Under which keyword would you try to find out more about the grouping behaviour which looks like this? Does anybody recognizes such behaviour, or know an animal species which uses it?

My tries: threshold group behaviour (threshold here is the group size - if bigger than 5 individuals == stable), or gravitational group behaviour?

enter image description here

After step 6 the group disintegrates (for example due to foraging) and mechanism starts by step 1. Important thing is that the location of forming stable group is always different.

  • $\begingroup$ You can check this paper out. I used the keyword "pack size threshold" $\endgroup$
    Feb 2, 2015 at 11:30
  • $\begingroup$ Hi @WYSIWYG! Close, but no cigar. Did you know any other examples? $\endgroup$ Feb 2, 2015 at 16:25

2 Answers 2


I believe it is a type of fission-fusion society or even some hybrid with agent based modeling.

In ethology, a fission–fusion society is one in which the size and composition of the social group change as time passes and animals move throughout the environment; animals merge (fusion)—e.g. sleeping in one place—or split (fission)—e.g. foraging in small groups during the day. For species that live in fission–fusion societies, group composition is a dynamic property [fission-fusion society].

A classical example is the Bonobo or pigmy chimp.

  1. Bonobo - Pigmy Chimpanzee
  2. Sociobiology
  3. The Social Behavior of Chimpanzees and Bonobos

While I don't think it's precisely the behaviour you are asking about, you might also be interested in slime molds (e.g., Dictyostelium or Myxogastria). They start out as motile single cell organisms, have limited 'vision' (chemical cues), coalesce into a large multicellular structure when resources become limiting, and then disperse from (usually) a fruiting body.This is an example of "emergent collective behaviour".

Other searches you may want to try are "Biased random walk models" and "Shoaling behaviour".

Although pretty heavy, these papers may contain a formal mathematical description of the model you are after:

  • Akira Okubo, 1986, doi: 10.1016/0065-227X(86)90003-1

  • Angela Stevens and Hans G. Othmer, 1997, doi: 10.1137/S0036139995288976

  • $\begingroup$ After reading about emergent behavior and previously shoaling(aka schooling), I don't think those apply. When I answered this question, I original did some digging into schooling as well but concluded it doesn't fit. A discrete agent based model can be represented as a random walk so you didn't add anything new. $\endgroup$
    – dustin
    Feb 7, 2015 at 20:17

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .