Optimal cue integration in ants

Abstract

In situations with redundant or competing sensory information, humans have been shown to perform cue integration, weighting different cues according to their certainty in a quantifiably optimal manner. Ants have been shown to merge the directional information available from their path integration (PI) and visual memory, but as yet it is not clear that they do so in a way that reflects the relative certainty of the cues. In this study, we manipulate the variance of the PI home vector by allowing ants (Cataglyphis velox) to run different distances and testing their directional choice when the PI vector direction is put in competition with visual memory. Ants show progressively stronger weighting of their PI direction as PI length increases. The weighting is quantitatively predicted by modelling the expected directional variance of home vectors of different lengths and assuming optimal cue integration. However, a subsequent experiment suggests ants may not actually compute an internal estimate of the PI certainty, but are using the PI home vector length as a proxy.

Principles of Insect Path Integration

“…In contrast, ants that live in slightly cluttered landscapes learn visual landmark information and establish individualistic routes that lead them to their food sources and back to the nest [39]. If they are displaced close to the familiar foraging corridor, they initially orient using visual landmarks (Figure 2A, 0 m) or chose an intermediate direction between the real home and the direction indicated by the path integrator (Figure 2A) [40,41], and subsequently find home by relying on visual landmarks along the route [40]. In contrast, when ants are displaced to previously unvisited locations but within their foraging range, they ignore compass information from the path integrator and exclusively use visual landmark information to directly head home [42,43] (Figure 2B).…”

“…Intuitively, when in conflict, a less certain estimate should be weighted less. However, [41] showed that positional certainty cannot explain behavior resulting from conflicts between path integration and visual cues (Figure 2A). Instead, the data matched better with a strategy during which the ants ignore the distance to the goal and only account for the angular component of the uncertainty distribution, i.e.…”

Sure enough: efficient Bayesian learning and choice

“…Additionally, many organisms are known to weight cues by their reliability (e.g. Vossel et al 2014; Yu and Dayan 2005; Yang and Shadlen 2007; Fischer and Peña, 2011; Wystrach et al 2015), in a manner compatible with Bayesian reasoning.…”

Compass cues used by a nocturnal bull ant, Myrmecia midas

“…We also tested foragers with the maximum accumulating vectors we could find at the field site (12.8-14.0 m), as the weighting of vector cues when in directional conflict with familiar terrestrial cues has been shown to increase with vector length, as a result of reduced directional uncertainty (Wystrach et al, 2015). Thus, foragers travelling to trees farther away may accumulate sufficiently strong or sufficiently reliable vectors to orient in unfamiliar locations.…”

“…However, when these cues are in conflict with each other, some ant species in some conditions choose a direction that is a compromise between the information dictated by the terrestrial cues and the path integrator (Narendra, 2007a,b;Collett, 2010;Legge et al, 2014;Wystrach et al, 2015;Wehner et al, 2016). Under the testing conditions in which M. midas foragers were displaced to an off-route site 5 m from the nest, either during the morning inbound section of the foraging trip or after a holding period of 15 min or overnight, individuals oriented toward the terrestrial cue direction and showed no effect of the conflicting vector even when tested with the largest (14 m) vector observed at the site (Fig.…”