Insect navigation: use of maps or Ariadne's thread?

Abstract

Animals which regularly depart from and return to fixed positions in their environment can resort to two principally different ways of navigation by using egocentric or geocentric systems of reference. In the first case positional information is obtained by dead reckoning (route integration). A route-integration system provides the animal with a continuous representation of its spatial position relative to its starting point (home). Like a rigidly tightened thread of Ariadne a homeward-bound vector always connects the animal with home. In the second case, the animal is expected to use a map-based system of navigation, i.e. to obtain positional information by virtue of the spatial position it occupies in the larger environmental framework. Rather than relying on information continuously collected en route (dead reckoning), an animal using the mental analogue of a topographical map can episodically take a positional fix, i.e. rely on information collected on site. In central-place foragers like social hymenopterans route-integration based on external systems of reference (e.g. celestial polarization patterns) is the predominant mechanism of navigation, but there is still substantial debate about the extent to which map-based information is used as well. The present account focuses on the computational and neural representations of the bee's and ant's way of integrating circuitous routes, and discusses the extent to which landmark-based information is used in addition.