Artificial Synapse to Mimic Brain Functions

Artificial Synapse, created by scientists, is the 1st water-based brain-like computer.

A physicists’ team from Sogang University in South Korea and Utrecht University in the Netherlands has constructed an artificial synapse, the biological basis of the human brain’s capacity for thought and reason. Numerous attempts have been made to replicate the complex information processing of the human brain, but these have typically relied on silicon and metal. In contrast, the synapse devised by the researchers operates using ordinary water and sale, much like the human brain. 

Modern binary computers perform calculations and graphics rendering at an astounding rate, but it isn’t easy to comprehend. Sophisticated artificial intelligence systems, such as ChatGPT and Gemini, may appear to possess critical thinking capabilities; however, their true nature is that of potent word processors. An analog of the human brain’s synaptic medium possesses an advantageous position in the endeavor to duplicate its operational capabilities.

The apparatus under investigation (or is it the brain?) is referred to as an intronic memristor. Its dimensions range from 150 to 200 micrometers, slightly larger than a human hair. The memristor’s microchannel is in the shape of a cone and is filled with a solution of dissolved potassium chloride in water. Ions in the water ascend the channel in response to an electrical signal received by the device; consequently, the conductivity and ion density of the channel are altered. 

The conductivity variation of the memristor is influenced by both the intensity and duration of the electric impulse. According to the research team, this closely resembles how synaptic connections between neurons can change as the brain processes information to form conscious thoughts. Additionally, the researchers highlight how variations in channel lengths can either prolong or abbreviate the period necessary for the induced charge changes to converge. As a result, it might be feasible to generate distinct channel types to accommodate diverse computational tasks. This again mirrors the arrangement of synapses in the brain.

Tim Kamsma, a doctoral candidate at Utrecht University, spearheaded the investigation. He had only recently begun describing the theoretical capabilities of the iontronic memristor he was developing. They immediately began constructing his memristor after establishing communication with a group at Sogang University who shared Kamsma’s enthusiasm for translating theory into practice. Surprisingly, the development of the minuscule device and demonstration that it operated precisely as Kamsma had predicted occurred in mere months. 

While the artificial synapse does not permit the team to perform complex calculations or ponder the meaning of existence, that is also impossible with a single biological synapse. This could be essential in developing computers that operate using the same principles as the human brain.