Precision Neuroscience demonstrates thought-based gaming in non-paralysed volunteers

By Inside BCI · July 21, 2026

In an ICU or step-down bed somewhere on Precision Neuroscience’s clinical network of fifteen or so US hospitals, a man named Jasaun sat and played a game of Pong against his own neurosurgeon. He won or he lost is not on the record. What is on the record is that Jasaun was moving the paddle with his thoughts, through a one-thousand-and-twenty-four-electrode thin-film subdural array that had been placed on the surface of his brain earlier that same hospital stay during a scheduled neurosurgical procedure, and that when he went home the array came out at his bedside in under five minutes. Precision made the video public on Tuesday 21 July 2026 through a MobiHealthNews Q&A with chief clinical officer Jayme Strauss and a same-day MassDevice write-up. Jasaun is the first-named participant on camera; Strauss says a “number of patients” have now done comparable sessions. She calls them the first able-bodied recipients of a brain-computer implant. That framing is Precision’s own, and it requires some careful unpacking.

What the announcement actually reports

Precision Neuroscience’s Layer 7 Cortical Interface is a thin-film array of 1,024 microelectrodes on a substrate roughly one-fifth the thickness of a human hair, designed to sit on the surface of the cerebral cortex without penetrating the tissue underneath. The device received FDA 510(k) clearance on 17 April 2025 for the recording, monitoring and stimulation of electrical activity on the brain’s surface, with implantation durations of up to thirty days. Since the clearance, Precision has been placing Layer 7 in patients who are already undergoing scheduled neurosurgery for another medical reason (a tumor resection, a long-term epilepsy monitoring workup, an awake craniotomy for cortical mapping) and inviting them to enrol in Precision’s research protocol alongside the primary procedure. The array goes in during that neurosurgery. Precision’s team works with the patient across the remainder of their hospital admission, running training sessions in the ICU or monitored bed multiple times a day. At discharge, the array comes out.

The 21 July announcement is Precision saying, publicly and on video, that participants in this cohort have used the Layer 7 array to perform sustained real-time thought-based control of consumer computing tasks that materially exceed the abstract cursor demonstrations that the surface-ECoG brain-computer interface literature has been producing since 2004. Strauss’s on-record inventory of tasks demonstrated across the cohort: grid selection (choosing boxes that light up), Pong (Jasaun’s match against his neurosurgeon), Mario Kart, Instagram scrolling, answering FaceTime calls, Fruit Ninja. MassDevice adds phone-call placement and message-sending to the list. The Precision homepage now reports 95-plus cumulative implanted clinical-study patients across all cohorts, a figure that pools paralysed and non-paralysed participants and is not specific to the July 21 milestone. Strauss did not disclose how many of those 95 are the non-paralysed participants relevant to the Tuesday announcement. Precision’s active hospital network runs to more than fifteen partners, anchored by Mount Sinai, WVU Medicine, Penn Medicine, Mass General Brigham and Johns Hopkins Medicine, among others.

Strauss’s own on-record summary quote, which is worth reading directly: “Our goal from day one has been to break down the walls of isolation that come with paralysis. By achieving real-time cursor control completely on the surface of the brain, we’ve proven that we don’t have to compromise safety to achieve high-performance results. This is a thrilling glimpse into the future of BCIs, and we are moving faster than ever to bring this technology to patients who need it most.”

What “first able-bodied” actually means, and does not mean

Precision’s framing that this is the “first able-bodied” cohort to receive a brain-computer implant is narrower than the headline reads. The published surface-electrocorticography brain-computer interface literature since 2004 has included patients who are neurosurgically undergoing epilepsy monitoring with subdural electrode grids and who, while implanted for their clinical epilepsy workup, agree to sit in a research chair and drive a cursor with their thoughts. Eric Leuthardt, Gerwin Schalk, Jonathan Wolpaw, Jeffrey Ojemann and Daniel Moran published exactly this at Washington University in 2004 in the Journal of Neural Engineering, demonstrating one-dimensional cursor control via subdural ECoG electrodes in epilepsy patients. Schalk and colleagues extended the approach to two-dimensional closed-loop cursor control in the same population in a follow-on 2008 Journal of Neural Engineering paper. Multiple follow-on studies through the 2000s and 2010s continued this line of work. The peer-reviewed literature routinely refers to epilepsy-monitoring patients as “able-bodied” specifically in contrast to the paralysed BrainGate consortium cohort.

What is genuinely new in Precision’s demonstration, and what survives a rigorous fact-check, is a narrower set of specifics that Precision is not the party best positioned to articulate in a press release. First, the electrode density: Layer 7 is a purpose-built 1,024-electrode high-density array on a thin-film substrate. The 2004-era ECoG BCI work used clinical-grade Ad-Tech-style subdural seizure monitoring grids with roughly sixty-four electrodes at one-centimetre spacing, designed for epilepsy diagnostics rather than BCI use. Second, the training window: Precision’s April 2025 510(k) permits up to thirty days of implantation, which supports multi-session in-hospital training that legacy epilepsy-monitoring protocols never had time to run. Third, the applications: Mario Kart, Instagram scrolling, and Fruit Ninja are meaningfully different from a one-dimensional cursor task, and the shift from research tasks to consumer applications is the demonstration the video actually makes. The precise defensible claim is that Precision has shown, for the first time on public video, sustained multi-day thought-based control of consumer computing tasks in non-paralysed neurosurgical patients using a purpose-built high-density surface array. That is a real advance. It is not the first ECoG BCI, it is not the first non-penetrating BCI, and it is not the first BCI cursor control in an able-bodied human, and Inside BCI is not going to write it as any of those.

There is a further important distinction that Precision’s own language does elide. Strauss confirmed to MobiHealthNews that the Layer 7 array is removed at bedside in under five minutes at the end of the participant’s hospital stay. Layer 7 in this protocol is a temporary intraoperative and perioperative research placement, not a chronic implant. Strauss stated that the first chronic implant, targeted through a separate Early Feasibility Study for a permanent version of Layer 7, will “probably” begin at the beginning of 2028. The gap between the video-marketing language of “receive brain-computer implant” and the reality of “have a research array placed for the duration of a scheduled hospital stay and then removed at bedside” is a real editorial gap that Precision’s own competitors (Neuralink, Paradromics, Synchron, and the Chinese cohort Neuracle, StairMed, NeuroXess) will note.

Who Precision Neuroscience is

The company was founded in 2021 in New York City with an anti-Neuralink structural thesis at its core. Ben Rapoport, one of the eight-member founding team at Neuralink, left Neuralink in 2018 and co-founded Precision alongside Michael Mager, Demetrios Papageorgiou, and Mark Hettick. Rapoport holds a PhD in electrical engineering and computer science from MIT, an MD from Harvard Medical School, and a neurosurgery residency at Weill Cornell Medical Center. He is currently Chief Science Officer at Precision, Assistant Professor of Neurosurgery at Icahn School of Medicine at Mount Sinai, and Scientific Director of the Mount Sinai BioDesign medical device incubator. Mager is CEO and comes from a business rather than scientific background. Papageorgiou and Hettick are the two co-founding engineers most publicly associated with the Layer 7 device architecture. Hettick and Elton Ho are the co-first authors on Precision’s October 2025 Nature Biomedical Engineering paper on the array’s implantation and multimodal decoding, with Papageorgiou among the senior authors.

Precision has assembled a broader executive team since. Craig Mermel joined in March 2022 to lead machine learning and software and stepped up to Chief AI and Data Officer on 8 April 2026. Brian Otis, a former Verily co-founder, came in as CTO in 2024. Jayme Strauss, the voice on Tuesday’s announcement, runs clinical and commercial. Mike Kaswan is CFO.

The company has raised approximately US$180 million cumulative as of January 2026 per Bloomberg. The largest disclosed round is the US$102 million Series C announced 16 December 2024, led by General Equity Holdings, with Mubadala Capital and Stanley Druckenmiller’s Duquesne Family Office among the co-investors alongside B Capital, Steadview Capital, and Forepont Capital Partners. Post-Series C post-money valuation was reported at approximately US$500 million per Bloomberg via MassDevice. A follow-on undisclosed investment from SCI Ventures in November 2025 brought Christopher and Dana Reeve Foundation, Wings for Life, and Spinal Research money into the cap table on the paralysis-advocacy side. Precision owns and operates a 22,000-square-foot MEMS fabrication facility in Addison, Texas, and maintains offices in New York City and Santa Clara.

First-in-human intraoperative use of Layer 7 was in June 2023 at West Virginia University’s Rockefeller Neuroscience Institute (not Baylor, per corrections landed during pre-draft verification). Subsequent clinical partnerships added Mount Sinai and the Hospital of the University of Pennsylvania in March 2024. In May 2024 the Mount Sinai team under Joshua Bederson placed four Layer 7 arrays simultaneously in a single tumor-resection patient for a total of 4,096 electrodes covering approximately eight square centimetres of cortex, which Precision described at the time as a world record for simultaneous cortical microelectrode recording. In October 2025 the team published its consolidated technical case in Nature Biomedical Engineering. In February 2026 the Johns Hopkins group led by Lehner, Luo and colleagues published Precision’s first peer-reviewed cursor and speech-classification results across four awake-craniotomy patients in Neurosurgical Focus.

The strategic partnership that most sharpened Precision’s competitive positioning against Neuralink landed on 12 January 2026: Medtronic and Precision announced a co-development deal to integrate Layer 7 with Medtronic’s StealthStation surgical navigation platform. Medtronic’s endorsement of Precision, rather than of Blackrock or of Synchron, is the industry-validation signal that positioned Precision as a serious commercial-BCI infrastructure play regardless of whether the 21 July announcement holds up under superlative scrutiny.

What to watch

Watch the EFS filing for the chronic version of Layer 7. Strauss’s “beginning of 2028” for the first chronic implant is the operationally binding milestone. Precision will need to file an Investigational Device Exemption for a chronically implanted variant, which is a materially different regulatory pathway than the April 2025 510(k) for temporary use. Whether the IDE lands in Q4 2026 or slips into 2027 is the FDA-side signal to track.

Watch how the Medtronic partnership shows up in the eventual chronic product. The StealthStation integration announced in January 2026 has not yet appeared in a live Layer 7 procedure. When it does, and when the industrial-anchor partnership becomes concrete rather than a press release, Precision’s position against Neuralink shifts from “the founder left” to “and one of the largest neurosurgical device companies in the world went with them.”

Watch how the “able-bodied” framing survives peer scrutiny. Precision’s press-release language will meet the Leuthardt 2004 counter-example within days of the video’s second week, and the company’s medium-term credibility as a scientific communicator depends on whether it walks the framing back or defends it precisely. The narrow defensible version (first sustained multi-day consumer-application control in non-paralysed neurosurgical patients using a purpose-built high-density surface array) is a real accomplishment. The broad falsifiable version (first able-bodied to receive brain-computer implant) is not, and BCI-community fact-checkers will notice.

Watch the paralysed-cohort milestone that Precision’s own strategic mission actually points at. Strauss’s opening quote is about breaking down the walls of isolation that come with paralysis. The able-bodied video is the intermediate proof point. The chronic implant in a paralysed volunteer, targeted for early 2028, is where the commercial story lives. Every intervening announcement is context for the 2028 milestone.

Sources

news Precision-Neuroscience Layer-7 Jayme-Strauss Benjamin-Rapoport Michael-Mager Mark-Hettick Demetrios-Papageorgiou Craig-Mermel surface-ECoG non-penetrating-BCI FDA-510k Mount-Sinai West-Virginia-University-Rockefeller Medtronic-StealthStation State-Patchwork-posture

What the announcement actually reports

What “first able-bodied” actually means, and does not mean

Who Precision Neuroscience is

What to watch

Weekly BCI Brief in your inbox