Oregon State Researchers Develop Memory Device That Mimics Human Brain’s Forgetting Ability

• June 17, 2026

Scientists at Oregon State University have engineered a novel digital memory device that emulates a fundamental characteristic of the human brain: the ability to retain essential information while gradually discarding unnecessary data. This innovation integrates sensors, memory storage, and signal processing within a single phototransistor, potentially heralding a new era of energy-efficient artificial intelligence (AI) hardware.

Brain-Inspired Memory Device Combines Multiple Functions

The newly developed memory system unites multiple critical functionalities traditionally handled by separate components. By incorporating sensing capabilities, data storage, and signal handling directly into one phototransistor, the device streamlines operations that are key to AI applications.

This integrated approach could substantially lower power consumption compared to conventional architectures that rely on discrete elements for sensing and memory. Such reductions are particularly valuable for emerging AI technologies, which demand increasingly efficient hardware solutions to manage vast amounts of data.

One of the most striking features of this device is its ability to mimic the brain’s selective memory retention. Much like humans naturally forget irrelevant information over time, this memory technology is designed to maintain important data while intentionally shedding less critical details. This capability contrasts with typical digital memories that indiscriminately store all information.

Implementing a forgetting mechanism in hardware addresses a significant challenge in AI processing, where managing data overload and maintaining relevant context are ongoing concerns. By adopting principles observed in biological systems, researchers hope to improve the performance and resource efficiency of AI platforms.

The integration of sensing and memory in a phototransistor framework also holds promise for more compact and versatile devices. As AI adoption continues to grow across sectors, from mobile gadgets to autonomous systems, innovations like this could facilitate smarter, more adaptive technologies with smaller footprints and lower energy demands.

While specific technical details and performance metrics have not been disclosed, this research marks an important step toward brain-inspired memory components that enhance the capabilities and sustainability of future AI equipment.

Researchers at Oregon State University created a memory system that integrates sensing, storage, and processing, capable of forgetting unneeded data like the human brain.