Task-performing dynamics in irregular, biomimetic networks

Abstract

Understanding self-organized collective dynamics-especially in sparsely connected, noisy, and imperfect networks-has important implications for designing and optimizing task-performing technological systems as well as for deciphering biological structures and functions. We note that stomatal arrays on plant leaves might provide an ideal example of task-performance in this context. Guided by observations of stomatal networks, we examined a simple model of task-performing, collective dynamics that included state noise, spatial rule heterogeneity, dynamic modules, and network rewiring. Our results indicate that task-performance in such networks can actually be enhanced by various kinds of spatial and temporal irregularity. © 2007 Wiley Periodicals, Inc.