Neurotechnology, explained

What is neurotechnology?

Neurotechnology is the set of devices and methods used to measure, interpret or influence the activity of the nervous system. The OECD Neurotechnology Toolkit (2024) defines it as “devices and procedures used to access, monitor, investigate, assess, manipulate and/or emulate the structure and function of the neural systems of natural persons”. In November 2025, UNESCO’s member states adopted a Recommendation on the Ethics of Neurotechnology, the first global framework for the field.

The word covers a wide range: an electrode finer than a hair that listens to a single neuron, a wearable headset that reads brain rhythms through the scalp, an implanted stimulator that eases tremor, a brain–computer interface that turns neural activity into a cursor movement, and the software that makes sense of all of these signals. What they share is a direct line between a technology and the nervous system.

The main kinds of neurotechnology

Most neurotechnology does one or more of four things.

• Recording. Measuring neural activity: the electrical impulses of individual neurons, the summed activity of populations, or slower chemical and metabolic signals. Electrophysiology with microelectrodes is the method with the finest resolution in time.
• Stimulation. Delivering energy, usually electrical current, to drive or quieten neural activity. Clinical examples include cochlear implants and deep-brain stimulation; in the laboratory, stimulation is used to test what a circuit does.
• Interfacing. A neural interface is a device that records from or stimulates the nervous system. The Royal Society’s 2019 report iHuman describes neural interfaces as devices “implanted in the body, or worn externally” that are “capable of recording or stimulating activity in the brain and peripheral nervous system”. A brain–computer interface (BCI) is a neural interface plus the computing that turns recorded activity into a command, or a command into stimulation.
• Interpreting. Signal processing, statistics and machine learning that turn raw measurements into something meaningful: which neuron fired, what the person intended, how tissue is behaving. Increasingly this includes neuromorphic hardware, which processes signals in ways modelled on the nervous system itself.

What experimental neuroscience needs from neurotechnology

Experimental neuroscience is the study of the nervous system through controlled experiments, often in animals, tissue or cells. Its progress is tied to its tools: a question about how neurons coordinate can only be asked with a device that records many neurons at once, with millisecond timing, without changing what they would have done anyway.

That last condition is the hard one. Every probe inserted into the brain displaces tissue, and the damage it causes scales with its size. A central problem in neural-interface engineering is therefore to make electrodes smaller without losing signal, and to record from more of them without multiplying the wiring. Our published work sits on exactly this problem:

• jULIEs: glass-insulated metal fibres under 30 µm across, with nanostructured tips that bring impedance low enough to record and stimulate single neurons, tested in mice (Journal of Neural Engineering, 2022).
• CHIME: bundles of such fibres pressed directly onto mass-produced CMOS amplifier chips, so recording can scale with the chip industry (Frontiers in Neuroscience, 2020).
• Microwire bundles on camera and display chips: up to 8,640 wires on a single chip, recording from retina and from awake, running mice (Science Advances, 2020, with collaborators at Stanford).

Each of these is experimental neurotechnology: peer-reviewed, demonstrated in laboratory animals, and not a clinical device. Our technology page labels every platform with its evidence status.

What is cancer neurotechnology?

Cancer neuroscience is the study of how cancers and the nervous system interact. Over the past decade it has shown that gliomas, the most common malignant primary brain tumours, form connections with neurons and respond to neural activity (Venkatesh et al., Nature, 2019; reviewed by Mancusi & Monje, Nature, 2023, and Winkler et al., Cell, 2023).

We use cancer neurotechnology to mean the application of neurotechnology to that interaction: measuring the electrical and physicochemical state of living tissue across the boundary between a brain tumour and the brain around it, with neural interfaces, electrochemistry, impedance, spectroscopy and computational analysis. The aim is to find out which measurements are reproducible, how signals from different physical domains relate, and whether together they add useful information to imaging and molecular pathology. This is an active research programme; it does not yet offer a diagnostic or surgical tool. Our research page sets out the five lines of enquiry.

Translational neurotechnology

Translational neurotechnology is the work of turning a laboratory method into something that can be used reliably in a clinical or commercial setting. It is slower than the headlines suggest. Between a published experiment and a clinical tool lie analytical validation, preclinical evidence, usability and workflow studies, quality management, regulatory approval, clinical trials and comparison with current practice. A granted patent, such as the two listed on our patents page, records that an invention was disclosed and examined for novelty; it is not evidence that the invention works in people.

Neuromorphic BCI

Neuromorphic computing uses hardware whose organisation is modelled on nervous systems: many simple units working in parallel, communicating through sparse events rather than a central clock, often with memory and processing in the same place. A neuromorphic brain–computer interface would process neural signals in such hardware, close to the electrodes, instead of streaming every raw sample to a conventional computer. The potential advantages are lower power, lower latency and less data to move. Whether those advantages are real for a given task has to be measured against a conventional baseline. For this lab it is a research direction, not a built platform.

Further reading

• OECD. OECD Neurotechnology Toolkit (2024), and the OECD’s neurotechnology topic page.
• UNESCO. Recommendation on the Ethics of Neurotechnology, adopted by the General Conference in November 2025.
• The Royal Society. iHuman: Blurring lines between mind and machine (2019).
• Mancusi R, Monje M. The neuroscience of cancer. Nature, 2023.
• Winkler F, et al. Cancer neuroscience: state of the field, emerging directions. Cell, 2023.

These are independent publications that inform our work; they are not the lab’s own.