François-Xavier Dechaume-Moncharmont1, Anna Dornhaus2, John M. McNamara3, Alasdair I. Houston1, Edmund J. Collins3, and Nigel R. Franks1. (1) School of Biological Sciences, University of Bristol, Woodland Road, Clifton, Bristol, BS8 1UG, United Kingdom, (2) Department of Ecology and Evolutionary Biology, University of Arizona, 1041 E. Lowell, Tucson, AZ 85721, (3) School of Mathematics, University of Bristol, University Walk, Clifton, Bristol, BS8 1TW, United Kingdom
For a social insect colony, emigration from one nest to another is a risky venture. When its nest is damaged, a colony of Temnothorax albipennis has to emigrate as quickly as possible to the best available nest site. Two recruitment behaviours are involved in this emigration. A recruiter can either lead with a tandem run a single nestmate from the old nest to the new nest, or it can simply pick up a nestmate and carry it to the new nest. In both cases they can summon only one nestmate at a time. The tandem process is three times slower than carrying, but the ants recruited by tandem runs can learn the location of the new nest and are able to recruit other nestmates while carried ants may not be able to recruit others. Thus there must be a trade-off between these two tasks. What is the optimal task allocation that would minimize the emigration time? In this theoretical study, we compare two strategies. The first one assumes constant proportions of tandem runners and carriers. The second one assumes that each ant is able to perform both tasks. They begin by leading tandem runs until the number of ants recruited reaches a quorum threshold. Then every tandem run leader switches to social carrying. We show that a task allocation mechanism based on non-specialized workers sequentially involved in the two tasks performs significantly better than a strategy based on a strict division of labour with specialized recruiters. This theoretical prediction about task allocation mechanisms is consistent with experimental observations.
See more of Symposium 15: Collective and Individual Intelligence
See more of Invited Symposia Presentations
See more of The IUSSI 2006 Congress