Cooperators Attract Cooperators, Non-Cooperators are Stuck with Each Other

In catching up on a back-log of articles people have emailed me, I’m absorbing what I think are probably obvious but nonetheless profound implications of a study by Coren Apicella, Frank Marlowe, James Fowler, & Nicholas Christakis that was published as a letter in the January 2012 issue of Nature & summarized in that same issue by Joseph Henrich.  As Henrich states

Through the 1970s and 1980s, many researchers assumed that hunter-gatherers tackle this core dilemma [of avoiding ‘free-riders’] by relying on a combination of kinship and direct reciprocity. By targeting kin on the basis of shared genetic inheritance, cooperators are more likely to deliver benefits to fellow cooperators. Similarly, by reciprocating help with help, unrelated individuals can sustain tit-for-tat cooperation. However, by the twenty-first century it had become clear that although kinship and direct reciprocity can each explain some aspects of human prosociality, many domains of cooperation, ranging from the sharing of meat within bands of hunter-gatherers to territorial defence, cannot be easily accounted for by these models.

Let’s be clear–we’re not talking one model over the other.  We’re talking about a way to explain the residual variance in cooperation. The models previously outlined, according to Henrich, are as follow:

  1. cooperation is sustained by a process of cultural learning and the sanctioning of norm violators, which leads to the continuous reassortment of groups. More cooperative groups tend to endure and expand, whereas less cooperative groups gradually break down.

  2. individuals cooperate competitively, as a means of attracting an inflow of partners who bring benefits

  3. cooperation can be sustained as individuals seek out those with different skills, resources or abilities…assortment based on complementarity rather than similarity.

Apicella & colleagues study investigated the parameters of cooperation among the Hadza, a hunter-gatherer population in Tanzania. They found that the Hadza

  1. do not preferentially pick more cooperative individuals

  2. do not preferentially network with those possessing complementary attributes…as indicated by age, food preferences or various physical measures

What they did find, according to Henrich, is that

there is substantially more variation among the bands, and substantially less variation within them, than would be expected by chance. Despite the fluidity of band membership, it seems that some combination of similarity-based association, social learning and sanctioning establishes differences in cooperative tendencies among different bands.

And in terms of like gravitating to like, the free-rider problem is reduced because, again according to Henrich’s summary,

high contributors associate with other high contributors, and low contributors choose other low contributors.

The implications, according to Apicella et al., are more profound though. Despite dramatic differences between Hadza lifeways & those of modern denizens of Euro-America & other cosmopolitan population centers, there is significant consistency in cooperation behavior, suggesting strong selective forces have molded these traits.  They conclude that “social distance appears to be as important as genetic relatedness and physical proximity in explaining assortativity in cooperation” & that “social networks may thus have contributed to the emergence of cooperation.”  This is possible

if individuals tend to interact with others of the same type (cooperators with cooperators and defectors with defectors).

cooperation can evolve if population structure permits clustering. This feature allows cooperators to increase in the population because they benefit from the public goods provided by fellow cooperators with whom they interact. A key prediction of some evolutionary models is thus that there should be relatively more variance in cooperative behaviour between groups as compared to within groups.

Donations in the public goods game are associated with social network characteristics.

a, A comparison of variance in observed donations with variance in 1,000 simulations in which donations were randomly shuffled between all individuals in the population shows that between-group variance in cooperation is significantly higher than expected, and within-group variance is significantly lower than expected, at the camp level. b, An analysis of cooperative behaviour across all camps shows that correlation in cooperation extends to one degree of separation in the campmate networks and two degrees (to one’s friend’s friends) in the gift networks. Moreover, there is anti-correlation at three degrees of separation in the campmate network, suggesting polarization between cooperators and non-cooperators. c, Correlation in behaviour cannot be explained by cooperators being more likely to form or attract social ties. Instead, subjects with similar levels of giving are significantly more likely to be connected at the dyadic level. d, Finally, several measures of proximity are independently associated with similarity in donations, but social proximity (the inverse of the degree of separation between two people in the network) appears to be just as important as genetic proximity (relatedness) and physical proximity (residence in the same camp) in a multivariate test. (Gift networks are defined only within camps and so are not presented for ‘camp’ and ‘geographic’ proximity in Fig. 2d.) Vertical lines indicate 95% confidence intervals and asterisks indicate estimates with P < 0.05. See the Supplementary Information for details of the models.

Which is what they found.

The authors also introduce us to some fun new jargon that I can see myself annoying readers with in the future:

  • degree distribution = number of social ties
  • transitivity = the likelihood that two of a person’s friends are in turn friends
  • degree assortativity = the tendency of popular people to befriend other popular people
  • homophily = the tendency of similar people to form ties

Assortativity is nothing new if you’ve been paying attention to the “pathogen-driven wedge” & “behavioral immune system” literature, to which these data & conclusions are strongly related.  As the authors point out,

degree assortativity may constrain the spread of pathogens, high transitivity may help reinforce social norms (although it can also reduce the flow of new information), and homophily may facilitate collective action.

Christopher Lynn

About Christopher Lynn

Christopher Dana Lynn is an associate professor of anthropology at the University of Alabama, where he directs the Evolutionary Studies program.  Chris teaches undergraduate and graduate courses in biological anthropology, human sexuality, evolution, biocultural medical anthropology, and neuroanthropology.  He received his Ph.D. in Biological Anthropology in 2009 from the University at Albany, SUNY, where his doctoral focus was on the influence of speaking in tongues on stress response among Pentecostals.  Chris runs a human behavioral ecology research group where the objectives include studying fun gimmicky things like trance, religious behavior, tattooing, and sex as a way of introducing students to the rigors of evolutionary science.  In all his “free” time, he breaks up fights among his triplet sons, enjoys marriage to the other Loretta Lynn, strokes his mustache, and has learned to be passionate about Alabama football (Roll Tide!).  Follow Chris on Twitter: @Chris_Ly
This entry was posted in Anthropology, Cultural Evolution and tagged , , , , , , , , , , , , , , , , . Bookmark the permalink.

2 Responses to Cooperators Attract Cooperators, Non-Cooperators are Stuck with Each Other

  1. Pingback: Personal Learning and Cooperation | Clyde Street

  2. Pingback: Reflections of Exploring Personal Learning Networks MOOC | Activate Learning Solutions

Comments are closed.