A gelatin memristor with 16 stable conductance states mimics biological pain perception, rating intensity, sensitizing after injury, and self-healing while directly controlling mouse muscle response.
(Nanowerk Spotlight) A sharp pinprick, a burn from a hot stove, the ache of a stubbed toe. Each sensation triggers an immediate, graded response calibrated to the severity of the threat. Biological nociceptors, the specialized nerve endings responsible for detecting harmful stimuli, do far more than simply signal “pain.” They rate intensity across multiple levels, become hypersensitive after injury to guard damaged tissue, and gradually return to normal as healing progresses. Engineers have been trying to replicate these capabilities in artificial systems, but electronic pain sensors have typically offered only binary detection, registering pain or its absence.
Building humanoid robots or advanced prosthetics demands something more nuanced. A robotic hand that cannot distinguish between a gentle touch and a dangerous crush, or that fails to protect itself after sustaining damage, will struggle in unpredictable real-world environments. Traditional silicon circuits lack the adaptive, memory-like properties needed to mimic biological pain response.
Memristors offered a promising alternative. These two-terminal electronic components have resistance that depends on the history of current passed through them, enabling a form of memory similar to how neurons retain traces of prior stimulation.
Previous work showed that memristors with threshold switching behavior could replicate some nociceptive features, including heightened sensitivity after repeated stimulation. Yet artificial nociceptors capable of rating pain across multiple intensity levels while also healing after injury remained out of reach.
A study published in Advanced Functional Materials (“Bioinspired Artificial Nociceptor Based on Quantized Conductance Memristor With Pain Rating, Self‐Healing, and Neuromodulation Capabilities”) now reports an artificial nociceptor that achieves both capabilities. Researchers at Northeast Normal University in China built a bio-electronic sensorimotor system combining a pressure sensor, a memristor functioning as the nociceptor, and the sciatic nerve of an anesthetized mouse. The system converts mechanical pressure into graded electrical signals that directly stimulate muscle contraction, forming a complete artificial reflex arc.
The key innovation is the memristor’s quantized conductance behavior. Rather than switching between just two states like conventional devices, this memristor exhibits 16 discrete conductance levels. Each corresponds to an integer multiple of the conductance quantum, approximately 77.5 µS. These stable states allow the device to encode pain intensity much as biological nociceptors classify sensations from absent to severe.
Biological and artificial nociceptive nerves with the capability of pain rating. (a) Schematic of human nociceptive reflex arc in corresponding to various degrees of pain perception. The pain intensity scales are normally rated as no pain, mild, moderate, and severe levels. (b) Schematic of artificial nociceptive sensorimotor nerve. It is constructed by a gelatin-based pressure sensor, a gelatin-based quantized memristive nociceptor, and the sciatic nerve. The multiple quantized states of the memristor represent the pain intensity scales. (Image: Reproduced with permission from Wiley-VCH Verlag)
The memristor consists of a thin gelatin layer, the familiar protein derived from collagen, sandwiched between magnesium and nickel electrodes. When current flows, magnesium atoms oxidize and migrate through the gelatin as ions. Carboxylic acid groups in gelatin’s molecular structure channel this migration along orderly paths, producing reproducible conductive filaments rather than random tangles. Imaging with conductive atomic force microscopy confirmed these filaments have a cone-shaped structure that explains the stepwise conductance changes.
The pressure sensor uses the same gelatin base, converting mechanical force into current at approximately 8.49 µA/kPa across a 7 to 53 kPa range. Treating the hydrogel with sodium citrate exploits the Hofmeister effect, where certain salts alter protein structure, more than doubling breaking stress from 0.11 to 0.23 MPa.
Connecting the sensor in series with the memristor creates the artificial nociceptor. Increasing pressure lowers the sensor’s resistance, raising voltage across the memristor and pushing it into higher conductance states. Four different pressure intensities produced four distinct quantized outputs, confirming pain-rating capability.
The device also mimics two features of injured biological nociceptors. Allodynia causes pain from normally harmless stimuli; hyperalgesia produces exaggerated responses to painful ones. After applying high “traumatic” pressure, the researchers found that the triggering threshold dropped and subsequent stimuli generated larger conductance changes. The memristor remembered its injury and grew more sensitive, just like damaged tissue.
Conductance states naturally relax over time. The highest state (16 G₀) persists beyond 10⁴ seconds, while states below 5 G₀ decay in roughly 30 seconds. This parallels how severe pain lingers longer than mild discomfort. Warming the environment from 25 °C to 45 °C shortened decay time threefold for the 14 G₀ state, consistent with thermal energy helping nanoscale filaments break apart.
The gelatin hydrogel also heals physical damage. Cuts of 8.3 µm and 50.7 µm width disappeared after treatment at 60 °C for 20 minutes, with full electrical function restored. Six cut-heal cycles caused no degradation.
To validate the system, the researchers wired their artificial nociceptor to a mouse sciatic nerve, testing three animals with three repetitions each. Electrical signals bypassed the central nervous system and directly triggered leg muscles. Action potentials rose from 6.2 to 35.5 mV with increasing pressure, moving the hind limb through angles from 0° to 15°. A 4 × 4 sensor array showed that activating more sensors simultaneously increased limb displacement from roughly 5° to 12° when eight sensors fired together.
The prototype has limitations. The 60 °C healing temperature exceeds biological tolerance, though it suits robotic applications. Energy consumption remains low at approximately 23.9 pJ to switch on and 14.2 pJ to reset. The device reaches high conductance in about 20 nanoseconds and returns to baseline in roughly 128 nanoseconds.
This artificial nociceptor brings electronic systems closer to the adaptive, self-protecting behavior of biological pain perception. Robots equipped with graded pain sensing, injury-induced sensitization, and gradual recovery could operate more safely in complex environments. The approach may also inform neuroprosthetic design and new strategies for managing chronic pain through engineered nerve interfaces.
For authors and communications departmentsclick to open
Lay summary
Prefilled posts
Nanowerk Newsletter
Get our Nanotechnology Spotlight updates to your inbox!
Thank you!
You have successfully joined our subscriber list.
Become a Spotlight guest author! Join our large and growing group of guest contributors. Have you just published a scientific paper or have other exciting developments to share with the nanotechnology community? Here is how to publish on nanowerk.com.
