progressive neurodegenerative disease that eventually causes total paralysis of voluntary muscles. Brad is completely non-verbal and relies on a ventilator to breathe, representing one of the most severely disabled patient populations.

Brad's primary use of Neuralink is communication. Using the neural interface to control a computer cursor, he can compose text, which is then synthesized into speech using an AI-generated voice model cloned from his own voice prior to losing the ability to speak. In a compelling video released on social media, Brad narrated his experience using his thoughts to control the mouse, select words, and compose messages. He expressed that he can now communicate faster and in more nuanced ways than he could before using eye-gaze tracking, his previous assistive technology. Brad's case demonstrates that Neuralink is not just about controlling computers for leisure but about restoring the fundamental human ability to communicate and express oneself.

Nick Wray: The First Robotic Arm Control

In October 2025, ALS patient Nick Wray demonstrated controlled manipulation of a robotic arm using his Neuralink implant as part of the CONVOY feasibility trial. This represented a major milestone because controlling an external robotic limb requires decoding not just simple cursor movements but complex, coordinated multi-degree-of-freedom motion.

In a single 24-hour session, Nick completed an array of tasks including feeding himself, drinking from a cup, putting on a hat, microwaving food, and performing dexterity assessments. On the Nine-Hole Peg Test, a standard clinical assessment of fine motor function, Nick achieved 39 peg placements in five minutes, significantly exceeding average performance for people with ALS by 95 percent and outperforming previous brain-computer interface systems by 56 to 160 percent. These results demonstrate that the neural decoding from motor cortex activity can be translated not just to cursor movements but to sophisticated robotic control with real-world utility.

Paul: The First UK Patient

In October 2025, the first patient in the United Kingdom received the N1 implant at University College London Hospitals (UCLH) National Hospital for Neurology and Neurosurgery in London as part of the GB-PRIME clinical trial. The patient, referred to as Paul, has motor neurone disease (the term used in the UK for ALS). Within hours of surgery, Paul was able to control a computer cursor with his thoughts. By the following day, he had returned home from the hospital and continued using his implant in his domestic environment, demonstrating that the recovery and integration of the device proceed remarkably quickly.

Paul's case marks the expansion of Neuralink's clinical program from the United States to the United Kingdom and represents an important validation that the surgical procedures and protocols can be successfully implemented by clinical teams outside of the pioneering US sites.

The Global Expansion

As of November 2025, Neuralink has expanded to four hospital sites: Barrow Neurological Institute in Phoenix, Arizona; an additional site in Miami, Florida; University Health Network in Toronto, Canada (as part of the CAN-PRIME trial); and UCLH in London, UK (as part of the GB-PRIME trial). The company has also announced plans to expand to additional countries including the United Arab Emirates. Neuralink stated in mid-2025 that it was planning to perform 20 to 30 additional implant procedures throughout 2025, substantially increasing the patient population and accelerating the gathering of clinical data.

The Clinical Trials: PRIME and CONVOY

Neuralink's current clinical investigations are organized into two main trials, each with specific objectives.

The PRIME Study

PRIME stands for "Precise Robotically Implanted Brain-Computer Interface for Motor Enhancement." This study focuses on patients with quadriplegia due to spinal cord injury or limb paralysis. The primary objective is to evaluate whether the N1 implant safely and effectively enables users to control external devices such as computers and smartphones using thought alone. The PRIME study is enrolling up to five patients at participating sites.

Patients in PRIME are assessed on their ability to move a cursor, click buttons, type, play video games, and interact with common computer applications. The study also collects detailed performance metrics such as bits per second, which measures the amount of information the user can reliably transmit through the neural interface.

The CONVOY Study

CONVOY stands for "Brain-Computer Interface for Operating a Robotic Arm." This feasibility trial builds upon the foundation of PRIME and extends the investigation to real-world assistive device control. The CONVOY study evaluates whether Neuralink patients can control an investigational assistive robotic arm (ARA) to perform activities of daily living such as eating, drinking, and object manipulation.

The CONVOY study is particularly important because it demonstrates that neural decoding can extend beyond simple cursor control to sophisticated multi-degree-of-freedom robotic systems that perform real functional tasks. This directly addresses one of the primary stated goals of Neuralink: restoring not just digital autonomy but physical autonomy through robotic assistance.

Additional Trials and Future Applications

Beyond PRIME and CONVOY, Neuralink is actively developing applications in other domains.

Speech Restoration Trials

Neuralink is launching a clinical trial called "Our Voice" designed to restore speech to patients with speech disabilities resulting from ALS, stroke, or other neurological conditions. This trial will use the same N1 implant but with decoders specifically trained to recognize neural patterns associated with speech production, even though the actual vocal output is disabled. The decoded speech signals can be rendered as synthesized speech or displayed as text.

Visual Restoration: The Blindsight Trial

Neuralink is developing an experimental implant called Blindsight designed to restore vision to people who are blind. Unlike the N1, which targets the motor cortex, Blindsight implants electrodes into the visual cortex, the area of the brain that processes visual information. By stimulating electrodes in the visual cortex, the system can create perceptual experiences of visual patterns that allow blind individuals to navigate their environment, recognize faces, and read.

In September 2025, the FDA granted Blindsight "breakthrough device designation," a status that accelerates regulatory review and reflects the potential of this technology to provide a significant advantage over current treatment options for blindness. A clinical trial with Blindsight is expected to launch in the near term.

Emerging Applications

Looking further ahead, Neuralink and other researchers are exploring applications including treatment of neurological conditions such as epilepsy (through real-time detection of seizure precursors and targeted stimulation), Parkinson's disease (through closed-loop neuromodulation), depression and other psychiatric conditions, and cognitive enhancement in neurodegenerative diseases. However, these applications remain in earlier stages of development and research.

The Surgical Procedure: What The Implantation Actually Involves

The Neuralink implantation procedure is an elective neurosurgery performed under general anesthesia and typically lasts three to four hours.

Pre-Surgical Imaging and Planning

Before surgery, the patient undergoes advanced imaging studies, typically high-resolution MRI or CT, to map the brain's anatomical structures and identify the optimal surgical target. For motor-focused applications like PRIME and CONVOY, the target is typically the motor cortex in the precentral gyrus, specifically the region that represents hand and arm movements. The surgeons and Neuralink engineers use this imaging to identify a suitable implantation site that maximizes access to neurons encoding movement while minimizing the risk of encountering blood vessels or critical brain structures.

The Surgical Steps

The surgery begins with a craniotomy, in which a coin-sized opening is made in the skull. The dura, the tough membrane surrounding the brain, is carefully opened. The surgical robot is then positioned and uses its guidance system to identify where each of the 64 electrode threads should be inserted.

One by one, the robot uses its 25-micrometer-diameter needle to penetrate the brain tissue and deliver each electrode thread. The robot can insert multiple threads simultaneously and can insert up to six threads per minute. As each thread is inserted, the system checks for any nearby blood vessels and automatically adjusts insertion angles to avoid them.

After all 64 threads are implanted, the dura is closed, the N1 device (which is attached to the electrode threads) is positioned against the inside of the skull, and the bone is replaced. The incision is closed with sutures or surgical adhesive.

Post-Surgical Recovery

Remarkably, patients report relatively quick recoveries. Several Neuralink patients have reported spending only one to three days in the hospital for observation. Typically, within a week, staples can be removed and patients are cleared to begin training with the device. According to patient testimonies, there are often no lingering pain or complications, just excitement to begin using the implant.

The rapid recovery is enabled by several factors: the robot's precision minimizes tissue damage, the electrode threads are extremely thin and flexible, and modern anesthetic and surgical techniques have become very effective for neurosurgery. Of course, brain surgery always carries inherent risks, which we discuss in the next section.

The Risks and Safety Considerations

While Neuralink's results so far have been encouraging, it is critical to acknowledge the real risks associated with implanting a device in the brain.

Surgical and Post-Surgical Risks

Brain surgery carries inherent risks including infection, bleeding or hemorrhage, stroke, cerebrospinal fluid leakage, and damage to critical brain structures. These risks exist even with the most advanced technology and experienced surgical teams. While Neuralink's robot and techniques aim to minimize these risks through precision and automation, they cannot eliminate them entirely.

One specific phenomenon observed in Noland Arbaugh's case was "thread retraction," in which some of the implanted electrode threads moved outward or retracted from brain tissue in the weeks following surgery. The exact mechanism is not entirely understood but may relate to brain tissue inflammation, scar formation, or air pockets (pneumocephalus) formed during surgery. While this did not pose an acute safety risk to Arbaugh, it did reduce the number of functional electrodes available.

Long-Term Biocompatibility

A critical unanswered question is long-term biocompatibility and device stability. Will the electrode threads remain functional for years or decades? Will the constant influx of immune cells and scar tissue formation eventually degrade the signal quality? Will the device itself degrade or fail mechanically?

Animal studies have provided some reassurance that signals can be stable for months to years, but human data is still limited. Neuralink patients in current trials have had implants for approximately one year or less, so we do not yet have decades of long-term outcome data.

FDA Regulatory Concerns

The FDA initially delayed approval of Neuralink's human trials in 2022, citing concerns about the lithium battery used in the implant (which could pose a risk if it leaks), the potential for electrode threads to migrate within the brain, uncertainty about the safety of device removal if needed, and lack of long-term biocompatibility data. Neuralink addressed these concerns through additional animal studies, battery redesign, and detailed safety protocols, eventually receiving FDA approval for human trials.

The regulatory pathway continues to evolve as new data becomes available and as the device is tested in larger patient populations.

Infection Risk

Any implanted device carries a risk of infection. Unlike external brain-computer interface systems, implantable devices like Neuralink create a permanent foreign body in the brain. While the hermetically sealed design and inductive charging eliminate the need for percutaneous wires (which are a major infection risk in older systems), the implant itself is still a foreign material that could theoretically harbor bacteria or trigger infection.

To date, reported Neuralink patients have not experienced infections, but this remains a possibility that future patients must understand and accept.

Psychological and Ethical Considerations

Beyond the physical medical risks, receiving a brain implant raises profound psychological and ethical questions. What are the long-term psychological effects of having a device recording your brain activity? What are the privacy implications? Could the system be hacked? Could someone remotely interfere with the device? What are the implications for cognitive autonomy and mental privacy?

These are legitimate concerns that are being actively discussed in bioethics literature and regulatory contexts. Neuralink has implemented security measures including encryption and authentication, but absolute certainty of security can never be guaranteed with any technology.

Clinical Applications in Paralysis and Neurological Disease

In the near term (next 3-5 years), Neuralink could become a standard clinical treatment option for people with severe paralysis from spinal cord injury, ALS, or other causes. As the technology matures, costs decrease, and surgical expertise spreads to more medical centers, the number of patients who could benefit from these implants could expand significantly. Some estimates suggest millions of people worldwide might potentially benefit from BCIs to restore or augment function.

Communication Restoration

For people with conditions that affect speech, restoring communication through neural interfaces represents a profound humanitarian goal. If speech restoration trials prove successful, Neuralink could offer people with ALS, stroke, or other conditions the ability to communicate naturally despite severe paralysis.

Sensory Restoration

The Blindsight project suggests that neural interfaces could eventually restore vision to people who are blind, and similar principles could apply to hearing restoration. These applications are further in the future but represent the long-term potential of the technology.

Enhancement and Optimization

Elon Musk has explicitly stated that a long-term goal of Neuralink is not just restoration of lost function but augmentation and enhancement of human capability. This raises profound questions about cognitive liberty, equity, and what it means to be human. These questions will become increasingly important as the technology advances.

Regulatory and Ethical Framework Development

As brain-computer interfaces move from laboratory curiosities to clinical tools and potentially consumer products, regulatory agencies and bioethicists will need to establish clear frameworks for device approval, privacy protection, cognitive autonomy, and equitable access. Chile has already passed neural rights legislation, and other countries are beginning to grapple with these questions.

Long-Term Signal Stability

The biggest remaining question is long-term performance. Will electrode threads remain functional for decades? What happens as the brain forms scar tissue around the implant? How often will patients need surgical replacement or revision procedures? These questions cannot be fully answered until we have many more years of patient data.

Device Scalability

Current Neuralink procedures require highly specialized surgical teams and infrastructure. Before this becomes a widely available treatment, the surgical and technical expertise will need to be distributed to many medical centers worldwide. This is not trivial and will require significant training and infrastructure investment. 

Cost and Access

Brain implant surgery, hospitalization, and device manufacturing are expensive. Who will have access to this technology? Will it be limited to wealthy patients in developed countries? How do we ensure equitable access?

Decoding Algorithm Improvement

Current decoders achieve impressive performance, but there is still significant room for improvement. Decoding speech with high accuracy, for example, remains a significant technical challenge. Ongoing research in decoding algorithms, neural signal processing, and machine learning will be needed to expand the applications and improve performance.

Public Perception and Acceptance

Many people find the idea of having an electrode array implanted in their brain deeply concerning. Building public understanding and acceptance of this technology will require transparent communication about risks and benefits as well as demonstrated success stories. Ethical concerns about privacy, autonomy, and equity will need to be thoughtfully addressed.

Conclusion

Neuralink represents a significant milestone in the development of brain-computer interface technology. The successful implantation of 13 patients with demonstrated recovery of digital and physical autonomy shows that the science and engineering that seemed like science fiction a few years ago is now clinical reality. Patients like Noland Arbaugh, Brad Smith, Nick Wray, and Paul are not just test subjects, they are pioneers who are demonstrating that people with severe paralysis can regain independence and engage with the world in meaningful ways.

However, we are still in the very early stages. Patients have had implants for less than a year in most cases. Long-term safety and efficacy data are still being gathered. Scaling the technology to many more patients and medical centers will require solving significant technical, logistical, and regulatory challenges.

The future of Neuralink and brain-computer interfaces more broadly will depend on continued technical innovation, careful safety monitoring, thoughtful ethical and regulatory frameworks, and maintaining focus on the fundamental goal: restoring independence and dignity to people with severe neurological disabilities. If these conditions are met, Neuralink has the potential to transform the lives of millions of people and to fundamentally reshape our understanding of what is possible when neurotechnology, clinical medicine, and human determination come together.

References

1. American Psychological Association. (2020). Publication manual of the American Psychological Association (7th ed.). American Psychological Association.

2. Arbaugh, N. (2024). One year later: First Neuralink implant patient shares story [Video]. YouTube.  https://www.youtube.com/watch?v=mIAsUkJZbow

3. Arbaugh, N. (2025, August). Neuralink's first study participant says his whole life has changed. Fortune. Retrieved from  https://fortune.com/2025/08/23/neuralink-participant-1-noland-arbaugh-18-months-post-surgery-life-changed-elon-musk/

4. Barrow Neurological Institute. (2024, April 12). PRIME Study Site Announcement [Press release]. Retrieved from  https://www.barrowneuro.org/about/news-and-articles/press-releases/prime-study-site-announcement/

5. Dravet, N., Dravet, C., Masnou, P., & Baylis, F. (2025). Exploring the ethical landscape of brain-computer interfaces: Insights from recent clinical trials. Neuroethics, 18(2), 115-127.

6. Electronic Transactions in Artificial Intelligence/Neuroscience Journal. (2025). Neuralink and its advantages: Advancements in brain-computer interface technology. IJSAT, 3(6), Article 6777. Retrieved from  https://www.ijsat.org/papers/2025/3/6777.pdf

7. Federal Drug Administration. (2023, May). Investigational Device Exemption approval: Neuralink PRIME Study [FDA approval]. U.S. Department of Health and Human Services.

8. Federal Drug Administration. (2024, September). Breakthrough Device Designation: Blindsight vision restoration device. U.S. Department of Health and Human Services.

9. Federal Drug Administration. (2025, May). Breakthrough Device Designation: Speech restoration device. U.S. Department of Health and Human Services.

10. Horne, A. K., Choudhury, S., & Hornsby, B. W. (2020). Recent advances in neural interfaces: Materials chemistry to clinical translation. Science Translational Medicine, 9(404), eaav3638.

11. Kunz, E. M., Ahmadi, N., Horne, A. K., & Vargas-Irwin, C. E. (2025). Inner speech in motor cortex and implications for brain-computer interface control. Nature Neuroscience, 28(8), 1234-1245.

12. Kumar, R., Sharma, A., & Patel, S. (2025). The potential power of Neuralink: How brain-machine interfaces are advancing clinical neurotechnology. Expert Review of Medical Devices, 8(3), 301-315.

13. Musk, E., Neuralink Team, & Contributors. (2019). An integrated brain-machine interface platform with thousands of channels. Journal of Medical Internet Research, 7(10), e16194.  https://www.jmir.org/2019/10/e16194/

14. National Institutes of Health. (2025). Control of assistive devices via brain-computer interface (CONVOY Study). Clinical Trials Database. Retrieved from  https://clinicaltrials.gov/study/NCT06710626

15. National Institutes of Health. (2024). Precise robotically implanted brain-computer interface (PRIME Study). Clinical Trials Database. Retrieved from https://clinicaltrials.gov/study/NCT06429735

16. National Institute of Neurological Disorders and Stroke. (2025). Recent progress on Neuralink's brain-computer interfaces. International Journal of Psychological Science, 27(5), 234-256.

17. Neuralink. (2025). Neuralink N1 implant technical specifications [Technical document]. Retrieved from  https://www.neuralink.com

18. Neurapod. (2025, October 26). What is Neuralink's PRIME Study? Essential details revealed. Retrieved from  https://www.neurapod.com/blog/neuralink-prime-study

19. The University of Miami. (2025, January 26). Miami Project to cure paralysis selected as site for Neuralink clinical trial [Press release]. Retrieved from https://news.med.miami.edu/the-miami-project-to-cure-paralysis-selected-as-site-for-neuralink-clinical-trial/

20. Truccolo, W., Friehs, G. M., Donoghue, J. P., & Hochberg, L. R. (2022). Decoding speech from human motor cortex using an intracortical brain-computer interface. Annals of Neurology, 91(2), 209-221.

21. University College London Hospitals NHS Foundation Trust. (2025, October 26). First UK patient uses thought to control computer hours after Neuralink implant [Press release]. Retrieved from  https://www.ucl.ac.uk/news/2025/oct/first-uk-patient-uses-thought-control-computer-hours-after-neuralink-implant

22. University College London Hospitals NHS Foundation Trust. (2025, July 30). UCLH to evaluate safety and functionality of Neuralink's brain-computer interface BCI technology [Press release]. Retrieved from https://www.uclh.nhs.uk/news/uclh-evaluate-safety-and-functionality-neuralinks-brain-computer-interface-bci-technology

23. University of Michigan. (2014). Flexible neural interfaces fabricated using microthread electrodes [Research article]. IEEE Transactions on Biomedical Engineering, 61(8), 2418-2429.

24. Vargas-Irwin, C. E., Shanechi, M. M., Skorjalek, J. B., Ting, J. T., Vitale, F., & Pandarinath, C. (2018). Decoding complete reach and grasp actions with simultaneous intracortical and intra-articular recordings. Journal of Neural Engineering, 15(3), 036002.

25. Waghmare, H., & Mukherjee, A. (2025). Architectural design and implantation strategies of brain-computer interfaces: A comparative analysis. Aucores Online Journal of Biomedical Engineering, 8(4), 156-178.

26. Wray, N. (2025, November). Barrow patient controls robotic arm with thought in CONVOY Study [Video]. YouTube. Retrieved from  https://www.youtube.com/watch?v=efeABhKBKg4

27. Ziaeefard, S., Berger, M. A., & Sanger, T. D. (2025). Electrode materials are key to brain-computer interface technology advancement. Materials Today, 42, 88-105.