An empirically parameterized individual based model of animal movement, perception, and memory

Authored by Tal Avgar, Rob Deardon, John M Fryxell

Date Published: 2013

DOI: 10.1016/j.ecolmodel.2012.12.002

Sponsors: National Science and Engineering Research Council of Canada (NSERC)

Platforms: MATLAB

Model Documentation: Other Narrative Mathematical description

Model Code URLs: https://ars-els-cdn-com.ezproxy1.lib.asu.edu/content/image/1-s2.0-S0304380012005686-mmc12.zip

Abstract

Our capacity to predict patterns of animal movement behavior is limited by our understanding of the underlying cognitive process. Determining what an animal knows about its environment, and how that information is translated into specific movement behaviors, is a conceptual challenge faced by movement ecologists. The modeling framework presented here is designed to evaluate the likelihood of alternative processes of perception, memory and decision making, based on readily available positional data and environmental metrics. The model is based on a flexible cognitive algorithm that provides the framework for an adaptive movement kernel. This enables a straightforward methodology for estimating key parameters for sensory perception, memory and movement while providing testable predictions of animal resource selection and space use patterns. In addition to describing the model and explaining the underlying logic, we demonstrate its parameterization potential using simulated data and investigate the robustness of its predictions over a wide range of temporal and spatial sampling scales. We show that the model can reproduce descriptive probes of movement paths with little sensitivity to the scale at which these paths were sampled and we discuss the merits of our approach in the context of movement-and cognitive-ecology and evolution. (C) 2013 Elsevier B.V. All rights reserved.

Tags

spatial memory mechanistic model Habitat selection Resource selection functions State-space models Incorporating home-range Ecological landscapes Bayesian-analysis Telemetry data Random-walks