Individual variation in behaviour

Mouchet, Alexia (2021): Individual variation in behaviour: adaptive mechanisms and ecological drivers. Dissertation, LMU München: Fakultät für Biologie

Abstract

Wild animals show remarkable phenotypic variation despite natural selection eroding it. Phenotypic variation within populations is intriguing because all individuals are expected to be adapted to the same environmental conditions, and thus, to present similar phenotypic traits. However, when repeatedly measured, individuals have been observed to differ in the average expression of various behaviours across time and contexts. Consistent among-individual variation (called “animal personality”) has been proposed to be adaptively maintained if the fitness costs and benefits of behaviour vary with the environment or other phenotypic traits. Theory postulates that two key adaptive mechanisms could play a role: life-history trade-offs and spatiotemporal variation in selection (or heterogeneous selection). Empirical tests of the role of these mechanisms in the maintenance of individual variation in behaviour remain scarce and findings are ambivalent. My PhD thesis aimed at shedding light on the mechanisms allowing the persistence of animal personalities, thereby advancing our understanding of how animals adapt to variable environments. I investigated the role of life-history trade-offs and heterogeneous selection in the coexistence of alternative personalities in the wild. I also examined potential ecological drivers of heterogeneous selection. I used a passerine bird breeding in the wild in nest boxes (the great tit Parus major) as model. Individuals must trade-off investment among various phenotypic traits because they have limited amount of energy and time to acquire resources, grow and reproduce. The optimal resolution of trade-offs may depend on ecological conditions and/or the phenotypic traits of the individuals. Individuals differing in their behavioural phenotypes may thus resolve trade-offs differently. In Chapter 1, my colleagues and I tested this hypothesis by focusing on the trade-off between current reproduction and reproductive senescence. Specifically, we asked whether behavioural phenotypes differed in patterns of senescence. We found that faster explorers increased and subsequently decreased their reproductive investment with age. This finding suggests that faster explorers reproductively senesced later in life. By contrast, slower explorers laid similar clutch sizes through their lifetime; that is, they did not show reproductive senescence. Different behavioural phenotypes, thus, resolved the trade-off between current reproduction and reproductive senescence differently, which may allow them to coexist. Spatial and temporal variation in the environment may cause natural selection to favour different phenotypes in different environments. Spatial variation in selection may maintain phenotypic variation across environments, whereas temporal variation in selection (or fluctuating selection) may maintain phenotypic variation within environments. Though these processes co-occur and may have counteracting effects on phenotypic variation, both processes have rarely been investigated simultaneously. The relative importance of spatial and temporal variation in selection, and thus, the evolutionary potential of phenotypic traits under heterogeneous selection, remains unexplored. In Chapter 2, I studied heterogeneous selection on behaviour within and among great tit populations. To this aim, I gathered longitudinal data from five West European wild great tit populations breeding in nest boxes. In all these populations, behaviour was assayed with the same experimental design. Selection on behaviour varied primarily spatially. Temporal variation in selection was also important. The existence of phenotypic variation in all populations suggests that temporal variation played a key role in counteracting local adaption promoted by spatial variation. Temporal variation in selection was population-specific, which suggests that local ecological conditions also played a role in the evolution of phenotypic variation. This study thereby demonstrated the importance of considering both large- and small-scale geographical and temporal variation to understand the ecological mechanisms maintaining variation in animal behaviour. Previous studies found that variation in the social environment induced by variation in population density caused selection on behaviour to vary. However, we did not find such evidence in great tit populations. Another ecological factor that varies ubiquitously and that is crucial for survival and reproduction is food availability. Food availability also generally positively correlates with population density. Therefore, the effects of population density on fitness may be indirect through food availability. Variation in food availability may cause selection pressures on behaviour to vary because behavioural phenotypes differ in competitive abilities and foraging tactics. In Chapter 3, I studied whether winter food availability drove heterogeneous selection on activity in a novel environment. I experimentally manipulated food abundance outside the breeding season by providing supplementary food in multiple great tit nest box plots. Against expectations, I did not find evidence for fecundity selection on behaviour to vary with the experimental manipulation of food availability. Food availability may drive variation in fecundity selection but simultaneous changes in breeding density may counteract its action. Food- and density-dependent selection on behaviour need to be estimated simultaneously to disentangle their effects. Interestingly, on average, individuals were more active in high than in low food availability context. Moreover, high food availability context increased behavioural variation among individuals. These findings suggest greater plasticity and/or higher survival, recruitment or immigration rate of more active individuals. Future studies should investigate whether viability rather than fecundity selection vary with food availability. In the different projects of this PhD work, I focused on behaviour scored in different “novel environments”, which are all generally labelled “exploration behaviour”. However, “exploration behaviour” was not assayed with the same experimental design in Chapter 2 compared to Chapter 1 and 3. In Chapter 1 and 3, behaviour was assayed in the field in a portable cage. In Chapter 2, behaviour was assayed in a standardized laboratory room. We assumed that birds expressed the same behaviour in both assays because laboratory- and field-based behaviours have been shown to each correlate with other field-based behaviours. In Chapter 4, I tested this assumption and found that laboratory- and field-based behaviour did not correlate. Both assays may present different contexts to the birds, which elicited the expression of different behaviours. I also showed that the population sampled for the laboratory test was biased toward fast explorers. This study highlights the difficulty assaying behaviour in an unbiased and reproducible manner. It is therefore important to cross-validate behavioural assays before making biological assumptions. Overall, this PhD thesis contributed to understanding the role of adaptive mechanisms in individual variation in behaviour and their ecological drivers. This work showed that behavioural phenotypes contribute differently to population dynamics and should thus be considered in ecological and evolutionary studies. This work also exemplified the importance of long-term and collaborative projects. For a comprehensive understanding of phenotypic variation, the next challenge would be to simultaneously consider multiple traits, ecological factors and species that all interact through eco-evolutionary dynamics. Such integrative studies will embrace the complexity of ecological interactions and allow us to better understand how populations adapt to variable environments.

• Dokumententyp: Dissertationen (Dissertation, LMU München)
• Themengebiete: 500 Naturwissenschaften und Mathematik ; 500 Naturwissenschaften und Mathematik > 570 Biowissenschaften, Biologie
• Fakultäten: Fakultät für Biologie
• Sprache der Hochschulschrift: Englisch
• Datum der mündlichen Prüfung: 2. November 2021
• 1. Bericht­erstatter:in: Dingemanse, Niels