Stigmergy

Stigmergy (/ˈstɪɡmərdʒi/ STIG-mər-jee) is a mechanism of indirect coordination, through the environment, between agents or actions. The principle is that the trace left in the environment by an action stimulates the performance of a next action, by the same or a different agent. In that way, subsequent actions tend to reinforce and build on each other, leading to the spontaneous emergence of coherent, apparently systematic activity. Stigmergy (/ˈstɪɡmərdʒi/ STIG-mər-jee) is a mechanism of indirect coordination, through the environment, between agents or actions. The principle is that the trace left in the environment by an action stimulates the performance of a next action, by the same or a different agent. In that way, subsequent actions tend to reinforce and build on each other, leading to the spontaneous emergence of coherent, apparently systematic activity. Stigmergy is a form of self-organization. It produces complex, seemingly intelligent structures, without need for any planning, control, or even direct communication between the agents. As such it supports efficient collaboration between extremely simple agents, who lack any memory, intelligence or even individual awareness of each other. The term 'stigmergy' was introduced by French biologist Pierre-Paul Grassé in 1959 to refer to termite behavior. He defined it as: 'Stimulation of workers by the performance they have achieved.' It is derived from the Greek words στίγμα stigma 'mark, sign' and ἔργον ergon 'work, action', and captures the notion that an agent’s actions leave signs in the environment, signs that it and other agents sense and that determine and incite their subsequent actions. Later on, a distinction was made between the stigmergic phenomenon, which is specific to the guidance of additional work, and the more general, non-work specific incitation, for which the term sematectonic communication was coined by E. O. Wilson, from the Greek words σῆμα sema 'sign, token', and τέκτων tecton 'craftsman, builder': 'There is a need for a more general, somewhat less clumsy expression to denote the evocation of any form of behavior or physiological change by the evidences of work performed by other animals, including the special case of the guidance of additional work.' Stigmergy is now one of the key concepts in the field of swarm intelligence. Stigmergy was first observed in social insects. For example, ants exchange information by laying down pheromones (the trace) on their way back to the nest when they have found food. In that way, they collectively develop a complex network of trails, connecting the nest in an efficient way to various food sources. When ants come out of the nest searching for food, they are stimulated by the pheromone to follow the trail towards the food source. The network of trails functions as a shared external memory for the ant colony. In computer science, this general method has been applied in a variety of techniques called ant colony optimization, which search for solutions to complex problems by depositing 'virtual pheromones' along paths that appear promising.In the field of artificial neural networks stigmergy can be used as a computational memory. F. Galatolo showed that a stigmergic memory can achieve the same performances of more complex and well established neural networks architectures like LSTM Other eusocial creatures, such as termites, use pheromones to build their complex nests by following a simple decentralized rule set. Each insect scoops up a 'mudball' or similar material from its environment, invests the ball with pheromones, and deposits it on the ground, initially in a random spot. However, termites are attracted to their nestmates' pheromones and are therefore more likely to drop their own mudballs on top of their neighbors'. The larger the heap of mud becomes, the more attractive it is, and therefore the more mud will be added to it (positive feedback). Over time this leads to the construction of pillars, arches, tunnels and chambers. Stigmergy has even been observed in bacteria, various species of which differentiate into distinct cell types and which participate in group behaviors that are guided by sophisticated temporal and spatial control systems. Spectacular examples of multicellular behavior can be found among the myxobacteria. Myxobacteria travel in swarms containing many cells kept together by intercellular molecular signals. Most myxobacteria are predatory: individuals benefit from aggregation as it allows accumulation of extracellular enzymes which are used to digest prey microorganisms. When nutrients are scarce, myxobacterial cells aggregate into fruiting bodies, within which the swarming cells transform themselves into dormant myxospores with thick cell walls. The fruiting process is thought to benefit myxobacteria by ensuring that cell growth is resumed with a group (swarm) of myxobacteria, rather than isolated cells. Similar life cycles have developed among the cellular slime molds. The best known of the myxobacteria, Myxococcus xanthus and Stigmatella aurantiaca, are studied in various laboratories as prokaryotic models of development.