An individual based model examining the emergence of cooperative recognition in a social insect (Isoptera: Rhinotermitidae).

11/2004; 46.
Source: OAI


We use an agent-based model to simulate the expression of recognition behavior in the form of aggression among individual workers within colonies of subterranean termites. Inter- and intraspecific recognition in the form of overt aggression varies in the genus Reticulitermes. Three patterns have emerged from nestmate recognition studies in termites: 1) interspecific aggression is often stronger and more immediate than intraspecific aggression, 2) a loss of genetic diversity may result in a loss of aggression, 3) with laboratory bioassays, single individuals don't show overt aggression, whereas groups do. We create an agent-based model that simulates these patterns to understand the mechanisms that create them. We assume that there are three components to successful recognition of alien individuals: the variation in recognition cues among different colonies, signaling among individuals in the same colony, and variation in response thresholds of individuals within the same colony. The results suggest that if cue difference is not significantly variable, and response thresholds are high, aggressive behavior will not be expressed. Under these situations, social signaling becomes increasingly important. The model provides a simple mechanistic description of how aggressive behavior emerges from simple worker-to-worker interactions, and explains a possible reason for a group effect in laboratory bioassays.

Download full-text


Available from: Kirsten A. Copren, Dec 23, 2013
1 Read
  • Source
    • "Agent Based Models (ABMs) [7] have been used to model a range of topics from trading [8] to battlefields [9]. Several such models have been constructed to study termites but these study only a particular behaviour involving a very limited number of agents – for example activations of resting termites [10] and defense of a colony [11]. A sophisticated model of a termite colony may provide further insights into their cooperative behaviour. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Termites exhibit complex and emergent behaviour that is still poorly understood in terms of both causes and mechanisms. We explore emergent cooperative behaviour amongst systems of simulated termites through a microscopic model capable of supporting realistic numbers of individual termites in a nest. We explore possible individual characteristics of a model ter-mite as they might respond to breaches in the nest wall. We report on how a simulated thermal gradient may suffice as a communications mechanisms that enables individual termites to cooperate in repair of a wall breach of their nest or mound on behalf of their macroscopic community. We review how intelligent agent-based models could explain some observed termite collective behaviours.
  • Source
    • "In termites there are a number of complex behaviours that have been studied using agent-based models. Examples include self-organised nest construction (Deneubourg 1977; Courtois et al. 1991; Bonabeau et al. 1997 ; O'Toole et al. 1999, 2003), social facilitated survival (Miramontes and DeSouza 1996), disease transmission (Pie et al. 2004) and individual recognition (Copren et al. 2005 ). Agent-based models have proved to be powerful tools for exploring aspects of these questions, because they incorporate basic rules of individual behaviour capturing the essence of the problem being explored. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Interactions among individuals in social groups lead to the emergence of collective behaviour at large scales by means of multiplicative non-linear effects. Group foraging, nest building and task allocation are just some well-known examples present in social insects. However the precise mechanisms at the individual level that trigger and amplify social phenomena are not fully understood. Here we show evidence of complex dynamics in groups of the termite, Cornitermes cumulans (Kollar) (Isoptera: Termitidae), of different sizes and qualitatively compare the behaviour observed with that exhibited by agent-based computer models. It is then concluded that certain aspects of social behaviour in insects have a universal basis common to interconnected systems and that this may be useful for understanding the temporal dynamics of systems displaying social behaviour in general.
    Journal of Insect Science 03/2008; 8(22):1-11. DOI:10.1673/031.008.2201 · 1.03 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Communication amongst multi-agents in a system gives rise to complex emergent phenomena that remain poorly understood in terms of the underpinning mechanism. We explore communications mechanisms between multi-agents in a many-agent spatial predator-prey model. We focus on solider-worker scenarios in a community-like model where collectives compete for resources and can gain an advantage by trading off numbers of unproductive soldiers against numbers of defenseless but productive workers. We show how even relatively simple cross-caste communications allows a particular collective to maintain a smaller and therefore less expensive army by making it more mobile and responsive to enemy incursions.