(ORDO NEWS) — Researchers have presented for the first time an artificial neuron based on electrochemical transistors printed on a substrate. It is capable of learning like real neurons and has been able to integrate with a living organism by causing a Venus flytrap to shut down in the absence of prey.
The development of research in the field of brain-machine interfaces, wearable electronics and next-generation prostheses requires the integration of artificial neuromorphic devices with biological systems.
Usually, silicon-based devices are used for this purpose, but they have a number of disadvantages: they are usually too rigid, not biocompatible enough, and their principle of operation is very different from the biological principles of signal transmission.
As an alternative, Swedish scientists have proposed organic semiconductors, the principle of which is much more similar to the work of real neurons.
Previously, the same group had created biocompatible electrochemical transistors printed on a substrate. They contained both p-type and n-type polymers in their composition, that is, they could carry both positive and negative charges.
This made it possible to create organic electrochemical transistors that could be printed on thin plastic foil. By combining a number of transistors into one circuit, the scientists created synthetic analogues of a neuron and a synapse, the point of contact between two neurons.
The resulting artificial neurons functioned at a voltage of less than 0.6 V, which is almost an order of magnitude lower than the voltages that previous analogues worked with, which created a problem when trying to integrate with living organisms.
To demonstrate the functionality of the created neuron, the scientists “connected” a Venus flytrap to it and successfully transmitted a signal to close the trapping apparatus, although no prey was caught in it.
Finally, the authors of the work showed that an artificial electrochemical synapse is capable of learning in accordance with the Hebb principle formulated for real neurons: according to this rule, simultaneous activation of neurons strengthens their synaptic connection.
Thus, scientists have managed to create localized artificial neural systems that can be integrated with the biosignal systems of plants, invertebrates and vertebrates. They can become the basis of wearable electronics and brain-computer interfaces of a new generation.
Contact us: [email protected]