- 50 views
- 77 downloads
Quantifying Links Between Group Size, Temperature Preference, and Behavioural Fever in a Gregarious Cyprinid Using a New Approach to Deconvolve Individual Contributions to Shoaling
-
- Author / Creator
- Proctor, Ethan
-
Shoaling and schooling play crucial roles in survival for over half of all vertebrate fish taxa. Social behaviour in fish reduces stress, increases collective vigilance towards predators, increases foraging productivity, and enhances communication between those electing to remain with the group. Some taxa are obligate schoolers, while others group up gregariously, occasionally shoaling in loose social groups.
Thermal stress caused by anthropogenic climate change may disrupt social behaviours by creating pressure on fish to alter their temperature preference, a behaviourally regulated range of temperature that optimizes physiological rates, fertility, growth, and maturity. Temperature preference also has a role in behavioural fever, where upon infection fish locate zones 2-4°C hotter to enhance innate defence pathways, similarly to metabolically induced fever in endotherms. This thesis presents an experimental approach to quantify the links between shoaling, thermal preference, and behavioural fever by combining a novel thermal enclosure and machine-learning multitracking protocol.
It is a challenge to properly assess shoaling behaviour and thermal preference within experimental aquaria. Previous studies have been limited in simulating realistic environmental gradients and avoiding systematic rheotactic and thigmotactic biases. This thesis aimed to develop and validate custom tank with an open rectangular arena that allows free thermoregulation within a finely controlled temperature gradient lacking physical barriers. The tank maintained a stable uniform temperature differential of 6.4°C (18.5-24.9°C ) for 48 hours.
Despite recent machine vision successes in applying conventional algorithmic and machine-learning approaches to the problem of tracking multiple targets in a visual scene, it remains a challenge to capture group behaviour from video while maintaining unique identities without physical markers. This thesis aimed to develop a markerless multitracking protocol which trains a convolutional neural network with the visual information of targets to create unique fingerprints that are matched to targets to correct tracking mistakes during complex social behaviour.
To demonstrate the application of this new approach, this thesis also presents two proof-of-concept trials examining how group size modulates the behaviour of goldfish (Carrasius auratus) shoals, and how febrile and afebrile fish mutually influence each other within a shoal. Larger shoals of goldfish preferred colder temperatures and swam slower than smaller shoals, while individual differences in group speed were more repeatable than differences in temperature preference. These preliminary findings argue that the methods developed herein are well-suited to provide insight into individual contributions to shoal movement and temperature selection. This work also represents a novel aquatic approach to study the contributions of fever to innate immunity and disease transmission.
-
- Graduation date
- Fall 2024
-
- Type of Item
- Thesis
-
- Degree
- Master of Science
-
- License
- This thesis is made available by the University of Alberta Library with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.