Positive and Negative Networks: Lessons from Ecology?

Let me pose a puzzle.  Mathematical sociologists like puzzles!

Jordi Bascompte recently published an article in Science (vol. 329, 13 August, 2010, p. 765-6) titled “Structure and Dynamics of Ecological Networks.”  The piece is in the “Perspectives” section — which are commentaries and explanations of the “real” articles (which are often a bit too discipline-specific to be accessible to the average reader).  Basecompte’s article is an explanation and appreciation of the piece by Elisa Thebault and Colin Fontaine (also vol. 329, 13 August 2020 p. 853-6) titled “Stability of Ecological Communities and the Architecture of Mutualistic and Trophic Networks.”

A thanks to Peter Burke of the SPYRL social psych lab, and Department of Sociology at UCR for passing these along to me. 

A very common use of graphs and networks in ecology is as a representation of web of relations among species.  A “food web” for example, is composed of nodes (species) and directed relations (e.g. “eats”) as in Bear -> Salmon.

The food web is an “antagonistic” relation:  one species feeds on the other (occasionally, such relations are reciprocated).  The relation results in a “gain” for one node and a “loss” for the other.

Ecologists also focus on “mutualistic” relations where two species interact, but both benefit.  Bees and flowers, for example are mutualistic — bees pollinate flowers, flowers provide food for bees.  These kinds of networks can be thought of as undirected (though maybe they would better be regarded as multiplex and directed).  Generally, mutualistic ecological relations form bi-partite networks.

Bear with me just a moment more…

The interesting results in the Science article are that networks of “antagonistic” relations in ecology tend to have a characteristic form of largely separated and rather densely connected clusters – something akin to “communities”.  Mutualistic networks, on the other hand have a characteristic form composed of “generalists” (nodes with high degree) and “specialists” (nodes with low degree) of the species in each mode, in a nested “hierarchy”.

Networks of individual humans, groups, roles, organizations, or populations often have “antagonistic” relations — “A” dominates “B”.  But, human networks are also connected by mutually beneficial ties (e.g. emotional support and identity confirmation).

In sociology, we tend to think of “antagonistic” networks as, perhaps, having an equilibrium tendency toward hierarchy.  “Mutualistic” human relations, we tend to see as likely to be organized as communities — often largely separated from one another.

Are these patterns of findings really contradictory?  They would seem to suggest that structures of inter-species networks follow rather different rules than the structures that emerge in relations among heterogeneous human agents (individuals, groups, organizations, communities, societies).

Advertisements

5 Responses to Positive and Negative Networks: Lessons from Ecology?

  1. Michael Bishop says:

    Hi Bob, (multimode = Bob Hanneman)

    I thought I’d provide the direct links to the specific articles under discussion:
    http://www.sciencemag.org/cgi/content/full/329/5993/765

    http://www.sciencemag.org/cgi/content/full/329/5993/853

    It will be easy for me to say something stupid because I only skimmed the articles and this is decidedly not my area of expertise, but here goes…

    First I’d like to inquire about your suggestion that
    “In sociology, we tend to think of “antagonistic” networks as, perhaps, having an equilibrium tendency toward hierarchy. “Mutualistic” human relations, we tend to see as likely to be organized as communities — often largely separated from one another.”

    I couldn’t tell if you were endorsing those claims or just saying they are common in sociology. But I would imagine that in people we could see diversity in network structures even if we restrict ourselves to antagonistic or mutualistic relations… at least in part due to the diversity of relations one might group under the label antagonistic or mutualistic (competition for political power, and competition for economic power may both be described as antagonistic, but they are often different in many ways including their network structures).

    Another idea… the choice/definition of the network has huge implications for what we find. In this example, it is the decision to consider antagonistic/mutualistic relations and ignore how these networks fit into much larger networks. Its just speculation, but I imagine that what we find in the larger networks may dramatically alter what we thought our understanding of the smaller (and more importantly, selectively defined) network.

  2. multimode says:

    Thanks for the thoughts!

    On the first point…
    no, I certainly wasn’t endorsing the claims that competition tends to be associated with hierarchy, and cooperation with community. Not in any sense that they must, or should. I do have the sense that many sociologists would probably suggest that there are such tendencies.

    I couldn’t agree more that “antagonistic” or “mutualistic” are pretty difficult to directly “port” into social networks. With animal species, not too tough: you eat some other species or you don’t; you pollinate or you don’t. Human “negative tie” networks and “positive tie” networks are certainly much more diverse classes of relations than what the ecologists work with.

    And, your last point is a good one, too. The thing about humans (and their organizations and institutions) is that the ties are often multiplex — some positive and some negative. Some networks are embedded in others. One problem I’m working on is competition in the airline industry. Here, airlines in “alliances” cooperate with one another in order to compete with other alliances.

    Thanks for the thoughts!

  3. […] Positive and Negative Networks: Lessons from Ecology? – via Permut – The interesting results in the Science article are that networks of “antagonistic” relations in ecology tend to have a characteristic form of largely separated and rather densely connected clusters – something akin to “communities”. Mutualistic networks, on the other hand have a characteristic form composed of “generalists” (nodes with high degree) and “specialists” (nodes with low degree) of the species in each mode, in a nested “hierarchy”. Networks of individual humans, groups, roles, organizations, or populations often have “antagonistic” relations — “A” dominates “B”. But, human networks are also connected by mutually beneficial ties (e.g. emotional support and identity confirmation). […]

  4. Excellent post, l quite agree with your conclusion. However lam having problem subscribing to your rss.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: