Canis lupus familiaris- otherwise known as domesticated dogs- have been studied by researchers of animal cognition and behavioral psychology due to their closeness with humans and the abilities that this intimacy may build. Dogs as we know them were not always obedient and gentle in nature but were once wild animals before the processes of social domestication and selective breeding occurred 15,000 years ago (Madsen & Persson, 2013). Human and dog cohabitation began when hunter-gatherer societies noticed that wild wolves were attracted to their settlements- likely because of the carcasses that hunters would discard after every hunt. Archeologists have shown that our ancestors began to form a bond with canines and kept them as pets because experts believe they used the wolves to help them hunt. This bond with humans created the selective pressure to be tamer and more submissive, but has this closeness also allowed other humanistic traits to form in dogs?
According to Romero, Ito, Saito, & Hasegawa (2014), empathy is “, the ability to share the feelings and sensations of others,” (p. 1) and “is essential to engage in successful social interactions, coordinated activity, and cooperation toward shared goals” (p. 1). When we think of empathy we picture celebrating a friend’s achievements or sympathizing their loss- but this is more than a one-sided emotional state experienced solely by humans. It is a complex yet unconscious interaction between social animals in which they assess their partner’s expression and match it to their own their disposition. This trait exists on an ‘empathic gradient’ because it requires the development of certain cognitive capabilities such as theory-of-mind and the ability to recognize others’ emotions (Madsen, et al., 2013). The most complex form of empathy is cognitive empathy and requires advanced mental abilities such as perspective-taking and self-other differentiation. Cognitive empathy is essential to sympathetic concern, which requires animals “, to have an appraisal of the others’ situation and showing concern-like behaviors,” (Quervel-Chaumette, Faerber, Farago, Marshall-Pescini, & Range, 2016, p. 1).
Empathy in its simplest form is affective empathy, which is an involuntary motor reaction to others’ emotions that encompasses emotional contagion. In order to be considered affective empathy, two interconnected processes must exist called non-conscious mimicry, which is the involuntary mirroring of an individual’s behavior, and afferent feedback, which is the emotional influence that the behavior elicits. These two processes interact when an individual catches a person’s yawn (non-conscious mimicry) and become tired as a result (afferent feedback). Since affective empathy is an extremely old and deep-rooted instinct, it normally develops before cognitive empathy. Because empathy is a phylogenetically ancient trait, it suggests its existence in other non-human animals such as domesticated dogs (Huber, Barber, Farago, Muller, & Huber, 2017). Any dog owner would tell you that their pet can sense when they are excited or in distress, but is there sufficient empirical evidence to support the notion of empathy in canines?
Researchers of animal cognition have discovered that instances of the most basic yet fundamental form of empathy, or affective empathy, may be found in social animals’ instinctual behaviors such as rapid facial mimicry. This is defined as a quick, involuntary mirroring of another individual’s facial expression that lasts no more than a single second (Palagi, Nicotra, & Cordoni, 2015). Rapid facial mimicry is a prime example of non-conscious mimicry at work. To be considered affective empathy, however, one must assess whether the animal was emotionally affected by the behavior presented. This is difficult considering animals cannot communicate a self-evaluation, so experimenters must rely on body cues and facial expressions to determine their affective state. When studying dogs and other social animals, the positioning of their tail and head as well as the expression of their eyes, teeth, and lips are most indicative of how they feel. For instance, it is universally known that when a dog wags its tail that it is excited, but when their tail stays between their legs they are afraid. In addition to being found in roughly 77% of canines, rapid mimicry can also be seen in non-human primates. Although this phenomenon does entail the perception and duplication of a physical action, its principal function is to fortify social relationships, support the coordination of members in a group, and create an attachment between individuals and conspecifics (Palagi & Scopa, 2017). Because rapid facial mimicry is an instantaneous reflex of basic empathy, it can be likened to the experience of emotional contagion.
A suite of social-cognitive traits has been identified in domesticated dogs, and some of their instincts may even be precursors to complex human behaviors. Animals have been shown to exhibit a wide array of emotional states that range from anxiety, fear, and frustration to pleasure and excitement (Quervel-Chaumette, et al., 2016). We are interested in the concept of emotional contagion in domesticated dogs, which is defined as the unconscious and innate phenomenon of adopting the emotional state of another individual (Palagi, et al., 2015). It is a state in which a person loses all self-awareness, which offers a blank canvas for the individual to experience the affect that underlies the other person’s actions (Yong & Ruffman, 2014). There are certain criteria that need to be met in order for a behavioral response to be considered emotional contagion. First, dogs should exhibit different responses to emotional verses non-emotional stimuli because we would expect the emotional condition to elicit a response (Huber, et al., 2017). Next, the behavioral reaction to positive and negative stimuli or sounds should differ- again because the negative sound would create a stronger response.
Finally, there should be an increased rate and intensity of behavior in reaction to negative affects. These requirements for emotional contagion were created in light of recent research and reflect the various dimensions of a dog’s behavior. The precursor of empathy, or emotional contagion, that is seen in countless animals is a trait that was selected for over many generations that eventually formed into true humanistic empathy. Empathizing is essential to social interactions and coordinated activities, which both secure the welfare of groups of animals by allowing them to hunt in large numbers for their food supply and to communicate if there is an impending danger. Social animals, such as rats and pigeons, have been found to react to a stimulus with changes in facial and vocal expression- which implies a primitive form of emotional contagion (Yong, et al., 2014). Because a sense of self is not required for emotional contagion, researchers have begun to study this empathetic phenomenon in non-human animals (Quervel-Chaumette, et al., 2016). Experts are attempting to prove if canines are capable of experiencing emotional contagion and whether that implies a rudimentary level of empathy. Because dogs were artificially selected for based on their level of sociability and emotional sensitivity, it is entirely possible.
A reliably accurate method of investigating emotional contagion in humans and other social animals is through the study of contagious yawning. This is a relatively familiar concept that most people have experienced themselves. Contagious yawning is the ability to ‘catch’ another individual’s yawn when it is perceived (Madsen, et al., 2013). Researchers have discovered that dogs are able to contagiously yawn from humans and conspecifics. Because canines are so unusually apt to social interaction and display an understanding of human behavior- even more so than our closest primate ancestors- it is plausible that “, dogs’ ability to yawn contagiously evolved with the capacity for reading human communicative signals,” (Romero, et al., 2014, p. 2). Roughly half of adults will yawn contagiously after observing a person yawn- and it only takes so much as the sound or even thought of yawning to elicit a response in some. But what makes contagious yawning so infectious? The answer lies in the phylogenetic tree of life where we find that empathy and the contagious yawning associated with it are connected by the biological evolution of mammals (Romero, et al., 2014). Humans that display a higher level of advanced empathy- which can be assessed using theory-of-mind and self-recognition tests- show a significantly increased aptitude for contagious yawning (Silva, Bessa, & de Sousa, 2012). We seek to answer the question of whether this capacity is stronger when observing a familiar individual’s yawn, such as their owner or their mother, due to more empathetic feelings towards them.
Behavioral psychologists Silva, Bessa, & de Sousa (2012) have conducted research that demonstrations the “effect of a social variable (familiarity/unfamiliarity) on the level of contagion, with dogs yawning more at familiar yawns than at unfamiliar ones” (p. 723). This phenomenon of social intimacy effecting the contagiousness of yawns has also been revealed in wolves, which shows a connection between carnivores and the stepping stones to empathy. Despite this evidence, some researchers have not found the same results. Some experimenters suggest that this yawning is due to a fixed motor sequence that does not require advanced cognitively abilities like empathy- which is known as the ‘contagion-only’ hypothesis (Silva, et al., 2012). The issue with these experiments is that they did not allow ample time for the dog to receive the signal and did not control for the dog’s attention to ensure that they are not distracted by their owner- which are two factors that our study will account for. We predict that domesticated dogs will have a higher frequency of contagious yawning with a person that is familiar to them, depending on their level of familiarity. We hypothesize that dogs will have a greater likelihood to contagiously yawn if they observe a person they are familiar with, suggesting that emotional empathy is the underlying mechanism.
Methods
Participants in this study included ninety-five domestic dogs with various breeds and a range of ages between four and fourteen months old. There were forty-five male dogs and fifty female dogs with a total mean age of eight months. To ensure the safety of the dogs, the owners, and the experimenters, the dogs must be healthy and up-to-date on all vaccinations. Dogs that have a previous history of food guarding or aggression of any kind will not be permitted to participate in order to guarantee that there are no acts of hostility during the experiment. This protects the researchers and participants as well as reduces the instance of tensions yawns, which are induced by a stressful situation and could be confused with natural yawns. Subjects that are too distracted or owners that cannot produce a convincing yawn are excluded from this study as well. All participants have been reported to be comfortable around strangers and have no previous experience with an experimental setting.
All dogs had resided with their current owners for at least six months to ensure that they are familiar with each other and have an emotional connection. The participants were recruited with flyers and posters placed in various puppy schools around Greenville, NC. Two dogs were excluded from the analysis because of being distracted or over-excited. This study will take place in the owner’s home- and specifically a room that they are most familiar with- to ensure that the dog is comfortable and engaged. Owners will sit behind their dogs and will have them on a leash to ensure the safety of everyone involved as well as to assure that they do not walk away from the testing site. The owners are also instructed not to interact with their dog in any way, but if the dog approaches them they are instructed to acknowledge their dog with touch and encourage them to participate with the experimenters. To prevent the owner from hearing each stimulus, owners were given headphones that played white noise. This is meant to prevent the owners from inadvertently signaling to their dog, which could alter the study’s results. We gained informed consent from the owner for every participant in the study.
The participants were exposed to four experimental stimuli that include familiar yawns, unfamiliar yawns, familiar control sounds, and unfamiliar control sounds. Control stimuli include familiar and unfamiliar control sounds and yawn stimuli included familiar and unfamiliar yawns. Unfamiliar yawns and unfamiliar sounds will be recorded by one researcher, while familiar yawns and familiar sounds will be recorded by the dog’s owner. Yawns were recorded at a 50 kHz sampling rate and had a mean duration of 5 seconds. A sequence of authentic pre-recorded yawns was administered through headphones to the researcher and the owner when recording their yawns to ensure that they produce a natural yawn. Control sounds consisted of natural yawns that are played in reverse that was created with Garage Band software. The dogs will be exposed to sequences of each condition that include 5 repetitions of the same sounds followed by 5 seconds of silence. All sequences had the same amplitude and frequency and were played at a constant volume of 6 decibels.
This experimental procedure uses a repeated measures design and is based on previous studies of contagious yawning. The dogs were given a 10-minute period of habituation of the experimenters and the study’s environment. After the dog has become familiar with the setting, the dog was situated in front of two speakers. One researcher was positioned behind the laptop in order to record the number of yawns observed during the audio sequences. They also documented the intensity of their reaction by recording the amount of attention given to each stimulus through looking-time and head cocking time. To verify that the yawns are logged correctly, every experimental session was video recorded. The four experimental sounds were exposed to each dog in a randomized order. There were two experimental sessions of sequences- the second of which occurs 7 days after the first. Every dog was given the same unfamiliar sounds. Two sequences of sounds were given with a 3-minute rest period to prevent dogs from becoming fatigued or disinterested. It is important to take preventative measures for stress and exhaustion by keeping an environment free from anxiety and giving the dogs sufficient rest breaks. A dog that is tired or unfocused could produce ‘tension yawns’ and could skew the results of the study. Some dogs become anxious if experiments are conducted in complete silence because it is an unusual situation for them and could imply danger. To prevent the dogs from becoming uneasy, the researcher filled the silence with conversation directed at the dog and their owner.
Statistical analyses of the data such as a Chi-square test and Pearson’s r tests were achieved using the computing software STATISTACA 7.0. The data were analyzed both at the group level and the individual level. A significance level of 0.05 was also used. To compare the within-participant differences of time head cocking and time looking, we used the Wilcoxon matched pairs test to examine whether the effect of attention could be an extraneous variable causing the results.
