Experimental investigation of stochastic resetting in a non-Markovian environment

Abstract Stochastic resetting (SR), in which a system intermittently returns to a fixed location, is a powerful strategy for optimizing search processes. While extensively studied in memoryless (Markovian) systems, its behavior in complex media with memory remains largely unexplored. Here, we experimentally investigate SR in a viscoelastic fluid by tracking a colloidal particle subjected to intermittent resets. In this non-Markovian environment, the fluid’s memory gives rise to elastic restoring forces that oppose the reset, pulling the particle back toward its prior position and hindering efficient exploration. We show that these memory effects can be actively controlled: by holding the particle at the trap center for a sufficient time, the fluid relaxes, erasing its memory and allowing the system to re-equilibrate. When introducing a fixed target site, we find that this memory control enables a significant reduction in the mean passage time, with optimal search performance emerging at intermediate resetting frequencies. In this regime, memory enhances performance through a ‘bunching’ effect, in which the particle rapidly revisits the target due to temporal correlations in its trajectory. These results highlight the dual role of memory in resetting dynamics-as both a hindrance and a resource-and suggest new strategies for optimizing search in non-Markovian systems, with potential applications in soft matter, biological transport, and stochastic control.