THE LINK BETWEEN BRAIN SIZE, COGNITIVE ABILITY, MATECHOICE AND SEXUAL BEHAVIOUR IN THE GUPPY (POECILIARETICULATA)

Alberto Corral López

The link between brain size, cognitiveability, mate choice and sexualbehaviour in the guppy (Poeciliareticulata)Alberto Corral López

©Alberto Corral López, Stockholm University 2017 ISBN print 978-91-7797-045-3ISBN PDF 978-91-7797-046-0 Cover artwork by Alba Cortázar Chinarro Printed in Sweden by Universitetsservice US-AB, Stockholm 2017Distributor: Department of Zoology

A María, Axel, mis padres ymi familia

The thesis is based on the following articles, which are referred to in the text by their Roman numerals:

I Corral-López, A., Bloch, N. I., Kotrschal, A., van der Bijl, W.,

Buechel, S. D., Mank, J. E. & Kolm, N. (2017). Female brain size affects the assessment of male attractiveness during mate choice. Science Advances, 3(3), e1601990.

II Corral-López, A., Kotrschal, A. & Kolm, N. (2017). Brain size af-fects the judgement of female quality during male mate choice. – Manuscript.

III Corral-López, A., Garate-Olaizola, M., Buechel, S. D., Kolm, N. § & Kotrschal, A. § (2017). On the role of body size, brain size and eye size in visual acuity. – Behavioral Ecology and Sociobiology. Provi-sionally accepted. § shared senior authorship

IVV Corral-López, A., Eckerström-Liedholm, S., van der Bijl, W. Kotrschal, A. & Kolm, N. (2015). No association between brain size and male sexual behavior in the guppy. Current Zoology 61, 265-273.

V Corral-López, A., Romensky, M., Kotrschal, A., Buechel, S. D. & Kolm, N. (2017). Brain size, environmental complexity and mating behaviour. – Manuscript.

Candidate’s contribution to the articles in this thesis*

I II III IV V

Conceived the study Substantial Substantial Significant Substantial Substantial

Designed the study Substantial Substantial Significant Substantial Substantial

Collected the data Substantial Substantial Significant Significant Substantial

Analysed the data Significant Substantial Substantial Substantial Substantial

Manuscript preparation Substantial Substantial Substantial Substantial Substantial

* Contribution explanation Minor: contributed in some way, but contribution was limited. Significant: provided a significant contribution to the work. Substantial: took the lead role and performed the majority of the work.

I am also a co-author in the following articles that were written during the course of my doctoral studies, but are not included in this thesis:

Bloch, N., Corral-López, A., Buechel, S. D., Kotrschal, A., Kolm, N. § & Mank, J. E. § Transcriptional networks reveal early neurogenomic response underlying variation in female preferences – Submitted Manuscript.

§ shared senior authorship

Outomuro, D., Angel-Giraldo, P. Corral-López, A. & Realpe, E. (2016). Multitrait aposematic signal in Batesian mimicry. Evolution 70, 1596-1608.

Kotrschal, A., Corral-López, A., Zajitschek, S., Immler, S., Maklakov, A. A. & Kolm, N. (2015). Positive genetic correlation between brain size and sexual traits in male guppies artificially selected for brain size. Journal of Evolutionary Biology 28, 841-850.

Kotrschal, A., Corral-López, A., Amcoff, M. & Kolm, N. (2015). A larger brain confers a benefit in a spatial mate search learning task in male guppies. Behavioral Ecology 26, 527-532.

Kotrschal, A., Buechel, S. D., Zala, S. M., Corral-López, A., Penn, D. J. & Kolm, N. (2015). Brain size affects female but not male survival under predation threat. Ecology Letters 18, 646-652.

Kotrschal, A., Corral-López, A., Szidat, S. & Kolm, N. (2015). The effect of brain size evolution on feeding propensity, digestive efficiency, and juvenile growth. Evolution 69(11), 3013–3020

CONTENTS INTRODUCTION 13

Brain size, cognitive ability and mate choice

Brain size, cognitive ability and sexual behaviour

Costs and benefits of evolving a larger brain: artificial selection as a tool

The Trinidadian guppy: a model system in sexual selection research

Aim

METHODS 25 Study system

Morphological traits

Visual capacity

Preference tests

Scoring of behaviours

RESULTS AND DISCUSSION 32 Brain size, cognitive ability and mate choice

Brain size, cognitive ability and sexual behaviour

CONCLUDING REMARKS AND FUTURE CHALLENGES 41

LITERATURE CITED 46

SVENSK SAMMANFATTNING 55

RESUMEN EN ESPAÑOL 59

ACKNOWLEDGEMENTS 63

ARTICLES OF THE THESIS 69

DOCTORAL THESES FROM THE ZOOLOGY DEPARTMENT 175

13

INTRODUCTION

A hike into South American cloud forests in the adequate time of the year can surprise you with exceptional dance moves of colourful man-akin birds. Likewise, with a very careful observation while walking through the Australian bushland, similar dance moves can be observed in tiny male jumping spiders flashing their iridescent abdominal colour-ation towards the more cryptic females. But there is no need to travel across the world, a simple walk through the parks and gardens of your city will likely be accompanied with the sight of the magnificent plum-age of the male peacock’s tail or the sound of an elaborated birdsong waiting for the response of a potential mating partner. Such astonishing examples of dramatic courtship displays, elaborated traits and differ-ences between sexes have aroused the interest of naturalists and laymen alike for centuries.

Darwin’s foundational work on the theory of sexual selection (1859; 1871) described two mechanisms that would explain the value of such behaviours and traits in the struggle for survival and reproduc-tion, male-male competition and female choice. In the first, direct com-petition among males for access to females would favor the evolution of traits such as greater size, strength or disproportionate attributes used as weapons. While these weapons could provide an advantage under direct competition, in the second mechanism female preferences for males that possess certain attributes would similarly provide a selective advantage over time to traits that advertise the quality of a male by means of elaborated ornaments or courtship displays. It is in the de-scription of both these mechanisms in The Descent of man (Darwin 1871) where the first references to the role that mental processes play in sexual selection appears in the literature:

14

“...When we behold two males fighting for the possession of the female, or several male birds displaying their gorgeous plumage, and performing the strangest antics before an assembled body of females, we cannot doubt that, though led by instinct, they know what they are about, and consciously exert their mental and bodily powers...”

“…We can judge, as already remarked, of choice being exerted, only from the analogy of our own minds; and the mental powers of birds, if reason be excluded, do not fundamentally differ from ours...”

Yet, like the little initial support of Darwin and Wallace’s revolu-tionary ideas in evolutionary theory, Darwin’s early ideas on the role of mental processes in sexual selection were also neglected among the sci-entific community. Female choice was highly in contempt because of the superior mental abilities of females over males that these ideas im-plied in a male-dominated society. Likewise, Darwin found practically no support among peers in the mental capacities attributed to certain non-human animals (Cronin 1991). Interestingly, these misconceptions halted scientific progress on how mate choice works in non-human or-ganisms for over a century, but did not impede subsequent major ad-vances in the potential role of mate choice as an agent of sexual selection (Bateson 1983, Andersson 1994). This is so because the focus in the study of sexual selection of evolutionary biologists and ethol-ogists during the 20th century was centered on how males were able to stimulate females in order to be chosen as mates, leaving completely aside the role of mental processes of the female in such mate choice (Brown et al. 2005; Milam 2010).

In the last four decades we have seen an outburst in studies of sex-ual selection. A major consequence of this is that the field is progres-sively abandoning the classically defined sexual roles of males as always the competitive sex and females as passively choosing among potential candidates, as a wide variety of sexual systems have been found across taxa (Dugatkin 2013). Moreover, it is now beyond any doubt that mate choice and mating behaviours are central in the study of evolutionary biology. During these decades major contributions have been provided to understand how mate choice evolves and its role in speciation (see Andersson and Simmons 2006 for an extended review).

15

Yet, given the technical challenges of acquiring experimental evidence on the adaptive value of mate choice (Rosenthal 2017), further investi-gation of how direct and indirect mechanisms interact and the relative roles of the different mate choice mechanisms in shaping adaptation and speciation will be necessary. A second major line of investigation in the sexual selection field is how variation in sexually selected traits is up-held despite the strength that both competition for mates and mate choice provide as evolutionary forces. Several studies have provided major contributions in the unravelling of this paradox. First, through the role of additive genetic variance in genes contributing to increased in-dividual fitness via traits non-related to mating (Rowe and Houle 1996). Second, through conflicting co-evolutionary patterns of sexually se-lected traits between males and females of the same species (Arnqvist and Rowe 2005). Finally, variation in external and intrinsic factors have also been suggested as key drivers for the maintenance of sexually se-lected traits through their effect on mating decisions (Jennions and Pe-trie 1997, Hunt et al. 2005, Witte and Nöbel 2011, Verzijden et al. 2012). Surprisingly, a much smaller portion of the studies in the sexual selection field have been dedicated to investigate the role of mental pro-cesses in mating decisions. Cognitive ability, the acquisition, pro-cessing, retention and use of information (Shettleworth 2010), is a key intrinsic factor driving differences in behavioural patterns and decision-making within and across species (Dukas and Ratcliffe 2009, Shettle-worth 2010).Yet, when it comes to mating decisions, classic sexual se-lection theory have used mating decisions only as an agent of sexual selection, commonly omitting the role of cognitive ability in decision-making (Rosenthal 2017). However, as cognitive ability can be funda-mental when integrating the information of potential mates and external cues, there is a critical need to more often incorporate the role of cog-nitive ability in studies of sexual selection.

Brain size, cognitive ability and mate choice

The brain is a key organ in the study of animal behaviour given its cen-tral role in storing, integrating and processing information. Such cogni-tive processes might have major influence on basic organismal

16

functions determining fitness across and within species. As such, be-havioural patterns and cognitive processes are major factors of interest to understand what drives the evolution of brain size (Allman 2000, Striedter 2005, Dukas and Ratcliffe 2009).

Mating decisions are central from a fitness point of view (Ryan et al. 2009). Deciding when, where and whom to mate are among the most important decisions that individuals face in their lifespan. And, as in all other decision-making processes, cognitive ability plays a key role in mate choice. This is for instance the case when addressing the ability of individuals to detect and perceive sexual displays and compare traits of potential mates. Theoretical and empirical evidence pinpoint the im-portance of biases in the sensory system of individuals in the decision-making process (Kirkpatrick et al. 2006, Ryan et al. 2009, Ryan and Cummings 2013). In order to exert choices, individuals need to be able to distinguish between traits present in potential mates (Ryan et al. 2007). Moreover, environmental factors can alter or obscure signals transferred in the communication process (Bradbury and Vehrencamp 1998). Like in the case of perception of colour patterns of fish in murky waters, or acoustic mating calls of birds and frogs in noisy environ-ments.

However, cognitive processes in the brain during decision-making do not end once a signal is perceived and detected, but the information is subsequently processed to produce an action (Mendelson et al. 2016). Understanding what rules govern information processing in mating de-cisions of animals is an exciting niche of research within sexual selec-tion that have grown in recent years. Most of the recent work in this regard have been devoted to explore the rationality of choice during mating decisions. Rationality have been assessed across a wide variety of taxa using two key concepts in micro economical decisions (Hurley and Nudds 2006): transitivity (if A > B and B > C, then A > C) and the decoy effect (if A > B in the absence of C, then A > B in the presence of C). However, both rational and irrational mating decisions have been demonstrated in different taxa suggesting that further clarification is still needed (e.g. rational mate choice: Dechaume-Moncharmont et al. 2013, Arbuthnott et al. 2017; e.g irrational mate choice: Lea and Ryan

17

2015, Griggio et al. 2016). Moreover, mirroring the work of cognitive psychology in humans, recent theory suggest that comparative evalua-tion of stimuli should be a normative information processing rule across taxa (Akre and Johnsen 2014). Despite the huge implications that this avenue of research offers to our current knowledge in mate choice, the role of brain size is to date virtually unexplored in the field of cognitive mate choice.

Brain size, cognitive ability and sexual behaviour

Unlike the relationship between brain size and mate choice, the role of brain size in sexual behaviour of non-human animals that Darwin hinted towards in his early texts have received more research attention. The theoretical ground for such an association was first brought up in a re-view by Lucia Jacobs (1996), where she suggested that observed differ-ences in brain anatomy between sexes in sexually dimorphic species might be mediated by sexual selection trough the required cognitive abilities coupled to competition for mates. Support for this hypothesis has recently been provided in a comparative study across pipefishes and seahorses, where females, the competing sex in this sex-role reversed family of teleosts, present larger brains, a pattern that furthermore in-creased under higher female intra-sexual competition levels (Tsuboi et al. 2017). Several studies on birds have also found a positive correlation between the complexity of brain anatomy of a species and the ability to perform complex behaviours related to mating success, such as the abil-ity of constructing more complex bowers across bowerbird species (Madden 2001, Day et al. 2005), the complexity of song across passer-ine species (Garamszegi et al. 2005), and the complexity of acrobatic courtship displays across manakin species (Lindsay et al. 2015).

Improved cognitive abilities, tightly linked to brain anatomy (Reader and Laland 2002, Garamszegui and Eens 2004, Emery and Clayton 2004, Gonzalez-Voyer et al. 2009, Brown 2012, Kotrschal et al. 2013; 2015a, Benson-Amram et al. 2016), are likely required to be important for such complex behaviours (Shettleworth, 2010). This is especially evident for the complex courtship displays studied across

18

bird species, as they strongly rely on a learning component (Boogert et al. 2011). Indeed, learning from the social environment and from avail-able public information might play a fundamental role in determining mating decisions and behaviours of individuals (Brown et al. 2006, Ver-zijden et al. 2012). These studies across bird species, although scarce, stress that brain anatomy and cognitive ability might be tightly linked to the evolution of certain sexual behaviours and again suggest a critical need for additional empirical data on this topic.

Costs and benefits of evolving a larger brain: artificial selection as a tool

Across species the variation in brain anatomy is vast. Vertebrate brains accurately illustrate this fact by showing that absolute brain volumes vary up to five orders of magnitude. And this variation is commonly associated with changes in internal organization of the brain (Niew-ienhuis 1998), for instance brain region size, neuron density or neuronal connectivity (Striedter 2005). The simplicity of measuring brain size in comparison to other more detailed aspects of brain anatomy, together with the fact that humans have one of the largest brains in relation to body size (Jerison 1979), has led to that relative brain size (brain size controlling for body size) remains a key trait in the study of brain evo-lution. Given the obvious cognitive capacities of our own species, the selective advantage that a larger brain and associated greater cognitive ability might confer has also always been central to this scientific field. Social interactions with other members of the species, the ability to in-novate and learn to use new tools, or the ability to survive under novel and/or challenging environments have all been proposed as mecha-nisms driving the evolution of brain variation across species (Lefevbre et al. 1997; 2004, Reader and Laland 2002, Tebbich and Bshary 2004, Sol 2009, Brown 2012, Fristoe et al. 2017). Although the exact mecha-nism is still under strong debate (DeCasien et al. 2017, Powell et al. 2017), all these hypothesis find a common ground in the key role that demands for cognitive ability plays in evolving a larger brain. Histori-cally, the major focus of these studies has been on primates and birds, taxa that rank among the highest in brain size body size allometric re-

19

lationships (Striedter 2005). However, developments in the animal be-haviour field have shown unequivocal evidence of great cognitive abil-ities also in lower vertebrates, and even insects (Bshary et al. 2014, Chittka 2017).

Then why do we observe such an enormous variation across species in both absolute and relative brain size? A primary constraint of evolving a larger brain is the energetic requirements involved in devel-oping and maintaining a large brain. In fact, brain tissue is among the most costly to produce and maintain in an organism (Mink et al. 1981, Aiello & Wheeler 1995, Niven 2016). Interestingly, it is the comparison between investment in brain and another costly trait, gut tissue, in hu-man and chimpanzee that established the theoretical foundations of the energetic constrains of evolving a larger brain (the expensive tissue hy-pothesis; Aiello and Wheeler 1995). Comparative studies across species have further demonstrated this hypothesis (Kozlovsky et al. 2014, Tsu-boi et al. 2015, Liao et al. 2016), and that these constraints are likely to influence other energetically demanding features such as fat storage (Navarrete et al. 2011, Tsuboi et al. 2016), metabolic rate (Isler and van Schaik 2006), or testis (Pitnick et al. 2006).

As in the case of the link between brain size and sexual behaviour, our knowledge about the benefits and costs of evolving a larger brain was until recently exclusively based on comparative analyses between species. Although comparative studies have always been important generators of hypotheses, they should be complemented with experi-mental studies that can establish causality. One such experimental ap-proach is the manipulation of brain anatomy through artificial selection (Mery and Kawecki 2005). A recent artificial selection experiment on relative brain size in the guppy (Poecilia reticulata) performed in Ni-clas Kolm’s lab (see methods for further information), have provided further knowledge concerning the benefits and costs of evolving a larger brain. The increased abilities of large-brained individuals over small-brained individuals in ecologically relevant cognitive tests have provided evidence for a direct link between cognitive ability and brain size (Kotrschal et al. 2013; 2015a, Buechel et al. unpublished data). Likewise, large-brained females had higher survival than small-brained

20

females in semi-natural conditions, including high levels of predation (Kotrschal et al. 2015b). A study of the behavioural response of large-brained and small-brained individuals towards predation threat sug-gested that the effect on survival was due to a higher efficiency of large-brained females to evaluate the risk of predation (van der Bijl et al. 2015). Additional studies on the brain size selected guppies have also provided empirical evidence on potential costs of evolving a larger brain. As predicted from the expensive tissue hypothesis, large-brained individuals suffer an important reduction in the investment on gut tissue (Kotrschal et al. 2013). Moreover, two key reproductive traits are re-duced in large brained guppies, fecundity (Kotrschal et al. 2013), and juvenile growth (Kotrschal et al. 2015c). Innate immune response has likewise been demonstrated as a cost of evolving a larger brain (Ko-trschal et al. 2016). Despite previous findings in this direction (Pitnick et al. 2006), studies in the guppy did not provide evidence for differ-ences between large-brained and small-brained males in sperm invest-ment (Kotrschal et al. 2015d). On the contrary, this study provided unexpected evidence of a positive correlation between brain size and secondary sexually selected traits (Kotrschal et al. 2015d). To date, the mechanism behind this correlation is unknown. Yet, given the key role that male sexually selected traits play in mating decisions and behav-iours in this species, this finding set the stage for further research on the link between brain size, cognitive ability and sexual selection in this thesis.

The Trinidadian guppy: a model system in sexual selection research

The vast increase in the number of studies on sexual selection in the last decades goes hand in hand with the use of model systems such as fruit flies and stalk-eyed flies, three-spined sticklebacks, and poeciliid fishes. All these species present a series of characteristics that facilitate investigations of key questions in this field. Conspicuous sexual dimor-phisms, short generation time and well-known genetic background are only a few examples of such characteristics (Andersson and Simmons 2006). Within poeciliid fishes, the guppy is one of the most important models in sexual selection research. This is for instance illustrated by

21

the production of extensive monographs with hundreds of examples of how work on the guppy has provided insights in the evolution of mate choice and sexual selection (Houde 1997), and in evolutionary ecology (Magurran 2005). The studies presented in this thesis have enormously benefited from this previous work and research that has followed in subsequent years. Here I briefly summarise biological aspects of this species that are relevant to understand the work presented.

Fig. 1. Male and female guppy (Poecilia reticulata). Sexual selection is central in the evolu-tion of the sexual dimorphism observed in this species, where females are larger while males present conspicuous colour patterns in their body. Photo edited by Jose Orrite.

The Trinidadian guppy is a small livebearer and fresh water fish that has a dramatic sexual dimorphism. While females are much larger than males, males present complex colour patterns on their bodies and tails with conspicuous and unique combinations of red-orange, black, yellow, green and iridescent colouration (Fig. 1). The colouration of the males is highly heritable (Houde 1992) and highly susceptible to mate choice by females (Endler 1983). The constant pursuit and courtship of males can be catalogued as another striking characteristic that have

22

aroused the interest of researchers in this species. Male guppies have a broad behavioural repertoire when attempting to inseminate females. Males perform several sexual behaviours with the aim to court females and seek mutual consent to complete a copulation and transfer sperm into the female. In the most characteristic courtship display of male guppies, the so-called sigmoid display, males positioned in front of fe-males display the colouration on their bodies and tails by bending their bodies into an s-shaped curvature (Liley 1966). Moreover, males often extend and swing the gonopodium, their external sexual organ, in front of females. Males also perform a series of sexual behaviours that seek insemination without female consent. Here, males chase and position themselves behind the females, often initiating an extremely quick movement in which they extend and attempt to insert their gonopodia into the female genital´s opening. However, coercive attempts of in-semination are less likely to be successful than consented copulations (Houde 1997). Although in general, female receptiveness towards sex-ual encounters are lower than those of the males, females are not passive spectators. First, they exert their choice by hiding, escaping and avoid-ing male harassment. Second, they may opt for showing their receptive-ness towards displaying males by positioning themselves in front of the male and observing, and by gently swimming towards the male while smoothly bending their bodies (the so-called female glide; Liley 1966). However, such female sexual behaviours are not synonyms of complete acceptance towards a particular male since both sexes might opt for ending a consented copulation attempt at any moment before full in-semination has proceeded.

Biological differences between sexes in the reproductive system have led to important divergences in life history traits between males and females. After sexual maturation, females store sperm and ova are matured internally to give birth to well-developed offspring in non-overlapping batches approximately every 4 weeks (Houde 1997). Fe-males are only sexually receptive within a few days of giving birth (Li-ley 1966). Females provide no parental care after giving birth. On the contrary, males only contribute with sperm to offspring development. Unlike females, males are constantly sexually receptive during sexual

23

maturity. This dichotomy between the sexes leads to a wide variation in the ratio of sexually receptive males and females across time and space in a population (operational sex ratio), often resulting in strong male biased sex ratios with high levels of male-male competition and sexual harassment towards females (Jirotkul 1999).

The role of predation on sexual selection and life history in this species has also inspired a large body of work in the study of evolution. This is heavily influenced by the characteristics of the natural habitats where this species is found in Trinidad. Because of the terrain compo-sition in the island, several streams present large variation in the distri-bution of the main predator of the guppy, a pike cichlid (Crenicichla alta). This fact has led to numerous studies on evolutionary patterns under low and high predation habitats. In short, predation threat corre-lates negatively with the expression of male conspicuous colouration (reviewed in Magurran 2005). Likewise, higher exposure to predation has profound influence on the sexual behaviour, as males reduce their rate of the more risky courtship display (Endler 1987, Magurran and Seghers 1990), and females have been found to be less choosy in such situations of high predation (Godin and Briggs 1996, Gong and Gibson 1996).

Aim

Guppies artificially selected for relative brain size offered the unprece-dented opportunity to empirically study the link between brain size, cognitive ability and sexual selection. I aimed to explore this link by comprehensively assessing mating preferences and mating behaviours of both sexes. For this, I assessed preferences towards traits previously described to provide higher fitness. In females, I evaluated the prefer-ences in large-brained and small-brained individuals towards pairs of males that differed in colouration and tail size (Paper I). In males, pref-erence in large- and small-brained individuals was evaluated when pre-sented with pairs of females that differed in body size (Paper II). Given the potential role of sensory biases in mating preferences and the use of

24

only visual cues in these preference tests, I assessed potential differ-ences in visual perception between large-brained and small-brained in-dividuals (Paper I & III).

The link between brain size and sexual behaviour was assessed in two different experimental setups: i) by quantifying the behavioural repertoire of single large-brained and small-brained males in a simple non-competitive situation with only one female (Paper IV), and ii) by quantifying intra-sexual competition and mating behaviours of large-brained and small-brained individuals in a more complex setting in groups with different predation threat and different operational sex ra-tios (Paper V).

25

METHODS

Study system

Experimental studies on the link between brain size, cognitive ability, mate choice and mating behaviours in this thesis were based on labora-tory reared descendants of wild Trinidadian guppies from high preda-tion areas of the Quare River. I used guppies that were exposed to two types of treatments in their captive rearing. First, I used guppies that were kept in the laboratory with equal sex ratios in large stocks (> 100 individuals), and that were allowed to reproduce freely (wild-type gup-pies; Papers I, II & IV). Second, I used guppies artificially selected for small and large relative brain size (brain size selected guppies; Papers I-V). Selection for relative brain size was based on the residuals of pa-rental brain weight on body weight of, as this trait was not possible to assess in large numbers of live fish. Wild-type males (n=225) and fe-males (n=225) were paired randomly to set up three experimental rep-licate populations of 75 breeding pairs each (F0). After offspring production, parents were sacrificed and measurements of body size and brain weight were obtained. Relative brain size of the breeding pairs was ranked using standardised residuals of the male and female regres-sions of brain weight on body size. For three generations, two male and two female offspring from the 15 highest and 15 lowest ranked pairs in each replicate population were used to form breeding pairs for the next generation and generate six populations of juveniles (i.e. three repli-cates of up- and down-selected lines respectively). Evaluation of brain anatomy with microcomputed tomography of the third parental genera-tion of artificial selection (F3) showed that large-brained and small-brained individuals differed in 12.5 % in brain volume but none of the 11 major regions of the teleost brain differed in relative size between

27

of sexual maturation and then kept in single-sex groups of 8-12 individ-uals in 7-12 l tanks containing java moss, 2 cm of gravel and biological filters. If the study required individual identification, fish were isolated in 4 l tanks prior to behavioural tests and/or morphological measure-ments. To avoid isolation stress, I allowed for visual contact between the tanks. The laboratory was maintained at 26 °C with a 12:12 light:dark schedule. Fish were fed a diet of alternating flake food and freshly hatched brine shrimp six days per week.

Morphological traits

I collected measurements of several physiological attributes of the fish that participated in the studies. For this, fish were anesthetised with a low dose of benzocaine and photographed. I used ImageJ image analy-sis software (Schneider et al. 2012) to quantify body size and eye size of the fish in the obtained photographs (Fig. 3). Likewise, we followed this procedure to quantify tail size and colouration patterns of male gup-pies.

Physiological attributes were used in the evaluation of preferences of large-brained, small-brained and wild-type guppies for traits previ-ously demonstrated to indicate the quality of potential mates (see Houde 1997). To evaluate preferences of females for attractive males (Paper I), I sought for males diverging in body colouration and tail size by quantifying large numbers of young sexually mature wild-type individ-uals. Next I selected males based on their measurements to form attrac-tive-unattractive male pairs which differed in an average of 26 % in total colouration and 27 % in tail size, but that were matched by body size. In the case of male choice for female quality, I measured body size in a large number of female guppies and formed female pairs with large, medium and small differences in body size.

Visual capacity

I studied potential differences in the visual capacity of large-brained, small-brained and wild-type guppies. As colouration patterns of males

28

are central to female preferences, I studied the sensitivity to colouration in female guppies. First, I measured the expression of vision-related genes and opsins in the eyes of a subset of females that were assessed in their preferences for more colourful males (Paper I). Opsins are pro-teins in the retina that mediate the initial steps of photon capture in the visual system and quantifying opsin expression is a widely used tool to characterise colour vision (e.g. Hart 2001, Hofmann et al. 2009). Sec-ond, I assessed the behavioural response of these females when exposed to stimuli consisting of rotating arrays of stripes of alternating colours in different saturation contrasts (Paper I). Fish orient their positions us-ing objects as references and therefore stereotypically respond to rotat-ing arrays of stripes by following the motion of such rotation. Such behaviour is called optomotor response and has previously been used to characterise visual capacity of a wide variety of fish species, including guppies (Douglas and Hawryshyn 1990, Anstis et al. 1998). The colours used to generate the stripes of the rotating stimuli which were projected to the walls of a circular arena, were selected to test sensitivity of the females to orange colouration, a key secondary sexually selected trait in males of this species (Houde 1997; see methods in Paper I).

Figure 3. Morphological trait quantification. Fish were photo-graphed together with a colour card and a millimeter scale to calibrate colour and size between photo-graphs. Next, I calculated the area of colour spots in ImageJ (Schneider et al. 2012). Bottom photograph illus-trates the quantification of body size and tail size (red dashed line), orange coloration (yellow dashed line), and black coloration (pink dashed line) in a male guppy.

29

In addition to colour sensitivity, I measured the ability of large-brained and small-brained males and females to resolve spatial detail (visual acuity; Land and Nilsson 2012). As the preference tests per-formed in this thesis relied in visual cues, assessment of perceptual bias altering the resulting preferences were necessary. For instance, visual acuity could have influenced the ability of females to discriminate be-tween the areas of secondary sexually selected traits of males, as well as the ability of males to discriminate differences in the body size of females presented in dichotomous choice tests. To study visual acuity, I measured the behaviour (optomotor response) of guppies when ex-posed to rotating arrays of black and white bands (Paper III). After pre-liminary tests on the response of guppies to different band widths and speeds of the rotating stimuli, I assessed the optomotor response of large-brained and small-brained guppies in the lower end of guppy vis-ual acuity.

Preference tests

To study the preferences of both males and females I measured the time spent associating with potential mates in dichotomous choice tests. De-spite its limitations (see Wagner 1998), an important advantage of side-association data is to remove potential confounding effects of intra-sex-ual competition of individuals of the chosen sex in the test (Houde 1997, Wagner 1998). Although commonly used to assess mating preferences in fish (e.g. Cummings and Mollaghan 2006, Lehtonen and Lindström 2008), this methodology has also proven to be a valid tool in a wide range of taxa such as birds and insects (Shackleton et al. 2005, Rutstein et al. 2007). In guppies this setup is widely used to measure preferences of both sexes (reviewed in Houde 1997, Dosen and Montgomerie 2004), and it has recently been validated for male choice preferences for female body size (Jeswiet and Godin 2011).

Here, all fish were placed in the experimental setup 24 hours prior to the tests for acclimation. The setup consisted of a plain glass tank (42x20x20 cm) where potential mates were presented in left and right position by adjoining additional plain glass tanks (11x10x20 cm; Fig.

30

S1 in Paper II). Visual interactions between potential mates were avoided by adding a plastic film in the side of the wall of adjoined tanks. I balanced the number of potential mates presented in the left and right sides of the experimental setup for the different treatments. Every trial was recorded with a camera placed on top of the setup and broadcasted live on a computer screen to avoid disturbances during quantification of behaviour. To study mating preferences of females for colourful males, during 15 minutes I scored the position of the female in relation to the males presented (Paper I). Analyses of female preferences re-gardless of the colour patterns and tail size of the males presented indi-cated a loss of preference for males in the last 5 minutes of the trial (see results in Paper I). As such, for the study of male preferences for female body size differences we scored the position of males during 10 minutes (Paper II). To score the position of males and females assessed for their preference we divided the experimental tank into three zones: (i) left choice zone, the area adjacent to the left male tank up to a maximum distance of 10 cm from it; (ii) right choice zone, the area adjacent to the right male tank up to a maximum distance of 10 cm from it; and (iii) no choice zone, the area between the left and right choice zones and all areas further away than 10 cm from the male tanks (Fig. S1 in Paper II). I quantified preference for potential mates in each trial by calculat-ing a preference ratio that controls for differences in the motivation to associate with potential mates (Houde 1997):

Scoring of behaviours

Behavioural patterns of fish during the experiments were scored in two ways: i) by means of live observation, and ii) by visualization of video recordings of the test. To obtain reliable data on temporal patterns of the behaviours scored, I used behavioural observation software. Posi-tion of large-brained and small-brained males and females in preference tests was scored using the live observation mode in Jwatcher version

31

1.0 (Blumstein and Daniel 2007; Paper I) and BORIS version 2.72 (Fri-ard and Gamba 2016; Paper II). Optomotor response of fish to rotating stimuli projected in a circular arena was scored in video recordings us-ing BORIS (Friard and Gamba 2016; Paper III). Behaviours of large-brained and small-brained males and receptiveness of non-virgin fe-males towards them in non-competitive situations was scored in Jwatcher (Blumstein and Daniel 2007) using video recordings from lat-eral and top positions (Paper IV). Finally, sexual behaviours and intra-sexual competition behaviours of large-brained and small-brained fish were scored in groups of six individuals under different predation threat and operational sex ratio scenarios. To quantify all behaviours per-formed by each individual in a group of fish, movement patterns of the fish were video recorded from a top positioned high-resolution camera and tracked using idTracker (Pérez-Escudero et al. 2014; Paper V). Next, I visualised each trial at slow motion (0.33x) projecting the track-ing data onto the video recordings using idPlayer. This methodology provided very reliable information on individual identity in groups of fish (see methods Paper V). Importantly, scoring of behaviours in these studies was performed blind to the brain size treatments of the fish (large-brained, small-brained or wild-type), since randomly assigned numbers identified individuals during the course of experiments.

32

RESULTS AND DISCUSSION

Brain size, cognitive ability and mate choice

I evaluated the preference of large-brained, small-brained and wild-type guppies for traits that indicate the quality of potential mates in dichoto-mous choice tests. In Paper I, I showed that large-brained females and wild-type females presented significant preferences for more colourful males that also had larger tails (Fig. 4). However, small-brained females did not show such preference (Fig. 4). Moreover, the preference values observed in large-brained females and wild-type females differed sig-nificantly from those of small-brained females (LMMpreference: small-brained versus large-brained: χ2 = 6.952, df = 1, p = 0.008; LMMprefer-

ence: small-brained versus wild-type: χ2 = 8.660, df = 1, p = 0.003), while no difference was observed between large-brained and wild-type females (LMMpreference: large-brained versus wildtype: χ2 = 0.662, df = 1, p = 0.414).

On the contrary, the evaluation of male preferences for female body size in Paper II did not show significant differences between small-brained, large-brained and wild-type males (LMMpreference: brain size: F2, 1.58 = 0.14, p = 0.87; Fig. 5). All relative brain size treatments showed an overall preference towards larger females (Means ± SE: small-brained: 0.20 ± 0.07, t = 2.82, p = 0.02; large-brained: 0.15 ± 0.07, t = 2.22, p = 0.04; non-selected: 0.25 ± 0.08, t = 3.14, p = 0.01). In dichotomous choice preference tests I exposed all males to three types of female pairs, pairs that presented a large, medium and small difference in body size. I found no effect of the difference between the females presented when evaluating male preferences for larger females (Means ± SE: small-brained: 0.20 ± 0.07, t = 2.82, p = 0.02; large-brained: 0.15 ± 0.07, t = 2.22, p = 0.04; wild-type: 0.25 ± 0.08, t = 3.14, p = 0.01). However, I found a significant interaction in the preference

33

for larger females between the relative brain size of the male and the difference between females presented (LMMpreference: female pair differ-ence * brain size: F2,72 = 4.16, p = 0.04). This was caused by an increase in the preference for larger females in large-brained males as the differ-ence in body size between females increased in the choice tests (Fig. 5).

Fig. 4. Female preference for colourful males. Average preference ± standard error for col-ourful males in large-brained, small-brained and wild-type female guppies. Preference ratio was calculated for each female as the difference in time spent with colourful and non-colourful males, divided by the total time spent in defined choice areas in a dichotomous choice test. A value of 1 would indicate that a female spent the total time of the trial in the colourful male choice area.

Given our seemingly contradictory results between sexes in their mating preferences for high quality mates, it is important to consider the cognitive processes involved in decision-making. Following the conceptual framework of animal decision-making (Mendelson et al. 2016), I here describe the expected cognitive processes in our experi-mental setup to asses mating preferences in the guppy. Using visual cues the studied individuals acquired information of potential mates. The experimental design that I used aimed to eliminate sources of in-formation other than the trait of study (male colouration for female choice, and female body size for male choice). Using the information

34

on the trait, the focal individual judged the quality of potential mates. Two main cognitive processes influence such judgments in our test: dis-crimination and assessment. Next, the focal individual used the pro-cessed information to make a decision resulting in an action, which in our case was measured as time associating with each potential mate. In dichotomous choice tests without full interaction with potential mates, those decisions can be interpreted as preferences. Previous studies of similar tests in guppies have validated the correlation between mating preferences in these types of association set-ups and mate choice when free interaction between males and females is allowed (Houde 1997, Jeswiet and Godin 2011).

Fig. 5. Male preference for larger females. Average preference ± confidence intervals for larger females in large-brained, small-brained and wild-type male guppies when exposed to large, medium and small differences in body size of females in dichotomous choice tests. Pref-erence ratio was calculated for each male as the difference in time spent with the larger and smaller female, divided by the total time spent in defined choice areas. A value of 1 would indicate that a male spent the total time of the trial in the larger female choice area.

Since every female was only tested once in the female preference tests, I am unable to disentangle the effect of all different cognitive pro-cesses of judgement that lead to preferences for colourful males in large-brained and wild-type females. However, the results obtained based on visual capacities of females suggest no differences between

35

large-brained and small-brained individuals in their ability to acquire information of the males and discriminate between them. Indeed, there was no effect of brain size in the optomotor response towards colour contrasts (Fig. 3 in Paper I), in the expression of colour-vision related genes in their retinas (Table S3 in Paper I), or in their visual ability to resolve spatial detail (Fig. 6; see also Paper III). Similarly, I found no difference between large-brained and small-brained males in their abil-ity to resolve spatial detail (Fig 6; see also Paper III). This suggests that males were able to discriminate which females were larger in dichoto-mous choice test regardless of how small the difference was between females in the presented pairs. Indeed, I found no differences between large- and small-brained males in their overall preference for larger fe-males. This finding suggests that discrimination between traits may not require complex cognitive processes, but only simple proportional com-parisons with low neuronal requirements (Dehaene 2003, Nieder and Miller 2003, Akre and Johnsen 2014). However, in the study of male preferences I tested every male more than once and found that, unlike small-brained and wild-type males, large-brained males increased their preference for the larger female as the body size difference between females increased. Since larger females generally are more fecund, and given that in our test there were no costs derived from intra-sexual com-petition, this behavior would potentially result in higher reproductive success for large-brained males in situations where female body size varies substantially. Put together with the effect of relative brain size on female preference for colourful males, it is likely that better cogni-tive ability assists large-brained individuals to make context-dependent optimal mate choice decisions through better judgements of potential mates.

It is important to note that in these studies of male and female pref-erences for different quality mates, I carefully examined whether intrin-sic differences between large-brained and small-brained individuals resulting from the artificial selection could have influenced the ob-served results. For instance, we know from previous studies that large-brained individuals have a more proactive personality (Kotrschal et al. 2014). However, I found no indication that physiological or behavioural

36

differences among them provided alternative explanations to the results (see Paper I & II for details). On the contrary, these results point to-wards novel empirical evidence on the key role that brain size and cog-nitive ability play in the assessment of mate-quality affecting mate choice.

Fig. 6. The effect of artificial selection for relative brain size on visual acuity. Average optomotor response ± standard error of large-brained and small-brained males (top) and females (bottom) towards rotational stimuli. These stimuli consisted of black and white alternating stripes of gradually decreasing band widths projected over the wall of a circular arena. These band widths correspond to stimuli at the lower end of guppy visual acuity (see methods in Paper III). Dashed lines indicate the average baseline optomotor response of fish when exposed to a static image of the same stimuli. I did not found significant differences in analyses of general optomotor response between large-brained and small-brained individuals, or in independent analyses for each of the six stimuli after correction for multiple testing (see results in Paper III).

Brain size, cognitive ability and sexual behaviour

The study of male sexual behaviour in a non-competitive situation showed a wide individual variation in the behavioural repertoire of male guppies, but no differences in the sexual behaviours that large-brained and small-brained males performed towards females (Paper IV). This pattern was consistent when analyzing every single sexual behaviour independently, when grouping courtship-type displays and coercive-type sexual behaviours, in analyses of total intensity of sexual behav-iour, and in the latency to perform sexual behaviours in the test (Fig. 7;

37

see also Table 1 in Paper IV). This finding suggests that the behavioural repertoire of male guppies in a simple social context might not require higher cognitive ability than that inherent in small-brained males. Courtship displays found to correlate with brain size in birds require long training until young males are successful in attracting females (Boogert et al. 2011). For instance, while young male guppies seem to have an innate capacity to perform sexual behaviours prior to any con-tact with females (Houde 1997, pers. obs.), perfecting vocal and motor displays can take over three breeding seasons in long-tailed manakins (Trainer et al. 2002).

Fig. 7. Sexual behaviour of male guppies in a non-competitive scenario. Principal compo-nent analysis of sexual behaviours of large-brained and small-brained male guppies. Negative values of PC1 describe higher values of coerced copulation-related sexual behaviours while positive values of PC2 describe higher values of display-related sexual behaviours, as indicated by the direction of the arrows for specific behavioural measures. Larger symbols represent cen-troid values for small- and large-brained males. No difference is observed in the set of sexual behaviours between small- brained (circles) and large-brained (squares) males (A), or in the average values retrieved from the first component (B) and the second component (C).

This thesis also evaluated the role of more complex social interac-tions and complex environments in the sexual behaviour of large-brained and small-brained guppies. For this, I studied the role of brain size in the sexual behaviour of individuals in groups of six fish under two operational sex ratios (4 males and 2 females; 2 males and 4 fe-males) and when exposed to low and high predation threat (Paper V). Unlike in the non-competitive scenario, large-brained and small-brained males showed differences in their behavioural patterns. First,

38

large-brained males showed a general reduction in the number of court-ship displays that they performed (LMMdisplays: brain size: F1,4 = 6.93, p = 0.010; Fig. 8A). Second, large-brained males also showed a general reduction in the number of dominance displays towards other males (LMMdominance: brain size: F1,57 = 8.15, p = 0.006; Fig. 8B). These find-ings suggest an additional energetic cost of developing a larger brain in this species. Indeed, dominance and courtship displays are often ener-getically demanding behaviours (Hunt et al. 2004, Mowles 2014). Yet, such reduction in the courtship display of large-brained males was not observed when males were studied in a single male-single female situ-ation (Paper IV), stressing the key role that more complex social envi-ronments can play in behavioural patterns. Indeed, the effect of complex environments on sexual behaviour was confirmed in our study. Analyses regardless of brain size of male sexual behaviour across the different treatments showed a striking increase of courtship under low predation threat (Fig. 8A), together with an increase of intra-sexual dominance interactions when there were stronger intra-sexual competi-tion for mates (Fig. 8B).

Fig. 8. The effect of predation threat and sex ratio in male sexual behaviour. Average num-ber ± confidence intervals of courtship displays (A) and male-male dominance interactions (B) of large-brained and small-brained males under different levels of predation threat and changes in the operational sex ratio (male biased: four males and two females; female biased: two males and four females).

39

Interestingly, the observation that large-brained males performed a lower number of courtship and dominance displays did not result in lower success to attract females as the receptiveness showed by fe-males, as well as number of copulations did not differed between large-brained and small-brained males (LMMreceptiveness: brain size: F1,145 = 1.18, p = 0.280; LMMcopulations: brain size: F1,4 = 0.04, p = 0.848). This contradicts general theory in this and other species, as higher rates of courtship display often lead to higher receptiveness in females (Ger-hardt and Watson 1995, Kodric-Brown and Nicoletto 2001, Baird et al. 2007). This observation is potentially explained by the fact that large-brained males have more colouration, larger tails and larger gonopodia (Kotrschal et al. 2015d), sexually selected traits important in female choice in this species. Alternatively, large-brained males might be more effective in the display per se, as well as in how they time their displays in relation to female behaviour to receive higher receptiveness from these females. It is unfortunate that this study does not allow for full elucidation of the mechanism behind this pattern, as the link between brain size, cognitive ability and behavioural effectiveness of courtship is to date a completely unexplored aspect of animal behaviour.

In Paper V, I complemented findings on male sexual behaviour with examination of the role of more complex social interactions and complex environments for female sexual behaviour. Females showed higher receptiveness towards males under higher intra-sexual competi-tion for mates and with low levels of predation threat (Table 1). Fur-thermore, analyses of sexual behaviour of female guppies artificially selected for relative brain size revealed higher flexibility in the sexual behaviour of large-brained females. While both small-brained and large-brained individuals showed a reduction in sexual behaviour under lower levels of competition for mates and increased sexual harassment towards them (male biased sex ratios), I found that large-brained fe-males changed more in copulation levels than small-brained females in situations with lower predation threat (LMMcopulations: predator*brain size: F1,134 = 4.04, p = 0.046; see Table 1). This result provide further evidence of the key role of higher cognitive ability to process multiple forms of information during mating decisions (Paper I & II). Moreover,

40

previous studies in this species suggest high costs of exposure to preda-tion during sexual encounters (Magurran and Novak 1991). This pattern is indeed common also among other taxa (Magnhagen 1991, Andersson 1994). As such, this finding suggest an important benefit complement-ing previous work on ecologically relevant survival benefits of evolving a larger brain through increased cognitive ability (Kotrschal et al. 2015b, van der Bijl et al. 2015).

Table 1. The effect of predation threat and sex ratio on female sexual behaviour. Percent-age of large-brained and small-brained females copulating and average number of copulations of sexually receptive females under different levels of predation threat and changes in the op-erational sex ratio (male biased: four males and two females; female biased: two males and four females).

% of females copulating

(females tested) Average number of

copulations ± SE

OSR Small-brained

Large-brained

Small-brained

Large-brained

Low predation

Male- biased

7.14% (14)

7.14% (14)

1.00 ± 0.00

1.00 ± 0.00

Female- biased

20.83% (24)

45.83% (24)

1.33 ± 0.21

2.91 ± 0.89

High predation

Male- biased

0% (14)

0% (14)

0 0

Female- biased

14.28% (28)

10.71% (28)

1.25 ± 0.25

1.33 ± 0.33

41

CONCLUDING REMARKS AND FUTURE CHALLENGES

The experimental setups used in this thesis to assess mating decisions and behaviours in the guppy are necessary simplifications of the condi-tions that these fish encounter in nature. Indeed, the study of behav-ioural patterns under more complex environments in this thesis provide further evidence of the key role that factors such as predation threat and changes in the social environment can have on decision-making. Preda-tion, competition and availability of potential mates are hence only ex-amples of the multiple sources of information that guppies face in wild populations. Better cognitive ability likely provides important benefits when individuals need to integrate information from different sources to make optimal decisions. It has also become more evident that envi-ronmental heterogeneity has a central role in maintaining diversity of sexual traits and can likewise act as a driver of selection (Cornwallis and Uller 2010, Miller and Svensson 2014). As such, this thesis pro-vides a framework for linking environmental variation and its effect on sexually selected traits via neural mechanisms (here brain size and cog-nitive ability).

Studies of mate choice and sexual behaviour, including the ones in this thesis, often show a great deal of individual variation. And this variability is not only influenced by extrinsic factors but also shaped by state-dependent factors. Interestingly, cognitive ability is closely asso-ciated with major state-dependent factors that influence mate choice and sexual behaviour. Condition is one of these factors that has been widely observed to influence mate choice and sexual behaviour (Jen-nions and Petrie 1997, Cotton et al. 2006). If we put the focus on the chosen sex, it is straightforward that better condition can allow you to invest more energy into morphological traits or courtship displays that signal your quality. However, much less attention have been given to the role of condition in individual variation of mating decisions (Cotton

42

et al. 2006). This fact draws some parallels with the study of the asso-ciation between cognitive ability, condition and sexual selection. Sev-eral studies have shown a correlation between performing better in cognitive tests and being preferred by individuals of the opposite sex (e.g. Shohet and Watts 2009, Snowberg and Benkman 2009). However, in these studies choosers were seldom allowed to observe the perfor-mance of potential mates in cognitive tests, suggesting that mating pref-erences might be mediated by the condition of individuals with better cognitive ability (Boogert et al. 2011). To date, there are no studies available on the association between cognitive ability, condition and mating preferences from the perspective of the individuals exerting mate choice. Yet, condition should have strong influence on the recep-tiveness of individuals performing mate a choice (Rosenthal 2017). Given the changes in receptiveness observed in this thesis (Paper V), it will be highly interesting to study this potential association.

Another intrinsic factor shaping individual variation in mating preferences and sexual behaviour is experience (Hebets 2003, Dukas 2005, Bailey and Zuk 2008, Macario et al. 2017). The literature on the role of experience in mating decisions is extensive. From simple deci-sion-making rules in relation to instantaneous encounters, to seasonal variations and experiences accumulated across the lifetime of an indi-vidual. Likewise, early experiences and the social environment shape sexual preferences and behaviours of choosers and courters alike. Sex-ual imprinting, mate choice copying, audience effects and eavesdrop-ping are common features of sexual selection across a wide variety of taxa (ten Cate and Vos 1999, Walling et al. 2008, Witte and Nöbel 2011). In summary, every piece of information that an individual gath-ers prior to mating decision can influence the outcome of such deci-sions. It is likely that cognition in general have a central role to delimit these experiences. However, two cognitive processes stand out as par-ticularly important to shape variation in sexual behaviours and choices through experiences, learning and memory (ten Cate and Rowe 2007, Verzijden et al. 2012 Ryan and Cummings 2013). This is likely true also for the model species used in this thesis, as a recent study in closely related fish found a correlation in the genetic pathways associated with

43

learning and mating decisions (Cummings 2015). Unfortunately, the experimental design used in the studies of the thesis do not provide in-depth answers in how a larger brain affect mate choice and sexual be-haviours via memory or learning. The lack of experimental evidence on the association between cognitive ability and sexual selection put the focus of this thesis to understand innate sexual preferences and behav-ioural patterns. For this, virgin individuals that lacked previous sexual experiences were always used. However, short-time memory could have influenced the results observed in preference tests (Paper I & II), as females and males choosing between pairs of potential partners could not see both individuals at the same time in our experimental setup. In-vestigations on guppies artificially selected for relative brain size demonstrate differences in memory between large-brained and small-brained individuals (Buechel et al. unpublished). In addition, studies on this species have demonstrated the influence that early experiences may have on female mate choice later in life (Macario et al. 2017). These recent findings suggest that investigations on how the memory aspect of cognition could shape mating decisions are important in future stud-ies. Likewise, in two of the studies in this thesis (Paper II & V), I stud-ied behaviour of individuals that were exposed to more than one treatment and hence experienced multiple sources of information sepa-rated in time. Interestingly, in both studies I found that large-brained individuals presented higher flexibility in their choice. Large-brained males increased their preference for larger females when the difference between the females presented was larger (Paper II), and large-brained females were more receptive towards males with lower levels of preda-tion threat (Paper V). Investigations on flexible (or plastic) mating de-cisions is an emerging and rapidly developing field in the study of sexual selection. Studies on several species of fish, spiders and insects have recently demonstrated how learning from previous experiences can interact with extrinsic factors to shape mating decisions (Tingithella et al. 2013, Atwell and Wagner 2014, Stoffer and Uetz 2015). Such flexibility in mating decisions can have profound influence on evolu-tionary processes such as local adaptation and speciation (Servedio and Boughmann 2017). The findings of this thesis in relation to flexible mating decisions suggest the need for further investigations of the role

44

of brain size, more detailed brain anatomy variation and different as-pects of cognitive ability for adaptive sexual behaviour across longer time spans in this model system, but also for further research on the role of cognitive ability in mating decisions via learning in other taxa.

Developing a larger brain imply important benefits as studies across and within species have shown. Experimental studies in guppies artificially selected for relative brain size illustrate this by showing that large-brained individuals present better cognitive abilities, greater ex-pression of sexually selected traits, and higher survival (Kotrschal et al. 2013; 2015a; 2015b; 2015d; van der Bijl et al. 2015). The findings of this thesis provide further evidence in this regard by showing that a larger brain can confer important benefits in the assessment of better quality mates, higher flexibility in mate choice, and potentially even higher behavioural effectiveness in securing access to mating. How-ever, there are associated costs of evolving a larger brain which are tightly linked to the high energetic investment that this organ requires. My finding of a general decrease in sexual behaviour in large-brained guppy males under competitive scenarios adds novel support for such costs. Previous studies in these fish also demonstrated associated costs to traits of key relevance in sexual selection, such as fecundity, innate immune response and juvenile growth (Kotrschal 2013;2015c;2016). This contrast raises a very interesting question that is central in brain evolution and most likely the driver of the great variation in brain anat-omy observed across species. Under what ecological circumstances is it beneficial to invest energy in developing a larger brain? Several com-parative studies across species point towards the ability to innovate and adapt to challenging social and ecological environments as key drivers of brain size variation (e.g. Lefebvre et al. 2004, Sol 2009, Fristoe et al. 2017). Despite the obvious technical challenge, complementing such studies with empirical evidence on the adaptive fitness landscapes of larger brains in stable and changing environments is a necessary step to deepen our knowledge in this area.

In summary, the work presented in this thesis offers novel insights regarding the role that brain size can play in mating decisions and mat-ing behaviours through its tight correlation with increased cognitive

45

abilities. Likewise, this thesis stresses the need of an integrated and mechanistic approach to mate choice and sexual behaviour of males and females when assessing fundamental questions in how sexual selection drive evolutionary patterns. Using guppies artificially selected for large and small relative brain size, this thesis provides empirical evidence of the benefits that a larger brain confer in the cognitive assessment and flexibility in the choice of potential partners with different quality. Moreover, my findings concerning sexual behaviour of male guppies identify and provide further evidence for the important energetic con-straints of evolving a larger brain when competing for access to mates. It is my hope that the novel empirical approach in my thesis concerning the study of the association between brain size, cognitive ability and sexual selection, might set the stage for more innovative research on sexually selected traits, sexual behaviours and mating preferences in wild populations.

46

LITERATURE CITED

Aiello, L. C., & Wheeler, P. (1995). The expensive-tissue hypothesis: the brain and the digestive system in human and primate evolution. Current anthropology, 36(2), 199-221.

Akre, K. L., & Johnsen, S. (2014). Psychophysics and the evolution of be-havior. Trends in ecology & evolution, 29(5), 291-300.

Allman, J. (2000). Evolving Brains. Scientific American Library. Andersson, M. B. (1994). Sexual selection. Princeton University Press. Andersson, M., & Simmons, L. W. (2006). Sexual selection and mate

choice. Trends in ecology & evolution, 21(6), 296-302. Anstis, S., Hutahajan, P., & Cavanagh, P. (1998). Optomotor test for wave-

length sensitivity in guppyfish (Poecilia reticulata). Vision re-search, 38(1), 45-53.

Arbuthnott, D., Fedina, T. Y., Pletcher, S. D., & Promislow, D. E. (2017). Mate choice in fruit flies is rational and adaptive. Nature communica-tions, 8: 13953.

Arnqvist, G., & Rowe, L. (2005). Sexual conflict. Princeton University Press.

Atwell, A., & Wagner, W. E. (2014). Female mate choice plasticity is af-fected by the interaction between male density and female age in a field cricket. Animal Behaviour, 98, 177-183.

Bailey, N. W., & Zuk, M. (2008). Acoustic experience shapes female mate choice in field crickets. Proceedings of the Royal Society of London B: Biological Sciences, 275(1651), 2645-2650.

Baird, T. A., Hranitz, J. M., Timanus, D. K., & Schwartz, A. M. (2007). Be-havioral attributes influence annual mating success more than morpholog-ical traits in male collared lizards. Behavioral Ecology, 18(6), 1146-1154.

Bateson, P. P. G. (1983). Mate choice. Cambridge University Press. Benson-Amram, S., Dantzer, B., Stricker, G., Swanson, E. M., & Holekamp,

K. E. (2016). Brain size predicts problem-solving ability in mammalian carnivores. Proceedings of the National Academy of Sciences, 113(9), 2532-2537.

47

Bijl, van der W., Thyselius, M., Kotrschal, A., & Kolm, N. (2015). Brain size affects the behavioural response to predators in female guppies (Poe-cilia reticulata). Proceedings of the Royal Society of London B: Biological Sciences 282, 20151132.

Blumstein, D. T., & Daniel, J. C. (2007). Quantifying behavior the JWatcher way. Sinauer Associates Incorporated.

Boogert, N. J., Fawcett, T. W., & Lefebvre, L. (2011). Mate choice for cog-nitive traits: a review of the evidence in nonhuman vertebrates. Behav-ioral Ecology, 22(3), 447-459.

Bradbury, J. W., & Vehrencamp, S. L. (1998). Principles of animal commu-nication. Sinauer Associate, Inc.

Brown, C. (2012). Tool use in fishes. Fish and Fisheries, 13(1), 105-115. Brown, C., Laland, K., & Krause, J. (2006). Fish cognition and behavior.

John Wiley & Sons. Bshary, R., Gingins, S., & Vail, A. L. (2014). Social cognition in

fishes. Trends in cognitive sciences, 18(9), 465-471. Cate, ten C., & Rowe, C. (2007). Biases in signal evolution: learning makes

a difference. Trends in ecology & evolution, 22(7), 380-387. Cate, ten C., & Vos, D. R. (1999). Sexual Imprinting and Evolutionary Pro-

cesses in. Advances in the Study of Behavior, 28, 1. Chittka, L. (2017). Bee cognition. Current Biology, 27(19), R1049-R1053. Cornwallis, C. K., & Uller, T. (2010). Towards an evolutionary ecology of

sexual traits. Trends in Ecology & Evolution, 25(3), 145-152. Cotton, S., Small, J., & Pomiankowski, A. (2006). Sexual selection and con-

dition-dependent mate preferences. Current Biology, 16(17), R755-R765. Cronin, H. (1991). The ant and the peacock: Altruism and sexual selection

from Darwin to today. Cambridge University Press. Cummings, M. E. (2015). The mate choice mind: Studying mate preference,

aversion and social cognition in the female poeciliid brain. Animal Behav-iour, 103, 249-258.

Cummings, M., & Mollaghan, D. (2006). Repeatability and consistency of female preference behaviours in a northern swordtail, Xiphophorus ni-grensis. Animal Behaviour, 72(1), 217-224.

Darwin, C. (1859). On the Origin of Species by Means of Natural Selection. John Murray.

Darwin, C. (1871). Sexual selection and the descent of man. John Murray. Day, L. B., Westcott, D. A., & Olster, D. H. (2005). Evolution of bower

complexity and cerebellum size in bowerbirds. Brain, Behavior and Evo-lution, 66(1), 62-72.

48

DeCasien, A. R., Williams, S. A., & Higham, J. P. (2017). Primate brain size is predicted by diet but not sociality. Nature Ecology & Evolution, 1, 0112.

Dechaume-Moncharmont, F. X., Freychet, M., Motreuil, S., & Cézilly, F. (2013). Female mate choice in convict cichlids is transitive and consistent with a self-referent directional preference. Frontiers in zoology, 10(1), 69.

Dehaene, S. (2003). The neural basis of the Weber–Fechner law: a logarith-mic mental number line. Trends in cognitive sciences, 7(4), 145-147.

Dosen, L. D., & Montgomerie, R. (2004). Female size influences mate pref-erences of male guppies. Ethology, 110(3), 245-255.

Douglas, R. H., & Hawryshyn, C. W. (1990). Behavioural studies of fish vi-sion: an analysis of visual capabilities. In The visual system of fish. Springer Netherlands.

Dugatkin, L. A. (2013). Principles of Animal Behavior: Third International Student Edition. WW Norton & Company.

Dukas, R. & Ratcliffe, J.M. (2009). Cognitive Ecology II. The University of Chicago Press.

Dukas, R. (2005). Learning affects mate choice in female fruit flies. Behav-ioral Ecology, 16(4), 800-804.

Emery, N. J., & Clayton, N. S. (2004). The mentality of crows: convergent evolution of intelligence in corvids and apes. Science, 306(5703), 1903-1907.

Endler, J. A. (1983). Natural and sexual selection on color patterns in poe-ciliid fishes. Environmental biology of Fishes, 9(2), 173-190.

Endler, J. A. (1987). Predation, light intensity and courtship behaviour in Poecilia reticulata (Pisces: Poeciliidae). Animal Behaviour, 35(5), 1376-1385.

Friard, O., & Gamba, M. (2016). BORIS: a free, versatile open‐source event‐logging software for video/audio coding and live observations. Methods in Ecology and Evolution, 7(11), 1325-1330.

Fristoe, T. S., Iwaniuk, A. N., & Botero, C. A. (2017). Big brains stabilize populations and facilitate colonization of variable habitats in birds. Na-ture ecology & evolution, 1, 1706–1715

Garamszegi, L. Z., & Eens, M. (2004). The evolution of hippocampus vol-ume and brain size in relation to food hoarding in birds. Ecology Letters, 7(12), 1216-1224.

Garamszegi, L. Z., Eens, M., Erritzøe, J., & Møller, A. P. (2005). Sexually size dimorphic brains and song complexity in passerine birds. Behavioral Ecology, 16(2), 335-345.

49

Gerhardt, H. C., & Watson, G. F. (1995). Within-male variability in call properties and female preference in the grey treefrog. Animal Behav-iour, 50(5), 1187-1191.

Godin, J. G. J., & Briggs, S. E. (1996). Female mate choice under predation risk in the guppy. Animal Behaviour, 51(1), 117-130.

Gong, A., & Gibson, R. M. (1996). Reversal of a female preference after vis-ual exposure to a predator in the guppy, Poecilia reticulata. Animal Be-haviour, 52(5), 1007-1015.

Gonzalez-Voyer A., Winberg S., & Kolm N. (2009) Social fishes and single mothers: brain evolution in African cichlids. Proceedings of the Royal So-ciety of London B: Biological Sciences, 276, 161–167.

Griggio, M., Hoi, H., Lukasch, B., & Pilastro, A. (2016). Context‐dependent female preference for multiple ornaments in the bearded reedling. Ecol-ogy and evolution, 6(2), 493-501.

Hart, N. S. (2001). The visual ecology of avian photoreceptors. Progress in retinal and eye research, 20(5), 675-703.

Hebets, E. A. (2003). Subadult experience influences adult mate choice in an arthropod: exposed female wolf spiders prefer males of a familiar pheno-type. Proceedings of the National Academy of Sciences, 100(23), 13390-13395.

Hofmann, C. M., O'Quin, K. E., Marshall, N. J., Cronin, T. W., Seehausen, O., & Carleton, K. L. (2009). The eyes have it: regulatory and structural changes both underlie cichlid visual pigment diversity. PLoS biol-ogy, 7(12), e1000266.

Houde, A. E. (1992). Sex-linked heritability of a sexually selected character in a natural population of Poediia. Heredity, 69, 229-235.

Houde, A. E. (1997). Sex, color, and mate choice in guppies. Princeton Uni-versity Press.

Hunt, J., Brooks, R., & Jennions, M. D. (2005). Female mate choice as a condition-dependent life-history trait. The American Naturalist, 166(1), 79-92.

Hunt, J., Brooks, R., Jennions, M. D., Smith, M. J., Bentsen, C. L., & Bus-siere, L. F. (2004). High-quality male field crickets invest heavily in sex-ual display but die young. Nature, 432(7020), 1024-1027.

Hurley, S. L., & Nudds, M. (2006). Rational animals? Oxford University Press.

Isler, K., & Van Schaik, C. P. (2006). Metabolic costs of brain size evolu-tion. Biology Letters, 2(4), 557-560.

50

Jacobs, L. F. (1996). Sexual selection and the brain. Trends in Ecology & Evolution, 11(2), 82-86.

Jennions, M. D., & Petrie, M. (1997). Variation in mate choice and mating preferences: a review of causes and consequences. Biological Re-views, 72(2), 283-327.

Jerison, H. J. (1979). The evolution of diversity in brain size. In Develop-ment and evolution of brain size: behavioral implications. Academic Press.

Jeswiet, S. B., & Godin, J. G. J. (2011). Validation of a method for quantify-ing male mating preferences in the guppy (Poecilia reticulata). Ethol-ogy, 117(5), 422-429.

Jirotkul, M. (1999). Operational sex ratio influences female preference and male–male competition in guppies. Animal Behaviour, 58(2), 287-294.

Kirkpatrick, M., Rand, A. S., & Ryan, M. J. (2006). Mate choice rules in ani-mals. Animal Behaviour, 71(5), 1215-1225.

Kodric-Brown, A., & Nicoletto, P. F. (2001). Female choice in the guppy (Poecilia reticulata): the interaction between male color and display. Be-havioral Ecology and Sociobiology, 50(4), 346-351.

Kotrschal, A., Buechel, S. D., Zala, S. M., Corral‐Lopez, A., Penn, D. J., & Kolm, N. (2015b). Brain size affects female but not male survival under predation threat. Ecology letters, 18(7), 646-652.

Kotrschal, A., Corral-López, A., Amcoff, M., & Kolm, N. (2015a). A larger brain confers a benefit in a spatial mate search learning task in male gup-pies. Behavioral Ecology, 26(2), 527-532.

Kotrschal, A., Corral‐López, A., Szidat, S., & Kolm, N. (2015c). The effect of brain size evolution on feeding propensity, digestive efficiency, and ju-venile growth. Evolution, 69(11), 3013-3020.

Kotrschal, A., Corral‐López, A., Zajitschek, S., Immler, S., Maklakov, A. A., & Kolm, N. (2015d). Positive genetic correlation between brain size and sexual traits in male guppies artificially selected for brain size. Jour-nal of evolutionary biology, 28(4), 841-850.

Kotrschal, A., Kolm, N., & Penn, D. J. (2016). Selection for brain size im-pairs innate, but not adaptive immune responses. In Proceedings of the Royal Society of London B: Biological Sciences, 283: 20152857.

Kotrschal, A., Lievens, E. J., Dahlbom, J., Bundsen, A., Semenova, S., Sundvik, M., ... & Kolm, N. (2014). Artificial selection on relative brain size reveals a positive genetic correlation between brain size and proac-tive personality in the guppy. Evolution, 68(4), 1139-1149.

51

Kotrschal, A., Rogell, B., Bundsen, A., Svensson, B., Zajitschek, S., Brännström, I. ... & Kolm, N. (2013). Artificial selection on relative brain size in the guppy reveals costs and benefits of evolving a larger brain. Current Biology, 23(2), 168-171.

Kotrschal, A., Zeng, H. L., der Bijl, W., Öhman‐Mägi, C., Kotrschal, K., Pelckmans, K., & Kolm, N. (2017). Evolution of brain region volumes during artificial selection for relative brain size. Evolution. In press.

Kozlovsky, D. Y., Brown, S. L., Branch, C. L., Roth II, T. C., & Pravosu-dov, V. V. (2014). Chickadees with bigger brains have smaller digestive tracts: a multipopulation comparison. Brain, behavior and evolu-tion, 84(3), 172-180.

Land, M. F., & Nilsson, D. E. (2012). Animal eyes. Oxford University Press. Lea, A. M., & Ryan, M. J. (2015). Irrationality in mate choice revealed by

túngara frogs. Science, 349(6251), 964-966. Lefebvre, L., Reader, S. M., & Sol, D. (2004). Brains, innovations and evo-

lution in birds and primates. Brain, behavior and evolution, 63(4), 233-246.

Lefebvre, L., Whittle, P., Lascaris, E., & Finkelstein, A. (1997). Feeding in-novations and forebrain size in birds. Animal Behaviour, 53(3), 549-560.

Lehtonen, T. K., & Lindström, K. (2008). Repeatability of mating prefer-ences in the sand goby. Animal Behaviour, 75(1), 55-61.

Liao, W. B., Lou, S. L., Zeng, Y., & Kotrschal, A. (2016). Large brains, small guts: the expensive tissue hypothesis supported within anurans. The American Naturalist, 188(6), 693-700.

Liley, N. R. (1966). Ethological isolating mechanisms in four sympatric spe-cies of poeciliid fishes. Behaviour. Supplement, III-197.

Lindsay, W. R., Houck, J. T., Giuliano, C. E., & Day, L. B. (2015). Acro-batic courtship display coevolves with brain size in manakins (Pipridae). Brain, behavior and evolution, 85(1), 29-36.

Macario, A., Croft, D. P., Endler, J. A., & Darden, S. K. (2017). Early social experience shapes female mate choice in guppies. Behavioral Ecology, 28(3), 833-843.

Madden, J. (2001). Sex, bowers and brains. Proceedings of the Royal Society of London B: Biological Sciences, 268(1469), 833-838.

Magnhagen, C. (1991). Predation risk as a cost of reproduction. Trends in Ecology & Evolution, 6(6), 183-186.

Magurran, A. E. (2005). Evolutionary ecology: the Trinidadian guppy. Ox-ford University Press on Demand.

52

Magurran, A. E., & Nowak, M. A. (1991). Another battle of the sexes: the consequences of sexual asymmetry in mating costs and predation risk in the guppy, Poecilia reticulata. Proceedings of the Royal Society of London B: Biological Sciences, 246(1315), 31-38.

Magurran, A. E., & Seghers, B. H. (1990). Risk sensitive courtship in the guppy (Poecilia reticulata). Behaviour, 112(3), 194-201.

Mendelson, T. C., Fitzpatrick, C. L., Hauber, M. E., Pence, C. H., Rodríguez, R. L., ... & Stevens, J. R. (2016). Cognitive phenotypes and the evolution of animal decisions. Trends in ecology & evolution, 31(11), 850-859.

Mery, F., & Kawecki, T. J. (2005). A cost of long-term memory in Drosoph-ila. Science, 308(5725), 1148-1148.

Milam, E. L. (2010). Looking for a few good males: female choice in evolu-tionary biology. JHU Press.

Miller, C. W., & Svensson, E. I. (2014). Sexual selection in complex envi-ronments. Annual Review of Entomology, 59, 427-445.

Mink, J. W., Blumenschine, R. J., & Adams, D. B. (1981). Ratio of central nervous system to body metabolism in vertebrates: its constancy and functional basis. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology, 241(3), R203-R212.

Mowles, S. L. (2014). The physiological cost of courtship: field cricket song results in anaerobic metabolism. Animal Behaviour, 89, 39-43.

Navarrete, A., van Schaik, C. P., & Isler, K. (2011). Energetics and the evo-lution of human brain size. Nature, 480(7375), 91-93.

Nieder, A., & Miller, E. K. (2003). Coding of cognitive magnitude: Com-pressed scaling of numerical information in the primate prefrontal cor-tex. Neuron, 37(1), 149-157.

Nieuwenhuys, R. (1998). Comparative neuroanatomy: Place, principles and programme. In The central nervous system of vertebrates, Vol. 1. Springer-Verlag, Berlin.

Niven, J. E. (2016). Neuronal energy consumption: biophysics, efficiency and evolution. Current opinion in neurobiology, 41, 129-135.

Pérez-Escudero, A., Vicente-Page, J., Hinz, R. C., Arganda, S., & De Po-lavieja, G. G. (2014). idTracker: tracking individuals in a group by auto-matic identification of unmarked animals. Nature methods, 11(7), 743-748.

Pitnick, S., Jones, K. E., & Wilkinson, G. S. (2006). Mating system and brain size in bats. Proceedings of the Royal Society of London B: Biologi-cal Sciences, 273(1587), 719-724.

53

Powell, L., Isler, K., Barton, R. A. (2017). Re-evaluating the link between brain size and behavioural ecology in primates. Proceedings of the Royal Society of London B: Biological Sciences 284, 20171765.

Reader, S. M., & Laland, K. N. (2002). Social intelligence, innovation, and enhanced brain size in primates. Proceedings of the National Academy of Sciences, 99(7), 4436-4441.

Rosenthal, G. G. (2017). Mate choice: the evolution of sexual decision mak-ing from microbes to humans. Princeton University Press.

Rutstein, A. N., Brazill-Boast, J., & Griffith, S. C. (2007). Evaluating mate choice in the zebra finch. Animal Behaviour, 74(5), 1277-1284.

Ryan, M. J., & Cummings, M. E. (2013). Perceptual biases and mate choice. Annual Review of Ecology, Evolution, and Systematics, 44, 437-459.

Ryan, M. J., Akre, K. L., & Kirkpatrick, M. (2007). Mate choice. Current Biology, 17(9), R313-R316.

Ryan, M. J., Akre, K. L., & Kirkpatrick, M. (2009). Cognitive mate choice. In Cognitive ecology II. The University of Chicago Press.

Schneider, C. A., Rasband, W. S., & Eliceiri, K. W. (2012). NIH Image to ImageJ: 25 years of image analysis. Nature methods, 9(7), 671-675.

Servedio, M. R., & Boughman, J. W. (2017). The Role of Sexual Selection in Local Adaptation and Speciation. Annual Review of Ecology, Evolu-tion, and Systematics, 48.

Shackleton, M. A., Jennions, M. D., & Hunt, J. (2005). Fighting success and attractiveness as predictors of male mating success in the black field cricket, Teleogryllus commodus: the effectiveness of no-choice tests. Be-havioral Ecology and Sociobiology, 58(1), 1-8.

Shettleworth, S. J. (2010). Cognition, evolution, and behavior. Oxford Uni-versity Press.

Shohet, A. J., & Watt, P. J. (2009). Female guppies Poecilia reticulata prefer males that can learn fast. Journal of Fish Biology, 75(6), 1323-1330.

Snowberg, L. K., & Benkman, C. W. (2009). Mate choice based on a key ecological performance trait. Journal of evolutionary biology, 22(4), 762-769.

Sol, D. (2009). The cognitive-buffer hypothesis for the evolution of large brains. In Cognitive ecology II. The University of Chicago Press.

Stoffer, B., & Uetz, G. W. (2015). The effects of social experience with var-ying male availability on female mate preferences in a wolf spider. Be-havioral Ecology and Sociobiology, 69(6), 927-937.

Striedter, G. F. (2005). Principles of brain evolution. Sinauer Associates.

54

Tebbich, S., & Bshary, R. (2004). Cognitive abilities related to tool use in the woodpecker finch, Cactospiza pallida. Animal behaviour, 67(4), 689-697.

Tinghitella, R. M., Weigel, E. G., Head, M., & Boughman, J. W. (2013). Flexible mate choice when mates are rare and time is short. Ecology and evolution, 3(9), 2820-2831.

Trainer, J. M., McDonald, D. B., & Learn, W. A. (2002). The development of coordinated singing in cooperatively displaying long-tailed mana-kins. Behavioral Ecology, 13(1), 65-69.

Tsuboi, M., Husby, A., Kotrschal, A., Hayward, A., Buechel, S. D., Zidar, J., ... & Kolm, N. (2015). Comparative support for the expensive tissue hy-pothesis: big brains are correlated with smaller gut and greater parental investment in Lake Tanganyika cichlids. Evolution, 69(1), 190-200.

Tsuboi, M., Lim, A. C. O., Ooi, B. L., Yip, M. Y., Chong, V. C., Ahnesjö, I., & Kolm, N. (2017). Brain size evolution in pipefishes and seahorses: the role of feeding ecology, life history and sexual selection. Journal of evo-lutionary biology, 30(1), 150-160.

Tsuboi, M., Shoji, J., Sogabe, A., Ahnesjö, I., & Kolm, N. (2016). Within species support for the expensive tissue hypothesis: a negative association between brain size and visceral fat storage in females of the Pacific sea-weed pipefish. Ecology and evolution, 6(3), 647-655.

Verzijden, M. N., Ten Cate, C., Servedio, M. R., Kozak, G. M., Boughman, J. W., & Svensson, E. I. (2012). The impact of learning on sexual selec-tion and speciation. Trends in Ecology & Evolution, 27(9), 511-519.

Walling, C. A., Royle, N. J., Lindström, J., & Metcalfe, N. B. (2008). Expe-rience-induced preference for short-sworded males in the green swordtail, Xiphophorus helleri. Animal Behaviour, 76(2), 271-276.

Wagner, W. E. (1998). Measuring female mating preferences. Animal Be-haviour, 55(4), 1029-1042.

Witte, K., & Nöbel, S. (2011). Learning and mate choice. In Fish cognition and behavior. Blackwell Publishing.

55

SVENSK SAMMANFATTNING

Sexuell selektion har en stor påverkan på evolutionen av olika egenskaper och beteenden som påverkar konkurrenskraft vid parning. Under de senaste fyra decennierna har det skett en kraftig ökning av studier på olika aspekter av sexuell selektion, som bekräftar och bygger vidare på Darwins tidiga upptäckter. Dock kvarstår frågan om varför, trots kraften av sexuell selektion, en stor variation ofta återfinns i sexuellt selekterade egenskaper och beteenden mellan och inom arter. Flera faktorer har lagts fram som potentiella förklaringar till bibehållandet av denna variation. En av dessa faktorer som visat sig ha effekt är skillnader i miljö mellan individer och arter. Till exempel har man sett att predationstryck och födotillgång påverkar beteenden och val som individer gör vid parning. Ett annat exempel är inneboende egenskaper hos individen, som till exempel hälsotillstånd, erfarenhet och ålder, vilka även verkar interagera med de externa miljöfaktorerna när parningsbeteenden och partnerval formas. En annan inneboende egenskap som tidigare föreslagits kunna påverka partnerval och beteende är kognitiv förmåga; förmågan att inhämta, bearbeta, minnas och använda information. Kognitiv förmåga är viktig vid bedömning av en potentiell partner, eller för att avgöra vilket beteende som är lämpligt i en viss situation. Trots detta har effekten av kognitiv förmåga på olika aspekter av sexuell selektion, såsom partnerval och parningsbeteenden, fram tills nu förblivit relativt outforskad.

Det nuvarande kunskapsläget rörande förhållandet mellan kognitiv förmåga, partnerval och parningsbeteenden bygger på jämförande studier mellan arter, där hjärnstorlek och specifika parningsbeteenden har korrelerats. Hjärnstorlek i förhållande till kroppsstorlek (relativ hjärnstorlek) har visat sig vara en god prediktor för kognitiv förmåga; som exempel på detta kan nämnas att människor både har en av de största relativa hjärnstorlekarna och en i genomsnitt

56

mycket hög kognitiv förmåga. Dessa jämförande studier antyder en korrelation mellan relativ hjärnstorlek och förmågan att utföra komplexa parningsbeteenden, som i sin tur ökar parningsframgången. Trots det stora värdet av jämförande studier som hypotesgeneratorer, så kan de inte påvisa kausalitet. Vidare har praktiskt taget inga tidigare studier utforskat sambandet mellan hjärnanatomi, kognitiv förmåga och sexuell selektion utifrån det kräsna könets perspektiv, vilket oftast är honan. Avsikten med denna avhandling är att fylla denna kunskapsbrist genom att empiriskt undersöka den potentiella länken mellan kognitiv förmåga, partnerval och parningsbeteende.

För att empiriskt undersöka dessa samband, har artificiella hjärnstorleksselektionslinjer av guppys, en sötvattenslevande fisk inom gruppen Teleostei, använts i denna avhandling. Selektionslinjerna har skapats genom flera generationers avel på relativ hjärnstorlek. Tidigare studier på dessa selektionslinjer har funnit experimentella bevis för det ofta implicita, men också kontroversiella, antagandet om ett samband mellan relativ hjärnstorlek och kognitiv förmåga, i och med att individer med stor relativ hjärnstorlek presterade bättre än de med liten relativ hjärnstorlek i kognitiva tester. I denna avhandling kvantifierade jag parningsbeteende och parningspreferenser hos både hanar och honor av guppys, och fann empiriskt stöd för att hjärnstorlek, genom bättre kognitiva förmågor, spelar en nyckelroll i evolutionen av sexuella karaktärer, sexuella beteenden och partnerval.

I experiment med binära partnerval baserade på visuella signaler hittades skillnader mellan storhjärnade och småhjärnade guppys i deras skattning av partnerkvalitet. Guppyhonor exponerades för olika par av hanar, där hanarna skilde sig markant i hur färggranna de var. Denna metod valdes eftersom hanens färgteckning är en tydlig indikator av hanens kvalitet hos guppys. I denna studie fann jag att storhjärnade guppyhonor tydligt föredrog mer färggranna hanar, samtidigt som småhjärnade guppyhonor inte uppvisade någon preferens för mer färggranna hanar (Artikel I). Därefter genomförde jag en liknande studie för att mäta hanens preferenser för större honor (Artikel II). Inom denna art korrelerar honans kroppsvikt starkt med antalet ungar som en hona kan producera, vilket antyder att hanliga preferenser för större

57

honor kan medföra fitnessfördelar för dessa hanar. Till skillnad från honornas preferenser för färgstarka guppyhanar, fann jag en generell preferens för större honor hos hanarna. Men jag fann inga skillnader mellan storhjärnade och småhjärnade hanar i deras preferenser. I denna studie exponerades hanar för flera par av honor med alternerande små, medelstora, och stora skillnader i honornas kroppsstorlek, och jag fann att storhjärnade hanar ökade preferens för större honor när skillnaden i honornas kroppsstorlek ökade. Denna upptäckt antyder ytterligare en fördel med att ha en stor hjärna; en bättre förmåga att göra optimala partnerval. Det är värt att notera att dessa resultat inte påverkats av skillnader i fiskarnas synförmåga. Hanarnas förmåga att uppfatta färg skiljde sig inte mellan storhjärnade och småhjärnade guppys (Artikel I). Resultaten kan heller inte förklaras av skillnader mellan storhjärnade och småhjärnade hanar och honor vad gäller synskärpa, och därav deras förmåga att se skillnad på potentiella partners (Artikel III). Det faktum att synförmågan inte skiljer sig mellan selektionslinjerna antyder att optimering av partnerval drivs av mer komplexa kognitiva processer i hjärnan, snarare än begränsningar i förmågan till informationsinhämtning.

I denna avhandling kvantifierade jag även parningsbeteende hos guppyhanar som selekterats för liten och stor relativ hjärnstorlek, men fann inga skillnader i beteenderepertoar mellan storhjärnade och småhjärnade hanar i experiment i enkla sociala miljöer med bara en hona och en hane (Artikel IV). Detta resultat tyder på att komplex kognition inte är nödvändigt för de normala parningsbeteenden som guppyhanar utför. Vidare utförde jag djupare analyser, i mer socialt komplexa situationer, av hanars parningsbeteende under predationsrisk och olika antal konkurrerande hanar (Artikel V). Till skillnad från i konkurrensfria scenarion fann jag ett minskat parningsbeteende mot honor, och minskat dominansbeteende mot andra hanar, hos storhjärnade hanar vid högre konkurrens. Detta resultat tyder på en avvägning mellan energin som läggs på en utökad hjärnstorlek och de höga energimässiga kostnaderna av att utföra parningsbeteenden och dominansbeteenden. Däremot verkar inte denna minskning av parningsbeteende leda till en minskad tillgång till parningstillfällen för

58

storhjärnade hanar, vilket kan betyda att storhjärnade hanars beteenden är mer effektiva. I Artikel V undersökte jag även honornas parningsbeteende vid exponering för olika nivåer av predationsrisk och olika antal konkurrerande honor. Storhjärnade honor hade större flexibilitet i deras uppvisade mottaglighet gentemot hanarna vid hög predationsrisk. En sådan flexibilitet kan bidra med fitnessfördelar för storhjärnade honor under naturliga förhållanden, eftersom överdrivet parningsbeteende med hanar under hög predationsrisk kan minska överlevnadschanserna kraftigt.

Sammanfattningsvis visar min avhandling att hjärnstorlek och kognitiv förmåga är tätt kopplade till parningspreferenser och parningsbeteende. Mina resultat kommer förhoppningsvis att öppna upp för liknande innovativ forskning på detta område hos andra arter. Sådan forskning kan öka vår generella förståelse för evolutionen av sexuellt selekterade egenskaper, parningsbeteenden och parningspreferenser.

59

RESUMEN EN ESPAÑOL

La selección sexual es un proceso de gran importancia en la evolución de caracteres y comportamientos que ofrecen ventajas adaptativas en la competencia por el acceso a la cópula. La explosión de estudios sobre selección sexual en las últimas cuatro décadas han confirmado y desa-rrollado en gran medida las hipótesis postuladas por Charles Darwin hace siglo y medio. Sin embargo, una importante incógnita sin resolver es por qué a pesar de la fuerza evolutiva que la selección sexual provee hacia ciertos caracteres, todavía hoy se puede observar gran variación en el desarrollo de tales caracteres entre especies e individuos. Estudios previos han demostrado que factores ambientales tales como el grado de depredación o la disponibilidad de recursos son importantes modu-ladores de las decisiones y comportamientos que los individuos realizan en el ámbito sexual. A su vez, factores intrínsecos de los organismos como la condición interna, la experiencia o la edad, interaccionan con estos factores ambientales y contribuyen a determinar tales comporta-mientos y preferencias sexuales. Otro factor intrínseco que puede tener una gran importancia en la modulación de comportamientos es la habi-lidad cognitiva, es decir, la habilidad para adquirir, procesar, memorizar y usar información. La habilidad cognitiva tiene una importancia fun-damental en la toma de decisiones que llevan a realizar diferentes com-portamientos o preferir ciertos individuos en el proceso de elección de pareja. Sin embargo, la influencia de la habilidad cognitiva en el estudio de la selección sexual es hasta nuestros días prácticamente desconocida.

Nuestro conocimiento actual de la relación entre habilidad cogni-tiva y selección sexual proviene de estudios comparativos entre espe-cies, en los cuales se correlaciona el tamaño del cerebro con ciertos comportamientos sexuales. El tamaño del cerebro es un indicador fiable de la habilidad cognitiva, tal y como ejemplifica el hecho que nuestra especie posea una de las mayores capacidades cerebrales (respecto al

60

cuerpo) de entre todos los vertebrados. Estudios comparativos previos han demostrado la correlación existente entre el tamaño del cerebro y la habilidad para realizar comportamientos sexuales más complejos, lo que a su vez incrementa el éxito en la competencia por el acceso al apa-reamiento. Sin embargo, la principal limitación de tales estudios corre-lativos es la de no poder demostrar causalidad. A esta limitación se añade el hecho de que no hay estudios previos que exploren la relación entre la habilidad cognitiva y la selección sexual en la elección de pa-reja. El objetivo de esta tesis es investigar, por primera vez de forma experimental, la asociación existente entre el tamaño del cerebro, la ha-bilidad cognitiva y la selección sexual, tanto desde la perspectiva de los comportamientos que se traducen en mayor éxito reproductivo como desde la perspectiva de su efecto en la elección de pareja. Para llevar a cabo el primer estudio experimental para comprobar esta asociación, se utilizaron líneas de guppys que han sido seleccionadas artificialmente por su tamaño de cerebro. Previamente, se demostró experimentalmente el nexo existente entre el tamaño del cerebro de los guppys y su respec-tiva habilidad cognitiva, utilizando pruebas de evaluación de capacidad cognitiva. Esta tesis avanza esta línea de investigación a través del es-tudio de comportamientos y preferencias sexuales en individuos selec-cionados artificialmente por su tamaño de cerebro.

En primer lugar, se realizó un estudio que demostró diferencias entre guppies de cerebro grande y de cerebro pequeño en la valoración óptima de la calidad de potenciales parejas, usando pruebas de prefe-rencia sexual basadas en referencias visuales. En estas pruebas, las hembras de guppy fueron expuestas a parejas de machos con diferencias importantes en sus patrones de coloración, ya que la coloración es un indicador importante de la calidad de un macho en esta especie. Los resultados de este estudio muestran que las hembras de cerebro grande tienen preferencias significativas por machos con mayor cantidad de coloración en el cuerpo, mientras que las hembras de cerebro pequeño no muestran dichas preferencias (Artículo I). En un posterior estudio, se evaluaron las preferencias de los machos seleccionados artificial-mente por el tamaño corporal de la hembra (Artículo II). En esta espe-cie, el apareamiento con hembras de mayor tamaño puede suponer una

61

ventaja adaptativa notable ya que existe una correlación manifiesta en-tre el tamaño de la hembra y su fecundidad. A diferencia de lo obser-vado en las preferencias de las hembras, los resultados de esta prueba no mostraron diferencias entre machos de cerebro grande y pequeño en sus preferencias sexuales por hembras de mayor tamaño. Sin embargo, en este estudio las preferencias de los machos fueron evaluadas con más detalle ya que cada macho fue expuesto ante tres tipos de diferencias de tamaño corporal entre hembras: grandes, medias y pequeñas. Única-mente los machos de cerebro grande incrementaron significativamente su preferencia por las hembras de mayor tamaño a medida que incre-mentaba la diferencia en tamaño corporal entre las hembras a las que fueron expuestos. Este hallazgo sugiere un nuevo beneficio de poseer un cerebro grande a la hora de tomar decisiones de carácter sexual. Es importante recalcar que dichos resultados no se vieron condicionados por sesgos perceptivos resultantes de la selección artificial que hubieran afectado la capacidad visual de los individuos. Una investigación deta-llada de la capacidad visual mostró de hecho que no existen diferencias entre las hembras de cerebro grande y pequeño en su capacidad de di-ferenciar colores (Artículo I), ni en la agudeza visual de hembras y ma-chos seleccionados artificialmente (Artículo III). Estos resultados sugieren que las diferencias observadas entre individuos de cerebro grande y pequeño en la toma de decisiones no son debidas a procesos cognitivos simples relacionado con la adquisición de la información, sino a procesos de mayor complejidad cognitiva en la posterior evalua-ción de la misma.

La cuantificación del comportamiento sexual de guppys macho seleccionados artificialmente por tamaño de cerebro no mostró diferen-cias en el repertorio sexual que los machos de cerebro grande y pequeño usan cuando son expuestos a una situación sencilla, sin competencia con otros machos y ante solo una hembra (Artículo IV). Este resultado sugiere que en el desarrollo de la serie de comportamientos sexuales observada en esta especie, no son necesarios procesos cognitivos com-plejos. A su vez, durante el desarrollo de esta tesis se cuantificó el com-portamiento sexual de los guppys macho ante diferentes escenarios de presión de depredación y de número de competidores (Artículo V). A

62

diferencia del escenario sin competidores, los machos de cerebro grande expuestos a competidores tanto a bajos como a altos niveles de depredación mostraron una reducción en el número de cortejos y mues-tras de dominancia hacia otros machos. Este resultado sugiere un com-promiso entre el alto requerimiento energético de desarrollar un cerebro mayor y a su vez la alta demanda energética de estos comportamientos sexuales. Sin embargo, tal reducción en el cortejo no se vio reflejada en una reducción en el interés sexual que las hembras mostraron por los individuos de cerebro grande, sugiriendo así un cortejo más eficaz en los machos de cerebro grande. En el artículo V de la tesis, también se evaluó el comportamiento sexual de las guppys hembra ante diferentes escenarios de peligro de depredación y número de competidores. Las hembras de cerebro grande presentaron mayor flexibilidad a la hora de modificar su receptividad hacia los machos en situaciones con mayor peligro de depredación. Tal flexibilidad puede proveer mayor eficacia biológica a las hembras en condiciones naturales, ya que las hembras sufren gran riesgo de depredación durante la actividad sexual.

En conclusión, los hallazgos realizados durante el transcurso de esta tesis aportan evidencia empírica de la estrecha asociación existente entre el tamaño del cerebro, la habilidad cognitiva, y las preferencias y comportamientos sexuales. Es muy probable, dado el innovador enfo-que utilizado, que esta tesis siente las bases para nuevos estudios en esta línea de investigación que permitan expandir nuestro conocimiento en el importante papel que esta asociación juega en la enorme variación observada en la naturaleza en cuanto a caracteres, comportamientos y preferencias sexuales.

63

ACKNOWLEDGEMENTS

First and foremost, Niclas, thanks for hosting me in your lab in my first contact with research during my master studies. It is thanks to that ex-perience that I decided to pursue a career in academia. I am very grate-ful that you gave me the opportunity to continue in this direction and trusted in my possibilities to bring this thesis forward. Moreover, as a supervisor it is very difficult to find something that you could have done better. You have always provided all the necessary tools for the work involved in each project, from buying extra on all the equipment needed for the lab, to facilitating the contact with other researchers. When it comes to experimental design you have always given me independence but guided me into the right direction making me see so many times that I was overcomplicating things. Despite your heavy work load, you have had the patience to always sit with me for as long as necessary, never let me go unprepared to any presentation, and return manuscripts and other drafts so fast that I did not have time to change the date in the new version (and yet you had the time to give me key suggestions and correct all my grammatical mistakes in them). These are only a few ex-amples of how much I have enjoyed working with you for the last five years and the problem that you have created when I will compare you to future bosses.

The work presented in this thesis would have never materialized without the development of the guppy brain artificial selection lines. Thanks to all people involved over the years and especially to Alex Ko-trschal who together with Niclas Kolm led such a great enterprise. Alex, I have learned so much from your advice on all the tasks involved in research, from working in the lab all the way to writing and presenting our findings, thank you so much for sharing with me your impressive knowledge and always directing me to the right solution. Moreover,

64

your passion and commitment to science has always been a great exam-ple for me and has enormously contributed to how much I have enjoyed working in all these projects.

I am indebted to two people whose contribution to this work have made my life much easier and without them around this thesis would probably look very different. Wouter, despite you have to deal with your own PhD problems, you always had the time to help me out and more often than not come with solutions that save me days in trying to find the solution myself. To a large extent this refers to your incredible skills using R, but also applies to your general willingness to help me out in all other issues such as experimental design and lab work. Sever-ine, at the beginning of the PhD I was struggling to balance all the work needed to keep the fish happy with all the work required to put scientific findings in a piece of paper. That completely turned when you started working in the group thanks to your great skills in having everything organized and all the hours that you devoted yourself to keep the lab and fish in perfect conditions. I have always been impressed by your capacity to do that work and still being able to do excellent research projects of your own.

Despite of the conventional use of “I” in the text of the compre-hensive summary of the thesis, science is a team work and I have had the luck to collaborate with very hard-working and great colleagues over the years. In addition to the abovementioned people, thanks to Ju-dith, Natasha, Maddi, Maksym and Simon for their important contribu-tion to some of the studies presented in this thesis. I would also like to thank Simon for kindly translating the thesis summary to Swedish.

I have found an incredible support over the years with many peo-ple working in our research group. It has been very easy and productive to work in this team both inside and outside the lab. Our weekly meet-ings have provided me a great opportunity to discuss the progress of the work and get excellent feedback on so many different aspects. I am es-pecially thankful to all those life-saving advice I have received before I started some project that make me realize some of the problems with my planning. Thanks for all this to former and current members of the

65

group: Henrik, Steph, Gabriella, Alex H., Teddy and Masahito. Like-wise, I would like to thank other collaborators outside Stockholm Uni-versity, Alexander S., David, Kristiaan, and Maksym, for their always welcome non-biologist input in this work. I am glad that I have been able to regularly participate in the meetings of the KAW project that you are part of, which has been very important to apply technologies of fish tracking in this project. In addition, I have benefited from discuss-ing with colleagues working on fish (or that used to work on fish) in the Department, especially if this involved learning from people with great experience in guppies, but also with colleagues working in non-guppy fish. Thanks for this to Alessandro, Mirjam, Simon and Will whom I have had more interaction over the years, but also to all colleagues in the Zoology fish labs.

The Zoology Department have provided an excellent environment to progress in my PhD studies. Thanks to everyone in the Department for insightful talks, organizing invited talks, workshops, journal clubs and departmental meetings. Thanks to the members of my follow-up committee, Olof Leimar, Sven Jakobsson and Dick Nässel, for their in-put on the progress of the thesis. Thanks to Bertil Borg for your valuable input in the articles of the thesis. To John Fitzpatrick for your great ideas on how to present the articles in a much nicer way. And to Björn Rogell for always having some time to share with me your impressive knowledge on stats and showing me the right path to present the results in the articles. During this years I have taken a few PhD courses which I have found extremely important for my academic formation. Thanks to Karl Gotthard and Roger Karlsson for your work in the organization of such courses. I would also like to thank all the people managing and working in Tovetorp field station for the great environment that they provided while I was assiting in the Ethology master course and for the great organization of our annual departmental meeting which I have al-ways enjoyed to be part of. Thank you to Siw Gustafsson, Annete Lorents, and Minna Mietinen for solving all the important logistical and administrative paperwork so effectively and Ulf Norberg for always helping me out with computer issues. The easy day to day that I have working on the Department was to a large extent thanks to a great work

66

of the people in charge of running it. Thanks to Birgitta Tulberg and Linda Laikre for their work in this respect. The loss of Birgitta is now very recent in my memory and despite our strictly professional relation-ship I will always consider her as a role model for her brilliance and her competence in the difficult task of heading a Department.

Despite this is quite often ignored in my home country, basic re-search benefits from financial support. For that I would like to acknowledge Vetenskapsrådet and Knut och Alice Wallenbergs Stiftelse who provided the support to the work in this thesis through funds granted to my supervisor N. Kolm. I likewise would like to acknowledge the funds received during the course of my studies to as-sist to international conferences in respective travel grants by Knut och Alice Wallenbergs Stiftelse, and Liljevalchs och Augustinssons Stiftelse. I would like to thank the Department of Zoology for comple-menting this financial support to assist to conferences through the funds allocated for PhD students. Finally, thanks to the Kungliga Vetenskap-sakademien for providing the necessary funds to cover an external re-search project in which I collaborated during the course of my studies.

I am grateful to David Outomuro for his important role in expand-ing my experience in research outside the boundaries of what a large- or small-brained guppy does. I am glad that helping out a friend with some analyses of a few damselfly videos led to a bigger collaboration and took me out of the fish lab for a month all the way to the Colombian piedmont. I have learned many things from your extensive scientific knowledge but also I have enjoyed a lot our time together both in Co-lombia and every time we met in Uppsala.

In a less formal note, I would like to thank to all my friends in the Department for creating such a nice environment that often felt difficult to associate with a workplace. Many of you have already been men-tioned above for your important impact in this work, but your impact in my life outside science has likewise being very important. Thanks for the great moments that we have shared within the cold department con-crete walls, but also every moment and laugh that we have shared after

67

work mostly with a beer in our hand at parties, dinners, concerts, con-ferences and more importantly in a bathtub. It is my laziness to com-mute also during the weekends the one to blame that these occasions did not happen as often as I would have liked to. Thanks also to all my friends in Uppsala, you all have been a great support every time I needed help and I have enjoyed so much celebrating everything with you when we had a good excuse for it and when it was not needed one. I am particularly grateful for having found such a strong connection with some friends here. Thanks Ana, Carmen, Fernando, Guille, Jose and Laura for always being there and making Uppsala feel like being in Spain at times, until they bring the bill or we look at the weather fore-cast.

I am so grateful for the endless love and support that I received from my family who have always showed me how proud they are of my accomplishments, no matter how small they were or my difficulties to explain them what I do in my mother tongue. I am indebted to Alba and Jose for your willingness to help me out with all aspects of this work that could benefit from your creativity. Thank you so much for all the help and in particular for the artwork in the cover of the thesis. Thank you to my parents for basically everything that you have done for me, but in particular for in that day a few years back, not trying to convince me that resigning a well-paid job in Madrid and coming to study a Mas-ter in Biology in Uppsala might not be a very bright idea. Finally, thank you María for sharing all this experience with me, you have been the perfect travelling companion in all these years. Your help has been so important, both directly, when I needed it in the lab or with your sug-gestions for improvements, and indirectly, by taking care of everything else when I needed to focus on the work. More importantly, your con-stant love, support and encouragement have been fundamental to achieve the goal of finishing this thesis, a goal that you will soon make your own and I hope to be there for you in the same way. I am so glad for our mutual mating decisions and for that by the time this thesis is already printed we will be our own family.