Fuzzy cognitive maps outmatch loop analysis in dynamic modeling of ecological systems

Abstract

Modeling natural systems is challenging due to their complexity in terms of variables, interactions, and dynamics. Much of this complexity is rooted in the existence of multiple ways through which acting variables affect each other. Besides the simple direct effects, numerous indirect effects emerge in ecological systems. Through an illustrative example, I exemplify here several advantages of fuzzy cognitive maps (FCM) over loop analysis (LA) in dynamic modeling of ecological systems. In addition to being able to incorporate information about the magnitude of variables interactions, FCM can make predictions about multiple simultaneous perturbations. Furthermore, FCM allow for the simulation of different magnitude of initial perturbations to the system. Last, FCM estimate the amount of variable increase/decrease, not just the likely direction of change. Hence, even if LA is still much more used than FCM in the scientific literature, FCM can be considered fitter than LA in modeling ecological systems.

Keywords

• fuzzy cognitive maps
• loop analysis
• interaction weights
• multiple perturbations
• perturbation strengths
• stock increase/decrease