Artificial synapses are able to mimic a brain’s work processes. Researchers at Stanford University demonstrated this with experimental components as early as 2017: the neuronal learning processes could be simulated efficiently and – particularly important – with low energy. In 2019, it was also shown how nine of these artificial synapses can be controlled simultaneously and communicate with each other.
Electrochemical contact between brain and machine
Now Stanford is announcing the next progress together with the Istituto Italiano di Tecnologia IIT and the TU Eindhoven: Biohybrid electrochemical communication is possible. Specifically: the artificially produced synapses can communicate with living cells. Devices used in the brain normally work with electrical signals to recognize and process the messages from the brain. This step could one day be eliminated with the new procedure.
Communication is a “first tiny step” towards improved brain-machine contact points, commented Alberto Salleo, professor of materials science at Stanford University and one of the authors of the paper. Co-author and Stanford PhD student Scott Keene notes: “In a biological synapse, essentially everything is controlled by chemical interactions at the synaptic junction. When cells communicate with each other, it’s always chemical.” The fact that the artificial synapses can interact with the natural chemistry of the brain now promises further benefits.
Construction of the artificial synapse
Instead of just electrical, the artificial synapses communicate electrochemically with the neurons. Two soft polymer electrodes are used for this purpose, between which there is a small “trench”. It is filled with an electrolyte solution – based on the synaptic gaps in the brain.
If living nerve cells are placed on these electrodes, the neurotransmitters of the brain cells react to them and produce ions. The ions travel across the trench to the second electrode and form a connection. These connections were even partially preserved in the experiment, which corresponds to the natural learning process in the brain. It works differently with digital computers: Here data is processed first and only then pushed into memory.
The paper on the study was published in the journal “Nature Materials”. The direction in which further research will be carried out with these findings is currently still open. Think of the development of brain-inspired computers or new interfaces between brain processes and machines, according to the researchers.