Conflict and co-adaptation : the evolution of parental care in a wild bird species.
22 January 2016
Campus Drie Eiken, Promotiezaal Q0.02 - Universiteitsplein 1 - 2610 Antwerpen-Wilrijk
Organization / co-organization:
Department of Biology
Wendt Müller & Marcel Eens
PhD defense Carsten Lucass - Department of Biology
Parental care is the transfer of resources to or protection of offspring by the mother, the father, or both (= biparental). It is a widespread behaviour in the animal kingdom as it increases parental fitness through improved offspring condition and survival. However, parental care is associated with costs for providing parents, rendering care an investment. As parents and their offspring are not perfectly related, both parties rate these costs differently. In short, offspring seek more care than parents are selected to provide, which leads to a conflict between both parties (parent-offspring conflict). Post-natal parental care further requires communication between offspring, begging for food, and parents, providing care in response. The expression of each trait depends – because of the reciprocal interaction – not only on an individual’s own genotype (direct genetic effects) but also on the genotype of the individuals it interacts with (indirect genetic effects). This form of social selection is interesting as not only the individual but also its (social) environment can evolve itself, as both contain genes. Hence, the behavioural traits involved are simultaneously agents and targets of selection and thus follow complex evolutionary trajectories.
One consequence of this reciprocal interplay is that both traits are expected to co-evolve, which has indeed been shown both theoretically as well as empirically. But most previous empirical studies were performed in captivity, where relevant selection pressures, that act under natural conditions (e.g. costs of foraging, predators, parasites), are not present. This renders conclusions on the adaptive significance difficult. Furthermore, the interplay between both parties in the context of parental care is highly dynamic. Parents feed according to changes in offspring begging and offspring change their begging according to changes in supply and the consequent changes in their hunger level. These dynamics are adequately captured using behavioural reaction norms (= responsiveness), that is the change of a behaviour over an environmental gradient. However, most previous studies used a rather static approach, that is mean levels of behaviour. In this thesis I therefore aimed to study parent-offspring co-adaptation (on the reaction norm level) in a wild bird species, the blue tit (Cyanistes caeruleus), and how this is shaped by environmental factors.
To this end, I calculated the reaction norm of parental provisioning as the difference between an individual’s provisioning rate towards an experimentally enlarged brood (mimicking increased demand) and an experimentally reduced brood (mimicking decreased demand). Similarly, the begging reaction norm was estimated as the change of begging to changes in hunger (i.e. subtracting begging scores of 60 minutes of food deprivation from that of 120 minutes). Following this behavioural reaction norm approach, I indeed found co-variation on the phenotypic level (chapter 2). The reaction norm of parental provisioning of blue tit fathers, but not mothers, was negatively linked to the offspring begging reaction norm. This interesting sex-specific pattern suggests that fathers that are relatively unresponsive to changes in offspring demand have offspring that strongly increase begging with higher hunger levels. The low responsiveness in males may thus have evolved to prevent exploitation. However, it remained unclear why this relationship is absent in females. A potential explanation why this pattern is found in males but not females, may be the fact that changes in brood size are likely to affect not only parental provisioning, but also other care traits such as nest sanitation behaviour, a trait exclusively expressed in females. Thus, mothers may have to trade-off nest sanitation against providing food, which may have masked the relationship in females.
When I estimated the provisioning responsiveness of parents to offspring begging via sequenced begging playback, I found that the responsiveness of both parents was positively linked to begging behaviour of their offspring that was raised by foster parents (chapter 3). But this became only apparent after correcting for exposure time to playback, because mothers spend more time in the nest, thus experienced the treatment more often. The positive correlation suggests that parents control supply and it is offspring that have to adjust begging to variation in provisioning. Parental control occurs more likely, when a species breeds more than once (as parents are expected to defend residual resources for future offspring) or when environmental conditions (e.g. food abundance) are unfavourable, thus setting limitations for foraging. When environmental conditions set limitations to parental supply, offspring remain relatively powerless to increase parental supply via begging. Indeed, in the year of study, mothers postponed egg laying, incubation and thus hatching of nestlings to later dates, suggesting unfavourable environmental conditions. However, whether parents or offspring (via begging) are in control of supply remains speculative as we still lack theoretical predictions on the co-evolution of behavioural reaction norms.
In a next step I investigated the costs that are thought to stabilize co-adaptation. When parental and offspring traits are mismatched (e.g. after cross-fostering clutches), it may be highly demanding offspring that suffer impaired growth when their costly begging remains (partly) unrewarded by low providing parents. Alternatively, parents may suffer reduced survival or lower future reproductive performance when responding to a brood that is more demanding than actual genetic offspring. By reciprocally cross-fostering clutches pre-hatching into a non-genetic family environment, I found that offspring growth depended not only on the foster parental phenotype, that is the rate of provisioning, but also on the offspring’s own phenotype, that is its begging intensity (chapter 4). Highly demanding offspring reached lowest fledgling mass when raised by low providing parents (due to partly unrewarded costly begging) but highest fledgling mass when raised by high providers. Fledgling mass of less demanding offspring was intermediate, both, when raised by high or low providing parents. Further, it appears that mothers do not pay fitness costs in terms of future reproduction success when mismatched to offspring. Although I was as yet not able to analyse whether fathers pay costs, it seems likely that blue tit parents have the upper hand (but probably not full control) over provisioning because costs are (largely) paid by the party that is not in control of provisioning.
As mentioned above parental care relies on parent-offspring communication. Offspring share private information about their nutritional status via vocal and visual begging displays, to which their parents respond by providing food. However, as noise is ever-increasing in extent, duration and intensity in modern anthropogenic environments, this may impair vocal communication (and ultimately alter co-evolutionary dynamics), which I investigated in another experiment (chapter 5). Playback of anthropogenic highway noise increased the likelihood that offspring fail to respond to the returning parent. Parents underestimated begging probably as a consequence of lower begging levels and decreased their provisioning. This suggests that anthropogenic noise may have a substantial effect on offspring development. As a consequence, the relative importance of vocal begging in noisy environments is likely to decrease over evolutionary time scales, and parents may have to rely more on other, e.g. visual, begging cues, that are not masked by noise.
From a mechanistic point of view, co-variation of parental and offspring traits could be due to a genetic correlation or driven by maternal effects (that is variation in offspring phenotype as a consequence of the maternal phenotype, Roff 1998). In particular pre-natal maternal effects may fine-tune offspring phenotype to parental capacity. The latter may be reflected by the expression of ornamental traits of parents that are thought to signal individual quality. Investigating a potential relationship between reproductive investment and trait expression of an individual parent, I found that maternal, but not paternal carotenoid-based breast colouration was positively linked to begging intensity of genetic offspring (chapter 6). I hypothesized that this could be achieved via maternal deposition of carotenoids into egg yolk. Interestingly, these maternally mediated changes in offspring phenotype (here begging) are only translated into growth benefits, when offspring were raised by their own parents, again suggesting a cost for offspring in this case when mismatched with their mothers. However, all other (negative) results supported the current stand of evidence that structural or pigment-based plumage colouration of blue tits are only inconsistently correlated with central life-history traits.
Thus, in a series of complementary experiments in a wild bird species, the blue tit, I could show that it is highly relevant to introduce the recently developed behavioural reaction norm concept to the study of parent-offspring co-adaptation, and that parent-offspring communication is shaped by environmental conditions – also beyond the family environment itself. The latter highlights the importance of experimental studies in the wild, where natural selection is acting on (experimentally manipulated) traits. But these experiments also highlight that further progress has to be made, in terms of conceptual theoretical framework, underlying mechanisms, as well as functional consequences in order to understand the (co-) evolutionary dynamics of parental provisioning and offspring begging.