Better together
Groenewoud, Frank
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from
it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Groenewoud, F. (2018). Better together: Cooperative breeding under environmental heterogeneity.
University of Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Chapter 1
General introduction
INTRODUCTION
How I have enjoyed watching the simple life of this happy and affectionate family. Perfect harmony reigns between all; I am beginning to believe these birds incapable of a show of anger toward each other. Better such a life in the open fields, on a diet of cockroaches and grasshoppers, than life in a palace where the board groans under the cream and honey of the land, with the constant disagreements and bickerings which so often disfigure the con-duct of the wealthy. From: “Groove-billed Ani – some reflections on their family
rela-tions,” Journal, Vol. 5, September 22, 1930 – Alexander F. Skutch
The passage above was written in 1930 by the then twenty six year old Alexander Skutch (who died in 2004 – just eight days short of his one hundredth birthday), where he describes the pleasantly simple life of a group of groove-billed anis (Crotophaga sulcirostris). A few years later, Skutch would formalize these and other observations in an article where he coined the term “helpers at the nest”, which is still in use today to refer to particular types of coop-erative breeding (Skutch 1935). In the broadest sense, coopcoop-erative breeding is an umbrella term for any system where more than two individuals are engaged in raising offspring. Even without the aid of modern molecular techniques to confirm relatedness, and the use of banding to distinguish between individuals, Skutch appreciated by careful observations that different categories of “helpers at the nest” existed, and made the distinction between (i) juvenile helpers, which are retained offspring that provided alloparental care to their younger siblings, (ii) unmated sexually mature helpers, which cannot breed themselves due to a lack of mates or breeding vacancies, and decide to help others, and (iii), mutual helpers, which are breeding birds that assist each other in rearing their own respective fam-ilies. While many of the cooperatively breeding birds, fishes, mammals and insects known today could be assigned to one (or a combination) of these categories, our understanding of the environmental, social and genetic factors that drive these behaviours within- and be-tween species, on both proximate and ultimate levels, has improved considerably (Koenig & Dickinson 2016). However, we are still far from a general “theory of cooperative breeding”, as it has proven difficult to find the right balance between simple, generalizable hypotheses and predictions, and the number of systems to which these would apply. Many different reasons have contributed to this conundrum. First, within-species variation (i.e. between individuals and/or between populations) is the rule rather than the norm. Different sub-ordinates in the same species could be providing care for different reasons depending on potential future fitness benefits that may depend on sex, age, relatedness or body condition. Between-population variation poses a particular problem for studies that try to develop insights into the evolution of cooperative breeding using comparative methods. These ap-proaches usually resort to taking some kind of average trait value, or using data only from
a single population, leading to a loss of valuable information on the ecological drivers of cooperative breeding. Second, while cooperative breeding was initially studied in birds, it also occurs in mammals, insects and fishes, further complicating generalization due to vastly different modes of reproduction (i.e. oviparity vs viviparity), and life histories (e.g. growth patterns, number of offspring; Wilson 1971; Cockburn 1998; Clutton‐Brock 2006; Ta-borsky 2016). Last, cooperative breeding encompasses a whole range of different breeding and social systems (Cockburn 1998; Hatchwell 1999; Cockburn 2006; Riehl 2013), such as groups that consist of unrelated coalitions (e.g. dunnocks) and those that form (primarily) through the delayed dispersal of offspring (e.g. Seychelles warblers), and these systems like-ly had very different evolutionary origins (Ligon & Burt 2004; Clutton-Brock 2009; Wong et
al. 2012; Riehl 2013). Despite all these reservations, the field of cooperative breeding, and the
study of its genetic, social and ecological drivers has come a long way since its first descrip-tions by early observers.
Providing a comprehensive overview of the cooperative breeding literature over the past decades is beyond the scope of this introduction. Instead I would like to highlight some important contributions that focus on the ecological drivers of cooperative breeding, spe-cifically the environmental conditions that affect the costs and benefits of delayed dispersal and helping behaviours (i.e. alloparental care). After this more general background, I will briefly introduce my two study species: the Seychelles warbler (Acrocephalus sechellensis) and the cooperatively breeding cichlid Neolamprologus pulcher. Lastly, I will summarize the questions that will be the focus of the remaining chapters of my thesis.
Cooperative breeding: a two-step process
While various evolutionary routes to cooperative breeding have been proposed, most forms of cooperation occur in family groups (Cockburn 2006; Riehl 2013). Two likely explanations exist for why this is so. First, delayed or limited dispersal of offspring is one of the primary modes by which cooperative breeding groups form (Emlen 1982; Koenig et al. 1992; Griesser
et al. 2017). Second, kin selection – selection on genes through its effects on others
carry-ing the same gene (Hamilton 1963) – is an important evolutionary driver of cooperation (West-Eberhard 1975; Foster, Wenseleers & Ratnieks 2006; Bourke 2014). It should be noted that these two statements relate to different processes: group formation in the former, and the benefits of cooperation, or helping, in the latter. This distinction is why it is generally accepted that the evolution of cooperative breeding is best approached as a two-step pro-cess where delayed dispersal is a nepro-cessary, but not a sufficient condition for the evolution of cooperative breeding. While it is difficult to overstate the importance of kin selection theory in the last half century of research in the field of cooperation (West-Eberhard 1975; Bourke 2014), its importance for the evolution of cooperative breeding has been put into
question (Clutton-Brock 2002, 2009). This has generally not been because of flaws in the theory itself, but rather because most studies that invoked kin selection were correlation-al, and other (direct) benefits could therefore be in place (e.g. Wright 2007; Taborsky 2013; Kingma et al. 2014). Despite these criticisms, support for kin selection theory is strong, both within (e.g. Emlen & Wrege 1989; Komdeur 1994b; Richardson, Komdeur & Burke 2003b; Wright et al. 2010), and between species (Hughes et al. 2008; Cornwallis et al. 2010; Lukas & Clutton-Brock 2012; Dillard & Westneat 2016). Furthermore, empirical tests of the benefits of cooperative breeding in any particular species only apply to the selective forces that are currently maintaining cooperative breeding, but not necessarily to its evolutionary origins. Kin selection might therefore have been a necessary initial driver of cooperation, but once these conditions were set, other direct benefits may have evolved that were not necessarily dependent on genetic relatedness (Cockburn 2013). This thesis does not focus on the indi-rect (kin selected) or diindi-rect benefits of cooperative breeding per se, but rather on the eco-logical and environmental factors that shape the costs and benefits of dispersal and group formation, cooperative breeding and the stability of groups.
Environmental heterogeneity and the evolution of delayed dispersal
Delayed dispersal is often seen as a necessary first step for the evolution of cooperative breeding, at least in so called “helpers at the nest” type systems, where the majority of help-ers are the offspring of previous breeding attempts (Emlen 1982; Koenig et al. 1992; Riehl 2013). Such delayed dispersal does not necessarily lead to cooperative breeding, if there are no further benefits of helping. Consequently, many family living species exist, where subor-dinates do not show alloparental care (Griesser et al. 2017). Two main hypotheses have been proposed that make predictions regarding the environmental conditions under which de-layed dispersal should be the preferred strategy by offspring. The “ecological constraints” hypothesis suggests that constraints on dispersal and independent breeding reduce the fitness benefits of leaving the natal territory, so that offspring achieve higher fitness by postponing breeding, and potentially staying at home (Selander 1964; Brown 1974; Emlen 1982). This hypothesis predicts that constraints on independent reproduction could arise through stable environments leading to habitat saturation and thus a lack of available suit-able breeding opportunities, or alternatively, fluctuating environmental conditions lead-ing to high costs of dispersal or rearlead-ing young in bad years (Emlen 1982). Habitat saturation has been criticized as being an insufficient explanation for the evolution of cooperative breeding, because of the observation that many, if not most species, experience severe com-petition for suitable breeding positions, yet do not show delayed dispersal (Stacey & Ligon 1991; Koenig et al. 1992). The “benefits of philopatry” hypothesis argues that subordinate individuals should forego dispersal if the benefits in their resident territory (e.g. food, pro-tection) exceed the benefits of dispersal (Stacey & Ligon 1991). This hypothesis assumes that
variation in quality between territories is required to offset the cost of delayed reproduc-tion by increased fecundity or survival when a high-quality territory is acquired. Although it has generally been agreed that these two hypotheses are effectively two sides of the same coin, and differ only in their focus on the benefits of staying versus the costs of leaving (Emlen 1994), there are notable differences. First, if the ecological constraints hypothesis only includes the costs and benefits of independent reproduction, but not dispersal, off-spring should not necessarily remain philopatric. In fact, unless there are some additional benefits of philopatry, offspring could disperse and roam through the population without association to any group or territory, a strategy generally referred to as “floating” (Koenig et
al. 1992; Ridley, Raihani & Nelson-Flower 2008; Kingma et al. 2016a). In contrast, the benefits
of philopatry hypothesis explicitly emphasizes the benefits that can be gained on the natal territory, thereby excluding the possibility of floating as a viable option. However, most individuals will experience some combination of ecological constraints and benefits in the natal territory, potentially leading to selection for natal philopatry, and delayed reproduc-tion (Koenig et al. 1992; Emlen 1994; Komdeur 1992).
The ecological constraints and benefits of philopatry hypotheses also differ in whether they emphasize temporal or spatial variation in environmental conditions as the primary driver of delayed dispersal. The ecological constraints” hypothesis derives its predictions mainly from environmental fluctuations (or lack thereof) through time that determine the costs and benefits for offspring to disperse or breed independently. For example, the number of breeding vacancies might be higher following a year with high breeder mortality, and years with high food availability might lead to lower costs of dispersal or independent breeding. In both examples, temporal variation in environmental conditions lead to decreased costs, or increased benefits, of dispersal. In contrast, when there is no variation in quality between territories, offspring have no prospect of obtaining a higher quality territory by initially delaying dispersal. Thus, the benefits of philopatry hypothesis requires consistent spatial variation in order for delayed dispersal to evolve.
Both the ecological constraints and the benefits of philopatry hypothesis emphasize the rel-ative costs and benefits of dispersal vs natal philopatry for subordinates, but do not include the costs and benefits of these subordinates for the rest of the group. These can be especial-ly important when (i) individuals have an effect on the fitness of other group members, and (ii) when a few group members (e.g. dominants) can control group membership (e.g. through eviction). While not further explored here, recent studies have taken such effects into account and have consequently provided important insights into the conditions un-der which groups form (Shen et al. 2017).
Environmental heterogeneity and the benefits of helping
In order for natal subordinates to be able to provide alloparental care, they have to be (i) present during a breeding attempt and (ii) capable of providing alloparental care. Obvi-ously, there are large differences in rates of maturation, and inter-reproductive intervals between different species and taxa that could affect these two conditions. For instance in the facultative eusocial hover wasp Liostenogaster flavolineata, nesting is year-round due to a lack of seasonality, and newly emerged females are almost immediately capable of provid-ing alloparental care. Conversely, white-wprovid-inged choughs (Corcorax melanorhamphos) only breed during the breeding season, which is a yearly occurrence. This necessitates offspring to remain with their parents for at least a year to be able to provide help. Subordinates can provide a wide range of alloparental care behaviours that are mostly an extension of the types of parental care provided by breeding pair. Help can consist of providing food to de-pendent young, egg-cleaning, incubation, nest-building, territory defense and territory maintenance (Brown 1987; Heinsohn & Legge 1999; Taborsky 2016).
Helping is not a necessary consequence of delayed dispersal: offspring could delay dis-persal and remain in the natal group, but not provide help (Drobniak et al. 2015; Griesser
et al. 2017). Thus, there need to be additional benefits that select for helping behaviour by
subordinates, and these benefits – which can be broadly categorized into indirect benefits (i.e. benefits accrued through kin selection), and direct benefits – have been extensively dis-cussed and reviewed (e.g. Cockburn 1998; Clutton-Brock 2002; Bergmüller et al. 2007; King-ma et al. 2014). The costs and benefits of subordinate help can be affected by abiotic and biotic conditions. For instance, when nestling starvation is the main cause of reproductive failure, subordinates in cooperatively breeding birds are expected to have a larger impact on reproductive success when food conditions are poor than when these are high (i.e. the “hard life” hypothesis; Koenig, Walters & Haydock 2011). The main reasoning behind this is that when conditions are good and food availability is high, the additional food provisioned by helpers is less valuable than when food conditions are poor and additional food might mean the difference between reproductive failure and success. A similar argument could be made with regards to any other ecological pressure that affects the reproductive success or survival or groups that can be modulated through subordinate help, such as reducing the risk of predation. One additional benefit of improving reproductive success under harsh environmental conditions is that offspring produced under such conditions are more valu-able, because fewer total offspring are produced in such years, and relative contribution to population growth (i.e. reproductive value; Fisher 1930; Taylor 1990) is therefore higher. In addition, there are other benefits of reducing variance in reproductive success that is in-duced by temporal variation in environmental conditions. Such strategies (i.e. rein-duced fe-cundity variance at the expense of mean fefe-cundity) are generally referred as “bet-hedging”
strategies (Gillespie 1977; Lehmann & Balloux 2007; Starrfelt & Kokko 2012). Several studies have recently suggested that cooperative breeding might similarly be a bet-hedging strat-egy that buffers against fecundity variance induced by temporal fluctuations in environ-mental conditions (Rubenstein 2011; Koenig & Walters 2015). Further support for coopera-tive breeding as a bet-hedging strategy comes from comparacoopera-tive studies that show that the occurrence of cooperative breeding is positively associated with climatic uncertainty and variability (Jetz & Rubenstein 2011; Cornwallis et al. 2017; Lukas & Clutton-Brock 2017). Thus, while it has been suggested that the benefits of cooperative breeding can depend on (fluc-tuations in) the environment, how and under what conditions subordinates in cooperative groups can improve reproductive success has received little attention.
The suppression of within-group conflict
Living in social groups provides many benefits for group members, but also comes with costs (Alexander 1974). One of the major costs of group living concerns conflict over the dis-tribution of limited resources (i.e. food, reproduction) between members of these groups. Such conflict leads to behaviours that aim to secure a larger share of resources, thereby negatively affecting the stability of groups and potentially negating the benefits of social-ity (Shen, Akçay & Rubenstein 2014). Investigating the factors that reduce conflict is thus important for understanding the evolutionary stability of groups (West et al. 2015). One important way by which conflict could be reduced between members of a social group, is when environmental conditions improve the fitness benefits of group living relative to leaving the group. These conditions should select for individuals refraining from using ag-gression to obtain a larger share of resources, because in doing so, they are reducing the fitness benefits they accrue through other group members (Alexander 1974; Brown 1982; Shen et al. 2014). While predation risk is often invoked as an important factor selecting for group-living (Inman & Krebs 1987; Krause & Ruxton 2002), its role in the evolution of more complex social organisation, such as that of cooperative breeders, has been largely neglect-ed. This is surprising, because predation risk can have substantial consequences for (i) off-spring survival, which might require cooperation between individuals and (ii) the costs of dispersal and independent reproduction, selecting for limited dispersal. Predation risk can therefore influence both the benefits of group-living and cooperation, as well as the costs of leaving the group and breeding independently. Consequently, increased levels of preda-tion risk should minimize the willingness of individuals to obtain a larger share of group resources at the expense of others, and reduce conflict between group members.
Ecological factors such as predation risk and food availability can have important conse-quences for group formation through delayed dispersal, the benefits of cooperation and levels of within-group conflict, all important aspects in the evolution of complex social
sys-tems and cooperative breeding. Investigating how these ecological factors shape the social systems and behaviours of cooperative breeders can therefore provide insights into the pro-cesses that underlie transitions from simple to complex social organization. In the coming paragraphs, I will introduce the species that I will use throughout the rest of my thesis to address these questions, the cooperatively breeding Seychelles warbler Acrocephalus
sechel-lensis and the social cichlid Neolamprologus pulcher.
Island life
The Seychelles warbler is a small (13-19 g) facultative cooperatively breeding passerine en-demic to several islands in the Seychelles archipelago (Fig. 1.1A; Komdeur et al. 2016, 2017). By the 1960’s, the last decimated population of Seychelles warblers (26-50 individuals) was confined to the island of Cousin (ca 29 ha; 04º20’S, 55º40’E). Habitat restorations and sev-eral translocations to other islands have restored the population to viable numbers and have provided a unique opportunity to study the cooperative breeding system of this once critically endangered species. Most Seychelles warblers on the island of Cousin (Fig. 1.1B), which is our main study population, live in pairs (ca 60% of all territories), but a proportion lives in groups consisting a dominant breeding pair and 1-5 subordinates (mean ± SE = 0.59 ± 0.02; 1996-2016; see chapter 3) of either sex. These helpers are usually, but not always the offspring of the dominant breeding pair (Kingma et al. 2016a; see chapter 3). Territories are defended year-round and breeding pairs often remain pair-bonded on the same territory throughout their lives. Seychelles warblers on Cousin Island typically produce single egg clutches, but around 13% of clutches contain 2-3 eggs (Richardson et al. 2001). Offspring can remain nutritionally dependent on their parents and other group members for a period of up to three months, which is extremely long for a passerine species.
The population of Seychelles warblers on Cousin Island is contained, with virtually no mi-gration on or off the island (Komdeur et al. 2004a). Additionally, annual resighting prob-abilities of birds on the island are extremely high: up to 0.98 for adult birds, and 0.92 for younger individuals (Brouwer et al. 2010). This combination makes the Seychelles warbler an excellent system to study the factors associated with natal dispersal, because these are not confounded by individuals dispersing from the study site. Seychelles warblers are strictly insectivorous, taking most of their arthropod prey from the underside of leaves (Komdeur 1991). They are thus highly dependent on the relatively short peaks in abundance of these prey following monsoon rains, which occur twice per year (Komdeur & Daan 2005). Reproduction therefore mostly occurs following such rains, during June-September (major breeding season), or January-March (minor breeding season).
FIGURE 1.1 An adult Seychelles warbler (A) and a bird’s eye view of our main study site – Cousin Island – with our research
sta-tion in the foreground and the neighbouring island of Cousine in the background (B). Photo (A) by Sjouke A. Kingma and (B) by Martin Harvey, courtesy of Nature Seychelles.
A
Detailed measurements of the abundance of prey have been collected since the start of the study on Cousin Island, and earlier studies have indicated that between-territory variation plays an important role in group formation, with offspring being more likely to delay dis-persal on high quality than on low quality territories (Komdeur 1992). However, there is substantial between-year variation in arthropod abundance, which could have consequenc-es for delayed dispersal and other aspects of cooperative breeding, which were until recent-ly unknown (see chapter 3).
Seychelles warblers on Cousin suffer virtually no adult predation, but predation of eggs by the endemic Seychelles fody Foudia sechellarum is an important cause of nest failure (Kom-deur & Kats 1999). However, several aspects of the Seychelles warbler breeding biology re-duce the risk of nest predation. First, (dominant) males perform nest guarding behaviour (i.e. remaining vigilant close to the nest during female off-bouts). Such behaviour reduces the likelihood of egg predation because eggs are only taken from unprotected nests. Nest guarding is probably costly for males since male investment into nest guarding behaviour increases with territory quality (see chapter 6; Komdeur & Kats 1999). Second, incubation by subordinate females reduces the time that the nest is unprotected, and should therefore also reduce the risk of nest predation by fodies. Third, Seychelles warblers will attack and chase off any fodies that come to close to their nest. However, several aspects of Seychelles warbler behaviour with regards to nest predation and the benefits of cooperative breeding remain elusive. For instance, the extent to which subordinate dispersal, dominant toler-ance of subordinates and co-breeding by subordinate females is associated with the risk of nest predation was hitherto unknown. Additionally, we knew very little about the tradeoffs that determine anti-predator behaviours for Seychelles warbler parents. These have now been addressed in box 1 and chapter 6, respectively.
On the southern shores of Lake Tanganyika
Neolamprologus pulcher is a highly social cooperatively breeding cichlid fish (Fig. 1.2A),
en-demic to Lake Tanganyika (Fig. 1.2B), where it inhabits the sublittoral zone between 2-40 meters depth (Konings 1998). It lives in groups that consist of a dominant pair and up to 25 subordinates of either sex (Taborsky & Limberger 1981; Groenewoud et al. 2016). Territories are centred on some kind of substrate, usually one or several small rocks, but fish have also been seen to use shells or crevices, which they use for both shelter and breeding (Josi et al. in prep.). Subordinates engage in various types of helping behaviours, including territory maintenance (i.e. removing sand and debris to create shelters), egg cleaning and fanning, defense against con- and heterospecifics space competitors and predators (Taborsky 1984).
FIGURE 1.2 A group of Neolamprologus pulcher in their territory (A) and the sunset over Lake Tanganyika as seen from the field
station at Kasakalawe bay (B). Photo (A) by Dario Josi and (B) by Arne Jungwirth.
A
Subordinates are of different sizes and show age-related task specialization: smaller, sexu-ally immature subordinates ususexu-ally engage in maintenance and defence against non-dan-gerous predators and space competitors, and large sexually mature subordinates defend against larger predators (Bruintjes & Taborsky 2011; chapter 5). Territories are aggregated into colonies, which are made up of several dozen, to several hundreds of territories (Heg
et al. 2008).
The evolution of cooperative breeding in N. pulcher – and other cooperative breeding fish – is likely distinct from the evolution of cooperative breeding in most other vertebrate systems, in that it is not based on habitat saturation or increased levels of within-group relatedness.
N. pulcher does not suffer from habitat saturation in the strict sense, since breeding
sub-strate appears to be plentiful. However, most of this free habitat occurs outside of the boundaries of the colony, and individuals seem unwilling to disperse to such habitat. When similar habitat was experimentally made available inside the colony, fish usually dispersed there within a few days (Heg et al. 2008). One of the main obstacles to dispersal, and the main drivers of delayed dispersal, is the risk of predation by large predatory cichlids such as Lepidiolamprologus elongatus and L. attenuatus. These fish are highly mobile predators, swimming through N. pulcher colonies in small groups (often together with Mastacembelid eels). Individuals that do not belong to a group and are devoid of protection, are easy prey, and several studies have shown that increased predation risk leads to delayed dispersal by larger, sexually mature subordinates (Heg et al. 2004a; see chapter 2). As a result of frequent dominant turn-over, subordinate-dominant relatedness decreases with increasing subor-dinate size (Dierkes et al. 2005). Larger, sexually mature helpers are therefore mostly not related to the dominants in whose territory they reside and to whom they provide help. As group membership is a prerequisite for survival and large subordinates impose costs on dominant breeders, large helpers have to compensate for these costs by providing help (i.e. “pay-to-stay”; Gaston 1978; Fischer et al. 2014).
The role of predation risk in the evolution of cooperative breeding, and other forms of so-cial complexity, have generally been understudied. The cooperative breeding system of N.
pulcher offers a unique opportunity to study such effects because fish can be easily observed
in the wild and there is high natural variation in predation risk (see chapter 2). The broader implications of predation risk on the structural complexity of social groups and the extent to which individuals show behavioural adjustments to increase the benefits of cooperative breeding under high predation risk were previously unexplored – but have now been ad-dressed in chapter 2 and 5, respectively.
Environmental heterogeneity and the evolution of cooperative breeding
Developing a hypothesis based on well-supported assumptions is the beginning of any in-vestigation. I hope I have been able to provide a strong argument for the hypothesis that the ecology of cooperative breeders is an important factor determining the costs and bene-fits of (delayed) dispersal, helping behaviour and behavioural strategies that maximize the benefits of cooperation. In the coming chapters, I will further explore these questions. In the second chapter, I will explore the consequences of predation risk for dispersal and the so-cial organization of N. pulcher, and discuss the general implications of our results for other cooperative breeding systems. In the third chapter, I will show that spatio temporal variation in food availability has important consequences for group formation in the Seychelles war-bler, but that the benefits that are obtained through cooperative breeding are not affected by such variation. In the fourth chapter, I will presents the results of an investigation into al-ternative dispersal strategies in the Seychelles warbler, and show that subordinate females can disperse into unrelated group to obtain reproductive benefits. In the fifth chapter, I will return to N. pulcher to show how groups reduce within-groups conflict to maximize the ben-efits of cooperative breeding under high predation risk, with wider implications for evo-lutionary transitions to higher social complexity. In my sixth chapter, I will present results of a field experimental study showing the social and environmental factors that mediate tradeoffs in anti-predator defenses in the Seychelles warbler. Box A contains an investiga-tion into the extent to which the risk of nest predainvestiga-tion drives the benefits of communal breeding for females in the Seychelles warbler. In the final chapter, I will present a synthesis the results of this thesis and their implications, and offer some perspectives for future stud-ies investigating the ecological and environmental drivers of cooperative breeding.
