Researchers have developed a method to control stochasticity, or randomness, in metal-insulator switching devices. These devices, fundamental in modern electronics, transition between conductive (metallic) and non-conductive (insulating) states in response to an external stimulus. The inherent stochasticity of this process has traditionally been a challenge for their reliability and reproducibility. The new approach is based on manipulating the jamming dynamics within the material, allowing for precise modulation of variability in switching behavior.

Controlling stochasticity is crucial for applications ranging from non-volatile memory to neuromorphic computing. In memory, randomness can lead to read/write errors, while in neuromorphic computing, controlled stochasticity could mimic the brain's synaptic variability, opening new avenues for information processing. This advance offers a tool to design devices with tailored switching characteristics, adapting to the specific requirements of each application.

The study details how engineering material properties and operating conditions influence the jamming of conductive phases, thereby allowing the degree of randomness to be adjusted. This control mechanism opens the door to a new generation of more robust and versatile electronic devices, with significant potential to improve the efficiency and functionality of computing and data storage systems.