Moving beyond structure to function

Abstract

For animal behaviorists,social network analysis is coming of age. Ever since John Crook’s early work on weavers and Peter Jarman’s on African ungulates, a central goal has been to understand why particular social structures take the forms that they do. This has typically meant identifying the adaptive value of particular social structures by searching for underlying ecological determinants. Unfortunately, if the structures themselves are only coarsely characterized, then identifying the environmental problems each has purportedly evolved to solve becomes challenging, if not impossible. Across a wide variety of disciplines ranging from mathematics to physics, from sociology to politics to epidemiology, social network analysis has become a powerful tool for characterizing the structures of complex systems of interacting particles or clever individuals. Given that animal social systems are also built upon individual interactions, it stands to reason that social network analysis could provide similar insights into their functioning. Because network structures are built upon all, rather than pairwise,interactions, networks offer nuanced characterizations of social structures that can reveal degrees of connectivity, both locally and globally, societal sub-structuring, ranging from cliques to modules to entire communities, and so much more, all of which can be quantitatively characterized by a series of network metrics. Armed with such graphic visualizations and corresponding metrics, animal social structures become precisely defined and changes over time driven by changes in physical or social environments can be monitored. For examples across a wide range of taxa,and fora primer on social network analytical techniques, see the volume Animal Social Networks edited by Krause, James, Franks and Croft (Oxford University Press, 2015). Published by Elsevier, Inc.