In the paper “From random walks to informed movement”, Emanuel Fronhofer and colleagues present a model showing that with “memorized” spatial information, an animal will boost it’s foraging success, as compared to random walk. Now on Early View.
Read Emanuel’s story about the model:
When animals move they frequently search for resources. This may, for example, be a female butterfly searching for proper host plants, a gnu exploiting grassland, or a male dragonfly searching for mating partners. Being an efficient searcher is thus an ability of great fitness implications. Indeed, movement is so fundamental to life that research on animal movement is highly relevant for many basic and applied issues.
Up to now movement is mostly modeled as a random process (random walk) and it is fascinating that such models, like the “Lévy walk”, are so capable in grasping the statistical attributes of animal movement. Yet, if we want to predict the influence of man made modifications of landscape structure on foraging success or the implications of global climatic change on animal dispersal we need a thorough understanding of the mechanisms governing movement decisions. Evidently, movement is controlled by an animal’s perception, memory, and its ability to infer the likely position of resource based on general knowledge about landscape attributes and the specific information at its hand. Further, animals should also be able to think more than one step ahead (anticipation), i.e. foresee future consequences of its moves.
Here we propose a model that accounts for all these elements (perception, memory, inference and anticipation). Our analysis shows that even a very basic implementation of these processes allows an enormous increase in foraging efficiency and results in movement patterns typical for systematic search within resource patches (e.g. of flowers of food plants; fig.1), straight movement between such patches (fig. 1) and even the emergence of foray loops (fig. 2) that have been observed in e.g. butterflies.
Our model is easily applied to insects like butterflies, wasps, or flies, searching for food or suitable plants to lay their eggs. The analysis of this model highlights the strength of mechanistic approaches to movement modeling and sets the stage for the development of more sophisticated models of perception and memory use invoked in movement decisions and dispersal.