Ruiz-Gomez, Maria De Lourdes
Decoupling aggression and risk-taking: patterns of variation in two species of freshwater fish.
PhD thesis, University of Glasgow.
Full text available as:
An extensive literature has documented the existence of suites of correlated behavioural traits (called behavioural syndromes) in a range of vertebrate species, as well as in some invertebrates. The existence and persistence of such behavioural syndromes is of both fundamental and applied interest and the main aim of the work described in this thesis was to examine sources of individual variation in risk-taking and aggression, as well as the circumstances under which those behaviours could be uncoupled. The study used two species of freshwater fish that have become something of a model to study behaviour: the three spine stickleback (Gasterosteus aculeatus) and the rainbow trout (Oncorhynchus mykiss).
In chapter 1 I give a background on the current research on individual variability in behaviour, behavioural syndromes and coping strategies, with special reference to fish; as well as considering the implications of behavioural syndromes for evolutionary biology and aquaculture. Chapter 2 describes a long-term experiment on boldness, aggression and the relationship between them in sticklebacks that grew at different rates under two different competitive regimes. In one treatment (the low interaction condition) food was dispersed, while in the other (the high interaction condition), food was clumped. Fish were fed to excess in both treatments. Analysis of the relationship between risk-taking and some morphological variables showed that, in general, shy fish were heavier and longer than both bold and behaviourally intermediate fish, independently of their body condition. Fish from the low interaction condition were more aggressive than those from the high interaction feeding regime. Boldness and aggression were positively associated only in the fast growing fish from the high interaction competitive regime.
Limited evidence suggests that individual personalities may influence reproductive success and other fitness-related traits in complex and context-specific ways. In the study described in Chapter 3, I used an indirect approach to relate fitness to personalities in sticklebacks. Specifically, I related hatching date of fry (used as an indirect measure of parental fitness) to their personalities (boldness and aggression). Individuals that hatched early were bolder than late hatched fish, whereas most of the shy individuals were found among the late bred fish. There were no detectable differences in aggression between early and late hatched fish, but there was a relationship between boldness and aggression independent of hatching date.
In chapters 4 and 5, I describe studies of rainbow trout from two lines selected for breeding for low (LR) or high (HR) post-stress plasma cortisol response that have become something of a model system for studies of coping strategies in fish. In addition to striking differences in cortisol responsiveness, LR and HR fish show patterns of brain biochemistry, risk-taking and aggression that are typical of so-called proactive and reactive animals respectively.
The results reported in chapter 5 strengthen this interpretation, by comparing behavioural flexibility and response to novelty in 3rd generation LR and HR rainbow trout. After being trained individually to find food in one arm of a T-maze, HR fish were able to found food strikingly faster than LR trout when the resource was moved to a different position. In contrast, LR fish were much less distracted by the presence of an unfamiliar object. Previous studies have shown that proactive animals develop and follow routines more strictly than do reactive animals, while the latter are more aware of changes in their environment. My results therefore give further support to the characterisation of LR and HR rainbow trout as showing proactive and reactive coping strategies.
In chapter 4 however, I complicate this interpretation by showing that the relationship between boldness and aggression is flexible. Following transport from the UK to Norway, HR and LR fish switched behavioural profiles. In contrast to the results of previous studies, HR fish fed sooner in a novel environment and became dominant over LR fish in pairwise aggressive interactions. One year after transport, HR fish still fed sooner than LR fish, but no difference in social dominance was found. Among offspring of transported fish, no differences in feeding rates were observed, but as in pre-transported 3rd generation fish, HR fish lost fights for social dominance against size matched LR opponents. Transported fish and their offspring retained their distinctive physiological profile throughout the study, with HR fish showing consistently higher post-stress cortisol levels at all sampling points. Therefore the striking difference in cortisol responsiveness in these two strains of trout is on its own not sufficient to maintain distinct behavioural phenotypes.
The work described in this thesis therefore extends current understanding of individual variability in behaviour and of behavioural syndromes by identifying circumstances under which risk-taking and aggression are uncoupled in two species of freshwater fish. It also suggests some potential consequences for fitness. In chapter 6 these results are discussed on the light of current research on animal personalities, behavioural syndromes, coping strategies and their implications for evolutionary biology. Particular reference is made to the existing literature on fish and the implications of those findings for aquaculture are also discussed.
Actions (login required)