Brain-Like Computers Coming Soon? Scientists Create Nano-Device That Mimics Function Of Human Synapse

A new type of nano-device developed by engineers at the University of Massachusetts Amherst can mimic the functioning of a biological synapse, reports Science Daily.

The human brain has about 100 billion neurons and approximately 1 quadrillion (1 million billion) synapses. A brain-inspired computer would ideally be designed to mimic the brain’s enormous computing power and efficiency

LiveScience reports that, with their latest device, the group has managed to mimic human neural functioning more faithfully than anyone that came before them. The report states that the new devices “pave the way for brain-like computers.”

“The brain-inspired computing component provides the most faithful emulation yet of connections among neurons in the human brain.”

null

The device the researchers developed is a new type of memristor. Memristors (the name is a portmanteau of “memory” and “resistor”) are electrical resistance switches that can alter their resistance based on the history of applied voltage and current. In other words, the amount of voltage and current that has previously traveled through them throughout their lifetime. The devices can store and process information and “have become a leading candidate to enable neuromorphic computing by reproducing the functions in biological synapses and neurons in a neural network system, while providing advantages in energy and size,” according to the group.

The new nano-device will be used in computer microprocessors. It can mimic the functioning of a biological synapse — the place where a signal passes from one nerve cell to another in the body. The researchers told reporters that their design closely parallels the way calcium ions behave at the junction between two neurons in the human brain.

Brain-inspired or “neuromorphic” computers would be much better at perceptual and learning tasks than traditional computers, and LiveScience reports that they would also be more energy efficient. The study’s leader, Joshua Yang, a professor of electrical and computer engineering at the University of Massachusetts Amherst, told reporters that past attempts to build devices that mimic the function of their synapse did not truly succeed. Past devices like brain-inspired transistors and capacitors had “very little resemblance to real biological systems,” says Yang.

“In the past, people have used devices like transistors and capacitors to simulate synaptic dynamics, which can work, but those devices have very little resemblance to real biological systems. So it’s not efficient to do it that way, and it results in a larger device area, larger energy consumption and less fidelity.”

Yang stated that the new memristor developed at the University of Massachusetts Amherst “can emulate the synapse in a more natural way, more direct way and with more fidelity.” The study leader said that many aspects of synaptic dynamics are mimicked, not just one.

“You don’t just simulate one type of synaptic function, but [also] other important features and actually get multiple synaptic functions together.”

null

Atomic diffusion — the movement of atoms from a region of high concentration to a region of low concentration — plays a critical role in allowing the memristor to mimic its bio-counterpart, the synapse.

“Specifically, we developed a diffusive-type memristor where diffusion of atoms offers a similar dynamics and the needed time-scales as its bio-counterpart, leading to a more faithful emulation of actual synapses, i.e., a true synaptic emulator.”

The new memristor is made up of silver nanoparticle clusters embedded in a silicon oxynitride film. The film is sandwiched between two electrodes. It is an insulator, but when a voltage pulse is applied, the silver nanoparticle clusters begin to break up and diffuse due to the effects of heating and electrical forces on the film.

The silver nanoparticles diffuse until they form a conductive filament that can carry the current from one electrode to the other. The nanoparticles coalesce back into clusters once the voltage is removed and the temperature drops.

This process is very similar to how calcium ions behave in biological synapses, and the device can mimic short-term plasticity in neurons, the researchers said.

“In biological systems, when a nerve impulse reaches a synapse, it causes channels to open, allowing calcium ions to flood into the synapse. This triggers the release of brain chemicals known as neurotransmitters that cross the gap between the two nerve cells, passing on the impulse to the next neuron.”

The “memory” element follows from the fact the conductivity of the new device will gradually increase if trains of low-voltage pulses at high frequencies are applied. If the pulses continue, this conductivity will eventually decline (ie. the device “remembers” its own voltage history.)

null

[Featured Image by Matt Cardy/Getty Images]