Presentation Details
| The position of identity information impacts its detection in synthetic vocalizations Ollia M.Miller1, Raneem Samman1, 2, Grethel A.Juárez1, Grace Smith-Vidaurre1, 2, 3. 1Department of Integrative Biology, East Lansing, MI, USA.2Department of Computational Mathematics, Science, and Engineering, East Lansing, MI, USA.3Ecology, Evolution, and Behavior Program, East Lansing, MI, USA |
Abstract
Vocal information encoding may track the complexity of the social environment. Some animals may encode multiple types of information in a single vocal signal. While computational tools can be used to detect complex identity information, it is difficult to validate such detection with empirical data. We used the paRsynth package to manipulate the detectability of group and individual identity information in synthetic vocalizations. We created datasets where we placed each type of information first or last in calls, then performed simulation experiments to test how position impacted information detection. To detect identity information, we applied a pipeline that included spectrographic cross-correlation and unsupervised clustering. We found a small, significant effect of position on identity information detection. Both group membership and individual identity information were more difficult to detect when positioned last. We are currently manipulating the distance between types of identity information to break information detectability. Our work provides a foundation for future simulation experiments and guidelines for detecting identity information in biological vocalizations.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author.
Content Locked. Log into a registered attendee account to access this presentation.