Perception of complex visual images
The observation that dogs can learn to discriminate various shapes, such as a circle and an ellipse, goes back to the experiments by Pavlov 1934 . Similarly, dogs can be trained to choose between objects that differ in shape, such as a cube or a prism Milgram et al. 2002 but systematic experiments are lacking, but see Range et al. 2007 . Dogs also show attraction to biologically meaningful but static visual images, such as the silhouette of a dog on a screen Fox 1971 , their own mirror images,...
The agonistic aspects of social relationships
In contrast to claims by experts in the field Bradshaw and Nott 1995 , modern ethological thought has had relatively little influence on the understanding of aggressive behaviour in dogs. Thus it seems timely to rethink dog aggression in terms of novel ideas that have been introduced by studying other animal species. There is general agreement among ethologists that the main function of aggression is to divide important but limited resources among group members. When the amount of available...
Heterochrony
The evolutionary change in the relative timing of developmental processes heterochrony has often been implicated as a source of phenotypic novelty Klingenberg 1998 . The idea that the transition from wolf to dog was made possible by such changes has been around for a long time Bolk 1926, Herre and Rohrs 1990 . The morphological and behavioural A ,' .- s r Figure 5.4 A schematic presentation of developmental changes based on Albrecht etal. 1979, Klingenberg 1998 . development of the ancestor...
Box continued Ikx
barks emitted in previously identified contexts by identified dogs. After the training phase the computer was provided with unfamiliar barks recorded in the same situations. The recognition rates found were high above chance level the algorithm could categorize the barks according to their recording situations and the barking individuals. Interestingly, the algorithm performed much better than humans. The program was successful both in categorizing the barks according to the predetermined...
Box Flexibility of the behavioural phenotype
Coppinger and Coppinger 2001 and Goodwin I et at. 1997 argued that the motor components i of predatory and aggressive behaviour show a mosaic pattern by being variably present and absent in certain genetically divergent breeds or i breed groups Table 8.2 . Frank 1980 noted i that the arbitrary relation between external stimuli and motor components of the behaviour j contributes to the behavioural flexibility in dogs which is advantageous in training. Altogether this I suggests that the...
Box One case for social competence the pedagogy hypothesis
Cognitive psychologists describe Leacher-learner interactions in humans as pedagogical knowledge transfer which can be defined as the explicit manifestation of generallzable knowledge by an Individual the teacher and interpretation of this manifestation in terms of knowledge content by another individual the learner Gergely and Cslbra 2006 . Teaching is often described as human-specific behaviour and is regarded as a primary, independent, and evolutionary earlier adaptation than many of our...
Dogs in the comparative psychology laboratory
One cannot avoid being emotionally touched on reading many of the papers published on dog behaviour in laboratories working on a Pavlovian model of learning. Professional scientists, often having a good 'personal' relationship with these dogs, often do not seem to realize what they are doing. There is no way that anyone today could or would do many experiments like these. The purpose of reviewing these experiments is to show how the lack of ethological thought can misdirect scientific efforts....
The archaeologists story looking at archaeological evidence
Two related but different kinds of evidence are usually collected to describe the process of dog domestication. When the interest is in the evolutionary aspect the emphasis is on the skeletal remains, but otherwise researchers look for possible indications of the relationship between humans and canids Morey 2006 . Most comparative archaezoologists agree that in general dogs can be discriminated from wolves on the basis of their generally reduced body size, shorter snout and facial part of the...
Evolution of the wolf
Today the wolf is recognized as a top predator throughout the northern hemisphere, but the situation was quite different even a few hundred thousand years ago Wang et al. 2004 . At that time herbivorous species were controlled by much larger predators on both continents. This was probably the result of a runaway evolutionary process in which there was a trend for increasing size in carnivore predators to outwit competitors. Their larger body size could be only sustained by a strongly...
Neural processing and visual ability
The dog's retina consists of two types of receptor cells that are non-uniformly distributed. The rods, which represent 97 of the receptor cells, are responsible for monochromatic vision in the dark. The maximum peak sensitivity of the visual pigment in the rods rhodopsin is at 506-510 nm, also indicating an adaptation to dim light. The remaining 3 of photoreceptors cones can be divided into two classes depending on their pigment content iopsin . Cones are responsible for colour vision, and the...
Box Wolf phenotypic plasticity
One reason why wolves may have been successful as the ancestor of dogs could be their phenotypic plasticity. Evolving and living in the temperate zone and surviving many glacial periods could have led to a species which has the means to adapt relatively rapidly to changing environments. To illustrate morphological and behavioural plasticity In wolves we combine data from various authors partially reported or cited by Mech and Boitani 2003 . Recent wolves follow the Bergman rule, thus in general...
Social relationships between and within wolf packs
In contrast to morphological and genetic research, comparative research on behavioural differences among wolf subspecies is lacking. Discussions on social behaviour always refer to 'the wolf' in general. However, the dynamic and variable social system of wolves probably played an important role in their survival in a range of different ecosystems. This ability to establish various group structures might have arisen as a consequence of being exposed to diverse environmental conditions during...
Visual communication
Unfortunately, very little is known about visual communication in wolves apart from the agonistic context see Harrington and Asa 2003 and even in this case little quantitative research has been done. In the case of the dog there have always been indications that dogs can rely on human visual communicative signals, and also that humans understand visual signals given by dogs. However, even in the case of dogs we have little knowledge of how they use visual signals among themselves. One general...
Is there an ethological description of aggressive behaviour in dogs
The short answer to this question is no. Various authors recognize the similarity between wolf and dog in the units of aggressive behaviour, and some texts provide shorter or longer lists of the behavioural units Feddersen-Petersen 1991, Packard 2003 . Importantly, behavioural analysis has been carried out at different levels of behavioural organization see also Chapter 2, Box 2.4 . For example, Feddersen-Petersen 2001a argued for seven facial regions muzzle posture, mouth corner, lips, nose...
Transition period days
Wolf cubs spend this period in the den too, mainly with their siblings and mother. Their motor abilities develop slowly and their exploratory behaviour is restricted to the immediate area underground. This period is characterized by increasing perceptual abilities. It starts with the opening of the eyes and ends with the opening of the ear canals. Interestingly, there is a large variation in the timing of both eye and ear opening which, at least at the level of the breed, seem to be...
Paws in hands
At the moment it seems that the future of dogs and humans is tightly coupled. Although there are no data, it is likely that the human population boom has also been paralleled by a huge increase in the dog population. A very crude estimate would put the size of the dog population somewhere between 0.5 and 1 billion individuals. A large part of the human economy, including veterinary medicine and the production of dog food, is devoted to the support of dogs. Sharing the environment not only means...
Geographic distribution and systematic relationships
Until 1800 the wolf was dispersed across Europe apart from the British Isles. Now, large populations 500 wolves survive only in Spain, Poland, Romania, Bulgaria, Serbia, the Baltic states, Ukraine, and central Russia Boitani 2003 . There are crude estimates of c.65 000 wolves living east of the Urals and in Asia, and probably a further 2000 living in Asia Minor and Egypt. The population in the Americas is judged to be about 60 000 individuals, of which only 10 are in the USA. Thus, based on...
Acoustic communication
Ethologists have spent many years collecting and analysing wolf vocalizations, both in the field and in captivity e.g. Theberge and Falls 1967, Harrington and Mech 1978, Schassburger 1993, Feddersen-Petersen 2000 , but there is little available data on dog vocalizations. Nevertheless there is a general consensus that the two species share most vocalizations Bleicher 1963, Cohen and Fox 1976, Tembrock 1976 , with the exception that dogs howl less frequently and are 'noisier' than wolves because...
Box A case study for dog personality research the Dog Mentality Assessment test
Recently, investigations led by Svartberg and others see text published a series of studies on dog personality based on the Swedish dog population that was subjected to the Dog Mentality Assessment DMA Test from 1997. This data set consists of more than 10 000 dogs belonging to a variety of breeds. Importantly, this test was designed not in order to investigate dog personality but to improve breeding standards in working dogs Svartberg and Forkman 2002 . The utilization of such a large data set...
Box The dog as a convergent behavioural model
Despite large morphological differences between humans and dogs, the notion of some sort of 'spiritual' similarity has been always around. Darwin 1872 also often refers to behavioural or mental parallels between dogs and humans, but it seems to depend from case to case whether the comparison is made on the basis of homology or convergence, Scott and Fuller's 1965 model of development of social behaviour in dogs was intended clearly as a homologous model for humans Chapter 9 , similarly to...
Box Representing the others state of mind
Gomez 2005 describes a method which seemed to be suitable to test for the ability to recognize knowledge or ignorance in others in species without language. Topal ef a . 2006a made only minor modifications to the procedure, which was originally used with an orang-utan, when testing a Belgian Tervueren dog Philip . The task of the subject is to get a piece of hidden food or a toy in the case of the dog by informing the helper human about the whereabouts of either the target object or a tool...
Descriptive approach to personality
A recent review on dog personality Jones and Gosling 2005 identified various goals of research, such as prediction of behaviour during development, description of behaviour traits to predict behavioural problems or individual suitability for certain training methods, or selection for preferred phenotypes. However, most recent reviews have concluded that many of these aims may be jeopardized by the lack of understanding of and more attention to theoretical and methodological problems. By...
An overview of wolves
Any argument on wolves Canis lupus as ancestors of dogs should rely on detailed knowledge of the species. Often the picture of wolves is oversimplified, hindering our understanding and interpretation of dog behaviour. Just as there is no such thing as 'the dog', there may be no such thing as 'the wolf'. We would argue that the range variability of the wolf phenotype covers many of the features that can be found compartmentalized in a mosaic pattern in other species of Canis. Up to the beginning...
Box Can a dog find its way home
One of the most highly praised abilities of dogs is finding their way home after getting lost. There are many anecdotal accounts of dogs returning home, which appear in more than one book. Writing about the intelligence of dogs, Menaull 1869 reports on a dog, Moffino, who returned home to Milan Italy after being lost somewhere in Russia after the Napoleonic wars. Dogs travelling on trains, or traversing huge areas to find their masters, were also among the most favourite anecdotes reported by...
Modelling of behaviour
Theories developed on the basis of modern biological, psychological, and even technical computational knowledge emphasize the possibility of interpreting behaviour in terms of inner states and processes of the mind. Accordingly, Shettleworth 1998 defines cognition as an array of mechanisms by which animals acquire, process, store, and act on information from the environment. The underlying framework for such views is based on the general assumption that the main function of the animal's mind is...
Changes in reproductive strategy and effects on generation times
An interesting consequence of dog domestication is the emergence of a diannual oestrus cycle. In contrast to wolves and with the exception of a few breeds females of domesticated canids can give birth to two litters per year. Tchernov and Horwitz 1991 argued that this trait could also be an adaptation to the anthropogenic environment, where large amounts of food could be utilized by a greater number of smaller animals, together with earlier maturity. Accordingly this would fit with the...
Top down or bottom up
Sometimes researchers have not much choice in formulating their models. Early cell biologists produced very crude models 'drawings' of the cell, which became more and more detailed as microscopes gained higher powers of resolution. Thus for mainly technical reasons cell biologists had followed a top-down approach to modelling. Meteorologists had to some extent the opposite fate. The modelling of wind systems probably started on a smaller scale, but as better technologies allowed for collecting...
Canon of parsimony
Morgan 1903 suggested that behaviour should be explained with reference to mental processes that stand lower on the scale of evolution and development, but he was also careful to add that 'the simplicity of explanation is no necessary criterion of its truth' Burghardt 1985 . Nevertheless, the first part of Morgan's suggestion reinforced approaches that interpret behaviour in terms of simple rules of association because this mechanism seems to be present even in very ancient organisms like the...
Comparing content and operation
Heyes 2000 suggested that we should distinguish between the content and the operation of the mind. She argued that the content of a mental representation depends on the species because ecological differences will determine 'what and when' is learned. In contrast, operational processes in the animal mind are based mainly on associative processes which do not differ markedly among animal taxa. This view shares many features with the general learning theory e.g. McPhail and Bolhuis 2001 ....
Box Contrasting alternative explanations how and why dogs learn to avoid eating
Solomon era . 1968 set out to examine the effect of delay of punishment on withholding some preferred action. The specific question was to find out the effectiveness of punishment if it coincides with the execution of the action. The subjects beagles were given a 'taboo training' when the dogs were punished for eating meat but were allowed to eat the same amount of dry laboratory chow. The experimenter punished the dogs by a hard blow on the snout with a tightly rolled-up newspaper. One group...
Box continued
touching the meat, and in parallel eating from the chow will be positively reinforced. Thus in the test the approach to meat arouses fear and delays approach. In the case of using long-delayed punishment dogs have the chance to experience the reinforcing effect of the meat which inhibits the effects of fear on approach behaviour. In the tests these dogs should approach food rapidly. 2 Cognitive 'A theory of conscience1 suggests that in both treated groups dogs know 'what they are not supposed...
Comparing intelligence
Unfortunately, the term intelligence has many different meanings, and is often used in a very superficial way. First, we should not forget that any kind of 'intelligence' reflects only the particular aspect of behaviour which was actually observed and tested under given conditions. Second, intelligence was originally invented as a measure for individual variability in flexible problem-solving abilities in humans . This means that it is questionable to use the concept of intelligence in a...
Epigenesis socialization and enculturation
Both bottom-up and top-down models often fail to recognize the complex ways in which genetic endowment can have an influence on mental processes. For example, genetic predisposition might orient the animal to certain aspects of the environment, which will determine what kind of experience is gained. Thus even small genetic differences can result in different kinds of mental representations through complex negative and positive feedback processes. In addition, the full potential of any organism...
An ethocognitive mental model for the dog
The model presented here is based on Csanyi's concept model 1989, 1993 but also includes ideas from both behaviour system and control structure models see above . The model assumes three different systems that 1 deal directly with environmental input perceptual system , 2 refer to aspects of the environment and inner state referential system , and 3 execute behavioural actions action system . All three systems function in a virtual two-dimensional space defined by a genetic and an environmental...
Box Scientific models of behaviour and dog training
Mills 2005 categorizes dog training technigues according to the two main behavioural models used in behavioural sciences, Accordingly associative training focuses on establishing a connection between two events, while more cognitive oriented approaches take into account the role of attention and the knowledge of the learner. In a similar vein, Lindsay 2005 assumes mental modules with abilities like 'prediction-control expectancy', 'emotional establishing operation', and 'goal direction'. From...
Introduction 1
The rediscovery of dogs for behavioural research is probably one of the most exciting developments in recent years. The fact that people with very different scientific training have started to study dogs has led to an increasingly confusing situation where a range of methods is applied, often without a clear understanding of their validity and limitation. Some researchers apply methods only because they seem to be simpler or faster, or because they have been used by others in the past. In some...
Finding phenomena and collecting data
De Waal 1991 argued that the 'real strength' of ethologists lies in the complementary use of different observational and experimental methods. Although his summary was based on primates it is clear that dogs offer an even better example because there is a wider range of possibilities. First of all, most observations on dogs take place 'in the wild' that is, in environments which are regularly inhabited by dogs. The environment could be the home of a human family, or even a laboratory which...
Qualitative description
People having regular and extensive contact with dogs often witness unique events. Anecdotes or qualitative descriptions of behaviour can be regarded as 'accidental observations' if the events are described in detail, in writing. The popular literature on dogs is filled with such stories, which not only serve as entertainment for the reader but are also presented as a sort of evidence in order to underline assumptions about the complex abilities of dogs. In scientific literature anecdotes are...
Quantitative description
Only the systematic collection of quantitative data allows scientific hypotheses to be tested. The explanatory value of such work often depends on the possibility of how well various variables can be controlled in the course of the observations. In the case of so-called uncontrolled observations, the main aim is to collect quantitative behavioural data with regard to some specific research question. For example, we might observe dogs sharing their life with inhabitants of a village, and by...
Box Do dogs show us what they want How to utilize anecdotes
Two well-known and experienced scientists and dog experts reported similar stories in their recent book on dogs. Due to space limitations both anecdotes are presented in a condensed form, together with a summary of the interpretations offered by the authors. Csanyi 2005, p. 38 After getting home from a walk in the rain, I had forgotten to dry him. Flip ran after me, got in front of me, stopped, and started to dry his head on the rug. Then he stopped and looked at me questioningly. 'Do you want...
Making behavioural comparisons
Researchers interested in the evolutionary effect of domestication have often based their arguments on the comparison between dogs and wolves. Although species comparisons seem to be quite a straightforward method for looking at adaptive processes in evolution, in reality nothing can be further from the truth. The main reason for this is that such comparisons often violate the basic condition for any comparative work that is, that only one independent variable can be changed at a time. Thus in...
Wolves and dogs
Unfortunately, dog-wolf studies are not exempt from the problems of comparative research. As a recent example it seems useful to refer to experimental investigations which were aimed at finding out whether dogs are able to rely on human pointing gestures Box 1.2 . In a study designed to find support for the hypothesis that domestication resulted in enhanced communicative skills in dogs Hare et al. 2002 , researchers found that dogs were superior to wolves in a test which involved choosing a...
Box continued 1
Figure to Box. 2.2 a, b Characteristic moments of the wolf socialization programme in Budapest, c Two-way choice test with a socialized wolf photos by Attila Molnar and EnlkoKubinyi Ludwig Huber . d The performance of dogs and Intensively socialized wolves-with the momentary pointing gesture. Dotted line, chance level significantly above chance performance. The percentages in the columns show the ratio of animals that choose significantly over chance binomial test, p lt 0.03, at least 15...
Dogs and children
Interestingly, from the beginning of dog research there have been proposals for comparative work with children. Menzel 1936 and Scott and Fuller 1965 argued for comparative ontogeny in dogs and children others Buytendijk and Fischel 1936 emphasized the similarity of the social relationship with humans. In spite of such theoretical discussion very little experimental work has been carried out. Importantly, in primate research such comparative work has long been performed, despite the fact that...
Sampling and the problem of single cases N
Comparative experimental work often raises the problem whether there are 'typical' dog breeds, or to put it in a different way 'What kind of sample can be said to be representative of dogs ' Unfortunately, there is no simple answer to this question because it would be hard to argue that one or a few breeds are more 'dog-like' than others. This question is also problematic if comparative work includes the wolf. The breeds cannot be ranked along a continuum of difference from the wolf, and it is...
Box Are there breed differences in humandirected communicative skills
Although it is generally assumed that dogs, as a species, have an advantage in communicating with humans, the selective environment might have affected different dog populations 'breeds' in different ways. For example, some dog breeds might have been under stronger human control for developing human-oriented communicative skilfe e.g. recent gundogs , whereas other breeds selected for different tasks might not display such abilities. Further, there are some arguments hare and Tomasello 2005...
ox continued
Figure to box 2.3 a Two representatives of the non-cooperative Independent hunters and cooperative hunting breeds. Hanover bloodhound non-cooperative, left , Welmaraner cooperative, right , b Cooperative hunting breeds are more successful in the two-way choice task than non-cooperative dogs, c Pedigree dogs achieve higher level of performance than mixed breed dogs, despite similar socialization history. Dotted line, chance level , significantly above chance performance significant differences...
A procedural problem in naturalistic observations the presence of humans
The ethological study of any animal aims for observations in the natural environment. This means that dogs should be observed under conditions that are 'natural' to them. The most significant compound of the environment for many dogs is the human s with whom they maintain a special relationship. Based on this reasoning, we have always observed the dogs in the presence of their owners e.g. Miklosi et al. 2000 in contrast, others avoid the presence of the owner and the dog is managed by a...
The mysterious laws of correlation
Obviously there are some trivial relationships between two or more phenotypic traits, and nobody is surprised to find that animals with longer long bones tend have longer skulls. A 'mystery' is involved when traits affecting very different aspects of the phenotype seem to be coupled in some way Box 5.8 . Such a correlation between fur colour and behaviour has often been implied and indeed verified to some extent Clutton-Brock 1984 . For example, solid-coloured cocker spaniels show a greater...
Box Behavioural coding in dogs an example
Various methods have been used to describe the of describing agonistic behaviour in dogs or wolves behaviour of dogs. The wide ranging possibilities are presented in the table below as an example. Method Short description Explanation of Behavioural Main reference 1. Single discontinuous categorical scale Scaling along a single dimension of aggressiveness The total score of whether the subject displays an item out of 10 aggressive behaviour elements Each category is characterized by a list of...