Benefits of Social Groups

There is an interesting article in Nature about group formation: the interactions between predators and prey (which was usually based on random mixing) is drastically changed when the predators and prey start forming social groups. Prey benefit from grouping by reducing the chances that predators will come across them, and predators benefit by being able to attack in numbers when they do encounter prey. Naturally, the question to ask is to what extent this could be true in other scenarios- (forming trust communities?) and can stability arise from being super-selective about who we interact with (thanks to Steve for the article).

9 Responses to “Benefits of Social Groups”

  1. In this book, Chapter 2 demonstrates that primates (including humans) are “first and foremost social creatures – this is our evolutionary strategy, the thing that enables our type of animals to continue to reproduce”

    Somewhat related: How shops benefit from clustering
    Why do some competing stores locate near one another, while others don’t?
    * Car dealers cluster because it is in their best interests to do so, given consumer behavior. Buyers would be less prone to buying if dealers were discrete and far away from one another, but they are more prone to shopping (and possibly buying) if dealers cluster around one another. Such, of course, is not generally the case with groceries.

    * Take the arrival of Bed Bath & Beyond on the 6th Avenue in NY. It generated a search-related information externality — whenever people search for information, their search behavior is generally influenced by what they can learn from the actions of others. This tends to generate herding phenomena, among other things. When stores saw the success of BB&Y, they relocated there, some because of the spillover effect of BB&Y’s success (it re-legitimated the area), and some because they wanted to take advantage of the new and larger crowds on lower Sixth Avenue.

  2. mike says:

    Scaling seems to work differently for predators and prey.

    Two reindeer are less than twice as easy to find as one reindeer (eg they leave one scent trail), and two wolves are more than twice as hard to avoid as one wolf (eg they can flank their prey). But while the benefits keep growing for the reindeer (for large n, n+1 reindeer are less than (n+1)/n times easier to find than n reindeer), they level off for the wolves (for large n, n+1 wolves aren’t more than (n+1)/n times harder to avoid than n wolves). So the size of the wolf pack reaches equilibrium, which further increases the scaling advantages for the reindeer herd – even if the wolves find the reindeer, they can only eat one each, so the reindeer face less individual risk in a larger herd.

    Does this apply to all predator/prey relationships, or are there any scalable predators (or unscalable prey)? :-)

  3. Neal Lathia says:

    I agree that wolves benefit less from being in packs, but their gain would also be measured in terms of how successful an attack would likely be- the more wolves, the easier the kill (even for the lazy ones)!

    Surely the benefits do not continue growing for deer? That would imply that the best tactic would be for all the deer to herd together, which does not happen in nature..?

  4. mike says:

    Good point about lazy wolves – I wonder how they deal with social loafing. Micropayments or reputations? ;-)

    You’re right about the deer of course – wildebeest herds can exceed a million members but there must be a limit somewhere – maybe geographical limits like the width of migration routes? And not all deer move in herds – maybe there’s less of an advantage from herding in dense woodland where individual animals are harder to track?

  5. On the benefits of grouping (or how to sell sex):

    Steven Levitt (economics professor and co-author of ‘Freakonomics’) presented preliminary findings from a study on the economics of street prostitution. Here is one of their findings (which I summarized here):

    Almost half of the city’s arrests for prostitution take place in just 0.3% of its street corners. The industry is concentrated in so few locations because prostitutes and their clients need to be able to find each other.

  6. mike says:

    Finally a use for location-based services?

  7. :-) Actually, What about a context-aware hooking up service? You are in a bar, you get dating profiles of people around you. You are in that 0.3% portion of Chicago, you get working profiles of lovely street workers. Of course, reputation models may help ;-) EPSRC often sponsors street-level social research. Shall we write a research proposal?

  8. mike says:

    Hmm, I guess it would count as “Physical Sciences” :-)

  9. [...] over exciting percom applications – if (context=space of interaction) as “my company premises” is in “my PDA is [...]