A single-pheromone model accounts for empirical patterns of ant colony foraging previously modeled using two pheromones

Abstract

In a 2009 paper, Dussutour et al. proposed that big headed ants (Pheidole megacephala) employ two attractant pheromones during foraging: one for exploration and another during food gathering. This claim was consistent with, and argued to be supported by, laboratory studies of ant exploration and food-gathering in a Y-maze apparatus. The authors measured foraging activity and colony foraging choice in terms of the number of ants choosing different branches over time, where experimental conditions modified the history of food availability at the end of each branch. They built a two-pheromone mathematical model to account for observed rates and proportions of ants traversing the left versus right branch. Here we show that the main reported experimental observations can be explained by a one-pheromone model. Our findings show that it is plausible, but unnecessary, to hypothesize that these ants employ two distinct pheromones in order to account for the two principal results of the Dussutour et al. study, and therefore, the study falls short of dispositive evidence for a two-pheromone model. More broadly, we highlight that patterns of animal behavior can be ambiguous with respect to sensory and cognitive mechanisms, hopefully motivating future modeling efforts that perform formal comparison across models with different structure.