In comparative psychology we draw inferences about the evolution of cognition by investigating the similarities and differences between human and non-human animals. I am especially interested in which cognitive skills have evolved in different species that allow them to be optimally adapted to their environment. Dogs, due to their high sociality and the fact that they were subject to a special domestication process, represent a highly promising model to investigate social cognition from a comparative perspective. Studying dog cognition not only sheds light on the question on what skills humans share with other animal species, but also what kind of selection pressures lead to human-like skills. Dogs are not simply pets that live in the human environment, but they also form a close relationship to humans and cooperate with them. Thus, studying dogs living in that special niche will not only inform us about their cognitive skills but might also help us to better understand the selection pressures that led to the unique cognition of humans. In this thesis I present a view on dog cognition that differentiates individual cognitive skills, pointing out how exactly they are adapted to their special human environment. Thus, I emphasize the unique closeness of the dog-human relationship. I also point out where current findings are incomplete or show limits of their paradigms and call for further research. Firstly, I criticize the fact that most data on dogs’ understanding of their social and physical environment is based on performance in the visual or sometimes in the auditory modality. As dogs’ olfaction is their most relevant sense, I therefore call for more dog studies that are based on olfaction. Secondly, regarding the dog-human relationship, there are many open questions that have not yet been considered well enough: for example, whether dogs are capable of skills like empathy, the human perspective on dogs, and cultural differences in dog-human interactions. Thirdly, to better understand the dog-human bond it is crucial to further investigate when, where and how domestication started. This is also needed in order to understand why dogs were domesticated and what made and makes them valuable for humans. To answer the above mentioned questions, an interdisciplinary approach is crucial, in which scientists from the fields of archaeology, linguistics, paleoclimatology, genetics, anatomy, ethology, psychology, sociology, and anthropology work together.
The aim of this habilitation thesis is to characterize the dog-human relationship taking diverse perspectives on dog cognition and the dog-human bond. During their long domestication process, dogs have evolved special cognitive skills that help them to function effectively in human societies. In this thesis, I present experimental evidence for these skills and I suggest that the skills have evolved in a domain-specific manner, independently from each other. Dogs show outstanding cognitive skills in the domain of (1.) communication, (2.) perspective taking, (3.) cooperation and (4.) olfaction processing, but perform poorly or average in other domains such as (5.) metacognition and (6.) behavioral matching.
Regarding (1.) communication I present experimental evidence that dogs without special training are able to successfully show a human a hidden object and that this showing behavior in dogs is a means to communicate the location of that hidden object. I argue that successful communication between dog and humans in general is the consequence of four preconditions in dogs: (i) they are extremely attentive and interested in what humans are doing, (ii) they have excellent learning abilities, (iii) they are able to read subtle cues of human behavior and (iv) they have extensive experience with different communicative situations.
Furthermore, I present an experiment about (2.) perspective taking – defined as the ability to
assess what others can perceive. Here I found that dogs are able conceal auditory but not visual information from humans when they approach forbidden food. Taken together with findings from previous studies, I conclude that dogs use certain strategies when they assess what a human can and cannot perceive. I studied (3.) cooperation both within dogs and between dogs and humans. Within dogs I used a problem-solving paradigm that involved aspects of a hunting-like situation. I compared the performances of dogs with those of wolves. My results suggest that the abilities needed to coordinate their actions were already present in the dog-wolf ancestor. Dogs and wolves may show similar cooperative skills when cooperating with their conspecifics, but dogs might cooperate better with humans than wolves do, as it is likely that during the domestication process dogs have been selected to cooperate specifically with humans. Consequently, I investigated the cognitive and motivational skills required for a dog to support a human. From the results I conclude that dogs display a number of prosocial behaviors towards a human when they are able to infer
the goal of the human and when they understand how to fullfill it.
In contrast to communicative, cooperative and perspective taking skills, (4.) the special olfactory skills of dogs probably did not evolve during domestication, but could be one of the reasons why dogs were domesticated. Here I present evidence that dogs can use olfactory information in an adaptable way: Dogs were presented with a violation-of-expectation paradigm in which they could track the odor trail of one target, but at the end of the trail, they found another target. I found that they are able to represent what they smell—that is, when they follow a trail they have an expectation of something or someone at the end of the trail. Thus, not only is dogs’ sense of smell itself quite outstanding, but so are also their related cognitive skills.
In contrast, in their (5.) metacognitive skills and in (6.) behavioral matching, dogs do not show unique skills but perform similarly to other social mammals. Regarding (5.) metacognition I investigated whether dogs were sensitive to the information that they themselves had or had not acquired. I found that dogs seek additional information in uncertain situations, but their behavior in these situations is less flexible compared to great apes or human children. Finally, I did not find evidence for (6.) behavioral matching, ie. whether dogs develop an increased affiliation towards a human who mimics them. Dogs in my study showed no increased preference for one of two human experimenter who matched the dogs’ walk.