The Dog as a Model for Understanding Human Social Behavior

József Topál, Ádám Miklósi, Márta Gácsi, Antal Dóka, Péter Pongrácz, Enikő Kubinyi, Zsófia Virányi, Vilmos Csányi

Chapter 3 The Dog as a Model for Understanding Human Social Behavior

Advances in the Study of Behavior, Volume 39, 2009, Pages 71–116

Abstract

The traditional approach for studying the evolutionary emergence of human social cognition is based on comparisons with apes and monkeys as model species with a homologous relationship to other primates and humans. Recently, however, research interest has focused on other species offering analogous models for the evolution of human social cognitive abilities. Here we suggest that convergent social evolution in dogs can be used to model the early state of human social evolution, suggesting that functionally analogous forms of many traits of the human behavioral complex are present in dogs. We argue that the dog as a model species is unique among domesticated species because (a) many aspects of dog behavior are functionally analogous to the corresponding human traits, (b) socialization to humans is a natural process in dogs, (c) comparison with the ancestor is important for convergent modeling, and (d) the dog represents a natural experimental model for studying human behavior.

A friend or an enemy? Dogs’ reaction to an unfamiliar person showing behavioural cues of threat and friendliness at different times

Dogs and Strangers

Judit Vas, József Topál, Márta Gácsi, Ádám Miklósi, Vilmos Csányi

A friend or an enemy? Dogs’ reaction to an unfamiliar person showing behavioural cues of threat and friendliness at different times

Applied Animal Behaviour Science, Volume 94, Issues 1–2, October 2005, Pages 99–115

Abstract

Responsiveness of adult pet dogs (Canis familiaris) to an unfamiliar human was observed in two studies. Subjects were faced with an approaching woman (Stranger) who showed definite signs of friendliness and threat during alternate approaches. Observations consisted of two episodes: the Stranger either approached the dog in normal speed of walk while talking to it and finally petted it gently (Friendly approach episode) or she moved slowly and haltingly and looked steadily into the eyes of the dog without any verbal communication (Threatening approach episode).

In the first study 30 dogs of 19 different breeds were tested in the two episodes in a balanced sequential order. The dogs acted appropriately according to the different human behaviour cues. The order of the Friendly/Threatening approaches had no significant effect on the dogs’ responsivity.

In the second experiment 60 dogs of three breed groups (20 Belgian shepherds, 20 retrievers and 20 sled dogs) were first ‘greeted friendly’ and then approached ‘threateningly’ by the same Stranger. Results show significant breed specific differences in the responsivity when dogs faced an apparent switch of the human behaviour cues. Compared to retrievers and sled dogs, Belgian shepherds more frequently changed their response, showing passive or active avoidance or sign of aggression when approached threateningly.

While sex differences were not found, breed comparisons suggest that selective breeding (i.e. for hunting or shepherd work) influenced the dogs’ sensitivity to human social cues in different ways. Results also support the hypothesis that human influence (domestication) has led to extreme flexibility of the dogs’ situation-relevant behaviour while interacting with an unfamiliar human.

Attachment to humans: a comparative study on hand-reared wolves and differently socialized dog puppies

Attachment Theory, Hand-Reared Wolves

József Topál, Márta Gácsi, Ádám Miklósi, Zsófia Virányi, Enikő Kubinyi, Vilmos Csányi

Attachment to humans: a comparative study on hand-reared wolves and differently socialized dog puppies

Animal Behaviour, Volume 70, Issue 6, December 2005, Pages 1367–1375

Using the Strange Situation Test originally developed for testing the mother–infant relationship in humans, we compared the attachment behaviour of extensively socialized (hand-reared) dog, Canis familiaris, and wolf, Canis lupus, puppies towards their human caregiver with that of pet dog puppies of the same age. The experiment was designed to study whether (1) dog puppies as young as 16 weeks show attachment to a human caregiver, (2) extensive socialization by human caregivers affects attachment behaviour of dog puppies and (3) evolutionary changes (in the form of species-specific differences between wolf and dog pups) affect the emergence of dog–human attachment. We found a characteristic selective responsiveness to the owner in young dogs, similar to that observed in adults. This finding supports the view that puppies show patterns of attachment towards their owners. Extensive socialization had only a minor effect on the attachment behaviour in dog puppies, as the behaviour of pet dogs and hand-reared dogs was basically similar. However, we found a significant species-specific difference between wolves and dogs: both extensively socialized and pet dog puppies were more responsive to the owner than to an unfamiliar human participant, whereas extensively socialized wolves were not. Behavioural differences could be best explained by assuming that selective processes took place in the course of domestication (genetic changes) that are related to the attachment system of the dog.

Domestication is generally viewed as an evolutionary process controlled by human influence (Price 1984). The symbiotic relationship between humans and nonhuman animals entails adaptational demands, which create new conditions of selection for the species to be domesticated and thus might result in a wide range of genetic modifications. The dog has a long history of adaptation to the human environment ( [Vilá et al., 1997] and [Savolainen et al., 2002]), and it is widely assumed that the selection process during domestication may have altered not only their morphological traits but also their behaviour and behaviour control systems ( [Belyaev, 1979] and [Coppinger and Coppinger, 2002]).

Recent studies have suggested an unusual competence of dogs in social interactions with humans (cooperation: [Topál et al., 1997], [Naderi et al., 2001] and [Naderi et al., 2002]; social learning: [Kubinyi et al., 2003], [Pongrácz et al., 2003a] and [Pongrácz et al., 2003b]; communication: [Miklósi et al., 1998], [Miklósi et al., 2000], [Agnetta et al., 2001], [Soproni et al., 2001] and [Soproni et al., 2002]). However, to understand the significance of domestication-related changes in the behaviour of dogs, we need to compare dogs with wolves (Miklósi et al. 2004). In line with this, recent comparisons of the social cognitive skills in dogs and socialized wolves within the context of the interspecific relationship with humans have shed light on some genetic divergences at the behavioural level ( [Hare et al., 2002] and [Miklósi et al., 2003]). Compared with wolves, the dogs' preferential looking at the human in problem-solving situations and their superior performance in using human directional gestures support the existence of genetic predispositions related to the domestication process in the emergence of social cognitive abilities in dogs.

In general, it is widely accepted that the evolutionary emergence of social cognition is closely related to the ‘social field’ (Kummer 1982), which often presents more complex adaptational demands for the animal than do physical characteristics of the environment (Tomasello & Call 1997). One of the basic behavioural phenomena of social relationships is attachment. The evolutionary approach to function and mechanism suggests that attachment is one of the main behaviour organizing systems in parent–offspring relationships, and it is also claimed to be the basic organizational factor for any species' social structure leading to group formation (Bowlby 1958). Attachment is an asymmetrical social relationship between two individuals, which can be tested experimentally in choice situations such as the Strange Situation Test (SST) originally developed to study the mother–infant relationship in humans (Ainsworth & Wittig 1969). The paradigmatic element of this procedure is that separation from the caregiver in an unfamiliar environment evokes anxiety (which is manifested behaviourally in proximity seeking), while the activated attachment system upon reunion with the caregiver manifests in different forms of contact-seeking behaviours. Importantly, attachment behaviour is oriented mainly towards the caregiver, in the sense that there is a significant difference in the level of proximity and contact seeking, and in the effort made to maintain contact, between the caregiver and an unfamiliar person in the same novel situation. Adult dogs show specific patterns of attachment behaviour towards their owner in the SST ( [Topál et al., 1998] and [Prato-Previde et al., 2003]), suggesting a case of functional analogy (evolutionary convergence) to the human infant–parent attachment. Gácsi et al. (2001) reported that attachment develops rapidly: a short period of interaction with humans evoked attachment behaviour towards the handler in adult dogs that had been deprived of human contact (shelter dogs) and the dogs differentiated between their handler and a stranger in the same way as adult pet dogs did.

We designed a comparative experiment to investigate the attachment behaviour of hand-reared and extensively socialized wolf and dog puppies and pet dog puppies that had received a normal socialization regimen from their owners. We investigated whether (1) pet dogs' attachment to humans is observable in the SST as early as 16 weeks of age, (2) extensive socialization by human caregivers causes any change in attachment behaviour of dog puppies and (3) there are species-specific differences between wolves and dogs in their attachment behaviour to humans.

Although one might assume that the ability to show attachment behaviour to individuals of another species (humans) in adulthood is a unique feature of the domestic dog, despite much interest ( [Scott, 1963], [Scott, 1992] and [Ginsburg and Hiestand, 1992]), there has been no clear theory explaining the emergence of the phenomenon. By comparing the emergence of attachment behaviour to humans in dogs and socialized wolves tested by the same experimental method we can examine whether inheritance (genetic background) or environmental effects (rearing history) are more important.

Two hypotheses can be formulated. The socialization hypothesis suggests that attachment could develop mainly as a result of extensive hand rearing and individual socialization to the human social environment (i.e. enculturation) during the ‘critical period’ of socialization (Freedman et al. 1961). The domestication hypothesis, however, claims that there could have been specific genetic changes (in the attachment behaviour organizing system) that have emerged as the result of selective breeding for dependency and attachment to humans (see also [Hare et al., 2002] and [Miklósi et al., 2003] for similar explanations regarding communicative abilities in dogs). The socialization hypothesis predicts that hand-reared wolf and dog puppies will show similar forms of attachment behaviour to their human caregivers, whereas pet dog puppies, being less extensively socialized, will show less attachment to their owners. In contrast, the domestication hypothesis predicts species-specific differences in attachment behaviour to humans between wolves and dogs reared in the same way (i.e. dogs should show more specific attachment behaviour than wolves towards humans). These explanations are not mutually exclusive, however, and both of the hypothesized mechanisms could affect the behaviour phenotype.

Puppy power! Using social cognition research tasks to improve socialization practices for domestic dogs (Canis familiaris)

Tiffani J. Howell, Pauleen C. Bennett

Puppy power! Using social cognition research tasks to improve socialization practices for domestic dogs (Canis familiaris)

Journal of Veterinary Behavior: Clinical Applications and Research, Volume 6, Issue 3, May–June 2011, Pages 195–204

Abstract

Understanding the psychology of the domestic dog is a key element in both research and clinical applications for veterinary behaviorists. This article summarizes social cognitive skills observed in dogs in recent cognition studies. The tasks are grouped into the following 2 categories: those which facilitate eye contact between dogs and human beings, and social learning tasks. We also describe easy-to-implement experimental paradigms, previously used in cognitive tests for adult dogs, which we recommend adapting for use with puppies as young as 3 weeks of age. We highlight specific ways by which breeders may alter setup of these tasks to further generalize the social cognitive benefits for the puppies. Furthermore, studies of social development in puppies would enable researchers to explore whether encouragement of the skills examined have a positive effect on behavior as adults, and should thus be specifically incorporated into socialization practices by all breeders and new owners. Thus, the practical applications of this knowledge could include informing best practices for early age socialization by breeders, while puppies are still in the litter, and by new owners when they initially acquire a puppy. This could potentially increase the number of dogs that are well adjusted for human society and thus reduce the number surrendered to shelters. Finally, we discuss the ethical implications of working with puppies in particular and with companion animals generally; the positioning of veterinary clinicians and researchers between the scientific and lay worlds can improve understanding within the community of the benefits that minimally invasive companion animal research can provide.

Dogs, Canis familiaris, communicate with humans to request but not to inform

Juliane Kaminski, Martina Neumann, Juliane Bräuer, Josep Call, Michael Tomasello

Dogs, Canis familiaris, communicate with humans to request but not to inform

Animal Behaviour, Volume 82, Issue 4, October 2011, Pages 651–658

ABSTRACT

Dogs are especially skilful at comprehending human communicative signals. This raises the question of whether they are also able to produce such signals flexibly, specifically, whether they helpfully produce indicative (‘showing’) behaviours to inform an ignorant human. In experiment 1, dogs indicated the location of an object more frequently when it was something they wanted themselves than when it was something the human wanted. There was some suggestion that this might be different when the human was their owner. So in experiment 2 we investigated whether dogs could understand when the owner needed helpful information to find a particular object (out of two) that they needed. They did not. Our findings, therefore, do not support the hypothesis that dogs communicate with humans to inform them of things they do not know.

Reevaluating canine perspective-taking behavior.

Udell MA, Wynne CD
Reevaluating canine perspective-taking behavior.

Learn Behav. 2011 Dec ;39(4):318-23. doi: 10.3758/s13420-011-0043-5.

Abstract

Udell, Dorey, and Wynne (2011) demonstrated that both domesticated and nondomesticated canids-specifically, gray wolves-have the capacity to succeed on perspective-taking tasks, suggesting that dogs' ability to respond to the human attentional state is not a by-product of domestication alone. Furthermore, not all dogs were successful on the task. Instead, the occluder type used was a strong predictor of performance, indicating the important role of environment and experience for tasks of this type. Here, we address several commentaries reflecting on the methods and design of that study, as well as the interpretation of the results. We also discuss the positive shift toward more interactive approaches in the field of canine behavior and cognition. Finally, we question the functionality of describing canine social behavior in terms of theory of mind.

Can your dog read your mind?: Understanding the causes of canine perspective taking.

Monique A. R. Udell, Nicole R. Dorey, Clive D. L. Wynne

Can your dog read your mind?: Understanding the causes of canine perspective taking.

Learn Behav. 2011 December; 39(4): 289–302. doi: 10.3758/s13420-011-0034-6

Abstract

Prior studies have documented the domestic dog's (Canis lupus familiaris) sensitivity to human attentional state, including a tendency to preferentially beg for food from attentive individuals and an ability to selectively perform forbidden behaviors when humans are not looking. Due to the success of dogs on perspective-taking tasks, some have hypothesized that domestic dogs may have theory of mind, or the ability to infer what other individuals know. Here we provide the first evidence that nondomesticated canids, grey wolves (Canis lupus), are also sensitive to human attentional state under some conditions. We also demonstrate that dogs do not display an undifferentiated sensitivity to all visual cues of attentional state. Rather, dogs are more sensitive to stimuli encountered in their home environment. Some dogs perform poorly on perspective-taking tasks. These findings have important implications for the interpretation of research designed to understand complex social cognition across species.

A comparison of dog owners' claims about their pets' guilt with evidence from dog behavior

Damon Horowitz, Alexandra Horowitz

A comparison of dog owners' claims about their pets' guilt with evidence from dog behavior

Journal of Veterinary Behavior: Clinical Applications and Research, Volume 4, Issue 2, March–April 2009, Pages 104

ABSTRACT

New research on the social cognition of the domestic dog has established the dog's skill at interpreting communicative behavior of humans. By contrast, human interpretations of dog behavior rely heavily on anthropomorphisms, claims generally unverified by scientific research. The current research aims to examine behavior associated with one claim dog owners make of their dogs: “guilt” at doing a behavior that has been disallowed.

Previous research asking owners what they believed their dogs “know” found “disallowance” (disobedient acts) as the most frequent response. This study looks at dogs' behavior after doing or not doing a disallowed act, and after being scolded or greeted. We hypothesized that the data will match a model that suggests that dogs are responding to human behavior, but not a model suggesting dogs know a disallowed act is “wrong.”

Dogs and their owners participated in 5 videotaped trials each in the owners’ homes. Only dogs who obeyed commands of “sit,” “stay,” and “no” were used. Using owner report and prior behavioral descriptions of “guilt,” 8 guilt-related behaviors (GRBs) were isolated: avoiding eye contact, rolling over, low wag, ears back, head down, licking, retreat, and paw-up. In each trial, the owner showed the dog a treat, forbade eating it, and left the room. The owner returned in 30 seconds, was told whether the dog ate the treat, and was asked to act normally to greet or chastise the dog, as appropriate. To test the hypothesis that GRBs do not appear exclusively to doing a disallowed act, the trials varied the availability of the treat: a confederate removed the treat; let it be; or gave it to the dog. To test the hypothesis that GRBs are a response to owner chastising behavior, the trials varied the owners' knowledge of the dog's disobedience: guilty, owner knows; guilty, owner misinformed; not guilty, owner knows; not guilty, owner misinformed.

In every trial in which the treat was available, the dogs disobeyed and ate the treat. Rate of GRBs in all trials was scored. Preliminary analysis found no difference in the number of GRBs between disobedient and non-disobedient trials. By contrast, more GRBs were seen in trials when the owner scolded the dog, whether the dog had disobeyed or not. Confederate permission of the treat resulted in more GRBs than confederate removal of the treat.

In the context of a disallowed treat, dogs’ GRBs corresponded to the human's scolding of the dog, following discovery of the disallowed act or the belief that the act had occurred. GRBs were not more common when the dog did a disallowed act than when it had not. GRBs are similar to behaviors indicating fear and submission. In this study, dogs’ behavior was associated with the human, not the disallowed act or object, perhaps in fearful anticipation of punishment or scolding by the owner.

Domestic dogs (Canis familiaris) use visual attention cues when play signaling

Alexandra Horowitz

Domestic dogs (Canis familiaris) use visual attention cues when play signaling

Journal of Veterinary Behavior: Clinical Applications and Research, Volume 4, Issue 2, March–April 2009, Pages 53–54

ABSTRACT

Recent research has shown that dogs are able to recognize varying attentional states of humans ( [Call et al., 2003] and [Schwab and Huber, 2006]) and to use visual attention cues as an indication of a human's perceptual access (Virányi et al., 2006). The present research examines dogs’ use of visual attention cues when communicating with conspecifics in play.

Given dogs’ success at producing and interpreting communicative and attention cues of humans, a logical next step is to examine dogs’ abilities to use attention cues when communicating to conspecifics. The present research looks at 39 bouts of dyadic play in a natural setting. Sequential behaviors and head direction of both dogs were noted throughout the bouts. After coding, analysis considered the use of those behaviors identified as “play signals” (invitations to play) or “attention-getting behaviors” (acts which disrupt the partner's sensory [visual, auditory, tactile] experience) with respect to the visual attention cues of the recipient.

Of the 397 observed play signals, 376 were to a forward-facing audience (z = 5.25, n = 37 bouts, P < 0.001). When play signals were used in combination with attention getters, the actor preceded the play signal with the attention getter more often than the inverse (χ² = 9.52, df = 1, P < 0.01).

Dogs used attention getters more often when their partners were looking away than throughout the bouts (P[away] = 0.82, P[bout] = 0.34; z = 4.93, n = 33 bouts, P < 0.001). Further, dogs used more visual attention getters when their partner had their side to them (z = 1.93, n = 19 bouts, P = 0.0536), and dogs directed more tactile attention getters to dogs who were looking away or distracted (z = 2.37, n = 33 bouts, P = 0.0178). When a partner was socially distracted, tactile attention getters were used more often than visual attention getters (z = 2.22, n = 12 bouts, P = 0.0264).

Play behaviors were differentially used according to the partner's posture. Play signals were sent nearly exclusively to forward-facing conspecifics; attention-getting behaviors were used most often when a playmate was facing away, and before signaling an interest to play. In addition, the mode of attention getter matched the degree of inattentiveness of the playmate; stronger attention getters were used when a playmate was looking away or distracted, and less forceful ones when the partner was facing forward or laterally. The current evidence provides support for the claim that dogs can use behavioral indications of elements of attention in others to guide their own behavior.(Horowitz et al. 2009)

Clever hounds: social cognition in the domestic dog (Canis familiaris)

Jonathan J Cooper, Clare Ashton, Sarah Bishop, Rebecca West, Daniel S Mills, Robert J Young

Clever hounds: social cognition in the domestic dog (Canis familiaris)

Applied Animal Behaviour Science, Volume 81, Issue 3, 2 May 2003, Pages 229–244

Abstract

This paper reviews the reasons why domestic dogs make good models to investigate cognitive processes related to social living and describes experimental approaches that can be adopted to investigate such processes in dogs. Domestic dogs are suitable models for investigating social cognition skills for three broad reasons. First, dogs originated from wolves, social animals that engage in a number of co-operative behaviours, such as hunting and that may have evolved cognitive abilities that help them predict and interpret the actions of other animals. Second, during domestication dogs are likely to have been selected for mental adaptations for their roles in human society such as herding or companionship. Third, domestic dogs live in a human world and “enculturation” may facilitate the development of relevant mental skills in dogs. Studies of social cognition in animals commonly use experimental paradigms originally developed for pre-verbal human infants. Preferential gaze, for example, can be used as a measure of attention or “surprise” in studies using expectancy violation. This approach has been used to demonstrate simple numerical competence in dogs. Dogs also readily use both conspecific and human social signals (e.g. looking or pointing) as information sources to locate hidden rewards such as food or favourite toys. Such abilities make dogs particularly good models for investigating perspective-taking tasks, where animals are required to discriminate between apparently knowledgeable and apparently ignorant informants.

Wolves outperform dogs in following human social cues

Monique A.R. Udell, Nicole R. Dorey, Clive D.L. Wynne

Wolves outperform dogs in following human social cues

Animal Behaviour, Volume 76, Issue 6, December 2008, Pages 1767–1773

Abstract

Domestic dogs, Canis familiaris, have been shown capable of finding hidden food by following pointing gestures made with different parts of the human body. However, previous studies have reported that hand-reared wolves, C. lupus, fail to locate hidden food in response to similar points in the absence of extensive training. The failure of wolves to perform this task has led to the proposal that the ability to understand others' intentions is a derived character in dogs, not present in the ancestral population (wolves). Here we show that wolves, given the right rearing environment and daily interaction with humans, can use momentary distal human pointing cues to find food without training, whereas dogs tested outdoors and dogs at an animal shelter do not follow the same human points. In line with past studies, pet dogs tested indoors were successful in following these points. We also show that the reported failure of wolves in some past studies may be due to differences in the testing environment. Our findings indicate that domestication is not a prerequisite for human-like social cognition in canids, and show the need for additional research on the role of rearing conditions and environmental factors in the development of higher-level cognitive abilities.

The performance of stray dogs (Canis familiaris) living in a shelter on human-guided object-choice tasks

Monique A.R. Udell, Nicole R. Dorey, Clive D.L. Wynne

The performance of stray dogs (Canis familiaris) living in a shelter on human-guided object-choice tasks

Animal Behaviour, Volume 79, Issue 3, March 2010, Pages 717–725

Abstract
A decade of research on domestic dogs' responsiveness to human actions has led some to believe that all members of the species Canis familiaris possess a human-like social cognition not shared by their nondomesticated relatives. However, comparative studies on diverse populations of domestic dog are lacking, making species-wide generalizations premature. In this study we present the performance of one under-represented population, stray dogs living in shelters, on a human-guided object-choice task. Unlike pet dogs, shelter dogs universally failed to follow a momentary distal point to a target location in initial tests, although they were able to follow a simpler form of human point on the same task. Furthermore, the majority of subjects learned to follow a momentary distal point to a target when given additional training trials (experiment 2). Dogs' sensitivity to human gestures may not be entirely explained by phylogenetic variables; rather, the interactions between genetic, developmental and experiential variables must be considered.