Mating systems are classifications of the outcome of individual strategies aimed at maximising reproductive success. These systems have two components; one describes how individuals socially relate and bond to mating partners and the other describes the genetic outcome of actual mating activities. Reproductive activity is under strong endocrine and behavioural regulation where inter-sexual discrepancies in the regulation of reproductive success have resulted in the majority of mammal species exhibiting polygynous mating systems, with only 5% of mammals being socially monogamous. However, in monogamous species there can be a discrepancy between social pair bonds and sexual mating activities. Aardwolves are extreme diet specialists ii on a temporally fluctuating resource, Trinervertermes spp. A number of studies have described them as socially monogamous but at the same time observed a high frequency of extra pair copulations (EPCs). A recent study also linked sex variation in den use to polygamous mating. However, it is unclear to what extent these EPC's contribute to individual fitness, and how these contrasting mating strategies influence space use and reproductive physiology. The aim of this project was to determine if there was seasonality in reproductive activity and to test if endocrine physiology, home range size and space use were similar between males and females as predicted for a socially monogamous species. We assessed reproductive endocrinology in zoo-housed aardwolves and quantified if physiological and behavioural data in wild aardwolves relate more to predictions based on social monogamy or polygamy. We found physiological support for previous behavioural observations of reproductive seasonality in both zoo-housed and wild aardwolves. We suggest that the seasonal breeding strategy in aardwolves is as a direct consequence of their strong dietary preference on a temporally fluctuating resource, where it is important for aardwolves to time periods of high physiological investment in reproduction with high seasonal abundance of these termites. Our data on zoo-housed aardwolves also showed that the social environment appeared to modify physiological responses to variation in environmental conditions. We suggest that due to strict seasonal breeding in wild aardwolves females are time constrained in receptivity which, combined with their largely solitary behaviour, implies that males have to be dynamic in the onset of their reproductive activity to closely match that of locally receptive females. Therefore, it may be adaptive for aardwolves to retain social receptivity even if resource distributions cause these animals to forage alone. We also found that physiological and behavioural traits correspond better to predictions based on social monogamy than polygamous mating in a population of wild aardwolves. However, earlier studies in the same population found that behavioural traits were more related to predictions based on polygamous mating. Therefore, our data and other studies show how a difference between traits can relate to either social monogamy or polygamous mating in wild aardwolves in the same study area. We suggest that social mating system components regulate the observed endocrine and behavioural parameters more than actual mating patterns, which implies that social components pose a stronger selective pressure on physiology and behaviour than sexual mating patterns. Overall, we conclude that due to the discrepancy in traits that correspond to predictions based on different mating systems, aardwolves do not fit discreetly into any current mating system classification. We suggest that the evolutionary causes for the potentially conflicting mating strategies as well as the fitness benefits of these strategies need to be further investigated. / Dissertation (MSc)--University of Pretoria, 2013. / gm2013 / Zoology and Entomology / unrestricted

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

Thèse numérisée par la Division de la gestion de documents et des archives de l'Université de Montréal.

Thèse numérisée par la Division de la gestion de documents et des archives de l'Université de Montréal

La résolution de problème est souvent utilisée comme outil de mesure des capacités cognitives individuelles chez les animaux non-humains, puisqu’elle permet d’étudier leurs capacités d’apprentissage opérant et d’innovation. Les capacités cognitives diffèrent selon les espèces et les individus, et leur permettent de maximiser leur aptitude. Par exemple, chez les oiseaux et les mammifères, les espèces avec de plus gros cerveaux relatifs possèderaient de meilleures capacités d’apprentissage sociale et d’innovation, qui leur permettent de mieux s’adapter à un environnement nouveau. Malgré l’utilisation répandue des diamants mandarins dans la recherche en écologie comportementale, l’impact du sexe sur la performance de résolution de problème et la répétabilité des performances ont souvent été négligés, et aucune étude, à notre connaissance, ne s’est intéressée à leur performance en contexte de résolution de problème coopérative. Dans le cadre de notre étude, nous avons décidé d’aborder ces différents aspects pour améliorer notre compréhension des capacités cognitives et de la coopération chez cette espèce modèle. Nous avons confronté des diamants mâles et femelles à trois tests de résolution de problème individuelle (test de la ficelle, test de la barre et test du couvercle) et un test de résolution de problème coopérative. Pour ce dernier, nous avons séparé les individus en deux groupes : couples sociaux et couples expérimentaux, afin de manipuler le lien social entre individus. Les couples sociaux étaient composés d’un mâle et d’une femelle ayant cohabité pendant 7 jours, alors que les couples expérimentaux étaient composés d’un mâle et d’une femelle ne s’étant jamais rencontrés auparavant. Bien que nous nous attendions à ce que le sexe n’ait pas d’impact sur la performance de résolution, nos résultats ont révélé que, en moyenne, les mâles étaient plus rapides pour résoudre un test de résolution de problème individuelle. Ces résultats étaient principalement attribuables à leur performance sur le deuxième test de résolution de problème individuelle (test de la barre). Nous pensons que cette différence pourrait être due à un niveau de persistance plus élevé chez les mâles et/ou à l’impact de la sélection sexuelle sur les performances cognitives, et que ces aspects devraient être davantage explorés dans de futures études. Ensuite, en accord avec nos attentes, nos résultats ont confirmé l’hypothèse de la répétabilité contextuelle, selon laquelle les performances de résolution sont constantes entre différentes tâches impliquant le même processus cognitif. Enfin, nous nous attendions à ce que les couples sociaux soient plus performants que les couples expérimentaux sur le test de résolution de problème coopérative, puisque nous les pensions davantage capables de synchronisation et de coordination comportementale. Or, nous avons trouvé des résultats suggérant le contraire, et nous avons constaté que la performance du couple au test de résolution de problème coopérative, était principalement expliquée par les performances de résolution individuelle de chaque partenaire. Des questions subsistent quant aux raisons pour lesquelles les couples expérimentaux ont mieux performé que les couples sociaux. Est-ce qu’une différence de motivation pourrait être à l’origine de ces résultats ? De futures études seront nécessaires pour nous éclairer. Malgré les questions en suspens, nos résultats contribuent à enrichir nos connaissances sur les capacités cognitives individuelles, la résolution de problème ainsi que la coopération. / Problem solving is often used as a tool for measuring individual cognitive abilities in non-human animals, since it enables us to study their operant learning and innovation capacities. Cognitive capacities differ between species and individuals, enabling them to maximize their abilities. For example, in birds and mammals, species with larger relative brains are said to have better social learning and innovation capacities, enabling them to adapt better to new environments. Despite the widespread use of zebra finches in behavioral ecology research, the impact of sex on problem-solving performance and the repeatability of performance have often been overlooked and, to our knowledge, no studies have investigated their performance in cooperative problem-solving contexts. In our study, we decided to address these different aspects to improve our understanding of cognitive abilities and cooperation in this model species. We tested male and female zebra finches in three individual problem-solving tests (string test, bar test and lid test) and one cooperative problem-solving test. For the latter, we separated the individuals into two groups: social pairs and experimental pairs, to manipulate the social bond between individuals. Social pairs were made up of a male and a female who had cohabited for 7 days, while experimental pairs were made up of a male and a female who had never met before. While we expected sex to have no impact on solving performance, our results revealed that, on average, males were faster at solving individual problem-solving tests. These results were mainly attributable to their performance on the second individual problem-solving test (bar test). We believe that this difference could be attributable to a higher level of persistence in males and/or the impact of sexual selection on cognitive performance, and that these aspects should be explored in future studies. Secondly, in line with our expectations, our results confirmed the contextual repeatability hypothesis, according to which solving performance is consistent across trials requiring the same cognitive process. Finally, we expected social pairs to perform better than experimental pairs on the cooperative problem-solving test, since we expected them to be more capable of synchronisation and behavioral coordination. . However, we found results suggesting the opposite, and the performance of the pairs on the cooperative problem-solving task was mainly explained by the ability of each partner to solve a problem. Questions remain as to why the experimental pairs performed better than the social pairs. Could a difference in motivation be behind these results? Future studies are needed to shed light on this. Despite the outstanding questions, we believe that our results contribute to our knowledge of individual cognitive abilities, problem solving and cooperation.