While cooperation is common at many levels of life, ranging all the way from micro-organisms to complex social structures, cooperative behaviors are fundamentally vulnerable to selfish defection. Hence, their prevalence in nature is often considered puzzling. In multilevel systems, individuals may act as cooperators or defectors depending on the level at which they meet their partners. We refer to mixed roles of this kind as the dual face of cooperation. Other mechanisms, like tag-recognition can also lead to dual-faced cooperation. In either case, a differentiation of cooperation levels between in-group and out-group may occur. Such complex interactions, involving multiple levels and the possibility of discriminating between interaction partners of different type, are present in cancers and have their counterparts in the social world. Here we investigate, first, when it is beneficial to invest into in-group and/or out-group cooperation, and second, whether in-group and out-group individuals should best be defined based on being neighbors or on sharing the same tag. Our research also contributes to the longstanding debate on the relative importance of kin selection versus group selection as driving forces of cooperative behaviors.
We combine two existing approaches to the study of cooperation: multilevel selection  and tag-based cooperation . In our model, individuals play continuous pairwise prisoner's dilemma games and reproduce based on their payoffs. While stochastic events constitute the individual-level dynamics, assuming infinitely many demes (assemblies of individuals), we express our population-level model using forward Kolmogorov equations. The evolving strategies are quantitative and multivariate, with elements describing the levels of cooperative investments into in-group and out-group games. On this basis, we allow the group-defining criterion to evolve, ranging from being neighbors to sharing the same tag. To investigate the resultant evolutionary dynamics of dual-faced cooperation, we apply the framework of adaptive dynamics theory.
First, if all individuals share the same tag, in-group cooperation is typically beneficial. Quite surprisingly, out-group investments may also evolve to be positive: this particularly occurs when not all neighbors are considered to be in-group members. These evolutionary outcomes change when we allow new tags to appear. Then, a mutant with a tag different from that of the resident may invade only if it sufficiently differs in its investment strategy from the resident, toward larger in-group or smaller out-group investments.
 Traulsen, A. and Nowak, M.A. 2006. Evolution of cooperation by multilevel selection. PNAS 103: 10952–10955.
 Traulsen, A. and Nowak, M.A. 2007. Chromodynamics of cooperation in finite populations. PLoS ONE 2: e270.
Anne Seppänen, of University of Turku, Finland, is a Finnish citizen. She was funded by IIASA's Finnish National Member Organization and worked in the Evolution and Ecology (EEP) Program while with the YSSP.
Please note these Proceedings have received limited or no review from supervisors and IIASA program directors, and the views and results expressed therein do not necessarily represent IIASA, its National Member Organizations, or other organizations supporting the work.
Last edited: 19 August 2015
International Institute for Applied Systems Analysis (IIASA)
Schlossplatz 1, A-2361 Laxenburg, Austria
Phone: (+43 2236) 807 0 Fax:(+43 2236) 71 313