April 30 (UPI) — Researchers have developed a computing device that is capable of learning by association, essentially merging storage and memory capacity.
Researchers at Northwestern University and the University of Hong Kong used organic electromagnetic chemical “synaptic transistors” to simultaneously store and process information, according to a study published Friday in Nature Communications.
The device mimics the short-term and long-term plasticity of synapses in the human brain to build memories to learn over time, the researchers said.
By connecting synaptic transistors into a neuromorphic circuit, they demonstrated that their device could simulate associative learning.
“Although the modern computer is outstanding, the human brain can easily outperform it in some complex and unstructured tasks, such as pattern recognition, motor control and multisensory integration,” study senior author Jonathan Rivnay said in a press release.
“This is thanks to the plasticity of the synapse, which is the basic building block of the brain’s computational power. These synapses enable the brain to work in a highly parallel, fault-tolerant and energy-efficient manner. In our work, we demonstrate an organic, plastic transistor that mimics key functions of a biological synapse,” said Rivnay, a professor of biomedical engineering at Northwestern’s McCormick School of Engineering.
Rivnay led the study alongside Paddy Chan, an associate professor of mechanical engineering at the University of Hong Kong. They were joined by Xudong Ji, a postdoctoral researcher in Rivnay’s group.
The researchers were able to successfully condition their circuit to associate light with pressure. The transistor and circuit can continue to operate smoothly even when some components fail, and could overcome limitations of traditional computing.
Computer systems physically separate memory and logic. To perform any task, the computer must retrieve information from a memory unit. By bringing the separate functions together, Ji said, “we can save space and save on energy costs.”
A memory resistor, or “memristor,” is the most well-developed technology that successfully does this, but it lags in energy-efficiency and biocompatibility, researchers said.
Organic synaptic transistors, another technology which operates with low voltages, have to be disconnected from the write process, resulting in complicated integration into circuits.
The research team created a conductive, plastic material within the organic synaptic transistor that can trap ions and help the transistor remember previous activities, thereby developing long-term plasticity.
Ions work like neurotransmitters in the brain, where a synapse is a structure through which a neuron can transmit signals to another neuron.
Researchers connected single synaptic transistors to a neuromorphic circuit to simulate associative learning. They integrated pressure and light sensors into the circuit, then trained the circuit to associate the two unrelated physical inputs with one another.
The most famous example of associative learning is Pavlov’s dog, which naturally drooled when it encountered food. After conditioning the dog to associate a bell ring with food, the dog began to drool once hearing the bell.
To condition the circuit, researchers applied pulsed light from an LED light bulb then applied pressure with a finger press to activate voltage. In this case, the pressure is the food and the light is the bell.
After just one training cycle, the circuit made an initial connection between light and pressure. After five cycles, the circuit significantly associated the two. Light alone was able to trigger a signal, or “unconditioned response.”
The circuit is made of soft polymers which can be readily fabricated on flexible sheets and integrated into soft, wearable electronics, smart robotics and implantable devices that directly interface with living tissue and even the brain.
“While our application is a proof of concept, our proposed circuit can be further extended to include more sensory inputs and integrated with other electronics to enable on-site, low-power computation,” Rivnay said.
“Because it is compatible with biological environments, the device can directly interface with living tissue, which is critical for next-generation bioelectronics,” Rivnay said.