Neuroprosthetics
Active FrontierNeuroprosthetics
Neuroprosthetics applies brain-computer interface technology to restore lost function — motor control, speech, and cognitive capabilities — for patients with neurological conditions. The field is transitioning from proof-of-concept demonstrations to sustained clinical use, driven by improvements in both the hardware (electrodes, implants) and software (decoding algorithms, adaptive interfaces) layers.
Motor function restoration is the most mature application. BCI-driven robotic arms, cursor control, and functional electrical stimulation of paralyzed limbs have been demonstrated across multiple research groups. Neuralink's PRIME study is the most visible implementation, with at least 24 patients implanted by May 2026 using thought-controlled computing for typing, gaming, and communication. Its CONVOY extension moved this from screen control into the physical world — in Oct 2025 an ALS participant used a BCI-controlled robotic arm to drink from a cup, microwave food, and open a refrigerator, setting records on dexterity tests used for stroke rehabilitation.
Speech restoration is the highest-impact frontier — and in 2026 it crossed from lab demonstration into durable, real-world clinical use plus dedicated commercial programs. Card et al. (Nature Medicine 2026) showed an ALS/dysarthria participant using a multimodal speech+cursor intracortical BCI for >3,800 hours at home over ~2 years with no researchers present (183,060 sentences; 56 wpm; 92% ≥mostly-correct) — the strongest evidence that speech neuroprosthetics can work as a sustained daily assistive device, not just in a controlled session. Two companies now have speech-first BCIs in patients: Paradromics (first Connexus implant, June 2026; 421 microelectrodes; >200 bps preclinical) and Neuralink VOICE (FDA Breakthrough Device Designation). For patients with locked-in syndrome or severe ALS, this is the closest the field has come to restoring fluent communication, though open-vocabulary natural-speed speech remains the unsolved target.
Neurorehabilitation uses BCI not as a permanent prosthetic but as a rehabilitation tool — providing neurofeedback during physical therapy to promote neuroplasticity and motor recovery after stroke or traumatic brain injury. The BCI measures whether the patient is generating appropriate motor intention signals, even if those signals can't yet produce movement, and provides real-time feedback to reinforce correct neural patterns.
Cognitive applications remain earlier-stage: BCI-assisted attention training, memory augmentation, and treatment of neuropsychiatric conditions (depression, PTSD, addiction) through targeted neuromodulation.
Key Claims
- Motor BCI reaching sustained clinical use — Neuralink PRIME patients (≥24 by May 2026) using thought-controlled computing for daily activities; CONVOY extends to robotic-arm control of real objects. Evidence: strong (Neuralink PRIME, Neuralink VOICE/CONVOY)
- Speech neuroprosthetics demonstrated as a durable at-home device — >3,800 hrs unsupervised use over ~2 years (ALS/dysarthria), 56 wpm, 92% sentences ≥mostly-correct — beyond one-session benchmarks. Evidence: strong (Nature Medicine — Card)
- Speech-first commercial BCIs now in humans — Paradromics Connexus (first implant June 2026, >200 bps preclinical) and Neuralink VOICE (FDA Breakthrough) target communication restoration directly. Evidence: moderate (Paradromics Connexus, Neuralink VOICE)
- Sensory neuroprosthetics opening a new track — Neuralink's Blindsight stimulates the visual cortex to create perception in the blind (FDA Breakthrough 2024; first trials 2026), extending neuroprosthetics from motor/speech output to sensory input. Evidence: weak (single tech-report/news) (Neuralink Blindsight)
- BCI for neurorehabilitation promotes neuroplasticity — Neurofeedback during therapy reinforces correct neural patterns for motor recovery. Evidence: moderate (BCI Neuroprosthetics)
- Cognitive BCI applications remain early-stage — Attention, memory, and neuropsychiatric treatment through targeted neuromodulation. Evidence: preliminary (BCI Neuroprosthetics)
Open Questions
- Can speech BCI achieve natural-speed, open-vocabulary communication?
- What is the minimum electrode count needed for effective motor prosthetics?
- How do neuroprosthetics interact with natural neuroplasticity over years of use?
- What ethical frameworks govern cognitive enhancement vs. restoration?
Related Concepts
- Neural Signal Decoding — The computational engine powering neuroprosthetic control
- Invasive vs. Non-Invasive BCI — Modality determines which applications are feasible
- BCI Clinical Applications — Broader clinical landscape: epilepsy, Parkinson's, depression, sensory loss
- Neural Decoding — AI/ML approaches for decoding neural signals in rehabilitation contexts
Backlinks
Pages that reference this concept:
Changelog
- 2026-04-05 — Created from Frontiers neuroprosthetics review and Neuralink PRIME sources
- 2026-04-14 — Updated sources to include ScienceDirect clinical review (Deng et al.) and VIT Chennai AI review; added cross-links to new concept pages
- 2026-06-24 — Added 2026 milestones: Card et al. at-home speech neuroprosthesis (>3,800 hrs, 56 wpm), Neuralink CONVOY robotic-arm + VOICE speech + Blindsight sensory tracks, and Paradromics Connexus first implant; new claims and Paradromics links
Related Concepts
BCI Clinical Applications
Active FrontierInvasive vs. Non-Invasive BCI
Active FrontierNeural Decoding
Active FrontierNeural Signal Decoding
Active FrontierTheses that depend on this concept
These research positions cite this concept in their evidence. If the concept changes materially, these theses may need re-scoring.