Scorpion navigation by chemo-textural familiarity: modeling the interplay between sensory and landscape parameters.

Abstract

The ability to return consistently to one point - e.g., a food source or shelter - is a poorly understood animal behavior. One hypothesis proposed to explain navigation in bees and ants is navigation by scene familiarity. In this paper, we propose that scorpions use ground-directed sensors (instead of vision) to augment navigation. Specifically, the paired pectines may detect matrices of chemical and textural information and compare it to stored memories, allowing the animal to step toward the most familiar "scene." We first show that this form of familiarity navigation is consistent with scorpion homing excursions. We then developed a familiarity-based computer-simulated agent and analyzed navigation success using multiple combinations of sensor sensitivity levels, sensor resolutions, and landscape characteristics. We also assessed the effects of additional sensory channels and of landscape disruption. Finally, we showed that the simulated agent can navigate across a landscape the size of a typical sand scorpions home range.