Influence of Partner Selection on Functional Differentiation: Emergence of Diversity by Isolated Interaction and Preference Change

Authors
Saori Iwanaga, Akira Namatame
Corresponding Author
Saori Iwanaga
Available Online 1 March 2017.
DOI
https://doi.org/10.2991/jrnal.2017.3.4.10
Keywords
collective behavior, diversity, payoff matrix, isolated interaction, preference.
Abstract
When examining collective behavior, we focus on the interactions of individuals. We must also consider the microscopic level, where individual agent decisions occur, and the macroscopic level, where collective behavior can be observed. Collective behavior is both interesting and difficult because aggregate outcomes must be evaluated within environmental constraints. The performance of a collective system strongly depends on the type of interaction and heterogeneity of agent preference. In this study, relative to the emergence of diversity, we found that two different learning processes are required. In addition, we found that isolated interactions are required for the emergence of diversity. However, the utility of less clustered interactions is greater than that of isolated interactions. We found that isolated interactions are effective for the emergence of diversity. Once diversity emerges sufficiently, less clustered interactions are effective relative to the utility of each agent and the entire population.

Copyright
© 2013, the Authors. Published by ALife Robotics Corp. Ltd.
Open Access
This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).


Download article (PDF)