COMING SOON! PQDT Open is getting a new home!

ProQuest Open Access Dissertations & Theses will remain freely available as part of a new and enhanced search experience at

Questions? Please refer to this FAQ.

Dissertation/Thesis Abstract

Animal Social Networks and Movement
by Traud, Amanda Lynn, Ph.D., North Carolina State University, 2017, 157; 10610665
Abstract (Summary)

Studying animal social systems can provide valuable insight into the mechanisms behind animal population behaviors and inform scientists about the evolution of social structures in humans. We use social network analysis to discover whether animal networks are capable of serving as a suitable proxy for human contact networks. Finding this information has many ramifications for its use, including the spread of pathogens or being able to stop such spreading. To this end, we study the interactions of colonial animals, or animals that live in colonies such as prairie dogs (Cytomis gunnisoni) and ants (Formica subsericea).

In our study of prairie dog social networks, we examine how social network analysis techniques can be used to find important individuals and uncover the macro-structure. This then allows for the comparison of current prairie dog study techniques to social network analysis findings. We discover a high correlation between macro-structure found with social network analysis and current prairie dog social group classification. The social network techniques require far fewer data. Comparing macro-structure to prairie dog behavior, we discover no correlation thus groups of prairie dogs contain a mixture of behaviors. Key individuals are identified in each prairie dog social network. These key individuals have significant implications for disease spread and communication channels.

One way we study ant social structure is by examining how the ant queen’s presence affects ant social networks. To ascertain whether the queen affects the structures of ant social networks, we compare networks with a queen to networks without, individual network measures for queens to individual network measures for workers, individual network measures for workers in a network with a queen to those for workers in a network without a queen, and then compare the networks with and without queens to standard network models. While the queen is highly important to colony survival, the queen does not significantly affect global ant network structure. The queen is found to have a local network that is significantly different from that of workers. Like many human contact networks, both networks with queens and networks without are classified as Small-World networks, with networks with queens having a higher similarity to Small World networks than the networks without.

We also study ant social structure by examining how ant network structure changes over time, i.e. collecting and analyzing dynamic ant social networks. These data are also used to ascertain whether ants have preference to their associations, or friends. With this data, we create and present a method for finding the appropriate network observation window. We discover ants have preferred associations and that the accumulation of ant interactions approach a stable density.

As movement affects location and therefore affects the availability of individuals for interaction, we model ant movement. Ant trajectories are collected using the methods outlined in Appendix A and analyzed to produce movement models. Ant movement is highly complex: ants are discovered to have multiple ant step length categories, multiple states of movement, and different movement behavior based on individual location.

Indexing (document details)
Advisor: Lloyd, Alun
School: North Carolina State University
Department: Biomathematics
School Location: United States -- North Carolina
Source: DAI-B 78/10(E), Dissertation Abstracts International
Subjects: Entomology, Mathematics, Bioinformatics
Keywords: Ants, Biomathematics, Movement models, Network science, Prairie dogs, Social networks
Publication Number: 10610665
ISBN: 978-1-369-85548-7
Copyright © 2021 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy