UNIVERSITY OF LJUBLJANA BIOTECHNICAL FACULTY Elena GOBBO MECHANISMS OF AGGRESSIVE BEHAVIOUR IN DOGS DOCTORAL DISSERTATION Ljubljana, 2022 UNIVERSITY OF LJUBLJANA BIOTECHNICAL FACULTY Elena GOBBO MECHANISMS OF AGGRESSIVE BEHAVIOUR IN DOGS DOCTORAL DISSERTATION MEHANIZMI AGRESIVNEGA OBNAŠANJA PRI PSIH DOKTORSKA DISERTACIJA Ljubljana, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. II Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Based on the Statute of the University of Ljubljana and the decision of the Biotechnical Faculty senate, as well as the decision of the Commission for Doctoral Studies of the University of Ljubljana adopted on 24 September 2019 it has been confirmed that the candidate meets the requirements for pursuing a PhD in the interdisciplinary doctoral programme in Biosciences, Scientific Field Animal Science. Assoc. Prof. Manja Zupan Šemrov is appointed as supervisor. Na podlagi Statuta Univerze v Ljubljani ter po sklepu Senata Biotehniške fakultete in sklepa Komisije za doktorski študij Univerze v Ljubljani z dne 24. 9. 2019 je bilo potrjeno, da kandidatka izpolnjuje pogoje za opravljanje doktorata znanosti na Interdisciplinarnem doktorskem študijskem programu Bioznanosti, znanstveno področje znanost o živalih. Za mentorico je bila imenovana izr. prof. dr. Manja Zupan Šemrov. Supervisor (mentorica): Assoc. Prof. Dr. Manja ZUPAN ŠEMROV University of Ljubljana, Biotechnical Faculty, Department of Animal Science Committee for evaluation and the defense (Komisija za oceno in zagovor): Chairman (predsednica): Assist. Prof. Dr. Dušanka JORDAN University of Ljubljana, Biotechnical Faculty, Department of Animal Science Member (članica): Prof. Dr. Marko KREFT University of Ljubljana, Biotechnical Faculty, Department of Biology Member (članica): Assoc. Prof. Dr. Friederike RANGE University of Veterinary Medicine Vienna, Department of Interdisciplinary Life Sciences Date of defense (datum zagovora): Elena Gobbo Gobbo E. Mechanisms of aggressive behaviour in dogs. III Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 KEY WORDS DOCUMENTATION ND Dd DC UDC 636.7:591.5(043.3)=111 CX dogs, animal behaviour, ethology, aggressive behaviour, cognition, personality, attachment styles, animal physiology AU GOBBO, Elena AA ZUPAN ŠEMROV, Manja (supervisor) PP SI-1000 Ljubljana, Jamnikarjeva 101 PB University of Ljubljana, Biotechnical Faculty, Interdisciplinary Doctoral Programme in Biosciences, Scientific Field Animal Science PY 2021 TI MECHANISMS OF AGGRESSIVE BEHAVIOUR IN DOGS DT Doctoral dissertation NO VIII, 106 p., 1 fig., 4 ann., 130 ref. LA en AL en / sl AB In order to investigate the mechanisms associated with dog aggression using a multidisciplinary approach, four separate studies were conducted as part of this thesis. In the first study, retrospective questionnaire data were used to determine the contextual aspects of dog bites and revealed a high incidence of bites during unintended interaction with the biting dog. The second study examined psychosocial factors using a combination of behavioural tests and questionnaires reported by dog owners. Low sociability in dogs and high neuroticism in owners, as well as high avoidant and low anxious attachments between owner and dog were the characteristics associated with dog aggression. In the third study, three behavioural tests were used to examine the association between two different aspects of inhibitory control and aggression. Impaired self-control, measured as performance on the delay of gratification test, but not cognitive inhibition, measured as performance on the reversal learning test, was found to be associated with aggressive biting behaviour. In the final study, behavioural responses and several physiological changes were examined simultaneously and in real-time. Aggressive dogs were found to have decreased salivary serotonin concentration and increased facial surface temperature. In addition, frequent tail wagging and a tendency to wag the tail to the left during aggression were observed for the first time. Through new or improved approaches, our findings contribute to a better understanding of dog aggression, and may benefit not only the scientific community, but also the general public. Gobbo E. Mechanisms of aggressive behaviour in dogs. IV Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 KLJUČNA DOKUMENTACIJSKA INFORMACIJA ŠD Dd DK UDK 636.7:591.5(043.3)=111 KG psi, obnašanje živali, etologija, agresivno obnašanje, kognicija, osebnost, navezanost, fiziologija živali AV GOBBO, Elena, mag. kog. zn. SA ZUPAN ŠEMROV, Manja (mentorica) PP SI-1000 Ljubljana, Jamnikarjeva 101 KZ Univerza v Ljubljani, Biotehniška fakulteta, Interdisciplinarni doktorski študijski program Bioznanosti, znanstveno področje znanost o živalih LI 2021 IN MEHANIZMI AGRESIVNEGA OBNAŠANJA PRI PSIH TD Doktorska disertacija OP VIII, 106 str., 1 sl., 4 pril., 130 vir. IJ en JI en / sl AI Za ocenjevanje mehanizmov, povezanih s pasjo agresijo, so bile z uporabo multidisciplinarnega pristopa v okviru te disertacije izvedene štiri ločene študije. V prvi študiji je bil za določitev kontekstualnih vidikov pasjih ugrizov uporabljen retrospektivni vprašalnik. Razkril je visoko incidenco ugrizov med nenamerno interakcijo z ugrizlim psom. Druga študija je s kombinacijo testiranja obnašanja in vprašalnikov, ki jih je izpolnil lastnik, raziskovala psihosocialne dejavnike. Nizka družabnost pri psih, visok nevroticizem pri lastnikih ter visoka izogibajoča navezanost lastnika in psa ter nizka anksiozna navezanost lastnika in psa so bile lastnosti, povezane z agresijo psov. V tretji študiji je bila s tremi testi obnašanja raziskana povezava med dvema različnima vidikoma inhibicijske kontrole in agresije. Izkazalo se je, da je oslabljen samonadzor, merjen kot uspešnost pri nalogi zapoznelega nagrajevanja, vendar ne tudi kognitivna inhibicija, merjena kot uspešnost pri nalogi obratnega učenja, povezana z agresivnim grizenjem. V zadnji študiji so bili sočasno in v realnem času preučeni obnašanje in več fizioloških sprememb. Ugotovljeno je bilo, da so imeli agresivni psi znižano koncentracijo serotonina v slini in povišano površinsko temperaturo obraza. Med to študijo sta bila tudi med agresijo prvič opažena pogosto mahanje z repom in nagnjenost k mahanju repa v levo stran. Z uporabo novih ali izboljšanih pristopov so te ugotovitve pomembne za nadaljnje razumevanje agresije pri psih. Z novimi ali izboljšanimi pristopi naše ugotovitve izpopolnjujejo razumevanje agresije psov in lahko koristijo ne le znanstveni skupnosti, ampak tudi širši javnosti. Gobbo E. Mechanisms of aggressive behaviour in dogs. V Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 TABLE OF CONTENTS KEY WORDS DOCUMENTATION III KLJUČNA DOKUMENTACIJSKA INFORMACIJA IV TABLE OF CONTENTS V TABLE OF CONTENTS OF SCIENTIFIC WORKS VI LIST OF FIGURES VII LIST OF ANNEXES VIII 1 PRESENTATION OF THE PROBLEM AND HYPOTHESES 1 1.1 SPECIFICATION OF THE RESEARCH PROBLEM 1 1.1.1 Contexts, dogs’ and dog bite victims’ characteristics 2 1.1.2 Psychosocial effects 3 1.1.3 Cognitive effects 4 1.1.4 Physiological effects 5 1.2 AIMS OF THE STUDY AND RESEARCH HYPOTHESES 6 2 SCIENTIFIC WORKS 8 2.1 PUBLISHED SCIENTIFIC WORKS 8 2.1.1 Factors affecting human-directed aggression resulting in dog bites: 8 contextual aspects of the biting incidents 2.1.2 Dogs’ sociability, owners’ neuroticism and attachment style to pets as 29 predictors of dog aggression 2.1.3 Dogs exhibiting high levels of aggressive reactivity show impaired 48 self-control abilities 2.1.4 Neuroendocrine and cardiovascular activation during aggressive 60 reactivity in dogs 3 DISCUSSION AND CONCLUSIONS 75 3.1 DISCUSSION 75 3.1.1 Contexts, dogs’ and dog bite victims’ characteristics 75 3.1.2 Psychosocial effects 76 3.1.3 Cognitive effects 78 3.1.4 Physiological effects 80 3.2 CONCLUSIONS 82 4 SUMMARY (POVZETEK) 84 4.1 SUMMARY 84 4.2 POVZETEK 91 5 REFERENCES 98 ACKNOWLEDGEMENTS ANNEXES Gobbo E. Mechanisms of aggressive behaviour in dogs. VI Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 TABLE OF CONTENTS OF SCIENTIFIC WORKS Gobbo E., Zupan Šemrov M. 2021. Factors affecting human-directed aggression resulting in dog bites: contextual aspects of the biting incidents. Society & Animals (published online ahead of print), doi: 10.1163/15685306-bja10066: 20 p. 8 Gobbo E., Zupan M. 2020. Dogs’ sociability, owners’ neuroticism and attachment style to pets as predictors of dog aggression. Animals, 10: 315, doi: 10.3390/ani10020315: 15 p. 29 Gobbo E., Zupan Šemrov M. 2022. Dogs exhibiting high levels of aggressive reactivity show impaired self-control abilities. Frontiers in Veterinary Science, 9: 869068, doi: 10.3389/fvets.2022.869068: 10 p. 48 Gobbo E., Zupan Šemrov M. 2021. Neuroendocrine and cardiovascular activation during aggressive reactivity in dogs. Frontiers in Veterinary Science, 8: 683858, doi: 10.3389/fvets.2021.683858, 12 p. 60 Gobbo E. Mechanisms of aggressive behaviour in dogs. VII Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 LIST OF FIGURES Figure 1: Schematic representation of the main findings 83 Gobbo E. Mechanisms of aggressive behaviour in dogs. VIII Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 LIST OF ANNEXES Annex A: Permission of Brill to use article: Gobbo E., Zupan Šemrov M. 2021. Factors affecting human-directed aggression resulting in dog bites: contextual aspects of the biting incidents. Society & Animals (published online ahead of print) in the electronic version of the doctoral dissertation. Annex B: Permission of MDPI to use article: Gobbo E., Zupan M. 2020. Dogs’ sociability, owners’ neuroticism and attachment style to pets as predictors of dog aggression. Animals, 10: 315 in the electronic version of the doctoral dissertation. Annex C: Permission of Frontiers to use article: Gobbo E., Zupan Šemrov M. 2022. Dogs exhibiting high levels of aggressive reactivity show impaired self-control abilities. Frontiers in Veterinary Science, 9: 869068 in the electronic version of the doctoral dissertation. Annex D: Permission of Frontiers to use article: Gobbo E., Zupan Šemrov M. 2021. Neuroendocrine and cardiovascular activation during aggressive reactivity in dogs. Frontiers in Veterinary Science, 8: 683858 in the electronic version of the doctoral dissertation. Gobbo E. Mechanisms of aggressive behaviour in dogs. 1 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 1 PRESENTATION OF THE PROBLEM AND HYPOTHESES 1.1 SPECIFICATION OF THE RESEARCH PROBLEM Aggression can be observed in a variety of animal species, including dogs. It is a behaviour that threatens or inflicts physical or psychological pain (Anderson and Bushman, 2002). In humans, it can be further categorized as physical aggression, when the harmful behaviour results in pain or injury, or verbal aggression, when there is a harmful threat (Buss, 1961). Similarly, dog aggression can be grouped and expressed as the aggressive biting behaviour, by snapping, attacking or attempting to bite, and the aggressive threatening behaviour, by growling, barking and baring teeth (Netto and Planta, 1997). Other behavioural possibly indicative of aggression include staring, freezing, tail lifting, closed mouth with pursed lips, etc. (Christensen et al., 2007). Aggression can also be categorized based on the assumed internal motivation of the dog, for example territorial-, fear-, possessiveness- and dominance related aggression or more objectively, by target, as stranger-, owner-, and dog-directed aggression (Houpt, 2006). Although it is part of the normal behaviour of dogs (Netto and Planta, 1997), the level of aggression in individual dogs may be so high that it is unacceptable to the immediate environment or to society in general, due to the close relationship and coexistence with humans. It is considered one of the most common and dangerous behavioural problems in dogs, especially when it is directed towards owners or other people (Casey et al., 2014; Flint et al., 2015). Dog owners often do not have sufficient knowledge regarding safe interaction with dogs and identification of warning signals of aggression (Reisner and Shofer, 2008). Often, parents do not provide appropriate education, supervision and intervention when their children interact with dogs (Arhant et al., 2016). For this reason, a significant number of people, mostly children, are bitten by a dog every year and bites can have different consequences. A large number of bitten people need to seek professional medical help, and more severe bites can lead to hospitalization, physical and social handicap or even victim’s death, making dog bites a public health problem (Súilleabháin, 2015; Mora et al., 2018). Aggressive behaviour is also a common reason for dogs being relinquished to animal shelters, abandoned, or euthanized (Salman et al., 2000; Diesel et al., 2008), making the consequences of dog aggression an animal welfare and human-animal interaction issue. There are individual differences in how dogs express or inhibit behavioural responses in the environment. These responses can be associated with a number of internal and external factors (Haug, 2008), as measured by physiological changes, behavioural measures and owner reports. Each methodological approach has its advantages and disadvantages. For example, owner reports, despite involving people that are familiar with the dog’s everyday behaviour (Hsu and Serpell, 2003), include owners that have limited knowledge regarding Gobbo E. Mechanisms of aggressive behaviour in dogs. 2 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 dog behaviour or its interpretation and can be influenced by owner’s bias (Wiener and Haskell, 2016). On the other hand, behavioural observations, especially behavioural tests are time consuming and allow smaller and less diverse sample (Hsu and Serpell, 2003). Moreover, while observing physiological changes, especially during real-time behaviour, it is important to measure multiple physiological parameters simultaneously (Reefmann et al., 2009) and non-invasively, to avoid the influence of measuring devices on the behavioural responses (Ermatinger et al., 2019). For these reasons, the investigation of mechanisms of aggression in this thesis involves a combination of different methodological approaches and simultaneous measurement of different physiological parameters. We started by exploring external factors, focusing on the contexts of dog bites, then moved to psychosocial and cognitive aspects and concluded with physiological aspects. The following chapters present these main aspects of dog aggression, studied in this thesis. 1.1.1 Contexts, dogs’ and dog bite victims’ characteristics While focusing on biting behaviour only, previous studies reported on the characteristics of the dogs involved, suggesting large, adult males, known to the victim to be the most common (Rosado et al., 2009; Sarcey et al., 2017; Oxley et al., 2018). There is a wide variety of breeds commonly involved in biting incidents, from larger breeds such as German Shepherds, Rottweilers (Sarcey et al., 2017; Oxley et al., 2018), and Belgian Shepherds (Cornelissen and Hopster, 2010), to smaller breeds such as Jack Russell Terriers (Cornelissen and Hopster, 2010), Shih Tzus (Messam et al., 2012) and English Cocker Spaniels (Fatjó et al., 2007). Bites from larger breeds are often reported when observing hospital data, while bites from smaller breeds may also be common, but victims often do not seek medical attention (Westgarth et al. 2018). Children (Súilleabháin, 2015), particularly boys (Basco et al., 2020), are the most common victims of dog bites and resulting injuries can range from superficial wounds, to severe injuries, as children are often bitten on the face, neck and head (Morgan and Palmer, 2007). The prevalence of dog bites decreases with age and the fewest bites are observed in older adults (Quirk, 2012). Similarly, as in children, males are more likely to be involved in a biting incident than females in adults (Súilleabháin, 2015; Westgarth et al., 2018). In terms of the environment in which biting incidents occur, previous data suggest that non-public areas, mainly outside or inside a residential area are the most common (Cornelissen and Hopster, 2010; Oxley et al., 2018). A large proportion of children under the age of four are bitten at their own homes (De Keuster et al., 2006). Positive and negative contact activities with the dog, such as petting, playing (Horisberger et al., 2004; Cornelissen and Hopster, 2010; Oxley et al., 2018) or manipulation of the dog in an aversive manner (Rosado et al., 2009) appear to be the most commonly reported contexts for dog bites. Often, even a Gobbo E. Mechanisms of aggressive behaviour in dogs. 3 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 change in body posture or eye contact can trigger an aggressive response (Reisner et al., 2007). Nevertheless, these descriptions of dog bite incidents are found in only few studies (Westgarth and Watkins, 2015; Oxley et al., 2018; Owczarczak-Garstecka et al., 2018a, Owczarczak-Garstecka et al., 2018b) based on a limited sample size. This means that most studies lack a thorough description of dog bites and detailed information about the circumstances of the interaction or attempted interaction, especially prior to the incidents, as well as a precise description of the location where the bite occurred. There is also a lack of information such as the restriction of movement of the dog and more detailed information about the dogs involved, such as their housing conditions, prior socialization, origin, training history, etc. 1.1.2 Psychosocial effects The study of psychosocial effects related to aggression has shown that some psychosocial aspects of dogs and owners may be associated with aggressive behaviour in dogs. The first aspect is dog personality, defined as inter-individual behavioural traits that are consistent over time and across contexts (Fratkin et al., 2013). Animal personality is usually assessed using standardized tests (Dingemanse and Wolf, 2010). One of the most used tests in dogs is the Dog Mentality Assessment (DMA) (Svartberg and Forkman, 2002). During DMA, by observing dog’s reaction to external stimuli, the dogs are scored on five personality traits; playfulness, curiosity/fearlessness, chase-proneness, sociability, aggressiveness, and a broader dimension named shyness/boldness (Svartberg and Forkman, 2002). Another commonly used methods for assessment of dog personality are owner-reported questionnaires (Jones and Gosling, 2005). Their use revealed that less sociable dogs tended to show higher levels of aggression towards children and strangers (Kaneko et al., 2013) and that more fearful dogs showed more fear-related aggression and dog-directed aggression (Haverbeke et al., 2009; Arata et al., 2014). Next to dog’s personality traits, the owner’s personality also has an important influence on dog’s behaviour, as the human factors have greater impact on the dog-human relationship than dog factors (Meyer and Forkman, 2014). This phenomenon may be due to cohabitation and shared activities, leading to emotional contagion, or due to owner’s selecting dogs that matches their personality and lifestyle, as seen in humans (Tidwell et al., 2013). Human personality is most commonly assessed using the Big Five factor taxonomy, which identifies five personality traits (extraversion, conscientiousness, neuroticism, agreeableness, and openness) (Rammstedt and John, 2007). Moreover, Turcsán et al. (2012) found a positive correlation between owners and dogs in all five investigated personality traits. Regarding dog aggression, previous studies have shown that dogs whose owners have lower scores for Gobbo E. Mechanisms of aggressive behaviour in dogs. 4 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 agreeableness, emotional stability, extraversion and conscientiousness show greater aggression toward their owners and fear of strangers (Dodman et al., 2018). Another human factor that can potentially influence dog behaviour is attachment. Attachment is a cognitive-emotional bond that was first used to describe the affectional bond between human children and caregivers and later between humans and places, objects and non-human animals (Bell and Spikins, 2017; Meehan et al., 2017), including dogs (Archer and Ireland, 2011). There are secure and insecure attachment types. When the attachment between two individuals is categorized by trust and comfort with intimacy, it can be described as secure. In contrast, two types of insecure attachment, anxious and avoidant attachment, can be defined by controlling behaviour and avoidance of intimacy, respectively (Beck and Madresh, 2008). Attachment has not yet been studied in the relation to dog aggression, but it has been previously reported that owners’ attachment styles play a role in the occurrence of different behavioural strategies in their dogs during aversive situations (Rehn et al., 2017) and occurrence of behavioural problems in dogs (Konok et al., 2015). To date, psychosocial factors, that play or may play a role in dog aggression, have only been assessed using owner-reported questionnaire data and attachment styles have not yet been studied in this context. As mentioned above, due to limitations of owner reports, for more objective and comprehensive assessment of psychosocial effect, the combination of different research methods, for example behavioural testing conducted by professionals in the field and questionnaire-based assessment by the owners, is needed. 1.1.3 Cognitive effects Higher-order cognitive processes, involved in the self-regulation of emotions and actions, including aggression, can be referred to as executive control (Séguin and Zelazo, 2005). One of the proposed processes is inhibitory control, defined as an individual ability to block an immediate response in favour of a delayed but more appropriate behaviour (Bray et al., 2014), and has previously been associated with aggression in human adults (Anderson and Bushman, 2002; Hsieh and Chen 2017) and children (Raaijmakers et al., 2008). In dogs, inhibitory control has not yet been studied in relation to dog aggression, but it has been previously reported that dogs have the ability to inhibit behavioural responses that are unwanted by their owners (Gácsi et al., 2009). In addition, it has been suggested that cognitive impairments (Denenberg et al., 2014) and owner-reported trait impulsivity (Wright et al., 2011) play a role in dog aggression. Inhibitory control is not a unitary mechanism, but a collection of distinct cognitive processes (Beran, 2015; Brucks et al., 2017). In dogs, it is usually assessed with different cognitive Gobbo E. Mechanisms of aggressive behaviour in dogs. 5 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 tests (Bray et al. 2014; Brucks et al., 2017; Vernouillet et al., 2018), each targeting separate aspect of this ability. Motor inhibition, cognitive inhibition and self-control are the three most commonly assessed aspects of inhibitory control in dogs (Brucks et al., 2017; Brucks et al., 2019). Two of these aspects, self-control and cognitive inhibition, have previously been associated with aggression in humans (Mitchell et al., 2006; Herndon et al., 2015), but research on the association with dog aggression is lacking. Self-control is defined as the ability to control an impulse response in a tempting situation (Beran, 2015). It can be measured using an exchange paradigm (e.g., delay of gratification) in which an individual must resist immediate gratification for the sake of delayed, but better quality reward (Mischel et al., 1989). This ability to wait is proposed to be an evidence of self-control, because it leads to receiving a better reward in a given situation (Beran, 2015). In addition, it has been suggested that better self-control influences the ability to respond or override the urge to react aggressively (Denson et al., 2012). While this has to yet been explored in dogs, it has been reported in humans (Herndon et al., 2015) and rats (Van den Bergh et al., 2006) that impaired self-control is associated with a more frequent occurrence of aggression. Cognitive inhibition is defined as the ability to regulate the content of working memory by removing insignificant information in a given situation (Hasher et al., 1999). It can be measured using an object discrimination paradigm, often referred to as reversal learning, during which after an initial discrimination, two stimuli change their reward contingencies (Milgram et al., 1994). During the paradigm, the ability to inhibit a learned response and avoid the previously rewarded option, as well as flexibility in relearning object-reward contingencies, are measured (Milgram et al., 1994; Brucks et al., 2017). Impairments in reversal learning has previously been associated with aggression in humans (Mitchell et al., 2006), whereas the same results have not yet been reported in dogs. 1.1.4 Physiological effects In the field of physiological research, the authors reported cardiovascular and neuroendocrine changes associated with aggression. Heart rate (HR), heart rate variability (HRV) and skin (surface) temperature are the main studied cardiovascular parameters. Dogs exhibiting reacting behaviour in response to threatening stimuli have been found to have increased HR and decreased HRV (Gácsi et al., 2013), while dogs with history of aggression have lower resting HRV (Craig et al., 2017). Fewer studies have focused on changes in surface temperature related to dog aggression and also other animals. Only Rigternik et al. (2018) observed surface temperature during human-directed aggression and reported no changes between aggressive and non-aggressive dogs. In another study, Boileau et al. (2019) Gobbo E. Mechanisms of aggressive behaviour in dogs. 6 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 found decrease in dorsal surface temperature in fighting pigs. The remaining animal studies focused on other negative affective states and reported decreased nasal surface temperature in monkeys during aversive stimuli (Kuraoka and Nakamura, 2011; Ermatinger et al., 2019), decreased periocular area and ocular bulb surface temperature in rabbits during stress (Ludwig et al., 2007) and decreased nasal surface temperature in kennelled dogs (Part et al., 2014). To assess cardiovascular changes during aggression, especially in moving animals, surface temperature measured with infrared thermography seems to be better compared to HR and HRV measures. This is due to many limitations associated with moving artefacts, including displaced HR electrodes leading to false signals (Essner et al., 2015), poor electrodes conduction (Lensen et al., 2017) and intrusive measuring devices, often strapped to the chest of the subject that often require prior training with a dummy monitor (Lensen et al., 2017). In contrast, infrared thermography is non-invasive and can measure physiological changes in real time without potentially altering behavioural responses, making it a suitable tool to for studying aggression in real time. Regarding the neuroendocrine activations that modulate coping, cognitive and behavioural functions during internal and external stressors, the primary stress hormone cortisol (Veissier and Boissy, 2007) and the inhibitory neurotransmitter serotonin (Summers and Winberg, 2006) are the two commonly observed parameters. The release of cortisol helps the body to remain on high alert and provides the body with energy (Lee et al., 2015), while serotonin plays a role in behavioural inhibition and appropriate behavioural adaptations (Bari and Robbins, 2013). Observing both parameters in humans, Montoya et al. (2012) reported that a high ratio of testosterone and cortisol concentration, along with low serotonin concentration modulates impulsive aggression. Similarly, dogs with owner-reported aggression have been reported to have significantly lower serum serotonin concentration (Çakiroǧlu et al., 2007; Rosado et al., 2010; León et al., 2012) and higher plasma cortisol concentration (Rosado et al., 2010) compared to dogs with no history of aggression. To avoid observation of cortisol and serotonin in serum and plasma that requires invasive blood sampling and causes additional stress (Cook, 2012), both parameters can be observed in highly comparable saliva samples showing short-term physiological changes (Lensen et al., 2015). To date, there is lack of simultaneous investigation of aggression related behaviour in dogs and neuroendocrine activation measured non-invasively in real time. 1.2 AIMS OF THE STUDY AND RESEARCH HYPOTHESES Although dog aggression is a widely recognized problem, there are major knowledge gaps regarding the contextual, psychosocial, physiological, and cognitive factors that mediate this behaviour. The main aim of this thesis is to attempt to provide a comprehensive picture of the mechanisms involved in dog aggression through a combination of different methodological approaches in four separate studies. Gobbo E. Mechanisms of aggressive behaviour in dogs. 7 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 In the first study, which used retrospective owner-reported dog biting incidents, we predicted that investigation of larger sample of dog bites that are not limited to a prior dog-human interaction will lead to a greater number of contexts in which dog bites occur, than previously thought (Oxley et al., 2018). In the second study, based on previous reports using questionnaires only (Kaneko et al., 2013; Arata et al., 2014; Dodman et al., 2018), we predicted that dogs with owner-reported past aggressive behaviour will have higher trait scores for aggressiveness, chase-proneness and lower trait scores for playfulness, curiosity/fearlessness and sociability. In addition, aggressive dogs will be associated with owners with lower trait scores agreeableness, extraversion and conscientiousness and higher neuroticism and attachment scores. In the third study, based on human studies (Mitchell et al., 2006; Herndon et al., 2015), we hypothesized that dogs with higher aggression level during behavioural testing will show poorer inhibitory control on delay of gratification and reversal learning test. Our hypothesis in the fourth study was that dogs exhibiting aggressive behaviour during behavioural test would have increased salivary cortisol concentration and decreased salivary serotonin concentration in real time, as has been observed in other studies of dogs with history of aggressive behaviour (Rosado et al., 2010). We also predicted decreased body and facial surface temperature during aggression as observed in similar situations in other species (rabbits: Ludwig et al., 2007; monkeys: Kuraoka and Nakamura, 2011; pigs: Boileau et al., 2019). Gobbo E. Mechanisms of aggressive behaviour in dogs. 8 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 2 SCIENTIFIC WORKS 2.1 PUBLISHED SCIENTIFIC WORKS 2.1.1 Factors affecting human-directed aggression resulting in dog bites: contextual aspects of the biting incidents Gobbo E., Zupan Šemrov M. 2021. Factors affecting human-directed aggression resulting in dog bites: the contextual aspects of the biting incidents. Society & Animals (published online ahead of print), doi: 10.1163/15685306-bja10066: 20 p. Using a web-based questionnaire with 29 close- and open-ended questions about 400 self-reported dog-biting incidents in Slovenia, this research investigated the contexts of dog bites, focusing on characteristics of the dogs and the descriptions of the situations to identify the main risk factors for the occurrence of dog bites. Even though it has been suggested that most dog bites occur during initially non-aggressive interactions with the dog (e.g., during petting, playing), in our study we discerned a wide variety of contexts, including those in which the person did not intend to interact with the dog. Most victims reported unprovoked bites during fast movements near the dog, while coming into close proximity, and during incidents without a reason. These incidents more likely occurred in public than private places and were associated with purebred dogs with a history of aggression. Gobbo E. Mechanisms of aggressive behaviour in dogs. 9 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 10 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 11 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 12 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 13 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 14 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 15 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 16 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 17 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 18 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 19 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 20 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 21 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 22 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 23 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 24 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 25 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 26 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 27 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 28 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 29 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 2.1.2 Dogs’ sociability, owners’ neuroticism and attachment style to pets as predictors of dog aggression Gobbo E., Zupan M. 2020. Dogs’ sociability, owners’ neuroticism and attachment style to pets as predictors of dog aggression. Animals, 10: 315, doi: 10.3390/ani10020315: 15 p. A dog’s aggressive behavior is influenced by external and internal factors, including its psychological profile. In this study, dogs’ and owners’ personalities and the owners’ attachment style to their dogs were identified and associated with owner-reported dog aggression towards humans and animals. Forty Slovenian owners participated with their dogs, of different breeds and aggression history, sorted into three groups (non-aggressive dogs, dogs aggressive towards humans, and dogs aggressive towards animals). The owners filled out three separate questionnaires that assessed dog aggression history towards different targets, owner’s personality and degree of insecure attachment styles to dogs; namely anxious and avoidant attachment. Dog personality was characterized using a standardized dog mentality assessment test, during which the dog was exposed to nine tasks, performed outside, and dogs were scored based on behaviors they exhibited. The results indicated that dogs which were aggressive towards humans were less sociable than non-aggressive dogs and this was associated with the higher neuroticism scores of their owners. We also found that dogs which were aggressive towards strangers had owners with lower scores for anxious attachment and that dogs which were aggressive towards owners had owners with higher scores for avoidant attachment. These results imply that the psychological profiles of both a dog and its owner influence dog aggression towards humans. Gobbo E. Mechanisms of aggressive behaviour in dogs. 30 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 31 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 32 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 33 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 34 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 35 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 36 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 37 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 38 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 39 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 40 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 41 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 42 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 43 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 44 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 45 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 46 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 47 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 48 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 2.1.3 Dogs exhibiting high levels of aggressive reactivity show impaired self-control abilities Gobbo E., Zupan Šemrov M. 2021. Dogs exhibiting high levels of aggressive reactivity show impaired self-control abilities. Frontiers in Veterinary Science, 9: 869068, doi: 10.3389/fvets.2022.869068: 10 p. Inhibitory control describes a multitude of cognitive processes that prevents an impulsive response and enables a more appropriate behavior in a given situation. The ability to inhibit undesirable behaviors, such as aggression, is particularly important in dogs for safe and successful interspecific interaction and cooperation. The present study investigated the associations between two aspects of inhibitory control in dogs, self-control and cognitive inhibition, and the tendency to respond aggressively when provoked. Sixteen police and fourteen privately owned dogs of the same sex, breed group and similar age participated. Self-control, often described as impulsivity, was measured with an exchange paradigm themed the delay of gratification test, and cognitive inhibition with an object discrimination paradigm called the reversal learning test. Aggressive reactivity was assessed with a standardized aggression-eliciting behavior test. When comparing police and privately owned dogs, police dogs showed higher aggression levels and poorer self-control, while the two groups did not differ in cognitive inhibition. Regardless of the dog group, the main results indicated impairments in self-control in dogs with high levels of aggressive reactivity. Dogs showing biting behavior had worse self-control abilities compared to dogs with no signs of aggression. No association between cognitive inhibition and aggression was found. We conclude that self-control, measured as the ability to tolerate delayed rewards, appears to be an important aspect of inhibitory control involved in the tendency to respond aggressively, particularly in police dogs. Gobbo E. Mechanisms of aggressive behaviour in dogs. 49 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 50 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 51 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 52 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 53 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 54 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 55 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 56 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 57 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 58 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 59 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 60 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 2.1.4 Neuroendocrine and cardiovascular activation during aggressive reactivity in dogs Gobbo E., Zupan Šemrov M. 2021. Neuroendocrine and cardiovascular activation during aggressive reactivity in dogs. Frontiers in Veterinary Science, 8: 683858, doi: 10.3389/fvets.2021.683858: 12 p. Our aim was to investigate cardiovascular activation by measuring changes in facial and body surface temperature using infrared thermography, and neuroendocrine activation using salivary cortisol (CORT) and serotonin concentration (SER) in dogs exhibiting aggressive reactivity in real time. Based on two factors, owner-reported past aggressive behaviors, and detailed behavioral observations collected during a Socially Acceptable Behavior test consisting of 16 subtests and, each individual was categorized as aggressive or non-aggressive. CORT and SER showed no difference in neuroendocrine activity between dogs, but aggressive dogs with higher levels of aggression were found to have lower SER. Aggressive dogs also had an increase in facial temperature from pre-test values. The discovery of a correlation between tail wagging and left tail wagging with aggression level and aggression-related behaviors in aggressive dogs is further evidence of the right hemisphere specialization for aggression previously reported in the literature. This study provides the first evidence that both cardiovascular and neuroendocrine systems are activated during an active act of aggression in dogs. Gobbo E. Mechanisms of aggressive behaviour in dogs. 61 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 62 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 63 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 64 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 65 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 66 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 67 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 68 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 69 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 70 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 71 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 72 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 73 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 74 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gobbo E. Mechanisms of aggressive behaviour in dogs. 75 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 3 DISCUSSION AND CONCLUSIONS 3.1 DISCUSSION Previous research has revealed several methodological limitations and knowledge gaps regarding factors involved in dog aggression. In this thesis, we sought to build a more comprehensive picture of the mechanisms involved in dog aggression in a multidisciplinary manner, by using a combination of different methodological approaches in four separate studies. 3.1.1 Contexts, dogs’ and dog bite victims’ characteristics In the first study, we confirmed our hypothesis that examining dog bites that are not limited to a prior dog-human interaction, reveals a greater number of contexts in which dog bites commonly occur. In contrast to previous findings, where contact activities were most common (Gautret et al., 2013; Sarcey et al., 2017; Oxley et al., 2018), our results showed that half of the described biting incidents occurred without prior interaction and often included bites during fast movements around the dog, while getting into close proximity to the dog and during incidents described as without a reason. Before the incidents, victims often reported they were walking towards or past a dog, or making fast movements by running or riding a bike on the street. This type of human behaviour, especially at the location the dogs perceive as territory, can lead to territorial aggression (Chávez and Opazo, 2012; Owczarczak-Garstecka et al., 2018a), even without actually presenting a threat. This is particularly common when it involves a person with whom the dog does not live with (Kuhne et al., 2014; Tuber, et al., 1996), which was also confirmed in our study. These bites, during which the victim unintentionally or passively interacted with the dog, involved purebred dogs with a history of aggression. They resulted mainly in one bite and occurred in a public space in the vicinity of an acquainted or unknown dog’s home. The finding that biting incidents without prior interaction are as important and as frequent as bites with prior interaction, suggests a more complex contextual aspect of dog aggression than previously reported. Other findings of this study are consistent with previous research. As reported by Bregman and Slavinski (2012) and Touré et al. (2015), the majority of bite victims were bitten when they were children or teenagers, under 19 years of age. Despite the fact that males are more likely to be bitten than females (Súilleabháin, 2015; Westgarth et al., 2018), we were unable to confirm this in our study due to the overrepresentation of female responders. Regarding the location of biting incidents, as previously shown (Babazadeh et al., 2016; Rosado et al., Gobbo E. Mechanisms of aggressive behaviour in dogs. 76 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 2009), low-populated areas appeared to be at higher risk, probably due to better control of dogs in urban areas with more people and traffic (Rosado et al., 2009). Next, as reported before, the dogs involved were primarily adult males, large in size, and originated from a breeder or known person (Oxley et al., 2018; Sarcey et al., 2017). The majority of our dogs were unnutered (Patronek et al., 2013; Shuler et al., 2008) and socialized as puppies, which does not appear to reduce aggression (Wormald et al., 2016). Similar to Gershman et al. (1994), living exclusively outdoors and being chained, appear to be associated with display of aggression. As there were a large number of different breeds seen in our study, this suggests that there is a high variation among breeds involved in human-directed aggressive behaviour (Duffy et al., 2008), but German Shepherds again appear to be the most commonly involved breed (Oxley et al., 2018; Sarcey et al., 2017). Finnaly, despite the inconsistent results regarding the association between biting incidents and dog’s history of aggression (Oxley et al., 2018; Sarcey et al., 2017), we found that most biting dogs have reports of previous biting behaviour. As it appears that any dog is capable of exhibiting biting behaviour in specific context (Bradshaw et al., 2009), even if it has never bitten before (De Keuster et al., 2006). Regarding the data collection, it is important to note, that this was a retrospective study and a lot of responders described events that occurred when they were children. That puts in question the validity of responses due to questionable recall accuracy and difficulty to remember details of the past event (Hipp et al., 2020), especially since we did not control for the time passed between the present and the event of interest. On this note, also the presented statistical data regarding breeds of dogs have to be interpreted with caution, as it may differ from the past data at the time of the attack. In addition, as most responders were children at the time of the attack, it is possible they did not have enough knowledge regarding dog behaviour and potentially misinterpreted the situation. Despite that, our results show a more complex contextual aspect of dog bites and shed a new light on this old problem, especially regarding the prevention. Most commonly used preventive methods, such as recognition of warning signs, is not applicable in bites without prior interaction. To include preventions of all bites, other approaches focusing on proper education of dog owners and parents regarding the supervision of dogs and children, especially in situations that present a risk factor, should be highlighted in the future. 3.1.2 Psychosocial effects By using a combination of methodological approaches, knowledge regarding psychosocial factors was also upgraded. We predicted several dog and owner personality traits, as well as owner attachment style to be associated with dog aggression. In terms of dog personality, we were able to partially confirm our hypotheses, as sociability was the only trait associated Gobbo E. Mechanisms of aggressive behaviour in dogs. 77 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 with aggression. Group of dogs categorized as aggressive toward humans generally had lower scores for trait sociability than non-aggressive dogs. In addition, for less sociable dogs, more frequent display of aggression towards strangers was reported by the owners. Our results deriving from the combination of behavioural observations and questionnaire data were consisted with previous questionnaire-based reports, linking low sociability with stranger- and child-directed aggression (Kaneko et al., 2013). Sociability seems to be the trait most strongly associated with aggression, because highly sociable dogs have better social control from meeting new people, and being in new environments. Therefore, such situation does not cause higher stress levels that can potentially trigger aggression (Yang et al., 2017). Similarly, while observing owners’ personality traits, we partially confirmed our hypotheses, as only owners with higher scores for neuroticism were associated with dogs aggressive towards strangers and other dogs. Positive associations between owners’ neuroticism and behavioural problems in their pets have been found in dogs (Podberscek and Serpell, 1997) and cats (Finka et al., 2019). There are several possible explanations for why neurotic owners tend to affect social behaviour of their dogs. First, neurotic owners have higher stress and cortisol levels (Schöberl et al., 2012) and tend to display unpredictable styles of caretaking, are more hostile and less warm, possibly decreasing social control and increasing stress levels of their pet (Finka et al., 2019). Second, more neurotic individuals generally tend to choose dog breeds that are considered more aggressive (Egan and MacKenzie, 2012). Overall, it appears that neurotic owners have a particular physiological and behavioural characterization that affects their dogs’ social behaviour. Another interesting finding related to owner personality traits was the fact that in the non-aggressive dog group, more extraverted owners had dogs with lower scores for chasing behaviour. Despite not necessary being directly associated with aggression, it is important to address this behaviour because chasing other, usually smaller animals, is a common unwanted and potentially dangerous behaviour in dogs (Zamora-Nasca et al., 2021). The explanation for the association between extroversion and chasing may be that more extroverted people enjoy being in the company of other people and often participate in various activities and social events (Lucas and Diener, 2001), which socializes their dogs and familiarizes them with strangers and other animals. For the last parameter observed, attachment, we were the first to confirm a relationship between owners’ attachment style to their dog and dog aggression. Although we found that both insecure attachment styles play a role in dog aggression, the direction of the association with aggression was consistent only with our predictions regarding the avoidant and not anxious attachment style. In our study, owners with higher scores for avoidant attachment had dogs with higher scores for owner-directed aggression. More avoidant individuals are known for distancing themselves from others and not providing enough intimate contact, Gobbo E. Mechanisms of aggressive behaviour in dogs. 78 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 affection and availability (Hazan and Shaver, 1987). Because the owner presents a secure base for their pet (Finka et al., 2019), especially during threatening situations (Rehn et al., 2017), the dogs of owners with this type of attachment may perceive the lack of a secure base as stressful and fear-provoking, possibly leading to often observed fear-related aggression (Flint et al., 2017). While for anxious attachment, contrary to our predictions, owners with lower scores were associated with dogs scoring high for stranger-directed aggression. Our findings are in contrast with reports in humans, in which child aggression was associated with anxious mother-infant attachment (Amani, 2016). It appears that behaviour associated with anxious attachment, such as clinginess, control, closeness and support seeking (Shaver and Mikulincer, 2002), do not contribute to the loss of secure base (Whipple et al., 2011) and do not play a role dog aggression. We can surmise that dogs of owners with anxious attachment style are accustomed to proximity and close relationship with people, and are therefore comfortable and calm in the company of strangers, reducing the likelihood of aggression. For better understanding of attachment as a whole, further studies should also use approaches that include assessment of both, insecure and secure attachment styles, as well as a larger sample size. Nevertheless, the knowledge regarding the parameters linked to aggression in this study may be important for both, the scientific community and the general public. The parameters (e.g., sociability, neuroticism) are easily observable by dog owners and better understanding of importance of certain traits could help owners identify potential risk factors for aggression, take appropriate actions early and improve the relationship with their dog. 3.1.3 Cognitive effects For the first time, different aspects of inhibitory control were assessed in relation to aggression. Predictions in the third study were partially confirmed, as only self-control, but not cognitive inhibition, was associated with dogs exhibiting biting behaviour during behavioural testing. Consistent with research findings in humans (Herndon et al., 2015) and rats (Van den Bergh et al., 2006), we found that dogs with the most severe display of aggression showed impairments in self-control, measured as poor performance in the delay of gratification test. Self-control is considered as one of the neuropsychological concepts of executive control (Séguin and Zelazo, 2005), which plays a role in the self-regulation of actions and emotions. Another mechanism mediated by executive control (Reynolds et al., 2019), often described in the context of dog aggression (Amat et al., 2009) is impulsivity. Previous research has suggested that delay of gratification presents an index of impulsive behaviour and that a lack of self-control can also be described as impulsivity (Wright et al., 2011). Impulsive dogs have been categorized as showing no or fewer warning signs before aggression (Fatjó et al., 2005). Based on this finding, we found that our dogs with poorer performance on the delay of gratification test were impulsive, as they also displayed Gobbo E. Mechanisms of aggressive behaviour in dogs. 79 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 aggressive biting behaviour. Furthermore, our results are consistent with Wright et al. (2011), who reported that dogs with higher owner-assessed trait impulsivity were more likely to be aggressive. Despite the methodological differences, the results are highly comparable, since it was assumed that subjective owner reports regarding impulsivity are consistent with performance during cognitive testing (Wright et al., 2011). In summary, the relationship between impulsivity and aggression found in our study appears to be consistent across mammals, since it reflects the findings in rats (Ferrari et al., 2005), humans (Blair, 2016) and non-human primates (Ferrari et al., 2005). However, it is important to note that the specific training and lifestyle of police dogs included in our study may influence their performance in the tests. In addition, low variation in the delay of gratification results could potentially limit the power of the results and thus our findings have to be interpreted with cation. To avoid that, further studies assessing self-control should consider including larger sample size, dogs with more similar behavioural and training background and larger number of cognitive tests per aspect. In contrast, we found no association between impairments in cognitive inhibition, measured as poor performance on the reversal learning test, and aggression, as is the case in humans (Mitchell et al. 2006). However, direct comparison of our results with human literature is somewhat problematic. The first problem presents the human population included in these studies. Reversal learning is often studied in humans with psychiatric disorders associated with increased aggression, such as psychopathy, attention deficits, and hyperactivity disorder (Turgay, 2004), making comparison with dogs and interpretation of the results very hard. The second problem is frustration, which is the main factor contributing to aggression in people with psychiatric disorders (Blair, 2010). This makes comparisons with dogs very difficult, as in dogs frustration has not been studied as a mechanism of aggression, but mostly in the relation to inaccessibility, decrease in value or absence of food (Jakovcevic et al., 2013; Bremhorst et al., 2019; Dzik et al., 2019). The lack of association between performance in both, reversal learning and delay of gratification with aggression, supports the context specificity of inhibitory control measures, proposed in previous dog studies (Bray et al., 2014; Brucks et al., 2017; Vernouillet et al., 2018). It appears that task demands, described as the set of skills the dogs must possess to successfully perform in the test (Bray et al., 2014; Vernouillet et al., 2018) may account for the lack of association between the tests. In conclusion, our findings suggest that particular behaviour, such as aggression, may be associated with limited aspects of inhibitory control. Dogs in our study displayed better inhibitory control in both tests, compared to previous research on pet dogs (Brucks et al., 2017), highlighting the importance of any kind of behavioural training for better cognitive performance. Also, the identification of impulse behaviour as a potential mechanism of aggression, may be useful for early recognition of unwanted behaviours and early application of appropriate preventive measures. Gobbo E. Mechanisms of aggressive behaviour in dogs. 80 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 3.1.4 Physiological effects In the fourth study, in which we examined neuroendocrine and cardiovascular parameters measured simultaneously and non-invasively during aggression, we could only partially confirm our predictions. As expected, we found lower salivary serotonin concentration (SER) in highly aggressive dogs. This is consistent with previous studies reporting lower plasma and serum SER in dogs with owner-reported history of aggression (Çakiroglu et al., 2007; Rosado et al., 2010; León et al., 2012). In humans (Duke et al., 2013) and non-human animals (Ferrari et al., 2005), this occurrence of inverse relationship between SER and display of aggression, has been referred to as the serotonin deficiency hypothesis. This phenomenon is evident in dogs, since dogs with lower SER are reported to have impaired impulse control (Reisner et al., 1996) and SER is known to be a neural regulator of aggression that acts as an inhibitor of aggressive reactivity (Summers and Winberg, 2004). Based on our results it appears that neuroendocrine activation, measured as decreased SER, is only evident in dogs during display of severe aggression. In contrast, we found no association between neuroendocrine activation, measured as increased salivary cortisol concentration (CORT) and aggression in real-time. This is in contrast to reports in humans (Haller et al., 2005) and dogs with an owner-reported history of aggression (Rosado et al., 2010). Nevertheless, our results may be difficult to compare due to methodological differences. CORT was measured during real-time aggression in our dogs, whereas other dog studies compared CORT of dogs with or without owner-reported past aggressive behaviour. Furthermore, we believe that physiological changes may be masked by the effects of the owner or handler during testing and saliva collection, as the presence of a person with an emotional attachment to the dogs can influence the behaviour and physiology during aversive situations (Gácsi et al., 2013). However, we found that aggressive threatening behaviours during the test (e.g., growling, barking) was associated with higher CORT in the home environment, but only in a group of dogs categorized as non-aggressive. Another finding regarding CORT included higher pre-test CORT compared to CORT in the home environment. This can be explained by the fact CORT can indicate emotional arousal, and not necessarily emotional valence (Lewandowski et al., 2014) and can be influenced by the anticipation of an activity (Horváth et al., 2008), arrival to a new place and meeting of new people (Ng et al., 2014). For the third physiological parameter observed, skin surface temperature, we were unable to confirm our initial hypotheses. Based on several animal studies including dogs (Part et al., 2014; Travain et al., 2015), monkeys (Kuraoka and Nakamura, 2011), rabbits (Ludwig et al., 2007) and pigs (Boileau et al., 2019), which reported decreased skin temperature during aversive situations, we predicted decrease in facial and body skin temperature during dog aggression in real-time. In contrast, we found no changes in post-test compared to pre-test Gobbo E. Mechanisms of aggressive behaviour in dogs. 81 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 values, but we did observe increase in facial temperature in aggressive dogs during testing. When comparing our results to a previous study that examined surface temperature in dogs during aggression (Rigterink et al., 2018), some similarities can be seen. Rigterink et al. (2018) reported an increased eye temperature in the aggressive group, but also in the non-aggressive dogs during veterinary visit. Other studies have also shown that surface temperature of dogs increases in both negative situations (Travain et al., 2015), and positive experiences (Travain et al., 2016) or decreases in both winning and losing pigs during social aggression (Boileau et al., 2019). Therefore it can be suggested that changes in surface temperature may reflect emotional arousal, and not necessarily emotional valence. Our further results show that thermal images taken during the aggression test are a better indicator of cardiovascular changes due to aggression, compared to observation of temperature change before and after the test. This result should be viewed with caution, as aggressive dogs in our study moved significantly more than non-aggressive dogs during the test, and the increase in surface temperature may be influenced by exercise (Cilulko et al., 2013). Another limitation presents the grouping selection. As all dogs in the aggressive group were police dogs, the physiological activation and behavioural reactions may be specific for police dogs and not aggressive dogs in general. Independent of the physiological observations, in this study we identified and described a new behaviour associated with aggression in addition to the usual aggression-related behaviours (e.g., snapping, attacking) (Netto and Planta, 1997; Haverbeke et al., 2009). For the first time, we observed an increased frequency of tail wagging and left tail wagging bias in aggressive dogs. Although tail wagging has been described mainly in relation to positive affective states in dogs (McGowan et al., 2014; Travain et al., 2016), Quaranta et al. (2007) reported that asymmetric tail wagging can be observed during stimuli with different emotional valence. Left tail wagging bias has been observed during stimuli the dog perceived as negative, and right tail wagging bias during stimuli that were perceived as positive (Quaranta et al., 2007). This asymmetry arises from different activation of right and left brain hemisphere (Buxton and Goodman, 1967). Because left tail wagging occurs during right hemisphere activation (Buxton and Goodman, 1967), our finding further indicates the right hemisphere is specialized for expression of intense emotions, as previously shown in dogs (Siniscalchi et al., 2019) and other animals (Rogers, 2010). Despite some methodological limitations, our study highlights the benefits and problems of real-time physiological and behavioural observations in the field and can be used as a stepping stone for further research. Gobbo E. Mechanisms of aggressive behaviour in dogs. 82 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 3.2 CONCLUSIONS In this thesis, four separate studies investigating contextual, psychosocial, cognitive and physiological aspects of dog aggression were conducted and several novel findings were reported (Figure 1). The following conclusions can be drawn from our work: - When considering the contexts of dog bites, we found that during only half of the reported bites the biting victim willingly interacted with the dog. During the other half, the victim had no intention of interacting with the biting dog. Bites during fast movements around the dog, bites occurring while passively being in a close proximity to the dog and bites without a known reason were the three novel contexts that appear to be common in bites without prior interaction with the dog. - In terms of owner and dog personality traits, neuroticism and sociability, respectively, are the two traits associated with human-directed dog aggression. The novel finding of this study is the association between insecure owner-dog attachment styles and human-directed dog aggression. Owners with high scores for avoidant attachment had dogs exhibiting owner-directed aggression and those with low scores for anxious attachment had dogs exhibiting stranger-directed aggression. It appears that both the dog and owner personality profiles play a role in the occurrence of dog aggression. - For the first time, several aspects of inhibitory control were examined in relation to dog aggression. Impaired self-control, measured as poor performance in the delay of gratification was associated with aggressive biting behaviour. There was no association between cognitive inhibition, measured as performance on reversal learning, and aggression. It appears that behaviour such as aggression is associated with limited aspects of inhibitory control. - During display of real-time aggression, simultaneous investigation of dog behavioural and physiological responses in the field revealed serotonergic activation, measured as decreased salivary SER, as well as cardiovascular activation, measured as increased facial surface temperature. In addition, a novel aggression-related lateralized behaviour, seen as more frequent tail wagging and left tail wagging bias, were observed and described for the first time in our study. Gobbo E. Mechanisms of aggressive behaviour in dogs. 83 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Figure 1: Schematic representation of the main findings Gobbo E. Mechanisms of aggressive behaviour in dogs. 84 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 4 SUMMARY (POVZETEK) 4.1 SUMMARY Aggression is part of normal behaviour of dogs and can be categorized as aggressive threatening behaviour (e.g., growling, baring teeth) or aggressive biting behaviour (e.g., snapping, attacking) (Netto and Planta, 1997). However, when aggression is directed toward owners or other people, it is considered one of the most unwanted and dangerous behaviours in dogs. Despite the fact that a considerable number of people, especially children (Súilleabháin, 2015), are bitten by a dog, and that aggression is a common reason for dogs to be abandoned, euthanized or relinquished to animal shelters (Salman et al., 2000; Diesel et al., 2008), there are still significant knowledge gaps regarding dog aggression. The aim of this thesis was to address these gaps by conducting four separate studies, each investigating a particular aspect of this behaviour. In order to assess aggression from different perspectives, we began with external factors, looking at the environment and interaction with humans, then moved to the psychosocial and cognitive factors, and concluded with the physiological factors. Our first study of contexts and interactions with humans focused exclusively on aggressive biting behaviour. Previous research on dog bites reported on the characteristics of the dogs involved, such as breed, age and sex, and the post bite implications for the dog (Salman et al., 2000; Diesel et al., 2008; Oxley et al., 2018). They often described the consequences and characteristics of victims, for example their age and gender, the location and severity of the bite (Rosado et al., 2009; Horisberger et al., 2004; Oxley et al., 2018). Authors reported that most bites occur during the direct interaction between humans and dogs, for example, during petting or play (Rosado et al., 2009; Horisberger et al., 2004; Oxley et al., 2018) and on private property (Oxley et al., 2018). However, most studies lack a thorough assessment of dog bites, particularly a detailed description of the circumstances surrounding the dog biting incident. We hypothesized that a thorough investigation of dog bites, not limited to direct interactions with the dog, will lead to a greater number of dog biting contexts than previously reported (Oxley et al., 2018). To obtain this information, an online questionnaire with 29 questions was used. Six open-ended and 23 closed-ended questions were divided into four sections. The first and the second one touched on victims’ and biting dogs’ demographics. The third section focused on a precise description of the biting incident, dogs’ and victims’ behaviour and location, and the last section was about post-bite implications for the dog. The results of the first two sections were consistent with previous research. Most individuals were bitten when they were children or adolescents (Bregman and Slavinski, 2012; Touré et al., 2015). The dogs Gobbo E. Mechanisms of aggressive behaviour in dogs. 85 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 involved were mainly large adult males that were known to the victim and came from a breeder or known person (Oxley et al., 2018; Sarcey et al., 2017), were unneutered (Patronek et al., 2013; Shuler et al., 2008), and lived exclusively outdoors (Gershman et al., 1994). Although wide variation was observed between breeds, German Shepherds were the most commonly reported breed (Oxley et al., 2018; Sarcey et al., 2017). With the results from the third section of the questionnaire we confirmed our hypothesis. While previous studies have described bites during direct interaction as the most common (Gautret et al., 2013; Oxley et al., 2018; Sarcey et al., 2017), in our study, only half of the reported bites occurred during voluntary interaction with the dog. The other half of bites was reported in situations without prior interaction with the dog, namely during fast movements around the dog, while getting into close proximity to the dog and during incidents described as without a reason. These incidents occurred in public spaces, mainly in the street or on the road and involved purebred dogs with history of aggression. Our findings suggest a more complex contextual aspect of dog bites and indicate that knowledge of bites without prior interaction is equally important, if not more important, as they are more difficult to prevent. In further examining the role of humans on dogs’ aggression, the second study assessed psychosocial factors, namely dog and owner personality traits and owner attachment styles. Previously, personality traits related to dog aggression have only been assessed using owner-reported questionnaire data. In dog studies, lower sociability has been associated with child-directed and stranger-directed aggression (Kaneko et al., 2013), and higher fearfulness with fear-directed and dog-directed aggression (Haverbeke et al., 2009; Arata et al., 2014). In human studies, owners with lower agreeableness, emotional stability, extraversion and conscientiousness have been associated with dogs aggressive towards owners (Dodman et al 2018). Attachment styles, namely insecure attachment styles, have not been studied in relation to dog aggression, but previous studies revealed that they play a role in the occurrence of behaviour problems in dogs (Konok et al., 2015) and in the formation of behavioural strategies in threatening situations (Rehn et al., 2017). Based on existing research we predicted that the above mentioned traits are associated with dog aggression. Data collection using owner-reported questionnaires has a limitation of owner bias and their possibly limited knowledge of dog behaviour (Wiener and Haskell, 2016). To be more objective, psychosocial factors were assessed in our study using a combination of questionnaires and behavioural test. Data on the history of dog aggression were collected using parts of the Canine Behavioral Assessment and a Research Questionnaire (C-BARQ) related to owner-directed, stranger-directed, dog-directed aggression and chasing (Hsu and Serpell, 2003). Dogs’ personality traits, defined as playfulness, curiosity, chase-proneness, sociability, aggressiveness and shyness/boldness, were assigned to each dog using a behavioural test called the Dog Mentality Assessment (DMA) (Svartberg and Forkman, Gobbo E. Mechanisms of aggressive behaviour in dogs. 86 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 2002). The test consisted of nine subtests performed outside in a specially set up testing area. The owner was present during the test. In the first subtest, a stranger made the first contact with the dog, and in the second, he she played with the dog. In the third subtest, the dog's response to a moving object was observed. In the fourth, the passive state of the dog and the owner was observed. In the fifth subtest, the dogs’ response to the unusual behaviour of a stranger was observed, and in the sixth, to the sudden appearance of a puppet. In the seventh subtest, the dogs’ response to the sudden sound of metal was observed, and in the eighth to the slow approach of two people covered with a sheet. In the last subtest, the stranger played with the dog again. Behaviour during the test was videotaped and analysed using a predefined ethogram. Owner personality traits, defined as extraversion, agreeableness, conscientiousness, neuroticism and openness, were assessed with an abbreviated version of the Big Five Inventory Questionnaire (BFI-10) (Rammstedt and John, 2007). Attachment styles, namely pet-related anxiety and avoidance, were assessed using the Experiences and Close Relationship-Revised Questionnaire - ECR-R (Beck and Madresh, 2008). The results on the dog personality showed sociability is the only trait associated with aggression, which partially confirms our hypothesis. It semms that highly sociable dogs have better social control and are more comfortable in situations that can increase stress levels and potentially trigger aggression (Yang et al., 2017). Also, in terms of owner personality, only owners with higher scores for neuroticism were associated with dogs aggressive towards strangers and other dogs, partially confirming our hypothesis. Neurotic owners appear to have a certain behaviours and physiological traits that influence their dog’s behaviour. For example, they may exhibit unpredictable styles of caretaking, potentially decreasing social control of their dogs, which can lead to increased stress levels (Finka et al., 2019). In addition, neurotic owners themselves have higher stress and cortisol levels (Schöberl et al., 2012). They also seem to choose dog breeds they classify as more aggressive (Egan and MacKenzie, 2012). Regarding attachment styles, both anxious and avoidant attachment styles have been associated with aggression. Anxious attachment style is manifested as clinginess, control, closeness and support seeking (Shaver and Mikulincer, 2002), whereas avoidant attachment style as distancing from the others, insufficient intimate contact, affection and availability (Hazan and Shaver, 1987). Owners with lower scores for anxious attachment were associated with dogs that had high scores for stranger-directed aggression. This result is in contrast to reports on mother-infant attachment (Amani, 2016) and it appears that owners’ behaviour related to anxious attachment do not influence their dog aggressive behaviour. In contrast, owners with higher scores on avoidant attachment were associated with dogs that had higher scores on owner-directed aggression. In this case, behaviours related to avoidant attachment, appear to play a role in the dog’s aggression. It seems that dogs of dogs of owners with avoidant style of attachment perceive the lack of a secure base (Finka et al., Gobbo E. Mechanisms of aggressive behaviour in dogs. 87 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 2019) stressful, especially in aversive situations (Rehn et al., 2017), possibly leading to more frequent display of aggression. Overall, we were the first to confirm the association between owners’ anxious and avoidant attachment styles and dog aggression. Another novel observation regarding aggression was made in our third study. For the first time, the association between various aspects of inhibitory control and dog aggression was examined. Inhibitory control is a part of executive control, a higher-order cognitive process, involved in self-regulation of actions and emotions (Séguin and Zelazo, 2005). It can be defined as the ability to block the execution of an immediate response in favour of a delayed but more rewarding behaviour (Bray et al. 2014) and is not a unitary mechanism, but a collection of separate cognitive processes (Beran, 2015; Brucks et al., 2017). Commonly described are self-control, cognitive inhibition and motor inhibition (Brucks et al., 2017; Brucks et al., 2019). Self-control can be defined as the ability to control impulse responses (Beran, 2015) and is usually assessed with an exchange paradigm, when an individual must resist immediate reward, in favour of a better, but delayed reward (Mischel et al., 1989). Cognitive inhibition is described as the ability to regulate the contents of the working memory, by removing information that is irrelevant in a given situation (Hasher et al., 1999). It is usually assessed using an object discrimination paradigm in which two stimuli change their reward contingencies, after an initial discrimination (Milgram et al., 1994). Both, impairments in self-control and cognitive inhibition, have previously been associated with aggression in humans (Mitchell et al., 2006; Herndon et al., 2015), but there has been no research on the association with dog aggression. We predicted that aggressive dogs will exhibit poor performance in both paradigms. To assess aggression in a controlled environment, a standardized behavioural test called Socially Acceptable Behavior (SAB) (Planta and De Meester, 2007) was used. The test was performed outside in a specially set up testing area and consisted of 16 subtests, known to elicit dog aggression. The owner or handler was present only for the first eight and for the last subtest. In the first subtest, a person approached the dog and tried to pet it with an artificial plastic hand and in the second, the dog was led past a moving sheet. In the third subtest, the silhouette of a big black cat suddenly appeared in front of the dog, in the fourth, the dog was led past a suddenly activated horn, and in the fifth, past a sudden sound of metal (rattling cans). In sixth and seventh subtests, three people approached the dog, first in slow and then in fast walking. In the eighth subtest, a toy dog of similar in size was placed in front of the dog and in the ninth, a person approached the dog and tried to pet it with an artificial hand. In the tenth subtest, a bell rang in front of the dog, and on the eleventh, an umbrella was opened and closed. In the twelfth subtest, a sled with a doll on it was pulled past the dog and in the thirteenth, a person approached the dog and tried to pet it with a doll. In fourteenth tasks, a person approached the dog in a threatening manner and stared at it, and in the fifteenth, the same person tried to touch the dog with an artificial hand. In the sixteenth Gobbo E. Mechanisms of aggressive behaviour in dogs. 88 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 subtest, the owner or handler approached the dog and tried to pet it with a doll. Behaviour during the test was video recorded and it was observed whether the dog showed no aggression, only threatening behaviour or aggressive biting behaviour. Self-control was measured using the delay of gratification test, modified after Brucks et al. (2017). In this test the dog had to resist eating an accessible but low-quality reward (LQR) and wait for delayed but high-quality reward (HQR). The test was conducted indoors using a wooden enclosure and consisted of three parts: food preference test (to determine LQR and HQR for each dog), training trails (to familiarize the dogs with the procedure), and test sessions. Rewards were presented on top of two plastic bowls that were simultaneously pushed towards the dog. The LQR bowl entered the enclosure first, and if the dog did not eat the reward, it was replaced by the HQR bow, after a 2-second delay. As the test progressed, the delay time was increased. The measures of inhibition were the number of successful trails and the maximum delay time achieved. Cognitive inhibition was measured using the reversal learning test, modified after Brucks et al. (2017). The test required the dog to inhibit a learned response and shift the behaviour to a new object-reward contingency. The test was conducted indoors in front of a wooden enclosure and consisted of two phases: the acquisition phase and the reversal phase. In the acquisition phase, the dog learned that one of the two presented differently shaped bowls always contained a reward. After the dog correctly identified the bowl containing the reward enough times and reached a learning criterion, the reversal phase followed. In this phase, the position of the reward was change, and the other bowl now contained the reward. At this point, the dog had to correctly identify the bowl containing the reward. The measures of inhibition were the number of correct choices during the last acquisition (last session during acquisition phase when the dog reached the criterion) and the reversal phase. The results confirmed our hypothesis for self-control but not for cognitive inhibition. Dogs that performed poorly on the delay of gratification test were those that showed severe signs of aggression during SAB. This result was observed for the first time in dogs and is consistent with research findings in humans (Herndon et al., 2015) and rats (Van den Bergh et al., 2006). Interestingly, the performance in the delay of gratification is also an indicator of impulsivity, and a lack of self-control can also be labelled as impulsivity (Wright et al. 2011). It could be that dogs with poor performance on the delay of gratification test in our study were impulsive, since it was suggested that impulsive dogs show fewer or no warning signs before acting aggressively (Fatjó et al., 2005). The found link between impulsivity and dog aggression is in line with reports in humans (Blair, 2016), non-human primates (Ferrari et al., 2005) and rats (Ferrari et al., 2005) and appears to be consistent across many mammalian species. Gobbo E. Mechanisms of aggressive behaviour in dogs. 89 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 No association was found between cognitive inhibition and aggression. We believe that reason for this lies in the population included in the human studies. Reversal learning in relation to aggression has been studied mainly among patients with psychiatric disorders (Turgay, 2004), which makes a comparison with dogs very difficult. It could also be because frustration is the main factor that contributes to aggression in humans with psychiatric disorders (Blair, 2010). Frustration in dogs has only been studied in the context of food (Jakovcevic et al., 2013; Bremhorst et al., 2019; Dzik et al., 2019), making the aforementioned studies difficult to compare. Furthermore, because performance in one paradigm and not both was associated with aggression, this further supports the context specificity of inhibitory control measures suggested previously (Bray et al., 2014; Brucks et al., 2017; Vernouillet et al., 2018) and implies that not all aspects of inhibitory control are associated with a particular behaviour, such as aggression. The final study addressed physiological activation during aggression, by simultaneously recording behaviour, cardiovascular and neuroendocrine parameters. Previous studies revealed several physiological parameters known to play a role in aggression. Two important neuroendocrine changes were proposed to be the increase in cortisol (Rosado et al., 2010) and the decrease in serotonin concentrations (Çakiroǧlu et al., 2007; Rosado et al., 2010; León et al., 2012). However, these studies had major shortcomings as they focused only on observing neuroendocrine changes between groups of dogs with or without a history of aggression, and not on observing changes during aggression in real-time. The most common cardiovascular parameter studied in the context of aggression is heart rate (Craig et al., 2017), but due to movement during aggression, and greater possibility of moving artefacts (Essner et al., 2015), non-contact observation of surface temperature has been suggested as a better alternative (Travain et al., 2015). The majority of studies assessing surface temperature in dogs and other animals, reported on decreased temperature during aversive situations (Ludwig et al., 2007; Kuraoka and Nakamura, 2011; Part et al., 2014; Travain et al., 2015; Boileau et al., 2019). In addition, it has been previously argued that it is important to measure multiple physiological parameters simultaneously while observing behaviour in real-time (Reefmann et al., 2009). Taking into account all the limitations of previous studies, the aim of our study was to assess cardiovascular and neuroendocrine changes during aggression in real-time, simultaneously and non-invasively. We predicted increased salivary cortisol concentration, decreased salivary serotonin concentration and decreased body and facial surface temperature during aggression. In this study, SAB test (Planta and De Meester, 2007) was also used. Behaviour during the test was videotaped and analysed using predefined ethogram. Dogs exhibiting or not exhibiting biting behaviour during the test were categorized as aggressive or non-aggressive, respectively. Cortisol and serotonin samples were obtained from dogs’ saliva by the owners or handlers, using commercially available cotton swabs. Saliva samples were collected three Gobbo E. Mechanisms of aggressive behaviour in dogs. 90 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 times, immediately before the participation in the SAB test (pre-test samples), immediately after for serotonin and 20 min after for cortisol samples (post-test samples) and in the home environment, when the dog was resting at home (home samples). Surface temperature, using infrared thermography was also measured at three occasions. Thermographic infrared images of the dogs’ face area (frontally) and body side (laterally) were taken immediately before the participation in the SAB test (pre-test images) and immediately after (post-test images). 16 images of the face area were also taken during SAB test, at the completion of each subtest (during the test images). The results only partially confirmed our hypotheses. As expected, we found association between dogs’ aggression in real time and lower salivary serotonin concentration. Our observations are consistent with the serotonin deficiency hypothesis, previously observed in humans (Duke et al., 2013) and non-human animals (Ferrari et al., 2005). Lower serotonin has also been associated with impaired impulse control (Reisner et al., 1996) and is known to be a neural regulator and inhibitor of aggressive reactivity (Summers and Winberg, 2004). Such association was not found between aggression and cortisol concentration (CORT), but we observed higher pre-test CORT compared to CORT in the home environment, indicating CORT may indicate emotional arousal, and not necessarily emotional valence (Lewandowski et al., 2014) due to arrival in a new environment, meeting of strangers (Ng et al., 2014) or anticipation of an activity (Horváth et al., 2008). The results regarding surface temperature also did not match our predictions. While we did not observe any changes in post-test values compared to pre-test values, we did observed increase in facial temperature in the aggressive dogs during the test. A similar increase in surface temperature during dog aggression was reported by Rigterink et al. (2018). However, in their study, only 27% of the dogs in the aggressive group actually displayed stranger-directed aggression, and the temperature increase was observed in both the aggressive and non-aggressive group. It appears that surface temperature also reflects emotional arousal and not necessary emotional valence, as its increase has previously been observed in dogs during positive situations (Travain et al., 2016) and negative situations (Travain et al., 2015). Although our initial hypotheses are not fully confirmed both the cardiovascular and neuroendocrine systems appear to be active during aggression in real-time. In this study, a novel aggression-related behaviour was observed in addition to standard aggression-related behaviours (e.g., biting, snapping) (Netto and Planta, 1997; Haverbeke et al., 2009). Dogs exhibiting aggression showed increased frequency of tail wagging and left tail wagging bias. Not related to aggression, Quaranta et al. (2007) observed asymmetrical tail wagging during stimuli with different emotional valence, with right tail wagging bias during positive stimuli, and left tail wagging bias during negative stimuli. Since left tail wagging is under the influence of the right brain hemisphere (Buxton and Goodman, 1967), Gobbo E. Mechanisms of aggressive behaviour in dogs. 91 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 our result further confirms the specialization of right brain hemisphere for expression of intense emotions in dogs and other animals (Rogers, 2010; Siniscalchi et al., 2019). 4.2 POVZETEK Agresivnost je del normalnega obnašanja psov in jo lahko kategoriziramo kot agresivno grožnjo (npr. renčanje, kazanje zob) ali agresivno grizenje (npr. šavsanje, napadanje) (Netto in Planta, 1997). Kadar je agresivnost usmerjena proti lastniku ali drugim ljudem, velja za eno izmed najbolj nezaželenih in nevarnih obnašanj pri psih. Kljub temu, da precejšnje število ljudi, zlasti otrok (Súilleabháin, 2015), ugrizne pes in da je agresivnost pogost razlog, da lastniki psa zapustijo, evtanazirajo ali odpeljejo v zavetišče (Salman in sod., 2000; Diesel in sod., 2008), še vedno obstajajo velike vrzeli v znanju o agresiji psov. Cilj te disertacije je bil zapolniti te vrzeli s pomočjo štirih ločenih študij, od katerih je vsaka raziskovala posamezen vidik tega obnašanja. Raziskovanje agresije iz različnih zornih kotov smo začeli z zunanjimi dejavniki, in sicer z raziskovanjem okolja in interakcije z ljudmi. Nato smo prešli na psihosocialne in kognitivne dejavnike ter zaključili s fiziološkimi dejavniki. V svoji prvi študiji o kontekstih in interakciji z ljudmi smo se osredotočili izključno na agresivno grizenje. Prejšnje raziskave v zvezi z ugrizi psov so večinoma poročale o značilnostih vpletenih psov, kot so pasma, starost in spol, ter o posledicah po ugrizu za psa (Salman in sod., 2000; Diesel in sod., 2008; Oxley in sod., 2018). Pogosto so tudi opisovali posledice in značilnosti žrtev, na primer njihovo starost in spol, lokacijo in resnost ugriza (Rosado in sod., 2009; Horisberger in sod., 2004; Oxley in sod., 2018). Avtorji so poročali, da se večina ugrizov zgodi med neposredno interakcijo človeka in psa, na primer med božanjem ali igranjem (Rosado in sod., 2009; Horisberger in sod., 2004; Oxley in sod., 2018) in na zasebni posesti (Oxley in sod., 2018). V večini študij primanjkuje natančnejši opis ugrizov psov, zlasti podrobnejši pregled okoliščin, v katerih je do ugriza prišlo. V študiji smo domnevali, da bo poglobljeno raziskovanje ugrizov psov, ki ni omejeno le na ugrize med neposredno interakcijo s psom, vodilo do večjega števila kontekstov ugrizov, kot so poročali v preteklost (Oxley in sod., 2018). Za pridobitev teh informacij smo uporabili spletni vprašalnik, ki je vseboval 29 vprašanj. Šest odprtih in 23 zaprtih vprašanj smo razdelili v štiri sklope. Prvi in drugi sklop sta se dotaknila demografije žrtev in udeleženih psov. Tretji se je osredotočil na natančen opis incidenta ugriza, obnašanje psa in žrtve in lokacijo ter zadnji na posledice za psa po ugrizu. Rezultati prvih dveh sklopov so bili v skladu s prejšnjimi raziskavami. Večino ljudi je pes ugriznil, ko so bili otroci ali najstniki (Bregman in Slavinski, 2012; Touré in sod., 2015). Psi so bili v glavnem veliki, žrtvi poznani odrasli samci. Prišli so od vzreditelja ali poznane osebe (Oxley in sod., 2018; Sarcey in sod., 2017), niso bili kastrirani ali sterilizirani Gobbo E. Mechanisms of aggressive behaviour in dogs. 92 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 (Patronek in sod., 2013; Shuler in sod., 2008) in živeli so izključno na prostem (Gershman in sod, 1994). Kljub veliki raznolikosti med pasmami so najpogosteje poročali o nemških ovčarjih (Oxley in sod., 2018; Sarcey et al., 2017). Z rezultati tretjega sklopa vprašalnika smo potrdili zastavljeno hipotezo. Medtem ko so prejšnje raziskave ugrize med neposredno interakcijo označile kot najpogostejše (Gautret in sod., 2013; Oxley in sod., 2018; Sarcey in sod., 2017), je v naši študiji prišlo le pri polovici opisanih ugrizov med prostovoljno interakcijo s psom. Za drugo polovico ugrizov so poročali, da so se zgodili v situacijah brez predhodne interakcije s psom, in sicer med hitrimi gibi okoli psa, medtem ko so se posredno približali psu in v primerih, opisanih kot brez razloga. Ti incidenti so se zgodili na javnem kraju, predvsem na ulici ali na cesti in so vključevali čistokrvne pse z zgodovino agresije. Naše ugotovitve nakazujejo na bolj zapleten kontekstualni vidik ugriza psov in da je znanje o ugrizih brez predhodne interakcije prav tako pomembno, če ne celo pomembnejše, saj jih je težje preprečiti. Za nadaljnje raziskovanje vplivov človeka na agresijo psa je bila v drugi študiji raziskana povezanost psihosocialnih dejavnikov, in sicer lastnosti osebnosti psov in lastnikov ter stilov navezanosti lastnikov. Doslej so bile v kontekstu agresije psov lastnosti osebnosti ocenjene le z uporabo vprašalnikov, ki so jih izpolnili lastniki. V raziskavah na psih je bila nižja stopnja družabnosti povezana z agresijo, usmerjeno proti otrokom in tujcem (Kaneko in sod., 2013), medtem ko je bila večja plašnost povezana z agresijo, usmerjeno proti psom in agresijo, povezano s strahom (Haverbeke in sod., 2009; Arata in sod., 2014). V študijah na ljudeh so bili lastniki z nižjo sprejemljivostjo, čustveno stabilnostjo, ekstravertiranostjo in vestnostjo povezani s psi, agresivnimi do lastnikov (Dodman in sod., 2018). Stili navezanosti, in sicer negotovi stili navezanosti, še niso bili raziskani v povezavi z agresijo psov, vendar so prejšnje študije pokazale, da igrajo vlogo pri pojavu nezaželenega obnašanja pri psih (Konok in sod., 2015) in oblikovanju strategij obnašanja v ogrožajočih situacijah (Rehn in sod., 2017). Na temelju obstoječih raziskav smo predvideli, da bodo vse zgoraj omenjene lastnosti povezane z agresijo psov. Zbiranje podatkov na podlagi poročanja lastnikov omejujeta lastnikova pristranskost in potencialno omejeno znanje o obnašanju psov (Wiener in Haskell, 2016), zato smo za večjo objektivnost psihosocialne dejavnike v svoji študiji ocenili s kombinacijo vprašalnikov in testiranja obnašanja. Podatki o zgodovini agresije psov so bili pridobljeni z uporabo delov vprašalnika C-BARQ, povezanih z agresijo do lastnikov, tujcev, psov in z lovljenjem (Hsu in Serpell, 2003). Lastnosti osebnosti psov, opredeljene kot igrivost, radovednost, nagnjenost k lovljenju, družabnost, agresivnost in plašnost/neustrašnost, so bile vsakemu psu dodeljene s testom obnašanja DMA (Svartberg in Forkman, 2002). Test je obsegal devet nalog, izvedenih zunaj na posebej pripravljanjem testnem polju. Lastnik je bil prisoten pri vseh nalogah. Pri prvi nalogi je neznana oseba vzpostavila prvi stik s psom, pri drugi pa se Gobbo E. Mechanisms of aggressive behaviour in dogs. 93 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 s psom igrala. Pri tretji nalogi se je opazoval odziv psa na premikajoči se predmet. Pri četrti nalogi se je opazovalo pasivno stanje psa in lastnika. Pri peti nalogi se je opazoval odziv psa na nenavadno obnašanje neznane osebe, pri šesti pa na nenaden pojav lutke. Pri sedmi nalogi se je opazoval odziv psa na nenaden zvok kovine, pri osmi pa na počasno približevanje dveh oseb, ki sta prekriti z rjuho. Pri zadnji nalogi se je neznanec ponovno igral s psom. Obnašanje med testom je bilo posneto in analizirano s pomočjo vnaprej pripravljenega etograma. Lastnosti osebnosti lastnikov, opredeljene kot ekstravertnost, sprejemljivost, vestnost, nevroticizem in odprtost, so bile ocenjene s skrajšano različico vprašalnika BFI-10 (Rammstedt in John, 2007). Stila navezanosti, in sicer anksiozna in izogibajoča navezanost do hišnih ljubljenčkov, sta bili določena z vprašalnikom ECR-R (Beck in Madresh, 2008). Rezultati osebnosti psov so pokazali, da je družabnost edina lastnost, povezana z agresijo, kar delno potrjuje našo hipotezo. Zdi se, da imajo zelo družabni psi boljši družbeni nadzor in so bolj sproščeni v situacijah, ki lahko povišajo raven stresa in potencialno povzročijo agresijo (Yang in sod., 2017). Pri proučevanju lastnosti osebnosti lastnikov so bili le lastniki z višjimi ocenami nevroticizma povezani s psi, agresivnimi do tujcev in drugih psov, kar je delno potrdilo našo hipotezo. Zdi se, da imajo nevrotični lastniki posebne vedenjske in fiziološke lastnosti, ki vplivajo na obnašanje njihovega psa. Na primer, kažejo lahko nepredvidljive sloge oskrbe, kar lahko zmanjša družbeni nadzor njihovih psov in potencialno povzroči višjo raven stresa (Finka in sod., 2019). Poleg tega imajo nevrotični lastniki psov tudi sami višjo raven stresa in kortizola (Schöberl in sod., 2012). Prav tako se zdi, da takšni lastniki izbirajo pasme psov, za katere menijo, da so agresivnejše (Egan in MacKenzie, 2012). Kar zadeva stila navezanosti lastnika na psa, sta bili tako anksiozna kakor tudi izogibajoča navezanost povezani z agresijo. Anksiozna navezanost se izraža kot oklepajoče obnašanje, iskanje nadzora, bližine in podpore (Shaver in Mikulincer, 2002), medtem ko se izogibajoča navezanost izraža kot distanciranje od drugih, pomanjkanje intimnih stikov, naklonjenosti in razpoložljivosti (Hazan in Shaver, 1987). Lastniki z nižjimi ocenami anksiozne navezanosti so bili povezani s psi, ki so dosegli visoko oceno za agresijo, usmerjeno proti tujcem. Ta ugotovitev je v nasprotju z raziskavami o navezanosti matere na dojenčka (Amani, 2016) in zdi se, da obnašanje lastnikov, povezano z anksiozno navezanostjo, ne vpliva na agresivno obnašanje njihovega psa. Nasprotno so bili lastniki z višjimi ocenami izogibajoče navezanosti povezani s psi z višjimi ocenami agresije, usmerjene proti lastnikom. V tem primeru se zdi, da obnašanje, povezano z izogibajočo navezanostjo, igra vlogo pri pojavnosti agresije psov. Zdi se, da psom lastnikov z izogibajočim stilom navezanosti primanjkuje varna baza (Finka in sod., 2019) in to dojemajo kot stresno, zlasti v negativnih situacijah (Rehn in sod., 2017), kar lahko povzroči pogostejše izkazovanje agresije. Na splošno smo bili prvi, ki smo potrdili odnos med anksiozno in izogibajočo navezanostjo lastnikov in njihovih psov ter agresijo psov. Gobbo E. Mechanisms of aggressive behaviour in dogs. 94 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Do novega odkritja v povezavi z agresijo je prišlo tudi v tretji študiji. Prvič je bila raziskana povezava med različnimi vidiki inhibicijske kontrole in agresijo psov. Inhibicijska kontrola je del izvršnega nadzora, kognitivnega procesa visokega reda, vključenega v samouravnavanje obnašanja in čustev (Séguin in Zelazo, 2005). Opredelimo jo lahko kot sposobnost zaustavitve takojšnjega odziva v korist zakasnjenega, a koristnejšega obnašanja (Bray in sod. 2014). Ne gre za enoten mehanizem, ampak za zbirko ločenih kognitivnih procesov (Beran, 2015; Brucks in sod., 2017). Pogosto opisani procesi so samonadzor, kognitivna in motorična inhibicija (Brucks in sod., 2017; Brucks in sod., 2019). Samonadzor lahko opišemo kot sposobnost nadzora impulzivnih odzivov (Beran, 2015) in se običajno ocenjuje s paradigmo izmenjavanja, med katero se mora posameznik upreti takojšnji nagradi v korist boljše, vendar zakasnjene nagrade (Mischel in sod., 1989). Kognitivno inhibicijo lahko opredelimo kot regulativno sposobnost v delovnem spominu, ki odstranjuje v dani situaciji nepomembne informacije (Hasher in sod., 1999). Običajno jo ocenjujemo s paradigmo razločevanja predmetov, med katero dva dražljaja po začetni razločitvi spremenita položaj nagrade (Milgram in sod., 1994). Obe, okvara samonadzora in kognitivna inhibicija, sta bili povezani z agresijo pri ljudeh (Mitchell in sod., 2006; Herndon in sod., 2015), vendar raziskav o povezanosti z agresijo psov še ni bilo. Mi smo predvidevali, da bodo agresivni psi v obeh paradigmah pokazali slabe rezultate. Za oceno agresivnosti v nadzorovanem okolju je bil uporabljen standardiziran test obnašanja SAB (Planta in De Meester, 2007). Test je bil izveden zunaj na posebej pripravljenem testnem polju in je obsegal 16 nalog, za katere je znano, pri psih izzovejo agresivne odzive. Lastnik ali vodnik je bil prisoten le pri prvih osmih nalogah in pri zadnji nalogi. Pri prvi nalogi se je oseba približala psu in ga poskušala pobožati s plastično umetno roko. Pri drugi nalogi se je psa peljalo mimo premikajoče se rjuhe. Pri tretji nalogi se je pred psom nenadoma prikazala silhueta velike črne mačke. Pri četrti nalogi se je psa peljalo mimo nenadoma aktivirane troblje, pri peti pa mimo nenadoma povzročenega zvoka kovine. Pri šesti in sedmi nalogi so se tri osebe približale privezanemu psu, najprej v počasni, nato pa v hitri hoji. Pri osmi nalogi se je pred psa postavilo lutko njemu podobno velikega psa. Pri deveti nalogi se je oseba približala psu in ga poskušala pobožati z umetno roko. Pri deseti nalogi se je pred psom zvonilo z zvončkom, pri enajsti pa odpiralo in zapiralo dežnik. Pri dvanajsti nalogi so bile mimo psa povlečene sani, na katerih je bila lutka. Pri trinajsti nalogi se je oseba približala psu in poskušala pobožati z lutko. Pri štirinajsti nalogi se je oseba grozeče približala psu in vanj strmela, pri petnajsti pa ga ta ista oseba poskušala pobožati z umetno roko. Pri šestnajsti nalogi se je lastnik ali vodnik prijazno približal psu in ga poskušal pobožati z lutko. Obnašanje med testom je bilo posneto in opazovano je bilo, ali pes ne kaže agresije, kaže samo grožnjo ali agresivno grizenje. Samonadzor smo merili s testom zapoznelega nagrajevanja, modificiranega po Brucks in sod. (2017). Med testom se je pes moral upreti vzgibu, da bi pojedel dostopno, a manj Gobbo E. Mechanisms of aggressive behaviour in dogs. 95 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 kakovostno nagrado (NKN) in počakal na zapoznelo, a bolj kakovostno nagrado (VKN). Test je bil izveden v lesenem ohišju in je bil sestavljen iz treh delov: testa izbire hrane (za določitev VKN in NKN za vsakega psa), treninga (za seznanitev psov s postopkom) in testnega dela. Nagrade so bile ponujene na dveh plastičnih posodah, ki sta bili hkrati potisnjeni proti psu. NKN-posoda je prva vstopila v ohišje in če pes ni pojedel nagrade, je bila po 2 sekundah zakasnitve zamenjana za VKN-posodo. Med testom se je čas zakasnitve povečeval. Merili inhibicije sta bili število uspešnih poskusov in najvišji doseženi čas zakasnitve. Kognitivna inhibicija je bila izmerjena s testom obratnega učenja, modificiranim po Brucks in sod. (2017), med katerim se je moral pes upreti naučenemu odziv in preusmeriti obnašanje v korist nove pozicije nagrade. Test je bil izveden v notranjih prostorih in je bil sestavljen iz dveh faz: faze usvojitve in faze obratnega učenja. V fazi usvojitve je bilo psu prikazano, da ena od dveh predstavljenih posod, različnih oblik in velikosti, vedno vsebuje nagrado. Po tem, ko je pes dovoljkrat pravilno identificiral posodo, ki vsebuje nagrado in dosegel merilo, je sledila faza obratnega učenja. V tej fazi se je položaj nagrade spremenil, saj je bil prestavljen pod drugo posodo. Na tej točki je moral pes znova pravilno prepoznati posodo z nagrado. Merilo inhibicijske kontrole je bilo število uspešnih izbir med zadnjo fazo usvojitve (ko je pes dosegel merilo) in fazo obratnega učenja. Rezultati so potrdili našo hipotezo o samonadzoru, vendar ne tudi o kognitivni inhibiciji. Psi, ki so se med testom zapoznelega nagrajevanja slabo odrezali, so bili tisti, ki so med SAB pokazali hude znake agresije. Ta ugotovitev je bila prvič opažena pri psih in je v skladu z raziskavami pri ljudeh (Herndon in sod., 2015) in podganah (Van den Bergh in sod., 2006). Zanimivo je, da je neuspešnost pri nalogi zapoznelega nagrajevanja tudi kazalnik impulzivnosti, medtem ko se pomanjkanje samonadzora lahko imenuje tudi impulzivnost (Wright in sod., 2011). Zdi se, da so bili psi, ki so bili v naši študiji slabi pri testu zapoznelega nagrajevanja, impulzivni, saj je glede na Fatjó in sod. (2005) za impulzivne pse značilno, da imajo pred agresijo odsotno ali zmanjšano kazanje opozorilnih znakov. Ugotovljena povezava med impulzivnostjo in agresijo pri psih je v skladu z raziskavami pri ljudeh (Blair, 2016), primatih (Ferrari in sod., 2005) in podganah (Ferrari in sod., 2005) in se zdi konsistentna pri več vrstah sesalcev. Med kognitivno inhibicijo in agresijo povezave nismo odkrili. Menimo, da je razlog za to populacija, vključena v študije na ljudeh. Obratno učenje v povezavi z agresijo so preučevali predvsem pri bolnikih s psihiatričnimi patologijami (Turgay, 2004), zaradi česar je primerjanje s psi zelo težko. Drugi razlog lahko prestavlja frustracija, ki velja za glavni dejavnik, ki prispeva k agresivnosti pri ljudeh s psihiatričnimi patologijami (Blair, 2010). Frustracija pri psih je bila raziskana le v kontekstu hrane (Jakovcevic in sod., 2013; Bremhorst in sod., 2019; Dzik in sod., 2019), zaradi česar je omenjene študije težko Gobbo E. Mechanisms of aggressive behaviour in dogs. 96 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 primerjati s študijami o agresiji. Ker smo odkrili povezanost med agresijo in samo eno od dveh kognitivnih paradigem, to dodatno podpira predlagano individualnost kontekstov pri merjenju inhibicijskega nadzora pri psih (Bray in sod., 2014; Brucks in sod., 2017; Vernouillet in sod., 2018) in nakazuje, da niso vsi vidiki inhibicijske kontrole povezani s posameznim obnašanjem, kot je na primer agresija. Četrta študija se je dotaknila fiziološke aktivacije med agresijo psov s hkratnim spremljanjem obnašanja, kardiovaskularnih in nevroendokrinih parametrov. Prejšnje raziskave so pokazale več fizioloških parametrov, za katere je znano, da igrajo vlogo pri agresiji. Dva glavna predlagana nevroendokrina parametra sta povišan kortizol (Rosado in sod., 2010) in znižan serotonin (Çakiroǧlu in sod., 2007; Rosado in sod., 2010; León in sod., 2012). Vendar so imele te študije veliko pomanjkljivost, saj so se osredotočile le na opazovanje nevroendokrinih sprememb med skupino psov z ali brez zgodovine agresije in ne na opazovanje sprememb med agresijo v realnem času. Najpogostejši kardiovaskularni parameter, raziskan v zvezi z agresijo, je srčni utrip (Craig in sod., 2017), vendar je zaradi gibanja med agresijo in večje možnosti merilnih napak ob premikanju (Essner in sod., 2015) brezstično merjenje temperature površine telesa boljša alternativa (Travain in sod., 2015). Večina študij, ki so raziskovale površinsko temperaturo pri psih in drugih živalih, je poročala o znižani temperaturi v negativnih situacijah (Ludwig in sod., 2007; Kuraoka in Nakamura, 2011; Part in sod., 2014; Travain in sod., 2015; Boileau in sod., 2019). Pred tem je bilo tudi predlagano, da je pomembno hkrati meriti več fizioloških parametrov ob opazovanju obnašanja v realnem času (Reefmann in sod., 2009). Glede na vse omejitve prejšnjih raziskav je bil cilj naše študije oceniti hkratne kardiovaskularne in nevroendokrine spremembe med agresijo v realnem času in neinvazivno. Predvideli smo povišano koncentracijo kortizola v slini, znižano koncentracijo serotonina v slini ter znižano površinsko telesno in obrazno temperaturo med agresijo. Med to študijo je bil uporabljen tudi standardiziran test obnašanja SAB (Planta in De Meester, 2007). Obnašanje med testom je bilo posneto in analizirano z uporabo vnaprej določenega etograma. Psi, ki so med preskusom prikazali ali niso prikazali agresivnega grizenja, so bili kategorizirani kot agresivni oziroma neagresivni. Lastniki ali vodniki so vzorce kortizola in serotonina s komercialno dostopnimi bombažnimi tamponi pridobili iz sline psov. Vzorci sline so bili odvzeti trikrat, tik pred udeležbo na testu SAB (vzorci pred testom), takoj po izvedbi testa za vzorce serotonina in 20 minut kasneje za vzorce kortizola (vzorci po testu) ter v domačem okolju, ko je pes počival doma (domači vzorci). Trikrat je bila z infrardečo termografijo merjena tudi površinska temperatura. Termografske infrardeče slike področja obraza psov (od spredaj) in strani telesa (bočno) so bile posnete tik pred udeležbo na testu SAB (slike pred testom) in takoj po tem (slike po testu). Med testom SAB je bilo ob zaključku vsake od šestnajstih naloge posneta tudi slika obraza (slike med testom). Gobbo E. Mechanisms of aggressive behaviour in dogs. 97 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Rezultati so le delno potrdili naše hipoteze. Kot pričakovano smo ugotovili povezavo med psi, agresivnimi v realnem času, in nižjo koncentracijo serotonina v slini. To, kar smo opazili, se imenuje hipoteza o pomanjkanju serotonina, ki jo lahko opazimo pri ljudeh (Duke in sod., 2013) in drugih živalih (Ferrari in sod., 2005). Znižan serotonin je povezan tudi s slabim nadzorom impulzivnosti (Reisner et al., 1996) in velja za regulator v možganih ter inhibitor agresivne reaktivnosti (Summers in Winberg, 2004). Takšne povezave med agresijo in koncentracijo kortizola (CORT) nismo ugotovili, vendar smo opazili višji domači CORT, v primerjavi s CORT pred testom, kar kaže, da lahko spremembe v CORT prikazujejo čustveno vzburjenje zaradi prihoda v novo okolje, srečanja tujcev (Ng in sod., 2014) ali pričakovanja dejavnosti (Horváth in sod., 2008) in ne nujno čustveno valenco (Lewandowski in sod., 2014), Tudi rezultati glede površinske temperature niso bili v skladu z našimi napovedmi. Čeprav nismo ugotovili nobenih sprememb ob primerjanju vrednosti pred in po testu, smo pri slikah, pridobljenih med testom, zabeležili zvišanje temperature obraza pri agresivnih psih. O podobnem zvišanju površinske temperature med agresijo pri psih so poročali Rigterink in sod. (2018), vendar je v njihovi študiji le 27 % psov v agresivni skupini dejansko pokazalo agresijo, usmerjeno proti tujcem, medtem ko so zvišanje temperature opazili tako v agresivni kot tudi v neagresivni skupini. Zdi se, da lahko tudi površinska temperatura odraža čustveno vzburjenje in ne nujno čustveno valenco, saj so o njenem zvišanju že poročali pri psih, tako v pozitivnih (Travain in sod., 2016) kot tudi v negativnih situacijah (Travain in sod., 2015). Čeprav naše začetne hipoteze niso bile v celoti potrjene, se zdi, da se med agresivnostjo pri psih aktivirata tako kardiovaskularni kakor tudi nevroendokrini sistem. Med to študijo smo poleg standardnega obnašanja, povezanega z agresijo (npr. grizenje, šavsanje) (Netto in Planta, 1997; Haverbeke in sod., 2009), opazili novo obnašanje, povezano z agresijo. Agresivni psi so prikazali pogostejše mahanje z repom in pogostejšo nagnjenost k mahanju repa v levo. Čeprav nepovezano z agresijo so Quaranta in sod. (2007) opisali, da lahko asimetrično mahanje z repom opazimo med dražljaji z različno čustveno valenco, s pristranskostjo mahanja z repom v desno med pozitivnimi dražljaji in pristranskostjo mahanja z repom v levo med dražljaji, ki jih pes dojema kot negativne. Ker je mahanje z repom v levo pod vplivom desne možganske poloble (Buxton in Goodman, 1967), naša ugotovitev dodatno potrjuje specializacijo desne možganske poloble za izražanje intenzivnih čustev pri psih in drugih živalih (Rogers, 2010; Siniscalchi in sod., 2019). Gobbo E. Mechanisms of aggressive behaviour in dogs. 98 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 5 REFERENCES Amani R. 2016. Mother-infant attachment styles as a predictor of aggression. Journal of Midwifery and Reproductive Health, 4: 506–512 Amat M., Manteca X., Mariotti V.M., Ruiz de la Torre J.L., Fatjó J. 2009. Aggressive behavior in the English cocker spaniel. Journal of Veterinary Behavior, 4: 111–117 Anderson C.A., Bushman B.J. 2002. Human aggression. Annual Review of Psychology, 53: 27–51 Arata S., Takeuchi Y., Inoue M., Mori Y. 2014. "Reactivity to stimuli’’ is a temperamental factor contributing to dog aggression. PLoS One, 9: e100767, doi: 10.1371/journal.pone.0100767: 11 p. Archer J., Ireland J.L. 2011. The development and factor structure of a questionnaire measure of the strength of attachment to pet dogs. Anthrozoös, 24: 249–261 Arhant C., Landenberger R., Beetz A., Troxler J. 2016. Attitudes of caregivers to supervision of child–family dog interactions in children up to 6 years - An exploratory study. Journal of Veterinary Behavior, 14: 10–16 Babazadeh T., Nikbakhat A., Daemi A., Yegane-kasgari M., Ghaffari-fam S., Banaye-Jeddi M. 2016. Epidemiology of acute animal bite and the direct cost of rabies vaccination. Journal of Acute Disease, 5: 488–492 Bari A., Robbins T.W. 2013. Inhibition and impulsivity: Behavioral and neural basis of response control. Progress in Neurobiology, 108: 44–79 Basco A.N., McCormac E.R., Basco W.T. 2020. Age- and sex-related differences in nonfatal dog bite injuries among persons aged 0-19 treated in hospital emergency departments, United States, 2001-2017. Public Health Reports, 135: 238–244 Beran M. 2015. The comparative science of ‘self-control’: What are we talking about? Frontiers in Psychology, 6: 51, doi: 10.3389/fpsyg.2015.00051: 4 p. Beck L., Madresh E.A. 2008. Romantic partners and four-legged friends: an extension of attachment theory to relationships with pets. Anthrozoös, 21: 43–56 Bell T., Spikins P. 2018. The object of my affection: attachment security and material culture. Time and Mind, 11: 23–39 Blair R.J. 2010. Psychopathy, frustration, and reactive aggression: the role of ventromedial prefrontal cortex. British Journal of Psychology, 101: 383–99 Boileau A., Farish M., Turner S.P., Camerlink I. 2019. Infrared thermography of agonistic behaviour in pigs. Physiology and Behavior, 210: 112637 Bradshaw J.W.S., Blackwell E.J., Casey R.A. 2009. Dominance in domestic dogs - Useful construct or bad habit? Journal of Veterinary Behavior, 4: 135–144 Bray E.E., MacLean E.L., Hare B.A. 2014. Context specificity of inhibitory control in dogs. Animal Cognition, 17: 15–31 Gobbo E. Mechanisms of aggressive behaviour in dogs. 99 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Bremhorst A., Sutter N.A., Würbel H., Mills D.S., Riemer S. 2019. Differences in facial expressions during positive anticipation and frustration in dogs awaiting a reward. Scientific Reports, 9: 19312, doi: 10.1038/s41598-019-55714-6: 13 p. Brucks D., Marshall-Pescini S., Range F. 2019. Dogs and wolves do not differ in their inhibitory control abilities in a non-social test battery. Animal Cognition, 22: 1–15 Brucks D., Marshall-Pescini S., Wallis L.J., Huber L., Range F. 2017. Measures of dogs’ inhibitory control abilities do not correlate across tasks. Frontiers in Psychology, 8: 849, doi: 10.3389/fpsyg.2017.00849: 17 p. Buss A.H., 1961. The psychology of aggression. New York, Wiley: 307 p. Buxton D.F., Goodman D.C. 1967. Motor function and the corticospinal tracts in the dog and raccoon. Journal of Comparative Neurology, 129: 341–360 Çakiroǧlu D., Meral Y., Sancak A.A., Çifti G. 2007. Relationship between the serum concentrations of serotonin and lipids and aggression in dogs. Veterinary Record, 161: 59–61 Casey R.A., Loftus B., Bolster C., Richards G.J., Blackwell E.J. 2014. Human directed aggression in domestic dogs (Canis familiaris): Occurrence in different contexts and risk factors. Applied Animal Behaviour Science, 152: 52–63 Christensen E., Scarlett J., Campagna M., Houpt K.A., 2007. Aggressive behaviour in adopted dogs that passed a temperament test. Applied Animal Behavioural Science, 106: 85–95 Chávez G.A., Opazo Á.J. 2012. Predatory aggression in a German shepherd dog. Journal of Veterinary Behavior, 7: 386–389 Cilulko J., Janiszewski P., Bogdaszewski M., Szczygielska E. 2013. Infrared thermal imaging in studies of wild animals. European Journal of Wildlife Research, 59: 17– 23 Cornelissen J.M.R., Hopster H. 2010. Dog bites in The Netherlands: A study of victims, injuries, circumstances and aggressors to support evaluation of breed specific legislation. Veterinary Journal, 186: 292–298 Craig L., Meyers-Manor J.E., Anders K., Sütterlin S., Miller H. 2017. The relationship between heart rate variability and dog aggression. Applied Animal Behaviour Science, 188: 59–67 De Keuster T., Lamoureux J., Kahn A. 2006. Epidemiology of dog bites: A Belgian experience of canine behaviour and public health concerns. Veterinary Journal, 172: 482–487 Denenberg S., Liebel F.X., Rose J. 2017. Behavioural and medical differentials of cognitive decline and dementia in dogs and cats. In: Canine and Feline Dementia. Landsberg G., Maďari A., Žilka N. (eds.) Springer International Publishing: 13–58 Denson T.F., Dewall C.N., Finkel E.J. 2012. Self-control and aggression. Current Directions in Psychological Science, 21: 20–25 Gobbo E. Mechanisms of aggressive behaviour in dogs. 100 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Diesel G., Pfeiffer D.U., Brodbelt D. 2008. Factors affecting the success of rehoming dogs in the UK during 2005. Preventive Veterinary Medicine, 84: 228–241 Dingemanse N.J., Wolf M. 2010. Recent models for adaptive personality differences: a review. Philosophical Transactions of the Royal Society B: Biological Sciences, 365: 3947–3958 Dodman N.H., Brown D.C., Serpell J.A. 2018. Associations between owner personality and psychological status and the prevalence of canine behavior problems. PLoS One, 13: e0192846, doi: 10.1371/journal.pone.0192846: 11 p. Duffy D.L., Hsu Y., Serpell J.A. 2008. Breed differences in canine aggression. Applied Animal Behavior Science, 114: 441–460 Duke A.A., Bègue L., Bell R., Eisenlohr-Moul T. 2013. Revisiting the serotonin-aggression relation in humans: a meta-analysis. Psychological Bulletin, 139: 1148–1172 Dzik V., Cavalli C., Iglesias M., Bentosela M. 2019. Do dogs experience frustration? Behavioural Processes, 162: 14–19 Ermatinger F.A., Brügger R.K., Burkart J.M. 2019. The use of infrared thermography to investigate emotions in common marmosets. Physiology and Behavior, 211: 112672 Essner A., Sjöström R., Ahlgren E., Gustås P., Edge-Hughes L., Zetterberg L., Hellström K. 2015. Comparison of Polar® RS800CX heart rate monitor and electrocardiogram for measuring inter-beat intervals in healthy dogs. Physiology and Behavior, 38: 247–253 Fatjó J., Amat M., Manteca X. 2005. Aggression and impulsivity in dogs. Veterinary Journal, 169: 150 Fatjó J., Amat M., Mariotti V.M., de la Torre J.L.R., Manteca X. 2007. Analysis of 1040 cases of canine aggression in a referral practice in Spain. Journal of Veterinary Behavior, 2: 158–165 Ferrari P.F., Palanza P., Parmigiani S., de Almeida R.M., Miczek K.A. 2005. Serotonin and aggressive behavior in rodents and nonhuman primates: predispositions and plasticity. European Journal of Pharmacology, 526: 259–273 Finka L.R., Ward J., Farnworth M.J., Mills D.S. 2019. Owner personality and the wellbeing of their cats share parallels with the parent-child relationship. PLoS One, 14: e0211862, doi: 10.1371/journal.pone.0211862: 26 p. Flint H.E., Coe J.B., Serpell J.A., Pearl D.L., Niel L. 2017. Risk factors associated with stranger-directed aggression in domestic dogs. Applied Animal Behaviour Science, 197: 45–54 Fratkin J.L., Sinn D.L., Patall E.A., Gosling S.D. 2013. Personality consistency in dogs: A meta–analysis. PLoS One, 8: e54907, doi: 10.1371/journal.pone.0054907: 19 p. Gácsi M., Gyoöri B., Virányi Z., Kubinyi E., Range F., Belényi B., Miklósi A. 2009. Explaining dog wolf differences in utilizing human pointing gestures: Selection for synergistic shifts in the development of some social skills. PLoS One, 4: e6584, doi:10.1371/journal.pone.0006584: 6 p. Gobbo E. Mechanisms of aggressive behaviour in dogs. 101 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Gácsi M., Maros K., Sernkvist S., Faragó T., Miklósi A. 2013. Human analogue safe haven effect of the owner: behavioural and heart rate response to stressful social stimuli in dogs. PLoS One, 8: e58475, doi: 10.1371/journal.pone.0058475: 9 p. Gautret P., Le Roux S., Faucher B., Gaudart J., Brouqui P., Parola P. 2013. Epidemiology of urban dog-related injuries requiring rabies post-exposure prophylaxis in Marseille, France. International Journal of Infectious Diseases, 17: e164–e167 Gershman K.A., Sacks J.J. Wright J.C. 1994. Which dogs bite? A case-control study of risk factors. Pediatrics, 93: 913–917 Haller J., Mikics E., Halász J., Tóth M. 2005. Mechanisms differentiating normal from abnormal aggression: glucocorticoids and serotonin. European Journal of Pharmacology, 526: 89–100 Haverbeke A., De Smet A., Depiereux E., Giffoy J.M., Diedrich C. 2009. Assessing undesired aggression in military working dogs. Applied Animal Behaviour Science, 117: 55–62 Haug L.I. 2011. Canine aggression toward unfamiliar people and dogs. Veterinary Clinics of North America: Small Animal Practice, 38: 1023–1041 Hazan C., Shaver P. 1987. Romantic love conceptualized as an attachment process. Journal of Personality, 52: 511–524 Herndon J.S., Bembenutty H., Gill M.G. 2015. The role of delay of gratification, substance abuse, and violent behavior on academic achievement of disciplinary alternative middle school students. Personality and Individual Differences, 86: 44–49 Hipp L., Bünning M., Munnes S., Sauermann A. 2020. Problems and pitfalls of retrospective survey questions in COVID-19 studies. Survey Research Methods, 14: 109–114 Horisberger U., Stärk K.D.C., Rüfenacht J., Pillonel C., Steiger A. 2004. The epidemiology of dog bite injuries in Switzerland – characteristics of victims, biting dogs and circumstances. Anthrozoös, 17: 320–339 Horváth Z., Dóka A., Miklósi Á. 2008. Affiliative and disciplinary behavior of human handlers during play with their dog affects cortisol concentrations in opposite directions. Hormones and Behavior, 54: 107–114 Houpt K.A. 2006. Terminology Think Tank: Terminology of aggressive behavior. Journal of Veterinary Behavior, 1: 39–41 Hsieh I-J., Chen Y.Y. 2017. Determinants of aggressive behavior: Interactive effects of emotional regulation and inhibitory control. PLoS One, 12: e0175651, doi: 10.1371/journal.pone.0175651: 9 p. Hsu Y., Serpell J.A. 2003. Development and validation of a questionnaire for measuring behavior and temperament traits in pet dogs. Journal of the American Veterinary Medical Association, 223: 1293–1300 Jakovcevic A., Elgier A.M., Mustaca A.E., Bentosela M. 2013. Frustration behaviors in domestic dogs. Journal of Applied Animal Welfare Science, 16: 19–34 Gobbo E. Mechanisms of aggressive behaviour in dogs. 102 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Jones A.C., Gosling S.D. 2005. Temperament and personality in dogs (Canis familiaris): A review and evaluation of past research. Applied Animal Behaviour Science, 95: 1– 53 Kaneko F., Arata S., Takeuchi Y., Mori Y. 2013. Analysis of association between behavioural traits and four types of aggression in shiba inu. Journal of Veterinary Medical Science, 75: 1297–1301 Konok V., Kosztolányi A., Rainer W., Mutschler B., Halsband U., Miklósi Á. 2015. Influence of owners’ attachment style and personality on their dogs’ (Canis familiaris) separation-related disorder. PLoS One, 10: e0118375, doi: 10.1371/journal.pone.0118375: 17 p. Kuhne F., Hößler J.C., Struwe R. 2014. Emotions in dogs being petted by a familiar or unfamiliar person: Validating behavioral indicators of emotional states using heart rate variability. Applied Animal Behavior Science, 161: 113–120 Kuraoka K., Nakamura K. 2011. The use of nasal skin temperature measurements in studying emotion in macaque monkeys. Physiology and Behavior, 102: 347–355 Lee D.Y., Kim E., Choi M.H. 2015. Technical and clinical aspects of cortisol as a biochemical marker of chronic stress. BMB Reports, 48: 209–216 Lensen R.C.M.M., Betremieux C., Bavegems V., Sys S.U., Moons C.P.H., Diederich C. 2017. Validity and reliability of cardiac measures during behavioural tests in pet dogs at home. Applied Animal Behaviour Science, 186: 56–63 Lensen C.M.M., Moons C.P.H., Diederich C. 2015. Saliva sampling in dogs: How to select the most appropriate procedure for your study. Journal of Veterinary Behavior, 10: 504–512 León M., Rosado B., García-Belenguer S., Chacón G., Villegas A., Palacio J. 2012. Assessment of serotonin in serum, plasma, and platelets of aggressive dogs. Journal of Veterinary Behavior, 7: 348–352 Lewandowski Jr G.W., Mattingly B.A., Pedreiro A. 2014. Under pressure: the effects of stress on positive and negative relationship behaviors. The Journal of Social Psychology, 154: 463–73 Ludwig N., Gargano M., Luzi F., Carenzi C., Verga M. 2007. Technical note: applicability of infrared thermography as a non invasive measurement of stress in rabbit. World Rabbit Science, 15: 199–206 McGowan R.T., Rehn T., Norling Y., Keeling L.J. 2014. Positive affect and learning: exploring the “Eureka Effect” in dogs. Animal Cognition, 17: 577–587 Meehan M., Massavelli B., Pachana N. 2017. Using attachment theory and social support theory to examine and measure pets as sources of social support and attachment figures. Anthrozoös, 30: 273–289 Messam L.L., Kass P.H., Chomel B.B., Hart L.A. 2012. Risk factors for dog bites occurring during and outside of play: are they different?. Preventive Veterinary Medicine, 107: 110–120 Gobbo E. Mechanisms of aggressive behaviour in dogs. 103 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Meyer I., Forkman B. 2014. Dog and owner characteristics affecting the dog-owner relationship. Veterinary Behavior, 9: 143–150 Milgram N.W., Head E., Weiner E., Thomas E. 1994. Cognitive functions and aging in the dog: Acquisition of nonspatial visual tasks. Behavioral Neuroscience, 108: 57–68 Mischel W., Shoda Y., Rodriguez M.I. 1989. Delay of gratification in children. Science, 244: 933–938, doi: 10.1126/science.2658056: 6 p. Mitchell D.G.V., Avny S.B., Blair R.J. 2006 Divergent patterns of aggressive and neurocognitive characteristics in acquired versus developmental psychopathy. Neurocase, 12: 164–178 Montoya E.R., Terburg D., Bos P.A., van Honk J. 2012. Testosterone, cortisol, and serotonin as key regulators of social aggression: A review and theoretical perspective. Motivation and Emotion, 36: 65–73 Mora E., Fonseca G.M., Navarro P., Castaño A., Lucena J. 2018. Fatal dog attacks in Spain under a breed specific legislation: a ten year retrospective study. Journal of Veterinary Behavior, 25: 76–84 Morgan M., Palmer J. 2007. Dog bites. British Medical Journal, 334: 413–417 Netto W.J., Planta J.U.D. 1997. Behavioural testing for aggression in the domestic dog. Applied Animal Behaviour Science, 52: 243–263 Ng Z.Y., Pierce B.J., Otto C.M., Buechner-Maxwell V.A., Siracusa C., Werre S.R. 2014. The effect of dog–human interaction on cortisol and behavior in registered animal-assisted activity dogs. Applied Animal Behaviour Science, 159: 69–81 Owczarczak-Garstecka S.C., Watkins F., Christley R., Westgarth C. 2018a. Online videos indicate human and dog behaviour preceding dog bites and the context in which bites occur. Scientific Reports, 8: 7147, doi: 10.1038/s41598-018-25671-7: 11 p. Owczarczak-Garstecka S.C., Watkins F., Christley R., Yang H., Westgarth C. 2018b. Exploration of perceptions of dog bites among YouTube™ viewers and attributions of blame. Anthrozoös, 31: 537–549 Oxley J.A., Christley R., Westgarth C. 2017. Contexts and consequences of dog bite incidents. Journal of Veterinary Behavior, 32: 33–39 Part C.E., Kiddie J.L., Hayes W.A., Mills D.S., Neville R.F., Morton D.B., Collins L.M. 2014. Physiological, physical and behavioural changes in dogs (Canis familiaris) when kennelled: testing the validity of stress parameters. Physiology and Behavior, 133: 260–271 Patronek G.J., Sacks J.J., Delise K.M., Cleary D.V., Marder A.R. 2013. Cooccurrence of potentially preventable factors in 256 dog bite-related fatalities in the United States (2000-2009). Journal of the American Veterinary Medical Association, 243: 1726– 1736 Planta J.U.D., De Meester R.H.W.M. 2007. Validity of the Socially Acceptable Behavior (SAB) test as a measure of aggression in dogs towards non-familiar humans. Vlaams Diergeneeskundig Tijdschrift, 76: 359–68 Gobbo E. Mechanisms of aggressive behaviour in dogs. 104 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Podberscek A.L., Serpell J.A. 1997. Aggressive behaviour in English cocker spaniels and the personality of their owners. Veterinary Record, 141: 73–76 Quaranta A., Siniscalchi M., Vallortigara G. 2007. Asymmetric tail-wagging responses by dogs to different emotive stimuli. Current Biology, 17: R199–201 Quirk J.T. 2012. Non-fatal dog bite injuries in the USA, 2005-2009. Public Health, 126: 300–302 Raaijmakers M.A., Smidts D.P., Sergeant J.A., Maassen G.H., Posthumus J.A., van Engeland H., Matthys W. 2008. Executive functions in preschool children with aggressive behavior: impairments in inhibitory control. Journal of Abnormal Child Psychology, 36: 1097–1107 Rammstedt B., John O.P. 2007. Measuring personality in one minute or less: A 10-item short version of the Big Five Inventory in English and German. Journal of Research in Personality, 41: 203–212 Reefmann N., Wechsler B., Gygax L. 2009. Behavioural and physiological assessment of positive and negative emotion in sheep. Animal Behaviour, 78: 651–659 Rehn T., Beetz A., Keeling L.J. 2017. Links between an owner’s adult attachment style and the support-seeking behavior of their dog. Frontiers in Psychology, 8: 2059, doi: 10.3389/fpsyg.2017.02059: 12 p. Reisner I.R., Mann J.J., Stanley M., Huang Y.Y., Houpt K.A. 1996. Comparison of cerebrospinal fluid monoamine metabolite levels in dominant aggressive and non-aggressive dogs. Brain Research, 714: 57–64 Reisner I.R., Shofer F.S. 2008. Effects of gender and parental status on knowledge and attitudes of dog owners regarding dog aggression toward children. Journal of the American Veterinary Medical Association, 233: 1412–1419 Reisner I. R., Shofer F.S., Nance ML. 2007. Behavioral assessment of child-directed canine aggression. Injury Prevention, 13: 348–351 Reynolds B.W., Basso M.R., Miller A.K., Whiteside D.M., Combs D. 2019. Executive function, impulsivity, and risky behaviors in young adults. Neuropsychology, 33: 212–221 Rigterink A., Moore G.E., Ogata N. 2018. Pilot study evaluating surface temperature in dogs with or without fear- based aggression. Journal of Veterinary Behavior, 28: 11–16 Rogers L.J. 2010. Relevance of brain and behavioural lateralization to animal welfare. Applied Animal Behaviour Science, 127: 1–11 Rosado B., Garcia-Belenguer S., Leon M., Chacon G., Villegas A., Palacio J. 2010. Blood concentrations of serotonin, cortisol and dehydroepiandrosterone in aggressive dogs. Applied Animal Behaviour Science, 123: 124-130 Rosado B., Garcia-Belenguer S., Leon M., Palacio J. 2009. A comprehensive study of dog bites in Spain, 1995- 2004. The Veterinary Journal, 179: 383–391 Gobbo E. Mechanisms of aggressive behaviour in dogs. 105 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Salman M.D., Hutchison J, Ruch-Gallie R., Kogan L., New Jr, J.C., Kass P.H., Scarlett J.M. 2000. Behavioral reasons for relinquishment of dogs and cats to 12 shelters. Journal of Applied Animal Welfare Science, 3: 93–106 Sarcey G., Ricard C., Thelot B., Beata C. 2017. Descriptive study of dog bites in France - Severity factors, factors of onset of sequelae, and circumstances. Results of a survey conducted by InVS and Zoopsy in 2009-2010. Journal of Veterinary Behavior, 22: 66–74 Séguin J.R., Zelazo P.D. 2005. Executive function in early physical aggression. In: Developmental origins of aggression. Tremblay R.E., Hartup W.W., Archer J. (eds.). New York, Guilford: 307–329 Shaver P. Mikulincer M. 2002. Attachment-related psychodynamics. Attachment & Human Development, 4: 133–161 Shuler C.M., DeBess E.E., Lapidus J.A., Hedberg K. 2008. Canine and human factors related to dog bite injuries. Journal of the American Veterinary Medical Association, 232: 542–546 Siniscalchi M., Bertino D., d’Ingeo S., Quaranta A. 2019. Relationship between motor laterality and aggressive behavior in sheepdogs. Symmetry, 11: 233, doi: 10.3390/sym11020233: 8 p. Súilleabháin P.O. 2015. Human hospitalisations due to dog bites in Ireland (1998–2013): Implications for current breed specific legislation. The Veterinary Journal, 204: 357– 359 Summers C.H., Winberg S. 2006. Interactions between the neural regulation of stress and aggression. Journal of Experimental Biology, 209: 4581–4589 Svartberg K., Forkman B. 2002. Personality traits in the domestic dog (Canis familiaris). Applied Animal Behavioral Science, 79: 133–155 Tidwell N.D., Eastwick P.W., Finkel E.J. 2013. Perceived, not actual, similarity predicts initial attraction in a live romantic context: Evidence from the speed dating paradigm. Personal Relationships, 20: 199–215 Touré G., Angoulangouli G., Méningaud J.P. 2015. Epidemiology and classification of dog bite injuries to the face: A prospective study of 108 patients. Journal of Plastic, Reconstructive & Aesthetic Surgery, 68: 654–658 Travain T., Colombo E.S., Grandi L.C., Heinzl E.U.L., Pelosi A., Prato-Previde E., Valsecci P. 2016. How good is this food? A study on dogs’ emotional responses to a potentially pleasant event using infrared thermography. Physiology and Behavior, 159: 80–87 Travain T., Colombo E.S., Heinzl E., Belluci D., Previde E.P., Valsecci P. 2015. Hot dogs: thermography in the assessment of stress in dogs (Canis familiaris) – a pilot study. Journal of Veterinary Behavior, 10: 17–23 Gobbo E. Mechanisms of aggressive behaviour in dogs. 106 Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Tuber D.S., Henessey M.B., Sanders S., Miller J.A. 1996. Behavioral and glucocorticoid responses of adult domestic dogs (Canis familiaris) to companionship and social separation. Journal of Comparative Psychology, 110: 103–108 Turgay A. 2004. Aggression and disruptive behavior disorders in children and adolescents. Expert Review of Neurotherapeutics, 4: 4623–632 Turcsán B., Range F., Virányi Z., Miklósia A., Kubinyia E. 2012. Birds of a feather flock together? Perceived personality matching in owner–dog dyads. Applied Animal Behaviour Science, 140: 154–160 Van den Bergh F., Spronk M., Ferreira L., Bloemarts E., Groenink L., Olivier B., Oosting R. 2006. Relationship of delay aversion and response inhibition to extinction learning, aggression, and sexual behaviour. Behavioural Brain Research, 175: 75– 81 Veissier I., Boissy A. 2007. Stress and welfare: two complementary concepts that are intrinsically related to the animal’s point of view. Physiology and Behavior, 92: 429– 433 Vernouillet A., Stiles L., Andrew McCausland J., Kelly D. 2018. Individual performance across motoric self- regulation tasks are not correlated for pet dogs. Learning & Behavior, 46: 522–536 Wiener, P., Haskell M.J. 2016. Use of questionnaire-based data to assess dog personality. Journal of Veterinary Behavior, 16: 81–85 Westgarth C., Brooke M., Christley R.M. 2018. How many people have been bitten by dogs? A cross-sectional survey of prevalence, incidence and factors associated with dog bites in a UK community. Journal of Epidemiology and Community Health, 72: 331– 336 Westgarth C., Watkins F. 2015. A qualitative investigation of the perceptions of female dog-bite victims and implications for the prevention of dog bites. Journal of Veterinary Behavior, 10: 479–488 Whipple N., Bernier A., Mageau G. 2011. A dimensional approach to maternal attachment state of mind: Relations to maternal sensitivity and maternal autonomy support. Developmental Psychology, 47: 396–403 Wright H.F., Mills D.S., Pollux P.M.J. 2011. Development and validation of a psychometric tool for assessing impulsivity in the domestic dog (Canis familiaris). International Journl of Comparative Psychology, 24: 210–225 Yang T., Yang C.F., Chizari M.D., Maheswaranathan N., Burke K.J., Jr, Borius M., Inoue S., Chiang M.C., Bender K.J., Ganguli S., Shah N.M. 2017. Social control of hypothalamus-mediated male aggression. Neuron, 95: 955–970 Zamora-Nasca L.B., di Virgilio A., Lambertucci, S.A. 2021. Online survey suggests that dog attacks on wildlife affect many species and every ecoregion of Argentina. Biological Conservation, 256: 10904 Gobbo E. Mechanisms of aggressive behaviour in dogs. Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 ACKNOWLEDGEMENTS I would like to thank my supervisor Assoc. Prof. Dr. Manja Zupan Šemrov for support, guidance, but also the freedom to explore. Thank you for everything you taught me and I'm looking forward to working together. Thanks to the members of my thesis committee - Assist. Prof. Dr. Dušanka Jordan, Prof. Dr. Marko Kreft and Assoc. Prof. Dr. Friederike Range for the comments on how to improve this dissertation. Special thanks go to Andrej Muhvič, Marko Medvešek, Primož Babič and the Service Dogs Training Section of the Slovenian Police and Prison Administration of the Republic of Slovenia for their willingness to be a part of our project. I would like to thank the Department of Animal Science of Biotechnical Faculty, University of Ljubljana for providing the test site and financial support for the neuroendocrine analysis. This work would not be possible without Alja Willenpart, Živa Logar, Živa Hernaus, Špela Zarnik, Lina Recer, Urša Blenkuš, Barbara Furdi, Helena Grbec, Viktorija Lipič and all the owners, handlers and dogs that participated in the studies. Thank you for your help! ˷ Posebej bi se zahvalila Mado, ki me je naučila brati in pisati. Zahvala gre tudi prijateljem in bratu za potrpežljivost in spodbude, še posebej takrat, ko sem se pritoževala, koliko imam za brati in pisati. Največja zahvala pa gre staršem. Za vse. To nalogo posvečam vama. Gobbo E. Mechanisms of aggressive behaviour in dogs. Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 ANNEXES Annex A Permission of Brill to use article: Gobbo E., Zupan Šemrov M. 2021. Factors affecting human-directed aggression resulting in dog bites: contextual aspects of the biting incidents. Society & Animals (published online ahead of print), doi: 10.1163/15685306-bja10066: 20 p. in the electronic version of the doctoral dissertation. Gobbo E. Mechanisms of aggressive behaviour in dogs. Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Annex B Permission of MDPI to use article: Gobbo E., Zupan M. 2020. Dogs’ sociability, owners’ neuroticism and attachment style to pets as predictors of dog aggression. Animals, 10: 315 in the electronic version of the doctoral dissertation. No special permission is required to reuse all or part of article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. Reuse of an article does not imply endorsement by the authors or MDPI. Gobbo E. Mechanisms of aggressive behaviour in dogs. Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Annex C Permission of Frontiers to use article: Gobbo E., Zupan Šemrov M. 2022. Dogs exhibiting high levels of aggressive reactivity show impaired self-control abilities. Frontiers in Veterinary Science, 9: 869068 in the electronic version of the doctoral dissertation. Copyright © 2021 Gobbo and Zupan Šemrov. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Gobbo E. Mechanisms of aggressive behaviour in dogs. Doct. dissertation. Ljubljana, Univ. of Ljubljana, Biotechnical Faculty, 2022 Annex D Permission of Frontiers to use article: Gobbo E., Zupan Šemrov M. 2021. Neuroendocrine and cardiovascular activation during aggressive reactivity in dogs. Frontiers in Veterinary Science, 8: 683858 in the electronic version of the doctoral dissertation. Copyright © 2021 Gobbo and Zupan Šemrov. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.