Home

Towards Realistic Models for Evolution of Cooperation

Summary of: Towards Realistic Models for Evolution of Cooperation

Author(s) / Editor(s)

Mui, Lik
The five major approaches to answering how cooperation emerges and becomes stable in nature (Group Selection, Kinship Theory, Direct Reciprocity, Indirect Reciprocity, and Social Learning) might be improved by not presuming asexual and non-overlapping generations, simultaneous-play for every interaction, dyadic interactions, mostly predetermined and mistake-free behavior, discrete actions (cooperate or defect), and the trivial role of social structure and social learning of individuals.

Disciplines

Biology, Cultural Evolution, Sociology

Keywords

trust, reputation, reciprocity, evolution, cultural evolution, cooperation, competition, bioeconomy, altruism, agent-based model

Publication Reference

Published in/by
MIT LCS Memorandum
Date
2002

Findings

  • Observer-based reciprocity relaxes the requirement that each individual's likelihood of cooperating be known globally by introducing randomly selected observers. Even though interactions are only visible to these observers cooperation can still evolve showing "that cooperation may evolve through indirect reciprocity with or without global knowledge about agents' image scores."
  • Darwin's notion of the "survival of the fittest" does not specify what "fittest" refers to, and for good reason: the outcome of a behavior in each contingent situation determines its fitness. Different interpretations of "fittest" lead to different models for how natural selection works and therefore offer different explanations for the evolution of cooperation.

Sociological and biological observations of humans and animals show that cooperation is an inherent part of human life and the life of many animals. This poses two questions: how do cooperative strategies become stable within evolution? And, how does cooperation emerge initially? Even though researchers have tried to answer these questions for at least a century, existing models do not fully explain why cooperation evolves. There are five major approaches: Group Selection, Kinship Theory, Direct Reciprocity, Indirect Reciprocity, and Social Learning. Each of these models explain only a few aspects of cooperation and might be improved by dropping some unrealistic assumptions: asexual and non-overlapping generations, simultaneous-play for every interaction, dyadic interactions, mostly predetermined and mistake-free behavior, discrete actions (cooperate or defect), and the trivial role of social structure and social learning of individuals.