Artificial neurons communicate with living brain cells

Northwestern University researchers said on April 15 that they had built printed artificial neurons that could send brain-like electrical signals to living brain cells in lab experiments. The team said the devices, made from printable electronic materials, activated neurons in slices of mouse brain tissue rather than only mimicking neural behavior in isolation. (news.northwestern.edu) The work was published in *Nature Nanotechnology* and was led by Mark C. Hersam, with Vinod K. Sangwan as a co-lead, according to Northwestern and the journal listing. (nature.com) The experiments are part of a broader push to build hardware that works more like biological nervous systems. Northwestern said the printed devices were flexible and low-cost, and that the signals they produced were close enough to biological firing patterns to trigger responses in real neurons. (news.northwestern.edu) Hersam said the work could inform both brain-interfacing electronics and lower-power computing hardware. ### What exactly did the researchers get the devices to do? Mouse brain tissue was the test bed for the central result. Northwestern said the artificial neurons were applied to slices of mouse brain and successfully triggered responses from living neurons, showing direct signal compatibility between the printed devices and biological tissue. (news.northwestern.edu) ScienceDaily, citing Northwestern, described the result as activation of biological brain tissue by lifelike electrical signals. Nature Nanotechnology’s article summary said the platform used printed MoS2 memristive nanosheet networks for spiking neurons with “multi-order complexity.” The journal summary added that thermally activated snap-back produced physiological waveforms that stimulated mouse Purkinje neurons. (news.northwestern.edu) ### What are these “artificial neurons” made from? Nature Nanotechnology identified the devices as printed MoS2 memristive nanosheet networks. Northwestern described the approach as based on “electronic inks” deposited onto a flexible polymer substrate, rather than conventional rigid silicon hardware. That matters because the researchers were trying to reproduce some of the heterogeneity and time-dependent behavior seen in biological neurons, not just build another fixed transistor array. (news.northwestern.edu) The journal’s summary said the printed networks yielded spiking neurons with behavior complex enough to resemble physiological neural signaling. Northwestern said the devices were designed to generate electrical signals “realistic enough” to activate living brain cells. (nature.com) ### Why are researchers pursuing this kind of hardware? Mark C. Hersam said the motivation includes the energy demands of modern artificial intelligence systems. In Northwestern’s account, Hersam said the brain is “five orders of magnitude more energy efficient than a digital computer,” and argued that brain-inspired hardware could help handle data-intensive AI workloads more efficiently. (nature.com) Northwestern also said possible applications include brain-machine interfaces and neuroprosthetics for hearing, vision and movement. Those uses remain prospective; the reported experiments were done in mouse brain slices in the laboratory, not in human patients or implanted systems. (nature.com) ### How far is this from a device that works in people? The April 15 paper is a laboratory study, not a clinical report. Northwestern’s description of the work centers on ex vivo mouse brain tissue, and the journal summary frames the result as a platform for bio-realistic neuromorphic hardware and brain-machine interfaces rather than a finished medical product. (news.northwestern.edu) That distinction matters because some social posts this week presented the result as a fresh development. The underlying peer-reviewed study, however, was published on April 15, 2026, in *Nature Nanotechnology*, with public summaries from Northwestern and other outlets appearing in mid-April. (news.northwestern.edu) ### Where can readers look next? The next concrete reference point is the April 15, 2026 *Nature Nanotechnology* paper, “Printed MoS2 memristive nanosheet networks for spiking neurons with multi-order complexity.” The named researchers in the public materials include Mark C. Hersam, Vinod K. Sangwan, Shreyash S. Hadke and Carol N. Klingler, according to Northwestern and Nature’s article listing. (news.northwestern.edu) (nature.com)