Scientific Program

The distribution of the golden jackal (Canis aureus) has dramatically changed in last decades. As a generalist and opportunistic species, jackal is capable to colonize wide range of habitats and exploit wide range of food resources. Its historic distribution was restricted to the southeastern Europe, mostly along the Adriatic and Black sea coasts and a few small isolated areas in the Balkans. Since the Second World War jackal population has been increasingly spreading through Europe and today reproductive populations are established in Central, East and North Europe with vagrant animals occurring also in parts of Western Europe. Factors affecting the jackal expansion across Europe are not completely understood. Likely the most important trigger that initiated the expansion was wide-spread persecution of wolf populations across Europe, which resulted in decline and local exterminations of this apex predator. Presence of wolves is still important factor affecting jackal distribution today, however, this effect can be weakened where wolf populations are under strong hunting pressure. Additional factors that have been suggested to influence jackal expansion include abundant and easy accessible anthropogenic food resources, changes in land use and wildlife management, and climate change. Although still understudied, jackal colonization of new areas can affect other wildlife,as well as people. These consequences can be negative, such aslivestock depredations, or positive due to ecosystem services provided by the jackal, but require further research to guide future management policy.

Among the social learning processes, imitation has always been challenging to study from a methodological perspective. Dogs trained with the Do as I Do method learn to match their behaviour to actions demonstrated by a human model upon request. The use of this method, combined with the two-action procedure provided evidence of imitation in dogs. By including a delay between the demonstration and the request to imitate, we investigated on dogs’ ability to form a mental representation of the demonstrated action and to recall it, to use it as the basis for performing a matching action (deferred imitation). In Do as I Do tasks, the observer’s behaviour depends on how it represents the model’s demonstration. Thus, this method can be also used to assess whether dogs represent the actions of others in a framework that includes their goals, as well as the means to reach those. Modified versions of the Do as I Do paradigm have also allowed to reveal episodic-like memory of others’ actions in dogs, a cognitive skill that was very elusive to test in non-human animals. In humans, episodic memory is thought to be linked to self-awareness. The findings about episodic-like memory in dogs pave the way to future studies on dogs’ ability to represent ‘the self’ and modified versions of the Do as I Do method may allow testing dogs capacity to represent and remember their own actions, a skill that may constitute a building block of the ability to represent ‘the self’.

Over the last decades dogs have become one of the most popular species in comparative psychology. There are several reasons why that is, but one is that because of dogs unique evolutionary history, dogs are a highly interesting model species for comparative research. Dogs have been compared to human infants, to primates and to other canids in a pursued to address different questions or to address similar questions, but from different angles. Here I will review dogs as a model species in comparative research. I will summarize some of the comparative research with a particular focus on dogs’ social cognitive skills. I will highlight the challenges and the benefits of the comparative approach and how it might help us investigate if dogs understanding of the world around them is in any way similar to ours or in fact quite different.

The differences in behavior of dogs and wolves strongly suggest an underlying genetic component. The genome sequencing identified several regions which differentiate the two species but it remains extremely difficult to pinpoint genes which are causally associated with domesticated behavior. Unlike modern dogs, which have been selected for any number of different traits including morphology and appearance since domestication began, the tame and aggressive fox strains (Vulpes vulpes) have been developed by 50 generations of selection solely for behavior. We sequenced a subset of foxes from the tame, aggressive, and conventional farm-bred populations and identified 103 regions with either significantly decreased heterozygosity in one of the three populations or increased divergence between the populations. Using experimental cross-breed pedigrees we highlighted a strong positional candidate gene for tame behavior: SorCS1, which encodes the main trafficking protein for AMPA glutamate receptors and neurexins and suggests a role for synaptic plasticity in fox domestication. Other regions identified as likely to have been under selection in foxes include genes implicated in human neurological disorders, in mouse behavior, and in dog domestication. The fox represents a powerful model for the genetic analysis of tame and aggressive behaviors that can benefit genetic studies of behavior in dogs and other domesticated species.

Despite numerous investigations leveraging both genetic and archaeological evidence, the geographic origins of dogs remain unknown. On the basis of an ancient Irish dog genome and an assessment of the spatiotemporal appearance of dogs in the archaeological record, a recent paper suggested that dogs may have been domesticated independently in Eastern and Western Eurasia from distinct wolf populations. Following those independent origins, a mitochondrial assessment suggested that the Mesolithic dog population in Western Europe may have been replaced by a population from the East. To test this hypothesis, we are generating nuclear genomes of ~10 Mesolithic dogs sampled from sites in Europe and Western and Eastern Russia, and mitochondrial genomes from ~400 dogs spanning the last 15,000 years across Eurasia. The results of this analysis will reveal the phylogenetic affinities of dogs that were present across the Old World prior to the introduction of dogs associated with farming communities. This study will also allow us to pinpoint the timing of the European mitochondrial turnover and to assess whether there was a commensurate turnover at the nuclear level, thus directly addressing whether dogs were domesticated from more than one population.

Vocal communication mediates a substantial proportion of interactions between domestic dogs and humans. Yet, our vocal communication systems are remarkably different. While humans have evolved one of the most complex systems of animal vocal communication (speech), like most nonhuman terrestrial mammals, dogs are considered to have very limited vocal flexibility and control. Moreover, the extent to which they can extract the complex information contained in human speech signals remains poorly understood. I will first review the key commonalities and differences in vocal production and perception between the two species. Then I will discuss recent research on humans’ perception of dogs’ vocal signals (focussing on size and emotions), as well as studies of dogs’ perception of human vocal signals (including nonverbal and verbal content). I will suggest that, while much of our vocal interactions appear to exploit the “size code” or “motivational-structural rules” that drive the evolution of vertebrate vocal communication, dogs may have been artificially selected to tune their perception abilities to extract relevant information from human speech.

An individual animal’s ability to cope with their environment, affects not only their wellbeing but also their potential working ability. Our studies of explosives search dogs show a significant link between the stress levels dogs experience upon entering training kennels and their subsequent performance after training. As the number of working dog roles increase, it is vital to determine optimal housing and husbandry, to implement welfare compatible breeding, rearing and training regimes, and to select dogs best suited specific tasks. I will describe a range of research studies, on search, medical assistance livestock guarding and racing dogs, aimed at elucidating the best ways to improve performance and welfare; prioritising those factors most important to the dogs. I will describe the importance of handler selection, based on attitude and behaviour and how training can improve handlers’ ability to detect the early signs of fear and hence avoid potential welfare problems. Finally, I describe our recent and ongoing research into racing greyhound welfare, prioritising and addressing some of the issues most important to the greyhound racing industry.

Each European country has, on average, four million households that cherish a canine companion. But not many people would welcome a wolf into their home. What makes dogs so uniquely friendly? I will present on the genetics that contributed to canine hyper-sociability, one of many traits central to their domestication. These variants occur in the same genes that cause Williams-Beuren Syndrome (WBS) in humans, a rare genetic disorder due to a chromosomal deletion that is characterized by overfriendliness. In our study, we found that dogs share very similar features in their hyper-social behaviors towards humans as people with WBS. When canines were presented with a problem-solving task in the presence of a human, wolves typically complete the task quickly. However, dogs displayed an attentional bias towards the human instead of solving their task, and engaged in prolonged greetings with humans relative to wolves. The genetic variants were identified as transposons and often were found inserted into the genes linked to WBS. We found that transposon insertions were strongly associated with the tendency for canines to seek out interactions with humans. Further, we found that these hypersocial variants also segregate at moderate frequencies in contemporary wolf populations. Our findings provide a possible story for how humans domesticated wolves, which is suggested to have been through “the survival of the friendliest”. As such, we present findings that give us insight into both the genetics of canine personality and how domestication shaped the genome of our canine companions.

Domestic dog breeds are characterized by an unrivaled diversity of morphologic traits, breed-associated behaviors, and disease susceptibilities. To identify the genetic underpinnings of such traits, we assembled 722 canine whole genome sequences (WGS), documenting over 91 million single nucleotide and small indels, thus creating the largest catalog of genomic variation for a companion animal species to date. We undertook genome wide association studies (GWAS), inclusive of over 130 breeds, to identify loci associated with morphologic features, behavior, and life span. Our results identify new variants of strong impact associated with body weight variation which, combined with existing data, explain greater than 90% of breed body size variability. We also demonstrate the relationship between body size variants and longevity, particularly via LCORL. Finally, we discuss our findings of genes under selection in the creation of terrier versus sport hunting breeds.

Non-conscious behavioural synchronization is evolutionary adaptive. It participates in fostering social cohesion; affiliation between interacting partners is associated with a high level of behavioural synchronization in many species, including humans. Conversely, humans also show increased affiliation towards people who synchronized with them. Surprisingly, until recently, little was known about these phenomena at interspecific level. Investigating the existence of similar behaviours and psychological processes in two different species is essential to better understand the respective roles of evolution and ontogeny in the skills. We therefore chose to investigate behavioural synchronization between two species that are closely associated and among which the effect of affiliation can be investigated. After presenting why dog-human dyads are a relevant biological model to study this field of social cognition, we review the recent findings about dog-human behavioural synchronization. For the first time, we present studies showing that behavioural synchronization of dogs is present between highly affiliated partners (i.e. pet dogs and their owners) in various situations. Pet dogs indeed synchronize their movement with their owners’ when freely walking inside or in open outside area, and when the owner adjusts differently to an unknown object or person. The effect of affiliation has been investigated by testing lower affiliated partners (i.e. shelter dogs and their caregiver). It has been found that shelter dogs synchronize their behaviour with their caregiver’s at lower rates than pet dogs with their owners when freely walking outside. More surprising, when encountering an unfamiliar person, shelter dogs do not synchronize their behavioural reaction with that of their caregiver contrarily to pet dogs relative to their owners. As between humans, affiliation thus modulates the degree of synchronization of dogs upon humans. This was the first time that the effect of affiliation on behavioural synchronization has been evidenced at interspecific level. More astonishing, it has been recently found that pet dogs present increased affiliation towards humans who synchronized their behaviour with them. It is the third species in which such an ability is evidenced, after humans and capuchins; and thus the first time it is found in canids. We conclude that, as in humans, behavioural synchronization acts as a social-glue between dogs and humans. This was the first time that such a human-like ability was evidenced at interspecific level, i.e. between humans and dogs. Mechanisms involved in dog behavioural synchronization, as well as implications of such findings in terms of pro-sociality, empathy, and theory of mind are discussed.