An agent-based model of collective nest choice by the ant Temnothorax albipennis

Stephen C. Pratt, David J.T. Sumpter, Eamonn B. Mallon, Nigel R. Franks

Research output: Contribution to journalArticle

95 Scopus citations

Abstract

Colonies of the ant Temnothorax (formerly Leptothorax) albipennis can collectively choose the best of several nest sites, even when many of the active ants who organize the move visit only one site. Previous studies have suggested that this ability stems from the ants' strategy of graded commitment to a potential home. On finding a site, an ant proceeds from independent assessment, to recruiting fellow active ants via slow tandem runs, to bringing the passive bulk of the colony via rapid transports. Assessment duration varies inversely with site quality, and the switch from tandem runs to transports requires that a quorum of ants first be summoned to the site. These rules may generate a collective decision, by creating and amplifying differential population growth rates among sites. We test the importance of these and other known behavioural rules by incorporating them into an agent-based model. All parameters governing individual behaviour were estimated from videotaped emigrations of individually marked ants given a single nest option of either good or mediocre quality. The time course of simulated emigrations and the distribution of behaviour across ants largely matched these observations, except for the speed with which the final transport phase was completed, and the overall emigration speed of one particularly large colony. The model also predicted the prevalence of splitting between sites when colonies had to choose between two sites of different quality, although it correctly predicted the degree of splitting in only four of six cases. It did not fully capture variance in colony performance, but it did predict the emergence of variation in individual behaviour, despite the use of identical parameter values for all ants. The model shows how, with adequate empirical data, the algorithmic form of a collective decision-making mechanism can be captured.

Original languageEnglish (US)
Pages (from-to)1023-1036
Number of pages14
JournalAnimal Behaviour
Volume70
Issue number5
DOIs
StatePublished - Nov 2005

    Fingerprint

ASJC Scopus subject areas

  • Ecology, Evolution, Behavior and Systematics
  • Animal Science and Zoology

Cite this