reciprocityReciprocity is a cooperative interchange of benefits. There are three types of reciprocity: direct, indirect, and generalized.  The image to the right from helps distinguish them.  Direct reciprocity is a tit-for-tat exchange of benefits by two actors.  Indirect reciprocity occurs when an altruist is rewarded by third parties for behaving generously towards others.  Generalized reciprocity is an anonymous, “pay it forward” model.

Many animals lack the cognitive skills required for the evolution of direct or indirect reciprocity.  Short term memory makes them incapable of identifying specific individuals and associating them with past acts of altruism.  But advanced memory isn’t necessary for the evolution of generalized reciprocity, which follows a simpler rule: help anyone if helped by someone.  It’s easier for a biologist to try to explain why animals sometimes incur fitness costs to help non-kin using the generalized reciprocity model.

In an article titled “The Evolution of Generalized Reciprocity on Social Interaction Networks,” Michael Taborsky and Gerrit Sander van Doorn of the Institute of Ecology and Evolution use theoretical models to identify types of social network structures that support evolution of generalized reciprocity. The results of Taborsky and van Doorn’s experiment suggest that when cooperators are rare, sparse networks with low connectivity are conducive to generalized reciprocity because the future probability of being rewarded for an altruistic act is greater.  In other words, your chances of being on the receiving end of an altruistic act go up when you interact frequently with a small number of social partners versus infrequently with a large number. The authors also found that a modular network structure supports generalized reciprocity for two reasons. First, a small “neighborhood” of 100% altruists is well-protected against exploitation by uncooperative defectors.  Second, in a population with a majority of altruists modularity increases the collective yield of “benefit” by increasing the average number of steps made during a random walk on the network before the “pay-it-forward” process is terminated by a defector.

The results of this study got the wheels in my head turning.  A theoretical computer model tells us generalized reciprocity is more likely to evolve when there is frequent interaction with a small number of social partners and modular network structure.  Three important questions follow.  First, can we use this knowledge to actively structure biomimetic human social networks that improve the probability of reciprocity, and thereby cooperation?  Second, when it comes to humans and high-level brain function how do factors such as a solid memory of people’s past behavior change the dynamics of cooperation?  And the third question, which the rest of this post attempts to answer: Can existing research from non-biology fields help us come up with new biological research questions about reciprocity, so that through further experimentation we might improve our biological understanding of the phenomenon?

At least three bodies of research outside the field of biology may help flesh out our biological understanding of reciprocity, and by consequence, improve our ability to build a biomimetic “environment” that supports cooperation. The first is research in the field of sociology on the influence of social networks.  The second is research by political theorists on rationality and cooperation in experimental public goods games.  The third is research by industrial-organizational psychologists on the effectiveness of team-based design.  Below I offer examples of references from each of these three academic fields that I feel could help biologists come up with a new set of research questions about the evolution and dynamics of reciprocity.


Social scientist (and physician) Nicholas Christakis is a leading researcher on social factors that affect health, health care, and longevity.  Research findings presented in his 2010 TED Talk, “The Hidden Influence of Social Networks,” suggest that an individual’s emotional and physical health are influenced by the emotional and physical health of members of their social network.  Influence extends through three degrees of separation such that your health is impacted by the health of a friend of a friend of a friend.  Research question: Which factor has more influence over speed of evolution of generalized reciprocity, percentage of altruists in a population, or the average degrees of separation between an altruist and any other member of the social network?

Australian criminologist John Braithwaite conducted research on factors influencing crime rates for his 1989 book Crime, Shame, and Reintegration. His research suggests most crime is committed by people with fewer connections in their social network, or by people who expect that social connections will not last. As a result, crime is committed disproportionately in areas with high residential mobility.  Research question: Is generalized reciprocity in nature more prevalent in stable, sedentary communities?

Political Theory

rationality-and-the-ideology-of-disconnectionLucky for me, as I was grappling with questions about reciprocity and conditions conducive to its evolution, I was simultaneously reading a political theory book recommended to me by UAkron Conflict Management professor Bill Lyons.  The 2006 book written by Michael Taylor is titled Rationality and the Ideology of Disconnection.  Part three of the book, titled “Living in unity, doing your part: rationality, recognition, and reciprocity,” was most helpful.  This section explored human behavior in experimental public goods games to determine conditions under which people appear motivated by a norm of reciprocity.

An example of an experimental public goods game is one in which each player is given a sum of money and invited to contribute some or all of it to the public good.  Contributions to the public good are multiplied by two, and that total is divided equally amongst all group members.  Clearly, there is a higher payoff for the player that contributes nothing, (the same way there is a higher fitness payoff in nature for a hungry animal that eats food shared by non-relatives in the community, but does not share his own); however, even when this experimental public goods game is played once among total strangers who do not communicate, each player contributes an average 40-60% of their money to the public good.  The behavior of human players in experimental public goods games defies common sense the same way some natural organisms defy common sense by cooperating with non-kin in nature.

I also find it interesting that in repeated public goods games rate of cooperation increases if 1) players are allowed to communicate with one another, 2) players are face-to-face when making decisions, 3) if players are allowed, for a small fee, to “punish” free-riders. Under conditions like these, rate of cooperation often rises until monetary contributions from most players near 100%.  I would be interested to learn whether the group achieves near perfect cooperation quicker if the experimenter “plants” a number of perceived strangers at the start of the game who go “all in” from the get go.  Research question: If the results of the experiment I just described show that level of generosity, not just whether or not other players are generous, is “contagious” – could we advance Taborsky and van Doorn’s experimental results by complicating their theoretical model with a new variable –scalable generosity?

Industrial-Organizational (I-O) Psychology

Literature in I-O psychology suggests team-based organizations are more efficient than individual-based, “every man for himself” organizations.  For example, in a 1977 Harvard Business Review article, “How Volvo adapts work to people,” then Volvo CEO Pehr Gyllenhammar demonstrates how his company’s shift from individuals in an assembly line to self-controlled work teams brought tremendous success.  Perhaps it is because a modular, team-based organization, where a small number of employees (ideally 7?)  interact frequently, supports the evolution of reciprocity…of cooperative workplace behavior.  In the same article, Gyllenhammar talks about the new role of managers in a team-based organization. Managers shift from directors to facilitators.  Instead of controlling employees, they create and maintain an environment where employees feel comfortable consulting with one another.  Research Question: In a non-human community, are their internal or external forces analogous to managers that function to facilitate evolution of generalized reciprocity?

The results of experiments built around these research questions may improve our biological understanding of reciprocity.  With a better understanding of the dynamics of reciprocity and conditions conducive to its evolution, we may be able to build human environments much more supportive of cooperation than those we already have in place.  High hopes for the future of biomimicry!


One comment

  1. Bill

    Thanks for sending this along, Emily. Reciprocity is the concept that most intrigued me about CAT when I was working with Michael at UW. In my first book, The Politics of Community Policing, I used this concept as a central analytical tool for differentiating between approaches to community policing that were democratizing and those that were more likely to be simply disciplinary. As I argued there, when the citizens in the Seattle neighborhoods I studied observed the hot spots along with the police supervisors this resulted in meetings where the supervisor reported on activity and progress and then the citizens reported, often identifying things the supervisor did not or criticizing the officers or supervisors.

    These meetings were not utopian; there was still a very real power differential. But during this period the police-community partnership was more reciprocal, and by doing so it invested in social capital in that neighborhood, and improved both formal and less formal mechanisms of social control. When this more reciprocal relationship faded (changing personnel on the citizens group, an effort by the police department to institutionalize and thereby domesticate this and other groups), the partnership was transformed…social capital dissipated, informal social controls atrophied, formal social controls became more just aggressive forms of traditional policing—made less accountable by being wrapped in a discourse about community.

    Anyway, I suggest linking reciprocity with social capital. And seeing both as constituent elements of community (as you have described in your blog). In addition to what you write, this highlights not only the importance of directness, but also overlapping/multifaceted relationships as supporting reciprocity, social capital, and community. Love the work you are doing. Thanks for your post, Bill

Leave a Reply

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

You are commenting using your 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