Skip navigation
Please use this identifier to cite or link to this item: http://arks.princeton.edu/ark:/88435/dsp0141687m144
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorLevin, Simon A-
dc.contributor.advisorCouzin, Iain D-
dc.contributor.authorLeblanc, Simon-
dc.contributor.otherApplied and Computational Mathematics Department-
dc.date.accessioned2018-04-26T18:46:27Z-
dc.date.available2018-04-26T18:46:27Z-
dc.date.issued2018-
dc.identifier.urihttp://arks.princeton.edu/ark:/88435/dsp0141687m144-
dc.description.abstractLiving in groups provides many benefits to individuals, like improved survival in the face of predators, increased ability to find food, or mates. In addition, group living animals, by pooling knowledge, can make better decisions than individuals in isolation. To gain such former benefits from being in a group, information has to be exchanged among group members, and how this is achieved when individuals are uncertain, have competing interests, exhibit individual differences, and must make decisions within complex habitats is not completely understood. While some animals communicate by exchanging signals, the majority of schooling fish predominantly use cues. They leak information through their actions: where, when and how they move. Fish have access to a lot of visual information, and yet how they use this has not been examined in depth. From recent knowledge of how fish employ visual information, interaction networks can be reconstructed revealing hidden pathways of communication (how the behavior of individuals influences others). In this thesis, I provide the tools and models necessary to generalize experimental techniques of visual field reconstruction to computer generated groups of individuals in arbitrary configurations. Then, I use the flexibility afforded by these methods to study multiple problems related to the anti-predatory behaviors of fish schools. A group has many eyes to keep watch of its surroundings, but how many? I show that the answer depends on multiple factors, like the global state of the group, its density, and the peripheral vision of the fish. Once a threat is detected, an effective response requires an alarm wave to spread. Although the capacity to detect predators can vary depending on their position around the group, alarms tend to spread equally well in all directions by virtue of the topology of the interaction network, itself a consequence of the embeddedness of the fish bodies in their environment. By studying how groups may adapt to perceived danger, I find that alarms spread best when individuals adjust their rules of motion, and their sensitivity to the movements of others, simultaneously. Finally, using a dynamical model, I show that predators can benefit from coordinating while hunting.-
dc.language.isoen-
dc.publisherPrinceton, NJ : Princeton University-
dc.relation.isformatofThe Mudd Manuscript Library retains one bound copy of each dissertation. Search for these copies in the library's main catalog: <a href=http://catalog.princeton.edu> catalog.princeton.edu </a>-
dc.subjectcollective behavior-
dc.subjectfish schooling-
dc.subjectinformation-
dc.subjectnetworks-
dc.subjectvisual fields-
dc.subject.classificationApplied mathematics-
dc.subject.classificationBiology-
dc.titleInformation Flow on Interaction Networks-
dc.typeAcademic dissertations (Ph.D.)-
pu.projectgrantnumber690-2143-
Appears in Collections:Applied and Computational Mathematics

Files in This Item:
File Description SizeFormat 
Leblanc_princeton_0181D_12460.pdf15.5 MBAdobe PDFView/Download


Items in Dataspace are protected by copyright, with all rights reserved, unless otherwise indicated.