Origin of Resource-Induced Phase Transitions in Biological Information Processing

Biological systems estimate and adapt to fluctuating environments by integrating sensory evidence with internal memory. Yet their information processing is constrained by limited resources such as energy and precision, which can alter their optimal estimation strategies. Understanding how resource limitations induce phase transitions in optimal estimation strategies is still elusive. In this presentation, we analytically derive the condition of resource-induced phase transitions in optimal estimation strategies. Our analysis reveals the origin of discontinuous, nonmonotonic, and scaling behaviors in these transitions. Moreover, we exemplify that our analytical results are consistent with recent experimental findings. The analytical framework presented here offers a theoretical foundation for understanding how limited resources shape biological information processing.