Seemingly complex group behavior can often result from simple rules repeated across a number of individuals.
There is a tendency to think that some of these highly coordinated manoeuvres must occur under the direction of a leader shouting orders that everyone obeys. The truth is that more often than we imagine the group behavior is just the result of a bunch of individuals each following some basic rules; when this is iterated dozens or even thousands of times the results are amazing.
Two of the most visually stunning examples of this self-organizing behavior is seen with flocking birds and schooling fish.
Luis Ibanez at the Kitware Blog (where I originally saw the starling video) writes; “In this particular case, Starlings joins in groups composed of hundreds-of-thousands of individuals, and in the absence of centralized control and devoid of a social hierarchical structure, display behaviours can easily be interpreted as those of a macro-organism whose cells are the individual birds.”
Both schooling fish and flocking birds follow the same basic rules which were first modelled in 1986 by Craig Reynolds. He described the behavior with 3 simple instructions:
His webpage has a lot more information as well as a java applet animation using the above algorithm.
A neat little example of emergence in humans happened during the opening of the Millennium Bridge. No one told all these people to start acting in unison, all you had to do was provide the appropriate stimulus and ’emerged.’
A Spanish-American team led by Cristina Muro of the Spanish Association for Assistance Dogs (AEPA-EUSKADI) developed a computer program that simulates wolf hunting strategies (for a single prey) using only two rules.
- Move towards the prey until a minimum safe distance to the prey is reached.
- When close enough to the prey, move away from the other wolves that are close to the safe distance to the prey.
Escobedo’s webpage provides several animated examples of these rules in action.
One cannot help but to notice the similarities between the two rules for hunting prey-chasing wolves and the 3 rules that result in flocking behavior. In both cases, group behavior is the result of multiple individuals applying the same simple rules. The broup behavior emergens without any planning and without any special knowledge from any individual..
The study concludes, “Our results suggest that wolf-pack hunting is an emergent collective behavior which does not necessarily rely on the presence of effective communication between the individuals participating in the hunt, and that no hierarchy is needed in the group to achieve the task properly.”
This model does not suggest that wolves lack intelligence; it just means that they don’t NEED to be intelligent, communicate or have a hierarchy to accomplish their task. And like the previous post, it also suggests that hunting is not the reason wolves form packs.
It also means that if someone wants to propose an alternate model they will need to provide evidence as to why it is more compelling.
[Note: I redacted the sections on leaf venation, power laws and fractals. On second reading I realized it added nothing to the article]
Muro C, Escobedo R, Spector L,Coppinger RP (2011) Wolf-pack (Canis lupus) hunting strategies emerge from simple rules in computational simulations. Behavioural Processes Volume 88, Issue 3, November 2011, Pages 192-197 http://www.sciencedirect.com/science/article/pii/S0376635711001884
Cavagna A, Cimarelli A, Giardina I, Parisi G, Santagati R, Stefanini F, Viale M. (2010) Scale-free correlations in starling flocks. PNAS June 29, 2010 vol. 107 no. 26 11865-11870. http://www.pnas.org/content/107/26/11865.full
Bialek W, Cavagna A, Giardina I, Mora T, Silvestri E, Viale M, Walczak AM. (2012) Statistical mechanics for natural flocks of birds. Proc Natl Acad Sci U S A. 2012 Mar 27;109(13):4786-91. Epub 2012 Mar 16. http://www.ncbi.nlm.nih.gov/pubmed/22427355