Scientists have developed a new electronic chip component that works more like the human brain than today’s computers.
The new device can both process information and store it at the same time, which could help create faster, smarter and more energy-efficient computers in the future.
The research was carried out by an international team of scientists, including Professor Victor López-Richard from the Federal University of São Carlos (UFSCar) in Brazil.
Their findings were published in the journal Nature Communications.
Most modern computers keep processing and memory separate. The processor performs calculations, while memory stores data. Because information constantly moves between these two parts, computers use extra time and energy.
The human brain works differently. Brain cells both process and store information through networks of neurons and synapses. Scientists have long wanted to copy this design, a field known as neuromorphic computing.
The new device brings researchers a step closer to that goal.
It is made from two special oxide materials called lanthanum aluminum oxide and strontium titanate. Where these two materials meet, they create a very thin layer that allows electricity to flow. This layer acts like a transistor, one of the basic building blocks found in almost every electronic device.
What makes this transistor different is that it can also remember what has happened to it. In other words, it behaves like both a memristor and a memcapacitor. These are electronic components whose behavior changes depending on the electrical signals they have received in the past. This built-in memory is similar to the way brain synapses become stronger or weaker as we learn.
The researchers also designed the transistor differently from traditional ones. Instead of placing the control gate above the electrical channel, they positioned control gates on the sides. These side gates slowly store electrical charges, allowing the device to adjust its behavior in a smooth, gradual way.
Unlike ordinary digital transistors that are simply “on” or “off,” this new device can operate in many different states. This analog behavior makes it better suited for tasks that involve learning, adapting and recognizing patterns, much like the human brain.
Another important feature is its flexibility. The same device can perform several different jobs simply by changing how it is connected electrically. This ability, known as electronic polymorphism, means one component can replace several different ones, reducing the number of parts needed in future computer chips.
The team tested the device on several brain-inspired computing tasks. It successfully recognized simple handwritten digits and demonstrated short-term memory and synaptic plasticity, which is the brain’s ability to strengthen connections after repeated activity. It could also perform logic operations while storing information inside the same device, removing the need for separate memory.
Perhaps most importantly, each operation used only a few nanojoules of energy, much less than many conventional computer designs.
Although the technology is still at the proof-of-concept stage, the researchers believe it has great potential.
More work is needed before it can be turned into commercial products, including improving large-scale manufacturing and ensuring consistent performance. Even so, this brain-inspired chip offers an exciting glimpse into the future of low-power artificial intelligence and next-generation computing.
