!
"# $

%&##


%# '       '
 
  
(
"#
)&

    
  
   
 
 
*  
 
 
 


 
 

   &   
  
 
+ 
  


 
  


 

 
  '

 
   
  '
 '
 

,  
  

  
   &
- 
  


   
   '





  
*  

  '
 
   

  .
   &
/ * '

 
  


  
  
 *
  

  


*
 &
0    



  


 


   

  
 
 &
-
* 
 
  

  1

  &
 
 
, 


  
  
 
   


  
  
  
 
  

 
  &
2
 
 


 
 
  '
 
 
* 
   

3 
 .

, 



  

  


 
  


 

 
  &
  

  


 
  
 ,  
 & 2

 '
2
  

   
 *
 
  

  
 

  
 *'
 
 &
-
 

   
* 
  
  
     '

  
  
 

 
  &
2
 *


 

  '
,  '
  


 
   '   

.


   
  
4


5'


  



 
  
 
* 


6
4 
6'

5


, 

*  

 
4
5&
7   
 
   
  


  

   '
* 

  
*




, 



4
5&
8
   
 
 

 



  
    '
2
 
 *


 4"

   5


 
 

  
+ 


*
 
 

  '
,   
  
4
5&
8
,  




*
 
   

%
* '
2
 *


   
 

  
4
5&
 '
2
 *


 
 
 
+ 
 

  



  
4
5&
  
   



 
   


  ' 
 

   
   

  

   
  

  


 *&
2 '

9   '


 

   '
 

  

  
    &
0 '

 
  
  1

  
  

  '

   


 
  
 
  
  
,  
  

 
  &

  
  
  
  
 


 
  
    
     
  
 
 
 


  
"# $  :;; && ;  < = :: : :  >"%>?

28@
A$?A $ >A?%?B 28@
A$?A $ >A?%A"

 

 !

  
  '
#
A
  

THE RELATIONSHIP BETWEEN PERSONALITY AND SOCIAL DOMINANCE IN THE DOMESTIC FOWL – A CRITICAL PERSPECTIVE

Anna Favati



The relationship between personality and social dominance in the domestic fowl – a critical perspective

Anna Favati



 Contents

List of papers………………….…………………………………………….ix Additional articles published during my doctoral studies ...... x Abstract ...... xi Introduction ...... 13 Social dominance ...... 14 Animal personality ...... 16 The aims of the thesis ...... 19 Methods...... 21 The study system: the domestic fowl ...... 21 Behavioural assays ...... 24 Results in short ...... 29 Paper I ...... 29 Paper II ...... 31 Paper III ...... 33 Paper IV ...... 33 Paper V ...... 35 Discussion ...... 39 Social status as a stable state ...... 39 Personality predicting dominance ...... 43 Recent contest experience ...... 46 Losers ...... 47 Concluding remarks ...... 48 References ...... 50 Sammanfattning på svenska ...... 59 Acknowledgements ...... 62



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

I Favati A, Leimar O, Radesäter T, Løvlie H. 2014 Social status and personality: stability in social state can promote consistency of behavioural responses. Proceedings of the Royal Society B. 281:20132531.

II Favati A, Udén E, Leimar O, Løvlie H. Personality remains: no effect of three-week social status experience on personality in male fowl. Submitted Manuscript.

III Favati A, Leimar O, Løvlie H. 2014 Personality predicts social dominance in male domestic fowl. PLOS ONE. 9:e103535.

IV Favati A, Løvlie H, Leimar O. 2017 Individual aggression, but not winner-loser effects, predicts social rank in male domestic fowl. Behavioral Ecology. In press. doi:10.1093/beheco/arx053

V Favati A, Løvlie H, Leimar O. Effects of social experience during development on competitive ability and personality traits in male domestic fowl. Manuscript.

Candidate contributions to thesis articles*

I II III IV V

Conceived the      study

Designed the study     

Collected the data     

Analysed the data     

Manuscript preparation     

* 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.

 The following articles were written during the course of my doctoral studies, but are not included in the thesis:

Rosher C, Favati A, Dean B, Løvlie H. 2017. Relatedness and age reduce aggressive male interactions over mating in domestic fowl. Behavioral Ecology. In press. https://doi.org/10.1093/beheco/arx024.

Zidar J, Sorato E, Malmqvist AM, Jansson E, Rosher C, Jensen P, Favati A, Løvlie H. 2017. Early experience affects adult personality in the red junglefowl: A role for cognitive stimulation? Behavioural Processes 134:78–86.

Favati A, Zidar J, Thorpe H, Jensen P, Løvlie H. 2016. The ontogeny of personality traits in red junglefowl. Behavioral Ecology 27:484–493.

Olofsson M, Wiklund C, Favati A. 2015. Ontogenetic change in a bird's response to a butterfly prey with eyespots. Current Zoology, 61:749-757.

  Abstract

Social dominance relationships are formed within numerous animal species and reduce costly fights over resources. Dominant individuals often enjoy greater access to important resources such as food and mating partners, and are generally more aggressive, bold, active and explorative compared to subordinate individuals. These behavioural traits can differ among individuals, but they can also be consistent within the individual, thereby describing the individual’s personality type. However, the causal direction of the observed correlation between dominance and personality is not well studied. One possibility is that some personality types have higher chances of obtaining a dominant social position. This would suggest that personality has consequences for fitness. Another possible explanation is that possessing different social positions gives rise to consistent behavioural differences among individuals on various timescales. If social status has a lasting effect on behaviour, social status would constitute a ‘stable state’ that explains some of the diversity of personality types that has been observed in a multitude of animal species. Dominance and personality may also share underlying proximate factors. In this thesis, I investigate the relationship between social dominance and personality using male domestic fowl, Gallus gallus domesticus. The species is group-living with pronounced dominance hierarchies, and dominance increases male access to mating partners. I show that some aspects of personality, exploration, vigilance and in particular aggressiveness, increased a male’s chances of obtaining dominance (paper III, IV, V), and that aggressiveness can be even more important than body weight and ornament size (comb size, paper V) or recent experience of winning or losing (paper IV). Winning a social interaction resulted in an increase in aggressiveness, while a decrease was seen in males that experienced a loss (paper IV). By observing behaviour before and after changes in male dominance relationships, I further show that a recent (2 days earlier) change in social status induced behavioural adjustments to the new social status in activity, exploration and vigilance (paper I). By extending the time of the new social relationship to 3 weeks, I show that such behavioural changes did not continue (paper II). Finally, I show that the social environment during juvenile development had little impact on adult male competitiveness (paper V). Social interactions appear to have several short- term effects on behaviour, but did not contribute significantly to variation and long-term consistency of personality in male fowl. Instead, a male's personality, and in particular his aggressiveness, affected the outcome of dominance interactions. Overall, my studies reveal important consequences of individual differences in behaviour, and contribute to the highly sought- after empirical testing of hypotheses explaining variation in animal personality.

   Introduction

In many animal species, social dominance relationships are formed when the outcome of agonistic interactions repeatedly end in favour of one of the participants. The winner is thereby defined as being dominant over the losing subordinate individual(s). Dominant individuals typically have priority access to important resources such as food and mating partners (Andersson 1994; Hardy & Briffa 2013). Dominant and subordinate individuals commonly differ quantitatively in behaviour: dominant individuals are often more aggressive, bold, active and explorative compared to subordinates (Colletér & Brown 2011; Verbeek et al. 1996). These behavioural traits have often been found to differ between individuals, and to be relatively consistent within the individual, thereby describing the individual’s personality type (Dall et al. 2004; Sih et al. 2004; Réale et al. 2007). However, the causal direction of the observed correlation between dominance and personality has received little research attention. This may partly be due to the fact that animal personality is a young research field where the large part of the causes and consequences of personality are yet to be discovered, but also because behavioural differences have received little attention during the long history of investigating factors that are important for obtaining dominance. Recent empirical data indicates that some personality types have higher chances of obtaining a dominant social position (e.g. Verbeek et al. 1996; Dahlbom et al. 2011). This would explain the observed correlation between personality and dominance, and further, suggest that personality has fitness consequences. Another possible explanation that has often been theorised, but less empirically tested, is that possessing different social positions gives rise to consistent behavioural differences between individuals (e.g. Wolf & Weissing 2010; Sih et al. 2015; Dingemanse & Araya-Ajoy 2015). If current social status gives rise to lasting effects on behaviour, this could potentially explain some of the variation of personality types that has been observed in a multitude of animal species (Gosling 2001; Carere & Mastripieri 2013). A relationship between dominance and personality may also arise if they share underlying proximate factors (e.g. Øverli et al. 2004; Briffa et al. 2015). By disentangling the relationship between personality and social dominance, several questions regarding causes and consequences of personality can be unravelled. Such

 research can also shed light on the longstanding question of what makes a winner.

Social dominance

Aggressive interactions are often driven by competition for limited resources such as food, territories and sexual partners, with a dominant social position typically bringing increased access to such desirable resources (Andersson 1994; Clutton-Brock and Huchard 2013; Hardy & Briffa 2013). Dominance relationships are commonly established through pair-wise interactions resulting in a dominant winner and a subordinate loser. Once the dominance relationship is established, aggression typically decreases. This lowers the intensity of fights and risk of injury, which suggests that subordinate individuals can also gain from the arrangement (Ricci et al. 2013).

Animal contest research contains a large body of literature investigating the pathways of decision making in agonistic interactions, for example when to continue and when to withdraw, and what factors underlie these decisions (e.g. Hardy and Briffa 2013). Disparities in physical attributes (such as body size), information and resources have often been used to explain behavioural decisions of whether to continue fighting or withdraw (Hardy & Briffa 2013). However, physical attributes seldom, if ever, entirely predict social dominance. Intrinsic behavioural qualities that vary among individuals, i.e. personality traits (Sih et al. 2004; Dall et al. 2004; Réale et al. 2007), may additionally play a role in the establishment of dominance hierarchies. Aggressiveness has a central role in dominance interactions, since conflicts are settled through aggressive interactions, and aggressiveness has consequently been found to predict and correlate with dominance (Verbeek et al. 1996; Kim & Zuk 2000; Kralj-Fiser et al. 2010; McGhee & Travis 2010; Ricci et al. 2013). Other personality traits found to predict dominance hierarchies include boldness, activity and exploration, each of which is generally positively associated with dominance (Kralj-Fiser et al. 2010; McGhee & Travis 2010; David et al. 2011). However, there are also examples of negative or missing relationship between these traits and dominance (Gomez-Laplaza 2002; Fox et al. 2009; Riebli et al. 2012; Taylor & Lattanzio 2016), suggesting that the relationship between personality and dominance may be species- or context-specific. When personality traits affect the probability of obtaining a dominant position, this constitutes a link between personality variation and fitness that could contribute to our understanding of the evolution of personality types.

  Besides individual qualities, earlier experience of social interactions can also influence both behaviour and the outcome in agonistic interactions. An individual may have to fight several opponents after one another, especially in group-living species, for example when transferring from one flock to another (Collias and Collias 1996). In general, winners become more aggressive while losers become more avoidant, influencing their chances of winning future interactions (Chase et al. 1994; Hsu et al. 2006). Such ‘winner- and loser effects’ contribute to the formation of dominance hierarchies (Dugatkin & Druen 2004; Dugatkin & Reeve 2014), and are possibly due to re-assessment of the own fighting ability (Rutte et al. 2006; Arnott & Elwood 2009; Hsu et al. 2009). The impact of recent contest experience on contest success has been compared with the importance of morphological traits (e.g., body size, Beaugrand and Cotnoir 1996; weight, Martin et al. 1997), whereas aggression and other personality traits have not been compared with experience to the same extent. Since personality may contribute to the outcome of duels, different personality types are likely to experience different ratios of wins and losses under natural conditions. It is therefore of interest to disentangle winner-loser effects from intrinsic behavioural qualities. Winner- and loser effects are well documented in a range of species, but they are considered to be fairly transient, generally dissipating within a few days or even hours (e.g. Chase et al. 1994; but see Lan & Hsu 2011). However, winner-loser studies generally investigate the impact of one or a few interactions, with the participants separated shortly after the winner has been identified. In social species, the same individuals often interact repeatedly, even after a winner is obtained. These agonistic interactions often end in favour of the same (dominant) individual, thus dominance hierarchies are formed. Members of solitary species are also likely to experience a number of interactions with the same opponents, when territory boundaries border or overlap. Repeated social interactions have the possibility to exert long-term impact on future competitiveness and other behaviour (Hsu et al 2006; Laskowski et al. 2016; Trannoy et al. 2016), but whether they may also change or shape personality has not been thoroughly investigated. Intuitively, the amount of time spent in a social relationship should be positively related to the impact on behaviour, and possibly even more so, if social relationships are already established at a young age (Sachser et al 2011; Laskowski et al. 2016; Bölting & Engelhardt 2017). In this thesis, I investigate the impact of dominance relationships on individual behaviour on various timescales, from single fights to year-long experiences across development.

 Animal personality

Personality in animals is commonly described as behavioural differences between individuals within a population, that are consistent across time or context (Dall et al. 2004, David & Dall 2016). This has also been termed temperaments (Réale et al. 2007), coping styles (Koolhaas et al. 1999) and behavioural syndromes (Sih et al. 2004). Put slightly differently, animal personality is a property of the population, where a significant proportion of the variation in behaviour is due to differences between individuals, compared to the variation within individuals, which can be quantified as the ‘repeatability’ of a behavioural trait (Réale et al. 2007; Nakagawa & Schielzeth 2010). Such consistent individual differences in behaviour have been observed in numerous animal species in various taxa, from insects and fish to birds and mammals (Gosling 2001). The interest in animal personality has increased tremendously over the last two decades, and has developed into a popular research field in behavioural ecology (Carere & Mastripieri 2013; David & Dall 2016; Hedrick 2017). The traditional view of animal behaviour has been that animals are selected to perform adaptively and optimally in each situation, and that behavioural flexibility permits this. Since widespread observations that animals show among-individual consistency in behaviour, there has been a shift in how we view behavioural variation. Behavioural differences between individuals are no longer seen as more or less uninteresting deviations from an evolved optimal mean, but instead studied with great interest (Wilson 1998). That individuals consistently respond differently from each other to the same situation, for example predator risk, is potentially important for fitness and raises question about why directional selection has not erased variation, giving rise to one ‘perfect’ personality type, or why individuals are not instead infinitely flexible in their responses (Sih et al. 2004). Animal personality offers a holistic view on behaviour by investigating correlatiosn among behavioural traits, and how behavioural traits co-vary with various other aspects of animal biology. The field aims to determine how and why individuals vary in behaviour, the ecological relevance of personality variation, and to explain how these differences can evolve (Wolf & Weissing 2010; Sih et al. 2012; David & Dall 2016; Hedrick 2017). Although animal personality is currently a very popular topic in animal behaviour research, some criticism has been raised. Terminological and methodological inconsistencies may have contributed to the field’s somewhat vague or confusing appearance (Réale et al. 2007; Carter et al. 2013), but more serious criticism has also been put forward, for example the lack of a robust conceptual framework (David & Dall, 2016). Moreover, it has been argued that there is no need to define variation and consistency of behaviour as a research field of its own, since variation in behaviour and its ecological consequences has since long

   been studied within behavioural ecology (Beekman & Jordan 2017). Variation provides the raw material evolution acts on, and intraspecific variation has therefore also more traditionally been appreciated within behavioural ecology. Nevertheless, the focus has rarely been on understanding variation and consistency at the level of the individual (Roche et al. 2016). Personality researchers therefore consider the concept of animal personality as described above as a useful tool to study variation in behaviour.

Commonly described personality traits include boldness (often synonymous with the willingness to take risks), activity, exploration (of unfamiliar environments), aggressiveness and, to some degree, sociability (Réale et al. 2007; Koski 2014). These traits describe a large part of an individual’s behaviour in many different contexts (e.g., mating, parental care, foraging, dispersal, dominance interactions, Careau & Garland 2012). To some degree, there is a general consensus how these traits should be measured and defined, which eases comparisons among studies (Réale et al. 2007), but there is also room for species-specific method uses and choice of responses. For example, ‘boldness’ is commonly used to describe risk-taking behaviour, that can range from latency to re-open tentacles after a startle (sea anemones, Actinia equina, Rudin & Briffa 2012), to time spent out of a shelter (zebra fish, Danio rerio, Dahlbom et al. 2011), latency to approach a novel object (raven, Corvus corax, Stöwe et al. 2006) or willingness to feed in the presence of a potential threat (threespined sticklebacks, Gasterosteus aculeatus, Bell & Sih 2007). Other behavioural traits are also, but less often, investigated within the framework of animal personality, for example parental behaviour (Albers et al. 1999) and social skills (Sih et al. 2014; Taborsky & Oliveira 2012). Selecting behavioural traits based on the ecology of the species in question, and adjusting the behavioural assays to suit its behavioural repertoire, is crucial to be able to make meaningful interpretation of the results, and to integrate them in a biologically relevant context (Koski 2014).

Personality traits can form ‘behavioural syndromes’, generally defined as correlations among behavioural responses among or within functional contexts (e.g. predation, foraging, mating, Sih et al. 2004; Carere & Mastripieri 2013). A commonly observed behavioural syndrome is that bolder individuals are also more aggressive and more explorative (e.g. Wilson & Godin 2009; Schuett et al. 2011; Santostefano et al. 2016). With inspiration from the coping style literature, these individuals can be described to be more proactive, compared to their reactive conspecifics (Koolhaas et al. 1999). However, the boldness-aggression-exploration syndrome is sometimes missing (e.g. Sinn et al. 2010; Ariyomo & Watt 2012; Favati et al. 2015), or varying results are obtained in the same species

 (Bell & Sih 2007). This indicates that behavioural syndromes are not universal, and may differ among species, populations and situations (Luttbeg & Sih 2010). This might be because behavioural syndromes are adaptive, e.g. behaviour is only correlated in contexts where it is advantageous (e.g. depending on predator pressure, Bell & Sih 2007; Dingemanse et al. 2007, 2010). Behavioural syndromes may otherwise represent genetic and/or physiological constraints, which would mean that individual behaviour may be adaptive in one situation, but have negative spill-over effects to other situations (Sih et al. 2004). For example, highly aggressive female fishing spiders (Dolomedes fimbriatus) successfully attack prey, but are also overly aggressive (cannibalistic) in a mating context (Arnqvist & Henriksson 1997). Further investigation of when and where behavioural syndromes form are necessary in order to understand their existence.

Overall across studies and species, personality shows a degree of heritability (Dochtermann et al. 2015). Several links between personality traits and fitness estimates have been found, indicating that at least some aspects of animal personality are adaptive (Dingemanse & Réale 2005; Smith and Blumstein 2008; Cote et al. 2008; Schuett et al. 2010). For example, and a central theme of this thesis, several personality traits have been found to correlate with social dominance (Briffa et al. 2015). Dominance governs the access to importance resources such as food and mating partners, and thereby survival and reproductive success (Andersson 1994; Clutton-Brock & Hutchard 2013). Despite promising links to fitness, little is known empirically about how personality variation arises or is maintained, and several key questions therefore remain to be answered both at proximate and ultimate levels (Bell 2007). The theoretical hypotheses of how variation in personality arises are plentiful, and often build on the idea that different personality types may have overall equal fitness, at least over a long enough time. Suggested mechanisms include selection regimes that fluctuate over time (Dingemanse et al. 2004), density dependent selection (Cote et al. 2008; Nicolaus et al. 2016), and various trade-offs, for example between survival and reproductive success (Smith & Blumstein 2008) or life-history trade-offs (Wolf et al. 2007; Stamps 2007; Biro & Stamps 2008). State-dependency of behaviour is another process that has been suggested to result in variation in personality (Houston and McNamara 1999, Dall et al. 2004, Dingemanse and Wolf 2010, Luttbeg & Sih 2010, Wolf & Weissing 2010; Sih et al. 2015). This hypothesis predicts that individuals vary in different ‘states’ that alter the costs and benefits of their behaviour (Wolf & Weissing 2010; Sih et al. 2015). This means that individuals of different states should differ consistently in behaviour from each other at least as long as the differences in state remain, resulting in different personality types (Wilson et al. 1994, Dall et al. 2004). Endogenous features such as metabolism and energy reserves have been suggested as potential states that alter behaviour.

   Features of the social environment have also been emphasized as key states (Dingemanse & Wolf 2010, Wolf & Weissing 2010), because of the selective pressure social interactions can have on various behaviours (Dingemanse & Araya-Ajoy 2015). Repeated social interactions have therefore recently been proposed to be a key component towards generating and maintaining individual differences in personality by state-dependency and positive feedback loops between status and behavior (Dall et al. 2012; Briffa et al. 2015; Dingemanse & Araya-Ajoy 2015; Niemelä & Santostefano 2015; Sih et al. 2015). However, this has rarely been empirically tested (but see Laskowski & Bell 2014; Laskowski & Pruitt 2014). Overall, empirical testing of the various proposed mechanisms of the origin and maintenance of personality variation is lacking.

To summarise, the literature on personality research has expanded enormously during the last decade, resulting in a wealth of empirical studies investigating the relationship between personality traits and most aspects of animal biology. There is also a substantial collection of theoretical models, both formal and verbal, aiming to explain the evolutionary mechanisms behind variation in personality types. The key aspects are to understand how variation between individuals arises, how consistency is maintained within an individual, and how the variation of personality types is maintained over evolutionary time. In order to advance the field, empirical testing of specific theoretical predictions is required. In this thesis, I ask if short- and long term experiences of social positions of a dominance hierarchy can give rise to, and maintain individual differences in aggression and other commonly studied personality traits, or, alternatively, if personality variation influences dominance positions.

The aims of the thesis

Given the large impact social relationships can have on behaviour, investigating the interplay between social context and variation in personality can provide insights into proximate and ultimate explanations of animal personality variation. It also provides possibilities to understand the role of personality in shaping social relationships. In this thesis, I investigate the causality of the relationship between social dominance and personality through a series of experimental studies of male domestic fowl (Gallus gallus domesticus). To achieve this, I study the influence of social dominance relationships on individual personality responses over various timescales, which can increase the understanding of why and how individuals differ in personality. I also investigate what behavioural characteristics are important for obtaining the desirable dominant position of

 a flock, and compare it with the impact of other individual characteristics and with the experience of earlier short- and long-term social interactions.

I start with the question of whether having a certain social role, for a shorter or longer period, gives rise to the observed differences in behaviour that we describe as personality (paper I and II). I continue by turning the perspective, and ask whether having a certain personality type affects an individual’s chances of obtaining a dominant social position (paper III). I then investigate factors that influence dominance acquisition, by comparing individual properties with the experience of a recent win or loss (paper IV). Finally, I study whether competitiveness and various personality traits are shaped by social experiences during development, or represent intrinsic properties of the individual (paper V).

  Methods

The study system: the domestic fowl

The domestic fowl, Gallus gallus domesticus, was an early model species for the study of social dominance hierarchies, starting with Schjelderup-Ebbe coining the concept 'peck order' in 1922. The specie’s accessibility as a common domestic species, as well as their clearly hierarchical social structure, likely contributed to the widespread use of domestic fowl in dominance studies in the 20th century.

Figure 1. Left: A group of domestic fowl, free-ranging at Tovetorp research station. Photo credit: Hanne Løvlie. Right: Male ornaments that signals health and dominance status. A large red comb crowning the head, accompanied by earlaps and wattles (below the cheeks).

The domestic fowl origins from the red junglefowl (Gallus gallus, Fumihito et al. 1994), with whom it still shares behavioural repertoire, sexual- and social systems (Kruijt 1964; Schütz & Jensen 2001). I will refer to both as the same species in this species description. As its wild ancestor, the domestic fowl is group living (Fig 1) and a sexually promiscuous species, which results in ample male competition over mating opportunities. Natural groups of wild or semi-wild fowl are typically constituted by a dominant male, a few subordinate males and several females, with group sizes ranging from 2-15, and often female-biased sex ratios (2:3-2:5, males: females,

  McBride et al. 1969, Collias and Collias 1996). When sexually mature (around 20-25 weeks of age, McBride et al. 1969), both sexes wear red fleshy ornaments on the head (comb, wattles and earlaps, Fig 1), that signals health and reproductive status (Zuk et al. 1990; Zuk & Johnsen 2000). These ornaments are enlarged in males, and may also function as an intra-sexual signal, since large-combed males have increased chances of winning dominance interactions, and the comb grows in males that become dominant (Zuk & Johnsen 2000; Parker et al. 2002; Cornwallis & Birkhead 2008). Both sexes form (most often linear) social hierarchies (Banks 1956, Collias et al. 1994). Dominance hierarchies quickly form whenever two unfamiliar fowl meet. The interaction generally starts with ritualised aggressive displays, including taking an aggressive stance with the shoulder turned downwards towards the intruder, crouching body posture, ground pecking (the latter similar to the behaviour of courtship feeding known as ‘tid- bitting’, but without vocalisation), raised hackles, waltzing and attacking (Kruijt 1964, Fig 2). Often, aggressive displays are enough to discourage one of the contestants, leaving the other to obtain the dominant position. In other cases, the conflict escalates to a physical stage with kicks, bites (in particular on the comb and wattles, Fig 1) and chases until one of the contestants surrender. The loser starts behaving in a subordinate fashion, with a lean and upright body posture, making alarm calls and avoiding the winner. The winner often crows, which is a territorial vocalisation (Leonard & Horn 1995).

Figure 2. Two males that engage in an aggressive interaction. Both males show an aggressive posture with crouched body, lowered wings and raised hackles.

  The hierarchy lasts on average one breeding season in free-ranging flocks, but a dominant male can sometimes keep his position for several years (Collias and Collias 1996). This is a considerable part of the lifespan of males, who lives up to 5.5 years in semi-natural flocks exposed to predation (Collias and Collias 1996). Dominant males enjoy increased access to females, and females often choose to copulate with dominant males (Pizzari 2001). Dominant males therefore obtain most matings which results in increased reproductive success (Cheng & Burns 1988; Froman et al. 2001; Dean et al 2010). Therefore, the position in the hierarchy can have evolutionary consequences. However, hierarchies can change more often, for example if the dominant male is injured or predated (McBride et al. 1969).

The study population of my thesis consisted of a population of ‘Old Swedish game fowl’, also known as ‘Swedish bantam’ (‘Gammalsvensk dvärghöna’ in Swedish) at Stockholm University’s research station Tovetorp, Gnesta, Sweden. This breed has the wild type colouration with vividly coloured males and cryptic brown females (Fig 1). The study population have been housed at the research station since the late 60’s without artificial selection for any particular traits. The population size ranged between approximately 90 and 120 birds over the years when the studies were performed (2007- 2015). These birds are habituated to human presence, enabling close observations of natural behaviour and natural interactions. The birds were divided into 6 large aviaries (5*10 meters) with access to indoor roosts. Groups were of mixed-sex and mixed-aged (1-11 years), with a sex ratio around 1:1 or slightly more female-biased, thus resembling natural group sizes and sex ratios of what is observed in the wild.

 

Novel object test Similarly as with the NA, the novel object test (‘NO’) is an often-used personality assay (Forkman et al. 2007). I used supposedly frightening objects that shared some elements with potential predators (eyes, fur), combined with presentation of desirable food (spaghetti or rice, that resembles worms or insect eggs which is part of the diet of free-ranging fowl) on a familiar plate near the object (Fig 4). To ensure that males had registered that there was food on the plate, all males were first singly trained to approach the plate without the object present. The latency to approach within a body length of the food plate while the object was present was used as a measure of boldness. Males that did not approach the food within the set time limit (10 min) were allowed to approach and eat the food after removal of the object, in order to exclude the possibility that the male had lost his interest in the food. All males approached the plate and ate the food.

Figure 4. A novel object test used in paper II. A male approaching food on a plate in front of a novel object; here a toy rat.

Startle test In paper II, a startle test was performed to estimate risk-taking behaviour (Forkman et al. 2007; Favati et al. 2016). A recorded alarm call from a conspecific male was played, which causes males to instantly adopt a vigilant posture with the head held high. Latency to resume feeding, or performing other activities that was interrupted by the startle, was then measured. Males that sooner resumed regular activities were considered to be less vigilant and more risk-taking compared to males that halted their activity for a longer period of time.

  Aggression test Measuring aggressiveness as an individual propensity is not a trivial task. Aggression is not a behaviour an individual can perform on its own, but is by definition expressed towards conspecifics in social contexts (Holekamp & Strauss 2016). Therefore, the appearance and behaviour of the antagonist can affect the behaviour of the focal individual. Several studies have nevertheless estimated individual aggressiveness by behaviours expressed in an aggressive interaction without controlling for the agonistic behaviour of the opponent, such as noting who initiate physical interaction, or simply by stating that the winner of a duel is more aggressive (Banks 1984; Francis 1988; Drickamer et al. 1999; Garcia et al. 2014). Measuring aggressiveness in the very same interaction where the outcome is scored may align with the aggressive tendencies of an individual in general, but there is a problem with circularity in this method if one aims to study the relationship between aggressiveness and dominance. In paper III, individual aggression was estimated by the occurrence of aggressive posture and aggressive approaches or attacks during the first 10 seconds of the duel, in an attempt to estimate initial aggressiveness before the males had evaluated their chances to win against the opponent. To independently estimate individual aggressiveness and prevent circularity, I later developed an aggression test (‘AGG’), where aggression towards a restrained intruder was scored in a standardised way. I used this assay to quantify aggressiveness in paper II, IV and V. Intruder tests are generally used to study the reaction of a resident individual towards an intruder, and the behaviour and outcome of the interaction are used as estimates for aggressiveness (Blanchard et al. 1988; Camerlink et al 2016). The scale I developed was based on often occurring agonistic behaviours, ranging from avoiding (least aggressive) to immediately attacking (most aggressive, table 1). The intruder was matched for comb size (similar or smaller) since a large comb might be deterring to the focal male. Behaviour of the intruding male was restrained by holding him in the hands of the observer (Fig 5). The male was presented for one minute or until the focal male attacked, whichever came first, starting approximately 1 meter in front of the focal male (Fig 5). The assay was performed in the home aviary of the focal male to minimise interference of stress. Focal males behaved similar to how males behave when confronting a free-ranging male (paper III). By repeating the aggression test with several intruders, I could conclude that the intruder identity did not have a significant effect on the focal male’s aggression (paper IV). Further, male initial aggressiveness in a duel correlated with aggressiveness measured in the aggression test (paper III), which strengthen the interpretation that the aggression test accurately measures the same type of aggressiveness that is expressed during free- ranging interactions.

 

Social dominance duels Social dominance interactions of male domestic fowl constitute the core of this thesis. To estimate the competitiveness of males, in other words their chances of obtaining a dominant position, experimentally controlled duels among pre-determined males were performed. I made sure that the two males performing a duel had not been housed together for at least 2 weeks, but often longer, reducing any effects earlier encounters may have had on the establishment of social status (Cloutier et al. 1995). Because several factors are known to influence the outcome of duels, males were matched for body size, comb size and when possible, age. Three to five avoidances were used to define one male as the losing subordinate, and the other as the winning dominant (Cloutier et al. 1996). Duels often lasted only for a few minutes, and physical fighting was often very brief if it happened at all (mean 12 seconds, paper III). In fact, many duels were resolved from start by one of the contestants immediately avoiding the other. In some cases, the contestants mutually avoided each other, and if the experimental setup allowed, the observation period was extended until a clear winner and loser could be distinguished. Once the dominance order was settled, it was very rarely reversed, which justify the use of a short interaction where a clear winner and loser were observed as a robust measure of their dominance relationship. All contests were closely monitored by live observation and the observers were ready to interrupt if one of the males was injured more than minor bleeding from the comb and wattles (Fig 1), or showed signs of fatigue (e.g. breathed heavily). Such interventions were however rarely needed.

  Results in short

Paper I

Social status and personality: stability in social state can promote consistency of behavioural responses

The aim of paper I was to investigate if social status influences behaviour in personality tests, which would indicate that variation in social status may be a state that gives rise to variation in personality. To minimise the impact of earlier dominance relationship, males were isolated from other males for 2-5 days. Isolated males soon start to act in a dominant fashion, including courtship of females and territorial crowing (Wilson et al. 2009). Thereafter I formed pairs of morphologically matched males (comb length, body weight) males (npairs = 42), and let them establish a dominance relationship where one became dominant and the other subordinate. After 2-3 days of acclimatisation to the dominance relationship, behaviour was measured in a novel arena test. Dominant males were observed to be bolder, more vigilant, active and explorative, and they uttered more territorial crows. The social status of males was ‘self-selected’ within the pairs and not experimentally randomly determined (i.e. it was not pre-determined which of the males would become dominant or subordinate in each pair). This means that I, at this stage, could not exclude the possibility that the differences in boldness, exploration, activity or vigilance were present before the staged duels, and that they had biased the outcomes of these duels. To disentangle a potential effect of possessing a certain social position because individual qualities lead to that position, I manipulated social status by re-organising the pairs so that each dominant male was now paired up with another dominant male, and subordinates were paired with subordinates. This setup forced one male in each pair to change social status, either from dominant to subordinate or vice versa, since dominance hierarchies inevitably forms when two unfamiliar male fowl meet. Thereafter I observed behaviour of males exposed to a second NA. Dominant males again differed in vigilance, activity and exploration depending on status (Fig 6), which means that the role of being dominant or subordinate affects behaviour, and further that these behavioural differences were carried over to the non-social test

 environment. At the same time, all behavioural responses showed among- individual consistency across the two NA trials, indicating that these behavioural responses represent personality traits (Fig 6). Boldness and territorial crowing were not strongly related to status (Fig 6), and correlated between the trials, across status manipulations. This indicates that variation in these traits is not mainly a product of current social status, but may have interfered with who became dominant in the first round when the first pairs were set up.

Paper II

Personality remains: no effect of three-week social status experience on personality in male fowl

To confirm that social status could act as a stabilising state, I continued in paper II by investigating the effects on behaviour of having a certain social position for a longer period of time, compared to what I used in paper I. I used a period of 3 weeks in which males stayed in pairs and possessed either a dominant or subordinate social position (npairs = 22), and measured behaviour in personality assays (AGG, NA, NO) before and after this period. Males were matched for comb size, body size and age within the pair. To minimise the impact of earlier social position, I only used males that previously possessed an intermediate social rank in their home aviaries. No behavioural traits predicted the outcome of the dominance interactions of the pairs, possibly due to the chosen sample, excluding males that were top and bottom ranked. More importantly, status did not have a significant effect on the behaviour measured (Fig 7), indicating that the observed differences in paper I did not remained when more time had passed since pair formation. However, all behavioural responses showed among-individual consistency, strengthening the view of consistent personality traits in male domestic fowl (Fig 7).

 

Paper III

Personality predicts social dominance in male domestic fowl

The aim of paper III was to investigate the predictive value of personality traits on social status. I particularly focused on traits previously found to correlate and/or change with social status in the species (paper I). To minimise the impact of earlier social interactions, males were isolated from other males for five days before I measured their behaviour in NA and a startle test. To investigate if any of the behavioural responses measured could predict social dominance, I performed duels of morphologically matched males (npairs = 25). Both initial aggressiveness at the beginning of the duel, and initiation of the interaction (aggressively approaching or attacking) were indicative of winning the duel. Some personality traits other than aggression also predicted the winner: males scoring high in exploration and wariness had increased chances of winning over more passive males.

Paper IV

Individual aggression, but not winner-loser effects, predicts social rank in male domestic fowl

In paper IV, I investigated short-term effects of winning and losing a social interaction to simulate the natural situation where a male has to conquer several opponents in a row in order to obtain a high dominance status. In the first part of the study, I investigated if winning or losing an interaction caused a change in aggressiveness. I scored aggressiveness of males using the aggression test and thereafter staged duels to expose males to an experience of either winning or losing (npairs = 20). To minimise the possibility that males obtained different social experiences depending on their aggressiveness (i.e. that more aggressive males win most duels, paper III), males were matched within pairs for aggression score, in addition to being matched for comb size, body weight and age. To investigate the impact of winning or losing on aggression, males were separated two minutes after the duel was resolved, and aggression was scored once again. Indeed, I observed short term winner and loser effects on behaviour: aggression increased in winners and decreased (but not as clearly) in losers (Fig 8).



Discussion

This thesis focuses on the interplay between social dominance and personality in the domestic fowl. I aimed to investigate this relationship critically and through an experimental approach. My first study (paper I) shows that social dominance status can have significant effect on some personality traits a few days after males had changed social status. However, when testing the durability of this effect both over several weeks as adults (paper II) and throughout development (paper V), social status did not significantly explain variation in the personality traits measured here. Instead, individuals showed consistency in their behavioural responses in personality assays, despite experimental changes in social status and exposure to experiences of winning and losing aggressive interactions (paper I-V). These differences proved to be important factors for the outcome of social interactions (paper III, IV, V). Together these five papers help to disentangle the causality of the relationship between social dominance and personality; social interactions and social positions have important effects on behaviour, but in terms of causality in the long run, the impact of personality (at least aggressiveness) on social dominance interactions appears to prevail. Based on these papers, I will focus on some important consequences of these findings and discuss the potential for social status as a state that generates variation in personality, and the impact of personality and social experience on the establishment of dominance status, in turn below.

Social status as a stable state

Individuals can differ between each other in their behaviour in a consistent way, thus each exhibiting a different personality type. Several explanations to this diversity have been put forward. The stable state hypothesis suggests that stability of ‘states’ such as social properties, including social status, can influence behaviour in such a way that variation and stability in personality are created (Wolf & Weissing 2010; Sih et al. 2015). Familiarity among group members has been shown to enhance repeatability of boldness (Laskowski & Pruitt 2014), which indicates that stable social relationships can give rise to consistent individual behaviour.



In paper I, I used an experimental design of manipulating social status that allowed us to study if the social position itself had an effect on behaviour, disentangled from intrinsic qualities that might otherwise separate different personality types into different status categories. Boldness (latency to enter the novel arena) and the number of crows uttered by a male differed when pairs were initially set up, but after the manipulating social status of males, these traits did not change according to males’ changes in social status. These traits may therefore represent individual differences that are less affected by social status, or affects the chances of becoming dominant (please see further discussion on this in the next section). In line with correlational studies, dominant males were more active and explorative (Colletér & Brown 2011; Verbeek et al. 1996). Possibly, the increased activity of dominant males and exploration in paper I may reflect increased territorial activities (Arakawa 2005). Other studies show that when animals of social species are observed together with their group members, dominant individuals generally spend more time scanning the surrounding for predators, while subordinate males engage more time in resource acquisition (great tits, Krams 1998; domestic fowl, Cornwallis & Birkhead 2008). This may be explained by different costs and benefits for dominant and subordinates with regard to engaging in activities other than resource acquisition (Krams 1998). For example, in order to protect their increased assets, dominant individuals may gain more from increased vigilance (i.e. scanning for predators) instead of allocating time to feeding, while subordinates have a different cost-benefit ratio, and gain more from feeding (Clark 1994; Wolf et al. 2007). In line with this when also observed singly in the novel arena assay, dominant male fowl were more vigilant in paper I, while subordinate males spent more time foraging. This suggests that behavioural differences based on social interactions can have carry-over effects to non-social contexts.

The observed behavioural differences between dominant and subordinate males in paper I indicated that social status may induce individual differences in behaviour. Previously, only short-term effects of social status on personality had been studied. To explain longer-term consistency of variation in personality, and test the prediction that social status could act as a stable state that creates between-individual variation and within-individual consistency in personality, studies stretching over a significant part of the animals’ life or breeding season are necessary. In paper II, I therefore extended the treatment period to 3 weeks. No dominance relationships were reversed during the treatment period, and the experimental aviaries were small enough (3 * 3 meters) for the dominant and subordinate male of each pair to interact frequently. Nevertheless, no longer-term effects of being in a certain social position were detected on behavioural responses in the

  personality tests these males later took part in. Personality differences therefore seems to be driven by other factors than social status. Indeed, in Seychelles warblers (Acrocephalus sechellensis) variation in explorative behaviour did not depend on social status, but on food abundance during the birth year, at least in young birds (Edwards et al. 2016).    !"early social environment may still shape an individual’s personality and competitiveness (Sachser et al. 2013; Ruploh et al. 2014). In paper V, males that were brought up with females only were less competitive in duels when adult, indicating that having any experience of same sex conspecifics is crucial for development of successful contest behaviour (Kim & Zuk 2000). Interestingly, dominance position did not have a significant impact on competitiveness and personality in males brought up in the presence of other males (paper V), similarly to what has been shown in female red junglefowl with different social experiences (Kim & Zuk 2000). Earlier studies of other group-living species have shown that males brought up without intra-sexual competition become more aggressive, possibly because they lack the experience of adjusting their behaviour to minimise the costs of competition and fighting (Tóth et al. 2008; Ruploh et al. 2013). However, no differences in aggressiveness or other behavioural responses were observed in male domestic fowl raised under different social regimes (paper V). It is possible that social dominance abilities and personality are determined by some underlying pleiotropic factors. Such pre-determined individual differences may lead to different social positions early in life and result in the observed personality-dominance correlations, in which only some personality components can be modified by social experience later in life (Benus et al. 1990; Ruiz-Gomez et al. 2008).

Male domestic fowl have shown status-dependent differences in behaviour, both in paper I and in an earlier study on the same population (Cornwallis & Birkhead 2008). This raises the question of why these differences were not observed in paper II and V. I propose two main explanations for this discrepancy: the time since group formation, and the social context in which the behavioural responses are observed. In paper I, behaviour was measured around two days after the males had been assembled as a pair, and it is possible that the differences observed between dominant and subordinate males were due to winner-loser effects. Winner-loser effects generally fade away rather quickly when contestants are separated after the resolution (Chase et al. 1994). The experimental setup in paper I, where males remained in the newly formed pair until behavioural observations were performed, might have increased the winner-loser effects, resulting in the observed behavioural differences. In paper II and V, the social status treatment period was much longer. Aggression by the dominant generally decreases with time after group formation (Issa et al. 1999; Ricci et al.

 2013), and the acute stress level caused by the initial aggressive interaction goes down (Summers & Winberg 2006). To investigate if time since group formation could explain the varying results of the relationship between dominance and behaviour, future studies should include potential physiological underlying mechanisms (e.g. hormonal or neuroendocrine factors) while integrating a time perspective on dominance relationships. Males may differ in behaviour also after a longer period of time, when observed in the social context in question (Cornwallis & Birkhead 2008), and not separated from the group as in a typical personality test. The presence of conspecifics may cause individuals to enhance or suppress threat-sensitive behaviour such as boldness, activity, exploration and feeding rate (Magnhagen & Staffan 2005; Magnhagen & Bunnefeld 2009; Schuett & Dall 2009). Indeed, when male domestic fowl were observed in the presence of conspecifics (Cornwallis & Birkhead 2008), they showed strong status- dependent differences in vigilance, and repeatability was low, indicating that social status has an even stronger effect than personality in the presence of conspecifics. Males readily adjust to the new situation when the dominant is removed, and soon start to crow and perform other dominant behaviours (Wilson et al. 2009). This indicate that subordinate males are generally prepared to obtain the dominant position when available, and that they are suppressed by more dominant males. Therefore, differences caused by the social context may be reduced when males are tested singly in personality assays. Additional observations of behaviour in different social contexts, with or without conspecifics present, could shed light on possible context- dependence of behavioural differences due to status.

It has been argued that only permanent and non-reversible changes due to differences in states should to be considered as ‘true personality’, while reversible behavioural changes should instead be referred to as ‘pseudo- personality’ (Niemelä & Santostefano 2015; Dingemanse & Araya-Ajoy 2015). Manipulating adult male dominance status for a couple of weeks was not expected to induce permanent changes in personality, given that behaviour readily changed in paper I and social status is a rather labile state. However, there is a possibility that altering male social context already from a young age can set males on a developmental trajectory resulting in the development of different personality types, or more or less permanent differences in competitive ability. Here (paper V) social status during juvenile development did not result in permanent or even moderately lasting effects on behaviours, and I therefore did not find support for that social status acts as a state that creates either ‘pseudo’ or ‘real’ personality variation in male domestic fowl. Overall, males remained rather consistent in behaviour also across such important events such as losing dominance (paper I, II, IV). Under controlled laboratory conditions, personality was found to emerge with age in Eastern mosquitofish (Gambusia holbrooki)

   even in absence of intra-specific competition and predation risk, extrinsic factors that otherwise are related to behavioural differentiation (Polverino et al. 2016). However, this does not exclude that other state differences stemming from small, early phenotypic differences over time may lead to larger differences in states, developing into personality differences (Stamps & Groothuis 2010). For example, male red junglefowl exposed to more cognitive stimulation when young show increased vigilance and less fearfulness as adults (Zidar et al. 2017). To summarise, the often-observed relationship between social dominance and personality likely relies on causal pathways other than social status acting as a mediating state. One possibility is that personality influences the outcome of dominance interactions, which will be further discussed below.

Personality predicting dominance

Competition for social dominance commonly follows stereotypic patterns that escalate from aggressive displays to physical attacks, ending at any stage when one of the competitors withdraws (e.g. domestic fowl, Collias 1943; Hardy & Briffa 2013). The behaviour of future winners and losers often differ already at an early stage of the interaction, indicating that contestants signal their internal state and/or intentions through their behaviour (Hardy & Briffa 2013; Camerlink et al. 2015). For example, initiating a fight can predict the outcome of the interaction (Ligon et al. 1990; paper III). Announcing the intention to escalate during the fight, for example by making the first move, gives information about the individual’s motivational state, which can be connected to their fighting ability (Arnott & Elwood 2009). By studying behaviour in the very first seconds of a duel, I could conclude that when there was a clear asymmetry of behaviour (males differed in aggressive display), duels were very quickly resolved in favour of the more aggressive male, often without physical fighting (paper III). This conforms to the honest signalling hypothesis, and saves both contestants from the risks an escalated fight could imply (Rutte et al. 2006). In line with this hypothesis, male domestic fowl with larger combs than their opponents often win (Zuk & Johnsen 2000; paper V). This is possibly because comb size may signal a hormonal (testosterone) state (Allee et al. 1939) that is related to the motivation to fight (Ligon et al. 1990).

Although fight initiation and attack rates are generally positively related to winning in various species, claiming that such aggressiveness predicts dominance, is reasoning that might suffer from circularity. To overcome this, I developed an aggression test that reduced the influence of the non-focal

 individual and enabled scoring individual intrinsic aggressiveness. The scores obtained in this test were strongly positively related to aggression in free-ranging agonistic interactions (paper III). Further, the aggression scores were very consistent across time (even over the timespan of a year) and across intruder identities (paper III, IV), and could therefore be used as a reliable measure of individual aggressiveness. Aggressiveness turned out to be a relatively strong predictor of the outcome of social interactions (paper IV, V). Aggressiveness, as an intrinsic and consistent individual quality, may therefore function as a cue of individual competitiveness, that influences an individual’s willingness to fight. Self-assessment of an individual’s own competitiveness is further shaped by information from contest experiences, resulting in a predetermined strategy for future encounters (Chase et al. 2002; Rutte et al. 2006). This self-assessment can continuously change depending on information from further sampling (Enquist & Leimar 1983; Arnott & Elwood 2009). Indeed, in several cases I observed a situation where two males both initially avoided the other and showed clear signs of submission, although after some time had passed, one of the contestants would slowly start to change behaviour and adopt a more aggressive approach. I interpret this as the males having a predetermined, but flexible conflict strategy.

Personality traits other than aggression may also predict the outcome of aggressive interactions and future social status. The general pattern is that more explorative, bold, or active individuals have higher chances of becoming dominant (e.g. Huntingford 1976; Verbeek et al. 1996; Colleter & Brown 2011; Dahlbom et al. 2011; David et al. 2011; Rudin & Briffa 2012, paper III), although the relationships between these personality traits and dominance are sometimes reversed (Gomez-Laplaza 2002; Fox et al. 2009). In some cases, these traits have been found to correlate with aggressiveness (e.g. Huntingford 1976; Verbeek et al. 1996; Kortet & Hedrick 2007; Santostefano et al. 2016), which could explain the success in fights of the more outgoing personality type. However, in some studies the association between aggressiveness and other personality traits that predicts status is weak or even absent (e.g. Bell 2005; Chang et al. 2012; paper III). This indicates that there are several aspects of personality that are important for winning, and offers the possibility that individuals with different personality types can reach the same social status. Since social status positively influences reproductive success, this could in turn potentially explain why there is not one ‘perfect type’ selected for by positive directional selection, and ultimately contribute to the conservation of variation of personality types in the population. The exact mechanism through which exploration, boldness and other non-aggressive personality traits sometimes act to

  increase the chances of winning, remains unknown and warrants further study.

The relationship between dominance and aggression can vary even across studies in the same species (cichlids, Neolamprologus pulcher, Riebli et al. 2011; Riebli et al. 2012). This has been suggested to depend on social context, i.e. the magnitude of difference in aggression between contestants. Likewise, exploration and aggressiveness predicted dominance in paper III, IV and V, but not in paper I and II, and boldness was not strongly related to dominance in any of the studies. In paper I and II, personality could not predict the outcome of duels, but the setups were not designed to specifically test this question (paper I, II). In paper I, behaviour in the novel arena test was probably affected by the recent social interaction, and therefore did not provide a neutral ground for measuring behaviour to be used as predictors of social conflicts. In paper II, I exclusively used males with an intermediate social position. If certain personality types are overrepresented in top and bottom dominance positions, using only the intermediately ranked males means that less contrast among males were obtained, and thus less opportunity for personality traits to influence the outcome. Indeed, strong results in predicting dominance by personality have been obtained in studies using a double duel setup, where individuals that have won twice against two different opponents are compared to individuals that have lost twice, while the ambiguous males that lost once and won once were not included in the analysis (Dahlbom et al. 2011).

Crowing in male fowl is an auditory signal of dominance that is uttered by the winner shortly a conflict, and as a part of a daily routine to signal his presence in the territory (McBride et al. 1969). In a social context, dominant males in general crow more often, and frequently punish subordinate males that crow (Leonard & Horn 1995; Johnsen et al. 2001). When separated from the flock in the novel arena test, the frequency of crowing appeared to have a strong individual component, being repeatable despite major changes in social status (paper I, II). Crowing frequency was not a reliable predictor of future chances of obtaining dominance (crowing predicted status in paper I, but not in paper II and III). Crowing is a highly labile and flexible acoustic signal status, and the frequency of crowing depend on the social context before and during the observations (Mench & Ottinger 1991). It is likely that crowing frequency predominantly functions as a signal of the current status rather than being a reliable cue of competitiveness.

At any given moment, an individual’s combination of individual qualities sets their odds of winning. Therefore, the scope for personality traits to influence the outcome of dominance interactions may depend on the

 presence of other dissimilarities between contestants in body size, armament size, recent contest experiences etc. In all contests in this thesis, except for paper V, the contestants were size-, comb- and age matched, and sometimes also matched for aggressiveness (paper IV), which obstructs a direct comparison of the effect of these factors. In paper V, the focal males and their opponents differed randomly in comb size, body size and personality, which enabled us to compare the effect of these traits on contest outcomes. Of these traits, aggression was the strongest predictor of dominance. Further, a male’s ability to win fights seems to be signalled through the size of the comb (Ligon et al. 1990), while other personality traits and body size had no strong effects on the outcome of duels.

Although not a main focus of this thesis, the variety of behavioural traits studied here offered an opportunity to investigate the relationship among personality traits. The presence of behavioural syndromes varies among studies and species, and defining when and where syndromes occur may give insight to the selection on personality traits (Sih et al. 2004; Sih et al. 2012). The different personality traits examined in the domestic fowl/red junglefowl (exploration/activity, boldness, aggressiveness, vigilance) do not seem to be strongly correlated (Favati et al. 2016; Zidar et al. 2017; paper I, II, III). This means that selective pressures can act on each trait independently, and that focus should be on each behavioural trait rather than a combined personality type construct.

Recent contest experience

Recent experience of winning or losing can have an impact on conflict resolution, such that winners tend to keep on winning, and losers keep on losing, when confronted with unfamiliar conspecifics. Individual traits like aggressiveness and body size can bias the chances of winning duels, which means that some individuals more often experience winning than losing and vice versa, making it hard to distinguish individual characteristics from the winner-loser experience (Cloutier et al. 1996). Paper IV was designed to disentangle individual aggressiveness from the experience of winning and losing. By pairing males matched for aggressiveness, this trait had minimal chance to affect the outcome of the duel, and I could thereby compare the effect of winning or losing between males of similar aggressiveness. The results showed that winners increased in aggressiveness while losers decreased, which aligns with the general pattern of winner-loser effects observed in other species (Hsu et al. 2006). Despite the observed changes in aggressiveness in winners and losers, earlier experiences do not always

   change the probability of obtaining a dominant position. Instead, original disparities in aggressiveness can be a stronger predictor of becoming dominant (Garcia et al. 2012; paper IV).

To sum up, individual qualities appear to be crucial for the settling of social dominance relationships, while winner-loser effects seem to be a possibly adaptive means to adjust behaviour to the outcome of duels, rather than determining future behaviour of an individual. Indeed, the loser effect, which appears to be stronger or more prevalent than the winner effect (Rutte et al. 2006), gives rise to submissive behaviours that prevent the loser from receiving further attacks (Parker & Ligon 2002). Different personality types are likely to respond differently to winning or losing a dominance interaction (Courtene-Jones & Briffa 2014; Briffa et al. 2015), but the potential interactive effect of personality and winner-loser effects on future probability of winning has not been tested. Further studies are needed to investigate if there are reaction norms of winner-loser effects depending on personality type.

Losers

Though it is beyond the scope of this thesis, it is intriguing to consider the fate of losers. I have shown here that individuals differ in personality traits that influence dominance (paper III, IV, V). In paper V, being subordinate from a very young age did not cause males to become ‘permanent losers’. This does not, however, exclude the possibility that stronger negative experiences, e.g. injury during fighting, can cause a male to develop a ‘loser character’. Alternatively, personality types are determined early in life, and influence adult competitive ability. Regardless of the origin of a poor competitive ability, failing to gain a dominant position can be detrimental to reproductive success. There are several options for a subordinate to overcome this situation. One is to focus on resource acquisition to increase growth, and possibly gain dominance in the future. In some species, however, less competitive individuals adopt alternative mating tactics to be able to reproduce (Andersson 1994; Oliveira et al. 2008). In male domestic fowl, some males are consistently less aggressive and seldom win fights over dominance (paper III, IV, V). Personal observations of duels between males with low aggression scores indicate that both males start the interaction very carefully, likely signalling their intent to not engage in a fight. To surrender from start might represent an alternative contest tactic. Many alternative tactics are plastic and/or reversible (Oliveira et al. 2008; Takegaki et al. 2012), showing that lower ranking individuals are waiting for an opportunity

 to gain dominance later on (e.g. when they have grown larger). After some time without anyone initiating a fight in the contests between nonaggressive males described above, one of them slowly start to take on a more confident body posture and slowly begin to threaten the other male. It therefore seems that his initial ‘loser tactic’ is a more-or-less temporary tactic he has adopted to minimise the risk of injury in fights against more confident males. Indeed, in groups where the dominant male is removed, subordinate males will soon display dominant behaviours (Mench & Ottinger 1991; Wilson et al. 2009). In other words, in the domestic fowl, subordinate males likely wait for an opportunity to step up in rank. Some personality types, i.e. less aggressive and more passive males, may be more likely to remain throughout life as subordinates. Will these personality types also adopt alternative mating tactics? Subordinate male fowl do attempt to mate with available females, but dominant males interrupt these matings (Pizzari 2001; Rosher et al. 2017). Subordinate males sometimes perform mating attempts when the dominant male is out of sight (personal observation), but it is untested if doing so represent an alternative mating tactic that depends on repeated position as subordinate, related to certain personality types. Investigating mating behaviour and reproductive success of males of different social status could show whether also personality, and not only morphological traits like body size, may lead to alternative tactics.

Concluding remarks

In this thesis, I have investigated the causality of the relationship between social dominance and personality in male domestic fowl. I conclude that the social environment can have pronounced effects on behaviour, but that these effects may be restricted both in time and to particular situations. Effects on behaviour of social status, and of experience of winning and losing, appear to diminish with time, but can nevertheless be important for an appropriate adjustment of behaviour during the turbulent periods when dominance hierarchies are established. I did not find support for social status as a state that generates and maintains consistent variation in personality. Thus, the general idea that social states explain personality variation is not supported in the domestic fowl. I do however not exclude the possibility that social position may have long-term effects on behaviour in a social context, where behaviour of subordinates is constantly suppressed by more dominant individuals. The opposite causal direction, that personality variation affects the chances of obtaining social dominance, received stronger support, both from previous studies and in this thesis. The relative importance of personality traits for obtaining dominance, compared to that of other traits

   such as body size, ornament size and age, is not completely resolved. However, in male domestic fowl, aggressiveness appears as an important individual trait that can override the effects of both previous experiences and differences in body size, which is an important contribution to the classical field of fighting behaviour and social dominance. Various other personality traits also predicted dominance, but the reliability was lower, and the exact pathway is not fully understood. To conclude, personality can affect social properties that are important for reproductive success and fitness, and this can guide future studies of the evolutionary understanding of personality variation.

 References

Albers PCH, Timmermans PJA, Vossen JHM. 1999. Evidence for the existence of mothering styles in guinea pigs (Cavia aperea f. porcellus). Behaviour. 136:469–479. Andersson M. 1994. Sexual selection. Princeton University Press, New Jersey. Arakawa H. 2005. Changes in the pattern of exploratory behavior are associated with the emergence of social dominance relationships in male rats. Dev Psychobiol. 48:39–47. Ariyomo TO, Watt PJ. 2012. The effect of variation in boldness and aggressiveness on the reproductive success of zebrafish. Anim Behav. 83:41–46. Arnold C, Taborsky B. 2010. Social experience in early ontogeny has lasting effects on 407 social skills in cooperatively breeding cichlids. Anim Behav. 79:621– 630. Arnott G, Elwood RW. 2009. Assessment of fighting ability in animal contests. Anim Behav. 77:991–1004. Arnqvist G, Henriksson S. 1997. Sexual cannibalism in the fishing spider and a model for the evolution of sexual cannibalism based on genetic constraints. Evol. Ecol. 11:255–273. Banks EM. 1956. Social organization in red jungle fowl hens (Gallus Gallus Subsp.). Ecology. 37:239–248. Banks EM. 1984. Relations between aggressive social dominance and resource utilization in domestic fowl. Polit Psychol. 5:617–622. Beaugrand JP, Cotnoir P-A. 1996. The role of individual differences in the formation of triadic dominance orders of male green swordtail fish (Xiphophorus helleri). Behav Process. 38:287–296. Beekman M, Jordan LA. 2017. Does the field of animal personality provide any new insights for behavioral ecology? Behav Ecol. In press. Bell AM. 2005. Behavioural differences between individuals and two populations of stickleback (Gasterosteus aculeatus). J Evol Biol. 18:464–473. Bell AM. 2007. Future directions in behavioural syndromes resch. Proc R Soc B. 274:755–761. Bell AM, Sih A. 2007. Exposure to predation generates personality in threespined sticklebacks (Gasterosteus aculeatus). Ecol Lett. 10:828–834. Benus RF, Dendaas S, Koolhaas JM, van Oortmerssen GA. 1990. Routine formation and flexibility in social and nonsocial behavior of aggressive and nonaggressive male–mice. Behaviour 112:176–193.

  Biro PA, Stamps JA. 2008. Are animal personality traits linked to life-history productivity? TREE. 23:361–368. Blanchard RJ, Hori K, Tom P, Blanchard DC. 1988. Social dominance and individual aggressiveness. Aggress Behav. 14:195–203. Briffa M, Sneddon LU, Wilson AJ. 2015. Animal personality as a cause and consequence of contest behaviour. Biol Lett. 11:20141007. Bölting S, von Engelhardt N. 2017. Effects of the social environment during adolescence on the development of social behaviour, hormones and morphology in male zebra finches (Taeniopygia guttata). Front Zool. 14:5. Camerlink I, Arnott G, Farish M, Turner SP. 2016. Complex contests and the influence of aggressiveness in pigs. Anim Behav. 121:71–78. Careau V, Garland T Jr. 2012. Performance, personality, and energetics: correlation, causation, and mechanism. Physiol Biochem Zool. 85:543–571. Carere C, Mastripieri D (ed). 2013. Animal personalities: behavior, physiology, and evolution. The University of Chicago Press, Chicago and London. Carter AJ, Feeney WE, Marshall HH, Cowlishaw G, Heinsohn R. 2013. Animal personality: what are behavioural ecologists measuring? Biol Rev. 88:465– 475. Chang C, Li CY, Earley RL, Hsu Y. 2012. Aggression and related behavioral traits: the impact of winning and losing and the role of hormones. Integr Comp Biol. 52:801–813. Chase ID, Bartolomeo D, Dugatkin LA. 1994. Aggressive interactions and inter- contest interval: how long do winners keep winning? Anim Behav. 48:393– 400. Chase ID, Tovey C, Spangler-Martin D, Manfredonia M. 2002. Individual differences versus social dynamics in the formation of animal dominance hierarchies. PNAS. 99:5744–5749. Cheng KM, Burns JT. 1988. Dominance relationship and mating behaviour of the domestic cocks: A model to study mate-guarding and sperm competition in birds. Condor. 90: 697–704. Clark CW. 1994. Antipredator behavior and the asset protection principle. Behav Ecol. 5:159–170. Cloutier S, Beaugrand JP, Lague PC. 1995. The effect of prior victory or defeat in the same site as that of subsequent encounter on the determination of dyadic dominance in the domestic hen. Behav Process. 34:293–298. Cloutier S, Beaugrand JP, Lague PC. 1996. The role of individual differences and patterns of resolution in the formation of dominance orders in domestic hen triads. Behav Process. 38:227–239. Clutton–Brock TH, Huchard E. 2013. Social competition and selection in males and females. Philos Trans R Soc Lond B Biol Sci. 368:20130074. Colleter M, Brown C. 2011. Personality traits predict hierarchy rank in male rainbowfish social groups. Anim Behav 81:1231–1237.

 Collias NE. 1943. Statistical analysis of factors which make for success in initial encounters between hens. Am Nat. 77:519–538. Collias NE, Collias EC. 1996. Social organization of a red junglefowl, Gallus gallus, population related to evolution theory. Anim Behav. 51:1337–1354. Collias NE Collias, RI Jennrich. 1994. Dominant red jungle fowl (Gallus-gallus) hens in an unconfined flock rear the most young over their lifetime. Auk 111:863–872. Cornwallis CK, Birkhead TR. 2008. Plasticity in reproductive phenotypes reveals status-specific correlations between behavioral, morphological, and physiological sexual traits. Evolution. 62:1149–1161. Costa PT, McCrae RR. 1992. Four ways five factors are basic. Pers Individ Differ 13:653–665. Cote J, Dreiss A, Clobert J. 2008. Social personality trait and fitness. Proc R Soc B. 275:2851e2858. Courtene-Jones W, Briffa M. 2014. Boldness and asymmetric contests: role and outcome-dependent effects of fighting in hermit crabs. Behav Ecol. 25:1073– 1082. Dahlbom SJ, Lagman D, Lundstedt-Enkel K, Sundstrom LF, Winberg S. 2011. Boldness predicts social status in zebrafish (Danio rerio). PLOSONE. 6:e23565. Dall SRX, Houston AI, McNamara JM. 2004. The behavioural ecology of personality: consistent individual differences from an adaptive perspective. Ecol Lett. 7:734–739. David M, Auclair Y, Cézilly F. 2011. Personality predicts social dominance in female zebra finches, Taeniopygia guttata, in a feeding context. Anim Behav. 81:219–224. David M, Dall SRX. 2016. Unravelling the philosophies underlying ‘animal personality' studies: a brief re-appraisal of the field. Ethology. 12:1–9. Dean R, Cornwallis CK, Løvlie H, Worley K, Richardson DS, Pizzari T. 2010. Male reproductive senescence causes potential for sexual conflict over mating. Curr Biol. 20:1192–1196. Dingemanse NJ, Araya-Ajoy YG. 2015. Interacting personalities: behavioural ecology meets quantitative genetics. TREE. 30:88–97. Dingemanse NJ, Both C, Drent PJ, Tinbergen JM. 2004. Fitness consequences of avian personalities in a fluctuating environment. Proc R Soc B. 271:847–852. Dingemanse NJ, Dochtermann N, Wright J. 2010. A method for exploring the structure of behavioural syndromes to allow formal comparison within and between data sets. Anim Behav. 79:439–450. Dingemanse NJ, Réale D. 2005. Natural selection and animal personality. Behaviour. 142:1165–1190. Dingemanse NJ, Wolf M. 2010. Recent models for adaptive personality differences: a review. Phil Trans R Soc B. 365:3947–3958.

   Dingemanse NJ, Wright J, Kazem AJN, Thomas DK, Hickling R, Dawnay N. 2007. Behavioural syndromes differ predictably between 12 populations of three- spined stickleback. J Anim Ecol. 76:1128–1138. Dochtermann NA, Schwab T, Sih A. 2015. The contribution of additive genetic variation to personality variation: heritability of personality. Proc R Soc B. 282:20142201. Drickamer LC, Arthur RD, Rosenthal TL. 1999. Predictors of social dominance and aggression in gilts. Appl Anim Behav Sci. 63:121–129. Dugatkin LA, Druen M. 2004. The social implications of winner and loser effects. Proc. R. Soc. Lond. B (Suppl.). 271:S488–S489. Dugatkin LA, Reeve HK. 2014. Winning, losing, and reaching out. Behav Ecol. 25:675–679. Edwards HA, Dugdale HL, Richardson DS, Komdeur J, Burke T. 2016. Exploration is dependent on reproductive state, not social state, in a cooperatively breeding bird. Behav Ecol. 27:1889–1896. Enqvist M, Leimar O. 1983. Evolution of fighting behaviour: decision rules and assessment of relative strength. J Theor Biol. 102:387–410. Favati A, Zidar J, Thorpe H, Jensen P, Løvlie H. 2016. The ontogeny of personality traits in red junglefowl. Behav Ecol. 27:484–493. Forkman B, Boissy A, Meunier-Salaün M-C, Canali E, Jones RB. A critical review of fear tests used on cattle, pigs, sheep, poultry and horses Physiol Behav. 92:340–374. Fox RA, Ladage LD, Roth TC, Pravosudov VV. 2009. Behavioural profile predicts dominance status in mountain chickadees, Poecile gambeli. Anim Behav. 77:1441–1448. Francis RC. 1988. On the relationship between aggression and social dominance. Ethology. 78:223–237. Froman DP, Pizzari T, Feltmann AJ, Castillo-Juarez H, Birkhead T. 2001. Sperm mobility: mechanisms of fertilizing efficiency, genetic variation and phenotypic relationship with male status in the domestic fowl, Gallus gallus domesticus. Proc R Soc Lond B. 269:607–612. Fumihito A, Miyake T, Sumi SI, Takada M, Ohno S, Kondo N. 1994. One subspecies of the Red junglefowl (Gallus gallus gallus) suffies as the matriarchic ancestor of all domestic breeds. PNAS. 91:12505–12509. Garcia MJ, Murphree J, Wilson J, Earley RL. 2014. Mechanisms of decision making during contests in green anole lizards: prior experience and assessment. Anim Behav. 92:45–54. Garcia MJ, Paiva L, Lennox M, Sivaraman B, Wong SC, Earley RL. 2012. Assessment strategies and the effects of fighting experience on future contest performance in the green anole (Anolis carolinensis). Ethology. 118:821– 834. Gomez-Laplaza, LM. 2002. Social status and investigatory behaviour in the angelfish (Pterophyllum scalare). Behaviour. 139:1469–1490.

 Gosling SD. 2001. From mice to men: What can we learn about personality from animal research? Psychol Bull. 127:45–86. Gosling SD, John PJ. 1999. Personality dimensions in nonhuman animals: a cross- species review. Curr Dir Psychol Sci. 8:69–75. Haller J, Halász J. 2000. Anxiolytic effects of repeated victory in male Wistar rats. Aggress Behav. 26:257–261. Hardy ICW, Briffa M (ed). 2013. Animal contests. New York: Cambridge University Press. Hedrick AV (ed). 2017. The development of animal personality. Frontiers Media, Lausanne. Holekamp KE, Strauss ED. 2016. Aggression and dominance: an interdisciplinary overview. Curr Opin Behav Sci. 12:44–51. Hsu Y, Earley RL, Wolf LL. 2006. Modulation of aggressive behaviour by fighting experience: mechanisms and contest outcomes. Biol Rev. 81: 33–74. Hsu Y, Lee I-H, Lu C-K. 2009. Prior contest information: mechanisms underlying winner and loser effects. Behav Ecol Sociobiol. 63:1247–1257. Huntingford FA. 1976. Relationship between anti-predator behavior and aggression among conspecifics in 3-spined stickleback, Gasterosteus aculeatus. Anim Behav 24:245–260. Issa FA, Adamson DJ, Edwards DH. 1999. Dominance hierarchy formation in juvenile crayfish procambarus clarkii. J Exp Biol. 202:3497–506. Johnsen TS, Zuk M, Fessler EA. 2001. Social dominance, male behaviour and mating in mixed-sex flocks of red jungle fowl. Behaviour. 138:1–18. Kim T, Zuk M. 2000. The effects of age and previous experience on social rank in female red junglefowl, Gallus gallus spadiceus. Anim Behav. 60:239–244. Koolhaas JM, Korte SM, De Boer SF, Van Der Vegt BJ, Van Reenen CG, Hopster H et al. 1999. Coping styles in animals: current status in behavior and stress- physiology. Neurosci Biobehav Rev. 23:925–935. Kortet R, Hedrick A. 2007. A behavioural syndrome in the field cricket Gryllus integer: intrasexual aggression is correlated with activity in a novel environment. Biol J Linn Soc. 91:475–482. Koski SE. 2014. Broader horizons for animal personality research. Front Ecol Evol. 2:70. Kralj-Fiser S, Weiss BM, Kotrschal K. 2010. Behavioural and physiological correlates of personality in greylag geese (Anser anser). J Ethol. 28:363–370. Krams I. 1998. Dominance-specific vigilance in the great tit. J Avian Biol. 29:55– 60. Kruijt JP. 1964. Ontogeny of social behaviour in Burmese red junglefowl (Gallus gallus Spadiceus) Bonnaterre. Behaviour. 12:1–201. Lan Y-T, Hsu Y. 2011. Prior contest experience exerts a long-term influence on subsequent winner and loser effects. Front Zool. 8:28.

  Laskowski KL, Bell A. 2014. Strong personalities, not social niches, drive individual differences in social behaviours in sticklebacks. Anim Behav. 90:287–295. Laskowski KL, Montiglio P-O, Pruitt JN. 2016. Individual and group performance suffers from social niche disruption. Am Nat. 187:776–785. Laskowski KL, Pruitt JN. 2014 Evidence of social niche construction: persistent and repeated social interactions generate stronger personalities in a social spider. Proc R Soc B. 281:20133166. Laskowski KL, Wolf M, Bierbach D. 2016. The making of winners (and losers): how early dominance interactions determine adult social structure in a clonal fish. Proc R Soc B 283:20160183. Leonard ML, Horn AG. 1995. Crowing in relation to status in roosters. Anim Behav. 49:1283–1290. Ligon JD, Thornhill R, Zuk M, Johnson K. 1990. Male–male competition, ornamentation and the role of testosterone in sexual selection in red jungle fowl. Anim Behav. 40:367–373. Luttbeg B, Sih A. 2010. Risk, resources and state-dependent adaptive behavioural syndromes. Phil Trans R Soc B. 365:3977–3990. Magnhagen C, Bunnefeld N. 2009. Express your personality or go along with the group: what determines the behaviour of shoaling perch? Proc R Soc B. 276:3369–3375. Magnhagen C, Staffan F. 2005. Is boldness affected by group composition in young- of-the-year perch (Perca fluviatilis)? Behav Ecol Sociobiol. 57:295–303. Martin F, Beaugrand JP, Laguë. 1997. The role of hen’s weight and recent experience on dyadic conflict outcome. Behav Process. 41:139–150. McBride G, Parer IP, Foenander F. 1969. The social organisation and behaviour of the feral domestic fowl. Baillière, Tindall & Cassell, London. McGhee KE, Travis J. 2010. Repeatable behavioural type and stable dominance rank in the bluefin killifish. Anim Behav. 79:497–507. Mench JA, Ottinger MA. 1991. Behavioral and hormonal correlates of social dominance in stable and disrupted groups of male domestic fowl. Horm Behav 25:112–122. Nakagawa S, Schielzeth H. 2010. Repetability for Gaussian and non-Gaussian data: a principal guide for biologists. Biol Rev Camb Philos Soc. 85:935–956. Nelson XJ, Wilson DR, Evans CS. 2008. Behavioral syndromes in stable social groups: an artifact of external constraints? Ethology. 114:1154–1165. Nicolaus M, Tinbergen JM, Ubels R, Both C, Dingemanse NJ. 2016. Density fluctuations represent a key process maintaining personality variation in a wild passerine bird. Ecol Lett. 19:478–486. Niemelä PT, Santostefano F. 2015. Social carry-over effects on non-social mechanisms and consequences. Front Ecol Evol. 3:24–35. Oliveira RF. 2012. Social plasticity in fish: integrating mechanisms and function. J Fish Biol. 81:2127–2150.

 Oliveira R, Taborsky M, Brockmann HJ (ed). 2008. Alternative reproductive tactics: an integrative approach. Cambridge University Press, New York. Parker TH, Knapp R, Rosenfield JA. 2002. Social mediation of sexually selected ornamentation and steroid hormone levels in male junglefowl. Anim Behav. 64:291–298. Parker TH, Ligon JD. 2002. Dominant male red junglefowl (Gallus gallus) test the dominance status of other males. Behav Ecol Sociobiol 53:20–24. Pizzari T. 2001. Indirect partner choice through manipulation of male behaviour by female fowl, Gallus gallus domesticus. Proc R Soc Lond B. 268:181–186. Polverino G, Cigliano C, Nakayama S, Mehner T. 2016. Emergence and development of personality over the ontogeny of fish in absence of environmental stress factors. Ecol Sociobiol. 70:2027–2037. Réale D, Reader SM, Sol D, McDougall PT, Dingemanse NJ. 2007. Integrating animal temperament within ecology and evolution. Biol Rev Camb Philos Soc. 82:291–318. Ricci L, Summers CH, Larson ET, O’Malley D, Melloni, RH Jr. 2013. Development of aggressive phenotypes in zebrafish: interactions of age, experience and social status. Anim Behav. 86:245–252. Riebli T, Avgan B, Bottini AM, Duc C, Taborsky M, Heg D. 2011. Behavioural type affects dominance and growth in staged encounters of cooperatively breeding cichlids. Anim Behav, 81:313e323. Riebli T, Taborsky M, Chervet N, Apolloni N, Zurcher Y, Heg D. 2012. Behavioural type, status and social context affect behaviour and resource allocation in cooperatively breeding cichlids. Anim Behav. 84:925–936. Roche DG, Careau V, Binning SA. 2016. Demystifying animal ‘personality’ (or not): why individual variation matters to experimental biologists. J Exp Biol. 219:3832–3843. Rosher C, Favati A, Dean R, Løvlie H. 2017. Relatedness and age reduce aggressive male interactions over mating in domestic fowl. Behav Ecol. https://doi.org/10.1093/beheco/arx024. Rudin FS, Briffa M. 2012. Is boldness a resource-holding potential trait? Fighting prowess and changes in startle response in the sea anemone, Actinia equina. Proc R Soc B. 279:1904–1910. Ruiz-Gomez M, Kittilsen S, Höglund E, Huntingford FA, Sørensen C, Pottinger TG et al. 2008. Behavioral plasticity in rainbow trout (Oncorhynchus mykiss) with divergent coping styles: When doves become hawks. Horm Behav. 54:534–538. Ruploh T, Bischof H-J, von Engelhardt N. 2013. Adolescent social environment shapes 483 sexual and aggressive behaviour of adult male zebrafinches (Taeniopygia guttata). 484 Behav Ecol Sociobiol. 67:175–184. Ruploh T, Bischof H-J, von Engelhardt N. 2014. Social experience during adolescence influences how male zebra finches (Taeniopygia guttata) group with conspecifics. Behav Ecol Sociobiol. 68:537–549.

   Rutte C, Taborsky M, Brinkhof MWG. 2006. What sets the odds of winning and losing? Front Ecol Evol. 21:16–21. Sachser N, Kaiser S, Hennessy MB. 2013. Behavioural profiles are shaped by social experience: when, how and why. Phil Trans R Soc B. 368:20120344. Santostefano F, Wilson AJ, Araya-Ajoy YG, Dingemanse NJ. 2016. Interacting with the enemy: indirect effects of personality on conspecific aggression in crickets. Behav Ecol. 27:1235–1246. Schjelderup-Ebbe T. 1922. Beiträge zur Sozialpsychologiedes Haushunhs. Z Psychol. 88:224–252. Schuett W, Dall SRX. 2009. Sex differences, social context and personality in zebra finches, Taeniopygia guttata. Anim Behav. 77:1041–1050. Schuett W, Godin J-G, Dall SRX. 2011. Do female zebra finches, Taeniopygia guttata, choose their mates based on their ‘personality’? Ethology. 117:908– 917. Schuett W, Treganza T, Dall SRX. 2010. Sexual selection and animal personality. Biol Rev. 85:217–246. Schütz K, Jensen P. 2001. Effects of resource allocation on behavioural strategies: a comparison of Red Junglefowl (Gallus gallus) and two domesticated breeds of poultry. Ethology 107:753–765. Sih A, Bell A, Johnson JC. 2004. Behavioral syndromes: an ecological and evolutionary overview. Trends Ecol Evol. 19:372–378. Sih A, Chang AT, Wey TW. 2014. Effects of behavioural type, social skill and the social environment on male mating success in water striders. Anim Behav. 94:9–17. Sih A, Cote J, Evans M, Fogarty S, Pruitt J. 2012. Ecological implications of behavioural syndromes. Ecol Lett. 15:278–289. Sih A, Mathot KJ, Moirón M, Montiglio P-O, Wolf M, Dingemanse NJ. 2015. Animal personality and state–behaviour feedbacks: a review and guide for empiricists. TREE. 30:50–60. Smith BR, Blumstein DT. 2008. Fitness consequences of personality: a metaanalysis. Behav Ecol. 19:448–455. Stamps JA. 2007. Growth-mortality tradeoffs and ‘personality traits’ in animals. Ecol Lett. 10:355-363. Stamps JA, Groothuis TGG. 2010. The development of animal personality: relevance, concepts and perspectives. Biol Rev. 85:301–325. Stöwe M, Bugnyar T, Loretto M-C, Schloegl C, Range F, Kotrschal K. 2006. Novel object exploration in ravens (Corvus corax): Effects of social relationships. Behav Process. 73:68–75. Summers CH, Winberg S. 2006. Interactions between the neural regulation of stress and aggression. J Exp Biol. 209:4581–4589. Taborsky B, Oliveira RF. 2012. Social competence: an evolutionary approach. Trends Ecol Evol. 27:679–688.

 Takegaki T, Svensson O, Kvarnemo C. 2012. Socially induced tactic change in 2 types of sand goby sneaker males. Behav Ecol. 23:742–750. Taylor JN, Lattanzio MS. 2016. Boldness, dominance, and territoriality in the color polymorphic tree lizard, Urosaurus ornatus. Ethology. 122:892–901. Tóth M, Halász J, Mikics É, Barsy B, Haller J. 2008. Early social deprivation induces disturbed social communication and violent aggression in adulthood. Behav Neurosci. 122:849-854. Trannoy S, Penn J, Lucey K, Popovic D, Kravitz EA. 2016. Short and long-lasting behavioral consequences of agonistic encounters between male Drosophila melanogaster. Proc Natl Acad Sci USA. 113:4818–4823. Verbeek MEM, Boon A, Drent PJ. 1996. Exploration, aggressive behavior and dominance in pair-wise confrontations of juvenile male great tits. Behaviour. 133:945–963. Verbeek MEM, Drent PJ, Wiepkema PR. 1994. Consistent individual differences in early exploratory behaviour of male great tits. Anim Behav. 48:1113–1121. Wilson DS. 1998. Adaptive individual differences within single populations. Philos Trans R Soc Lond B Biol Sci. 353:199–205. Wilson DS, Clark AB, Coleman K, Dearstyne T. 1994. Shyness and boldness in humans and other animals. Trends Ecol Evol. 9:442–446. Wilson ADM, Godin J-GJ. 2009. Boldness and behavioral syndromes in the bluegill sunfish, Lepomis macrochirus. Behav Ecol. 20:231–237. Wilson DR, Nelson XJ, Evans CS. 2009. Seizing the opportunity: subordinate male fowl respond rapidly to variation in social context. Ethology 115:996–1004. Wolf M, van Doorn GS, Leimar O, Weissing FJ. 2007. Life-history trade-offs favour the evolution of animal personalities. Nature 447: 581–584. Wolf M, Weissing FJ. 2010. An explanatory framework for adaptive personality differences. Phil Trans R Soc B. 365:3959–3968. Zidar J, Sorato E, Malmqvist AM, Jansson E, Rosher C, Jensen P, Favati A, Løvlie H. 2017. Early experience affects adult personality in the red junglefowl: a role for cognitive stimulation? Behav Process. 134:78–86. Zuk M, Johnsen TS. 2000. The role of hen’s weight and recent experience on dyadic conflict outcome. Behav Ecol. 11:146–153. Zuk M, Thornhill R, Ligon DJ, Johnson K, Austad S. et al. 1990. The role of male ornaments and courtship behaviour in female mate choice of red jungle fowl. Am Nat. 136:459–473. Øverli Ø, Korzan WJ, Hoglund E, Winberg S, Bollig H, et al. 2004. Stress coping style predicts aggression and social dominance in rainbow trout. Horm Behav. 45:235–241. 

   Sammanfattning på svenska

I djurvärlden pågår en ständig konkurrens om begränsade resurser såsom föda, revir och partners för fortplantning. Som ett sätt att minska risken för kostsamma och riskfyllda konflikter upprättas inom många arter dominansförhållanden genom aggressiva interaktioner som inkluderar hot och ibland fysiska attacker. Interaktionen slutar med att en av deltagarna ger upp, och därefter konsekvent undviker den andre deltagaren. Den undvikande individen klassas därmed som underlägsen, och den andre som dominant. Den sociala dominanspositionen kan ha stor betydelse för individens överlevnad och/eller fortplantning, då dominanta individer ofta åtnjuter större tillgång till viktiga resurser. Flera olika faktorer är positivt kopplade till chansen att vinna en social interaktion, såsom kroppsstorlek, storlek på horn osv, men även beteendedrag (t.ex. aggressivitet) och tidigare erfarenheter av vinst (och motsatt effekt av tidigare erfarenhet av förlust). Man har nyligen upptäckt att dominanta individer inte bara är mer aggressiva, utan även mer risktagande, utforskande och aktiva jämfört med underlägsna individer. Dessa beteendedrag kan också kan beskriva en individs personlighetstyp. Personlighet hos djur definieras ofta som att individer av samma art skiljer sig åt i beteenderesponser på ett konsekvent sätt. Sedan början av 2000-talet har man inom beteendebiologin börjat intressera sig alltmer för personlighet hos djur, och personlighetsvariation har nu beskrivits hos en rad arter, från olika insekter och fiskar till fåglar och däggdjur. Det kausala sambandet mellan vissa personlighetsdrag och social dominans har dock sällan undersökts och därmed än inte fastställts. Detta samband skulle kunna bero på gemensamma underliggande mekanismer. Andra möjliga förklaringar till sambandet mellan personlighet och social dominans är att vissa personlighetstyper har större chans att vinna en social interaktion, eller så kan innehavandet av en social position leda till att individer beter sig olika. Det senare är en intressant fråga som potentiellt skulle kunna förklara en del av den stora variationen i personlighetstyper som observerats. Man vet nämligen inte säkert hur beteendemässig variation mellan individer, och stabilitet inom individen, bibehålls över tid och mellan generationer. Flera orsaker har diskuterats, däribland att stabil variation mellan individer i yttre eller inre tillstånd (t.ex energireserver, metabolism, social miljö) skulle kunna leda till stabil variation även i beteende. Att ha samma sociala position under en längre tid skulle potentiellt kunna utgöra ett

 sådant stabilt tillstånd som leder till långvariga effekter på beteendet, och därigenom ge upphov till beteendeskillnader. Detta har teoretiskt föreslagits, men empiriskt är denna hypotes ännu inte utförligt undersökt.

I den här avhandlingen undersöker jag förhållandet mellan social dominans och personlighet hos tuppar, Gallus gallus domesticus. Höns är en grupplevande art. Under naturliga förhållanden bildar höns grupper på 2-15 individer, med en dominant tupp, en eller flera underlägsna tuppar och flera hönor. Det är en sexuellt promiskuös art, vilket betyder att de inte bildar par, utan både hanar och honor har många olika parningspartners. Detta innebär att framförallt hanarna konkurrerar om tillgången till honor att para sig med. För att minska antalet konflikter bildas dominanshierarkier. Genom hot, och om det behövs, attacker gör tupparna upp om den sociala rangen. När den väl är etablerad kan tupparna leva samman utan att ideligen behöva slåss. Den mest dominanta tuppen i gruppen har störst tillgång till hönorna, och därmed även en större reproduktiv framgång och högre fitness. Genom en rad experimentella studier undersöker jag här det kausala orsakssambandet mellan social dominans och personlighet; är en tupp tuff för att han är dominant, eller är han dominant för att han är tuff? Jag testar om social dominansposition och erfarenhet av att vinna eller förlora (så kallade vinnar- och förlorareffekter där vinst gör att man fortsätter vinna och förlust leder till vidare förluster) påverkar beteendet på kortare och längre sikt, samt undersöker om personlighetstyp och sociala erfarenheter påverkar chansen att vinna en aggressiv interaktion och bli dominant.

Inledningsvis visar jag att tuppar som nyligen intagit en dominant position är mer aktiva, utforskande och vaksamma i en ny miljö, jämfört med tuppar som nyligen blivit underlägsna (artikel I). Genom att experimentellt manipulera social status kunde jag visa att dessa egenskaper var både flexibla och följde med ändringen i tuppens status, samtidigt som en del av variationen i beteende mellan individerna förklarades av deras personlighetstyp – det fanns en signifikant grad av konsekvens i deras beteenderesponser. Risktagande och galande (en territoriell vokalisering) däremot, var stabilt inom individen, och ändrades inte nämnvärt av ett byte av social status. Att status till viss del förklarade tupparnas beteende tydde på att social status kunde vara ett tillstånd som orsakar och cementerar skillnader i beteenderespons mellan individer, åtminstone så pass länge som den sociala statusen bibehålls. För att undersöka om så var fallet utfördes en liknande studie (artikel II), där tupparna tilläts leva samman i sina respektive sociala roller under en längre tid – 3 veckor – innan beteendet utvärderades i olika personlighetstest. I denna studie kunde jag inte spåra några större skillnader i beteende mellan dominanta och underlägsna tuppar, men däremot uppvisade tupparna återigen konsekvens i sitt beteende. Detta

  tyder sammantaget på att personlighet är det som bäst förklarar tupparnas beteende i stabila grupper och att social status inte verkar orsaka skillnader i personlighet på längre sikt. Vidare utförde jag en studie (artikel V) där tuppar fick växa upp under olika sociala betingelser: som den enda tuppen, som dominant i ett par av tuppar eller som intermediärt rankad tupp i en grupp av tuppar. Även denna studie tydde på att den sociala rollen inte hade någon större inverkan på tuppens personlighet.

Kan den observerade skillnaden mellan dominanta och underlägsna individer istället bero på att vissa personlighetstyper har större chanser att tillskansa sig en dominant position än andra? Genom att först mäta tuppars respons i olika personlighetstest, och därefter sätta ihop dem i par för att se vem som vinner och blir dominant, kunde jag visa att tuppar som är mer aggressiva, utforskande eller vaksamma hade större chanser att vinna (artikel III, IV, V). Då dessa personlighetsdrag inte var starkt korrelerade tyder det på att flera olika personlighetstyper kan uppnå en hög social status. Eftersom en hög social status är positivt kopplad till reproduktiv framgång skulle detta i förlängningen kunna utgöra en koppling mellan personlighet och vilka personlighetstyper som är framgångsrika över generationer. Särskilt aggressivitet mellan tuppar har stor betydelse i sociala konflikter, och kan överspela effekten av att nyligen ha upplevt en vinst eller förlust (artikel IV), av att ha vuxit upp under olika sociala betingelser (artikel V) samt ha större betydelse än kroppsliga egenskaper såsom storlek och kamstorlek (kammens storlek signalerar vanligtvis tuppens kvalitet och/eller status, artikel V). Sammantaget visar dessa studier att social status kan ha starka kortvariga effekter på beteendet, men att status inte kan förklara varför individer skiljer sig åt i beteende på längre sikt. Det motsatta kausala sambandet, att tuppens personlighetstyp kan förutsäga vilken status han får, fick större stöd. Denna avhandling belyser viktiga konsekvenser av individuella skillnader i beteende, och bidrar till den snabbt växande litteraturen som undersöker orsaker och konsekvenser av variation i personlighet hos djur.

  Acknowledgements

It is with the deepest gratitude I wish to thank my supervisors Olof Leimar, Hanne Løvlie and Tommy Radesäter. You offered me a wonderful opportunity to dive into the world of science, and guided me with great skill during the process to become a doctor in biology. Tommy, you retired before I got half way here (yes, I did produce more babies than articles the first years, so things did go a bit slow at first), but you provided stable support, thank you for that. Olle, it has been the greatest pleasure to be your student. I am truly honoured. You have generously shared your wisdom, great stories and good laughs. You have pushed me to develop my own research skills in a very efficient way. I remember when I had some statistical query for my first analysis. You put a statistics book in my hands and said, ‘let’s meet again in two weeks’. Hanne, thank you so much for guiding me through academia, pointing at all the important stuff all along the way. You have been a great scientific role model, and you have taught me so much about writing by commenting (and slaughtering) my manuscripts. Always with fast replies (‘Blixten’!). Besides that, you have been the most generous supervisor, inviting me to your home in Linköping, and buying me sweet gifts.

Dear Zootis colleagues, I have enjoyed every day in your company. What an atmosphere - inspirational, friendly and fun. Thank you, Siw, Anette and Berit for all your help and advice. A special thanks to Minna for the good times during teaching and environmental work. I was lucky enough to get a spot in the best office, beautifully decorated (thanks David and Josefin!) and with awesome roomies: Martin O and Karolina T, I will always cherish the times we shared, talking about life, science, cookies and politics. My new roomies Simon and Mariana, thank you for good company, and in between, some serious work-promoting (non-awkvard) silence. Thank you Baharan, Sandra S, Alexander Marianne and Johanna for good times at work and after it. A big hug to Kristin, Sandra, Lovisa, Alexandra B, Marianne, Aleksandra, Jessica and all other old and new friends at the department, PhD students as well as seniors. Thank you, Josefina Z, for being a super nice partner in crime over at Linköping. I have really enjoyed your company during work journeys. Laura, thank you for proof-reading the manuscript.

  I want to thank the staff at Tovetorp with all my heart. Working and hanging out at Tovetorp during the lovely summer seasons, and sometimes also in the winter, has been like having a second home. Including around 120 adorable pets (yes, chickens). Thank you, Sven, for moral support when things got tough at the beginning of the studies, and for being a warm and welcoming manager at the research station. Thank you, Nisse, for all our heartily talks and for taking care of our dear chickens over the years. Thank you, Thomas, for solving every practical matter with a screwdriver and good spirits. And thank you Gunilla and the rest of the kitchen staff for all the good food!

I am very grateful for the financial support given by Alice and Lars Silén’s fund, Helge Ax:son Johnson foundation, Knut and Alice Wallenberg foundation, Rhodins minne, the Lars Hierta Memorial Foundation, the Längmanska Cultural Foundation, the Royal Swedish Academy of Agriculture and Forestry, the Royal Swedish Academy of Sciences, Tullberg foundation and Wilhelm Leche’s foundation.

Mamma – tack för att du alltid funnits där för mig, och alltid trott på mig, vad jag än tar mig för. Du har lärt mig så mycket om livet, om människor och allt möjligt. Jag vore inget utan dig. Pappa – ditt intresse för naturen och din nyfikenhet på hur allt fungerar smittade tidigt av sig, och jag har dig även att tacka för din inspirerande uppfinningsrikedom. Mitt varmaste tack till hela tjocka släkten, på både östkust, västkust och i Finland, för att ni är den bästa supporterklubb en människa kan ha. Ni är guld!

Mitt allra, allra, varmaste tack går till min älskade make Magnus, mitt största och mest ointresserade stöd. Tack för att du har stöttat och peppat mig, och backat upp där hemma när jag i perioder har slukats av jobb. Vide, Grim, Björn och Lovis. Ni är allt för mig. Tack för att ni dagligen påmint mig om att det riktigt viktiga livet pågår där hemma.