Owner Perceptions of Dog Emotions

PAGE 1

OWNER PERCEPTIONS OF DOG EMOTIONS BY KIRSTIN OHLSEN A Thesis Submitted to the Division of Psychology New College of Florida in partial fulfillment of the requirements for the degree Bachelor of Arts Under th e sponsorship of Dr. Gord on Bauer Sarasota, Florida April 2012

PAGE 2

! ii Dedication I dedicate this amalgamation of my blood, sweat, and tears to my little brother, Jacob Ohlsen. May he remember this when he gets famous.

PAGE 3

! iii Acknowledgments I would like to start by thanking the 3 B's of my thesis committee: Dr. Bauer, Dr. Barton, and Dr. Beulig. All three of you have played incredibly important roles in my academic career here at New College and I cannot thank you enough. I would like to extend a special thanks to Dr. Bauer, who was both my thesis sponsor and academic advisor. I will always cherish the support, guidance, and sports conversations. I would also like to thank my family and friends who helped me along this journey. I couldn't have done it without you. Thank you to my father Michael Ohlsen for always being there for me and staying up really late to watch terrible movies. To my stepmom Jennifer and the shining beacon of hope that is my little brother Jacob, thank you for helping me keep my sanity. And to my mom Anita and my stepfather Frank, thank you for all of your support. I should also mention my four best four legged friends: Toby, Maggie, Maximus, and Loki. They are the best dogs I could ask for and the inspiration for this study. Also a special than k you goes out to all of the humans and dogs that participated in this study. This really was a fun and rewarding experience. I'd now like to thank myself. That is to say, my other half: Mary Dunkelberger. Her friendship has lifted me to new heights on w ings of tender rectitude. My rock, my muse, my sparkle, my Mary. Thank you. And last but not least, thank you Chad Lowe.

PAGE 4

! iv Table of Contents Dedication..ii Acknowledgements.......iii Table of Contents ......iv List of Tables..v List of Figuresvi Abstract.vii Introduction1 Natural History of Dogs .2 Recognition o f Communicative Signals .6 Recognition of Human Attention ..10 Recognition of Emotion.....11 Animal & Human Emotion ....14 Human Guilt..19 Anthropomorphism. ...23 Current Study.25 Method...26 Results....29 Discussion..37 References..46 Appendix A....5 7 Appendix B....58 Appendix C....60

PAGE 5

! v List of Tables Table 1.31 Table 2.32 Table 3.33 Table 4.34

PAGE 6

! vi List of Figures Figure 153 Figure 254 Figure 355 Figure 456

PAGE 7

! vii OWNER PERCEPTIONS OF DOG EMOTIONS Kirstin Ohlsen New College of Florida, 2012 ABSTRACT This study investigated owner perceptions of dog emotions and its relations to anthropomorphic tendencies. Participants answered questionnaires about particular emotions, anthropomorphic attitudes, and beliefs about an i nteraction that could possibly involve guilt. Some also participated in a filmed interaction where the dog was left alone with a forbidden treat. There was a positive correlation between anthropomorphic attitudes and beliefs about dog emotions. There were also gender differences. Women were more likely than men to both anthropomorphize their dogs and perceive more complex emotions. The possibility of complex emotions such as guilt in dogs was discussed. Understanding the motivations and emotions of dogs cou ld help to strengthen the dog human relationship. _________________________ Dr. Gordon Bauer Division of Psychology

PAGE 8

! 1 Owner Perceptions of Dog Emotions Of all creatures the one nearest to man in the neness of its perceptions and in its capacity to render true friendship is a bitch Konrad Lorenz Humans have a long and unique relationship with dogs. Often regarded as man's best friend, dogs have occupied many roles in the lives of their human compatriots They have been used as tools to aid in hunting, herding, guiding, and even sniffing out dangerous materials (Topal, Miklosi, & Csanyi, 1997). Some breeds have even been trained to pull heavy loads through rough terrain where alternate forms of transporta tion are too costly and risky. They have also been used for entertainment purposes. Who hasn't heard of Lassie saving her beloved Timmy from the treacherous well? Dogs also bring comfort and joy in the form of companionship. They can provide a warm body to cuddle and a kind ear. I suspect that there has never been a dog that has looked at its owner and said "shut up, you're problems are boring me." In fact, whether or not they can even think this is a matter currently up for debate. But in any case, a well treated dog will almost always greet its owner with a sparkle in its eye and a wag of its tail as though it had never been happier. But does this seemingly joyful action show that the dog is feeling the emotion of joy upon the return of the human? It is hard for many owners to avoid attributing "human emotions" to their canine friends. But are these emotions really only reserved for humans? This study seeks to investigate owners' perceptions of their dogs' emotions, especially in an instance that could in volve the more complex emotion of guilt. Fully

PAGE 9

! 2 understanding dogs and their possible emotions can lead to better overall treatment of dogs and can increase our knowledge of the unique relationship between dogs and humans. Natural History of Dogs A folk t ale from the Kato Indians of California tells of the creation of the Earth. They speak of Nagaicho, the creator, and how he erected four pillars at the corners of the world in order to hold up the sky and expose the ground beneath. He created man and woman from the dirt, and placed animals in their proper spots. When he saw that they needed water, he dragged his feet and from his footsteps rivers sprang forth. He had created everything in the world, except for one thing, the dog. When he set out to create t he world, Nagaicho already had his dog walking beside him. The dog had always been there. Of course the scientific community is not quick to believe in folklore, but it would certainly be easier than the scientific explanation of the domestication of dog s. Although there is some controversy, the most common idea is that dogs evolved from wolves. Current morphological, behavioral, and genetic evidence supports this idea (Vila et al., 1997). Savolainen, Zhang, Luo, Lundeberg, and Leitner (2002) set out to pinpoint where domestication first occurred. They analyzed the genetic variation of mitochondrial DNA from a variety of domestic dogs from around the world. They found that over 95% of all of the sequences belonged to three phylogenic groups, which were di stributed throughout Eurasia with the largest genetic variation in East Asia. They also found that there was no clear division of the main

PAGE 10

! 3 morphological differences between breeds into the three groups suggesting that the differences are not necessarily fr om geographically distinct domestications. In a follow up to this study, entire mitochondrial genomes were analyzed in order to obtain evidence of a single origin in time and space (Pang et al., 2009). They found that dogs universally share a common gene pool that consists of 10 major genetic groups. They also found that the only place that contained all 10 of these groups was in South East Asia, specifically south of the Yangtze River, and that diversity decreased in the other surrounding areas (only 5 gr oups were seen in Central and North China and only 4 groups in Europe). The study went a step further by dating the genetic distance to the universal most recent common ancestor. They determined that the domestic dog likely originated about 5,400 16,000 ye ars ago from several hundred wolves in southern China. Further evidence suggests that dogs accompanied humans as they crossed the Bering Strait 10,000 to 15,000 years ago (Leonard et al., 2002). Dogs were the only domesticated species that was distribut ed across Eurasia and the Americas before transoceanic travel was developed. This would suggest that they either crossed the Bering Strait with humans or they were independently domesticated. Mitochondrial DNA sequences show that many of the native America n dogs have lineages from the Old World but that a few have unique lineages to the New World. The results support the idea that American and Eurasian dogs share a common Old World ancestor suggesting that domesticated dogs crossed the Bering Strait. Examin ations of the genetic make up of the Mexican hairless dog also support this claim (Vila et al., 1999). This breed of dog was considered to be

PAGE 11

! 4 one of the oldest breeds in America, although genetic analyses show that it shared no relation to the American gra y wolf and instead was more closely related to modern European breeds. Despite all of the uncertainty and controversy, what is known is that as far as domesticated animals go, dogs seem to be associated with humans the longest (Benecke, 1987; Savolainen e t al., 2002). Bones of wolves found in relation to hominids date back as far as 400,000 years ago, but one of the earliest definite records of domesticated dogs comes from a jaw fragment found in Central Europe (Benecke, 1987). The fragment was uncovered i n a grave of a man and a woman and dates back 14,000 years ago to the Magdalenian period. So somewhere in between 400,000 years ago and 14,000 years ago, wolves, and possibly other canid species, became domesticated. The process of domestication likely in volved both natural and artificial selection (Udell, Dory, & Wynne, 2010). As wolves or other canids adapted, individuals that were more tolerant of humans were more successful at exploiting human food sources, which in turn led to changes in the morpholo gy and behavior (Coppinger & Coppinger, 2001). These behaviors could include reduced fear and aggression when in the presence of humans. As individuals became more adapted to the human population, artificial selection and human involvement may have occurre d leading to animals with explicitly desired traits. A long term experiment with grey foxes showed that artificial selection for behavior could lead to changes in physical appearance (Trut, Plyusnina, & Oksina, 2004). With each litter of fox, the least ag gressive and fearful were

PAGE 12

! 5 removed and then bred together. At the same time, the more aggressive and fearful were removed and bred together and this process repeated itself for many generations. During selection for these behaviors, the tamer foxes began to show changes in their appearance that were observed during some of the early history of the dog. These included specifically localized depigmentation and yellow brown mottling in the coat of many of the foxes as well as altered skull shapes. The domestica ted foxes had more shortened and widened faces than their wild counterparts. The results successfully show that artificial selection could lead to changes in appearance that were likely involved in the domestication of the modern dog. There are also sugg estions that the evolution of dogs can be described as a coevolution of cooperation between humans and canids (Schleidt & Shalter, 2003). This theory proposes that any initial contact between early humans and canids was mutual and that any changes in morph ology and behavior can be explained through coevolution and not necessarily as a deliberate attempt to domesticate the animals. Schleidt and Shalter (2003) suggest that there may have even been a hypothetical "lupification" of human behavior and that human s may have adopted some wolf like behaviors. Wolves are a very social species that show a lot of cooperation between its members. They live in communal packs where all members share food and affection. The pack usually only has one pair breeding, but food and care are shared generously between members. Individuals that exert more effort and risk in taking down prey receive the "lion's share," but the rest of the food will be divided and shared throughout the pack. Some of these

PAGE 13

! 6 characteristics can also be s een in human life. This begs the question "is humaneness canid?" (Schleidt, 1998). The evolution of dogs is likely described by both coevolution and artificial selection. As wolves became closer to humans they may have helped to shape some human behaviors Humans were then able to breed wolves and select for certain traits until they were able to act as humans' alarm systems, hunting aides, and babysitters and eventually resembled modern day dogs. Currently there seems to be a shift between humans' reasons for owning dogs from a tool to more of a companion (Staats, Wallace, & Anderson, 2008). It seems as though during the evolution of the human dog relationship, "humans domesticated dogs, and dogs domesticated humans" (Groves, 1999). Recognition of Comm unicative Signals Most people who have taken a class in animal behavior have heard the story of Clever Hans. Clever Hans was a horse that had a real knack for solving math problems. His owner would ask a question and Hans would stomp out the correct answer each time. Hans had many believers, but there were also many skeptics. In the early 1900s, the psychologist Oskar Pfungst put to rest any dreams of a horse mathematician. He found that Hans was picking up on subtle clues from his human observers and he us ed those cues to arrive at the correct answer (Pfungst, 1911). So maybe Hans was not a math genius, but perhaps recognizing and utilizing human cues made him just as "Clever." Dogs are also very sensitive to human actions and behaviors. This is perhaps be cause of the possible coevolution between dogs and humans, but in any

PAGE 14

! 7 case dogs have adapted to several aspects of human behavior. Miklosi, Polgardi, Topal, and Csanyi (1998) were one of the first to investigate whether dogs are able to use human gestures as cues. The experimenters used two bowls to hide bait. Both bowls had a cloth that had been previously soaked in food in order to eliminate any possible odor cues. The experimenter sat about 0.5 meters away from the bowls. He/she then made eye contact wit h the dog and gestured toward the correct bowl using a predetermined gesture. All gestures lasted 1 2 seconds and consisted of pointing, bowing, nodding, head turning, and glancing all in the direction of the correct bowl. Two types of dogs were used: dogs that were training to be assistant dogs and were kept together in a large garden and dogs that had lived a long time in human families. They found that all of the dogs were familiar with the pointing gesture, but that training dogs initially had some diff iculty with nodding. There were also some individual differences in the success of bowing as a communicative gesture. The dogs that were raised in families were very successful at pointing, bowing, and nodding, perhaps because they were familiar gestures f rom their family homes. Head turning was not as initially successful, but the dogs were eventually able to use it. The results from this study show that dogs are successful at using human gestures and can even learn new ones. Dogs can use many of the gestu res that were presented and utilized by apes and humans (Itakura & Tanaka, 1998). It is possible that dogs only have this ability because they have more actual experience with humans. The human exposure hypothesis suggests that dogs only acquire the skil ls of utilizing human social cues from this experience.

PAGE 15

! 8 This predicts that variations in task performance will be associated with the individual dogs' experiences with humans, suggesting that young dogs should have relatively poor skills and that other soc ialized animals should also be equally successful. However, it has been found that domestic puppies as young as a few weeks old are successful at reading human communicative signals to locate hidden food and that horses and wolves are far less competent ev en when they have been socialized by humans (Hare, Brown, Williamson, & Tomasello, 2002; McKinley & Sambrook, 2000; Miklosi et al., 2003). Hare et al. (2002) performed the choice task with puppies of all ages and with various amounts of exposure to humans The puppies varied in age from 9 26 weeks with some living with families and some being raised in kennels with very little human contact. The puppies were tested with two cues: gaze, where the experimenter turned his head and stared at the correct answer or gaze and point, where the experimenter also pointed to the correct answer. It was found that the puppies performed above chance at both cues and that there was no effect of age. The results of this study suggest that individual experience with humans i s not necessary for dogs to learn social cues because young puppies with very little human contact are able to utilize these cues. Dogs are also more successful at utilizing communicative tools than horses (another domesticated animal) and wolves (possibl y the dogs' closest relatives). McKinley and Sambrook (2000) tested both dogs and horses and found that dogs were much more successful. The dogs were especially skilled at using the pointing gesture, even when the distance between the owner's hand and the

PAGE 16

! 9 correct answer was increased. This possibly shows that the dog understood that it was the direction of the hand that was communicating the correct answer and not just the proximity of the owner to the target. The trials from the horse were mixed, with only half of the horses finding the food if the owner touched the target, and only one was able to use the pointing cue. Miklosi et al. (2003) found that socialized wolves could find the location of hidden food with the touch of a hand and in even the pointin g gestures in some cases, but they were far less successful than dogs. All of the wolves could use the touching cue, but only two were able to use the pointing cue. The results show that wolves need human socialization in order to utilize communicative ski lls but that there are still many individual differences. The researchers also compared dogs' and wolves' tendencies to look for cues when a task was insoluble. Both species were given the opportunity to learn and master a problem situation (either rope pu lling or bin opening). Afterwards the animals were presented with the same problem but this time it was impossible to solve. They found that dogs were quicker to look back at humans and to hold their gaze longer than the socialized wolves. A total of seven out of the nine dogs looked back at the human while only two of the seven wolves did. It is possible that the dogs were using eye contact as a communicative gesture in order to get the human's attention or perhaps even to get cues about how to solve the p roblem. It appears as though dogs may have a genetic predisposition for preferential looking at humans. Since dogs know to make eye contact with a human, it is possible that this shows that they recognize attention.

PAGE 17

! 10 Recognition of Human Attention Not onl y can dogs follow human social cues such as pointing, but they can also use behavioral/facial cues for the detection of human attention. In an experiment by Call, Brauer, Kaminski, and Tomasello (2003) dogs were forbidden to take a piece of visible food wh ile the human either looked at the dog during the entire trial, left the room, turned his/her back, engaged in a distracting activity, or had her eyes closed. When the human was looking at them, dogs did not retrieve as much food and approached it in more indirect ways. The results suggest that the dogs were able to recognize human attentional cues, and were more likely to commit a transgression (eating forbidden food) if the owner was not looking. Gacsi, Miklosi, Varga, Topal, and Csanyi (2004) looked at the effect of body orientation and eye visibility on fetching actions. A group of 17 dogs played fetch with their owner. During this game, the owner was in one of four situations: facing the dog without a blindfold, facing away from the dog without a blind fold, facing towards the dog while blindfolded, or facing away from the dog while blindfolded. The game was also played with owners sitting in chairs or on the floor. They found that dogs performed slightly worse if the blindfold covered the person's eyes. It was also found that the dogs were more hesitant in bringing the object if the owner was facing back. In another experiment they looked at the effect of body orientation and eye visibility on begging actions. In this experiment, dogs had to choose betwe en two eating humans based on the orientation of their faces and eye visibility. It was found that dogs begged more

PAGE 18

! 11 from the attentive person when they were sitting at a table and there was a difference in the direction of their faces. Results from the stu dies show that dogs can use both behavioral and facial cues in the detection of human attention. Viranyi, Topal, Gacsi, Miklosi, and Csanyi (2004) also demonstrated that dogs could recognize human visual attention. A recording of the owner's verbal comman ds (e.g., Down!) was played for the dog while the owner was either facing the dog, facing another human, or was visually separated from the dog. They found that dogs were much more responsive to the command when the attention of the owner was directed towa rds the dog. Afterwards the same dogs were given the opportunity to beg for food from two unfamiliar humans with varied visual attention (facing the dog or turning away). The dogs showed a preference for the attentive person. The findings suggest that dogs can gather information about humans from facial and body cues, which could have implications on the human dog relationship. Recognition of Emotion If dogs are able to recognize owner attention from their facial cues, then it is also possible that they ca n gather emotional information from facial cues. Recognizing other's emotional states is extremely important for social animals. Knowing when a conspecific is angry can be very useful and may even allow for an animal to escape aggression. Animals show a lo t of their emotion in their body language, but when it comes to humans, a lot of the emotional information can be gathered just by looking at the human's face. Human emotion can be categorized into six archetypal categories: happiness, sadness, fear, disgu st, surprise, and

PAGE 19

! 12 anger are constant across cultures (Ekman & Friesen, 1971). All of these emotions have distinct facial expressions that are easy for humans to recognize and categorize. Dogs can also discriminate among some of these human facial expressio ns (Nagasawa, Murai, Mogi, & Kikusui, 2011). Dogs were shown photographs of their owners and strangers in a two choice discrimination task. They were able to select between photos of people that were smiling or were showing no emotion (blank face). They fo und that dogs selected the photos of the people smiling more than photos of a blank face. When it comes to looking at faces, humans have a left gaze bias (Sackeim, Gur, & Saucy, 1978). This means that people have a tendency to spend more time observing th e right side of the person facing them. This bias only occurs while looking at human faces, which suggests that the right side of the human face may contain important emotional information. Experimental evidence shows that the two sides of the human face are indeed asymmetrical. When photos of the two sides of a human face are split down the middle and then mirrored, the resulting pictures are shockingly different. The right side of the face seems to convey more emotional information. People are able to re cognize the intended emotion quicker if they are looking at a right side composite as opposed to a left side composite. When scanning other people's faces, humans have a natural tendency to look towards the left. It has been shown that this left gaze bia s is not unique to adult humans (Guo, Meints, Hall, Hall, & Mills, 2009). The experimenters looked at the left gaze bias in adult and infant humans towards different stimuli. They found that

PAGE 20

! 13 the 6 month old human infants had greater tendencies for left gaz e biases towards faces of varying species and orientations while adults only showed a bias towards upright human faces. This shows that humans are born with the bias in a much broader sense but become more attuned to human faces as they develop. Rhesus mon keys also had biases towards upright human and monkey faces, but showed no side preference towards inverted faces (Guo et al., 2009). The experimenters also investigated the left gaze bias of dogs. A group of 17 dogs looked at varying images that included upright and inverted faces of human, monkeys, and dogs. Eye movements of the dogs were recorded and later coded. They found that the dogs demonstrated a left gaze bias only towards human faces and not towards monkey or dog faces. Results suggest that the l eft gaze bias is not unique to humans. The fact that dogs have a left gaze bias towards human faces could suggest that they may be able to gather information from human faces. Humans can also gather emotional information from dogs. Pongracz, Molnar, Miklo si, and Csanyi (2005) found that humans are able to categorize dog barks and can associate them with emotional ratings. Dog bark recordings from a Mudi, a breed of Hungarian sheepdog, were played to three groups of people: Mudi owners, other dog owners, an d non owners. Dog barks were collected from six different situations. They found that the groups categorized the situations similarly and well above chance (with chance being at 16.7%): Mudi owners were at 40.75% and both other dog owners and non owners we re at 39.35%. The results showed that humans are capable of categorizing dog barks by sound alone.

PAGE 21

! 14 The fact that humans are successful at categorizing the dog barks shows that dogs have context specific barks. Dogs are also able to discriminate barks of c onspecifics from different situations suggesting that dog barks not only convey information to humans but to conspecifics as well (Maros et al., 2008). In a habituation dishabituation paradigm, dogs listened to barks of conspecifics from two situations: wh en a stranger entered the property where the dog lived and when the dog was alone and tethered to a tree. The dogs were equipped with heart rate monitors in order to measure habituation and mechanical noises were used for controls. It was found that the he art rates of the dogs decreased across playbacks of the same bark situation showing successful habituation and that upon hearing the new situation, heart rates increased. Results show that dogs also have the capacity to discriminate between barks from diff erent situations. Both humans and dogs can gather and utilize emotional content from dog barks. Animal & Human Emotion Currently there is a lot of debate concerning the nature of animal emotion. Some researchers view animals as machines (e.g. Descartes, 1646 1649/1976) and believe that they become conditioned to stimuli and respond automatically while others believe that animals are capable of having and recognizing emotions (Bekoff, 2000; Masson & McCarthy, 1995). The study of emotion has hit many snags in large part because it is difficult to acquire empirical evidence about subjective states. Early behaviorists, such as John B. Watson and B.F. Skinner, proposed that only observable behavior constitutes valid scientific data and

PAGE 22

! 15 therefore considered emot ions and mental states to be unscientific. Researchers from other fields including ethology and psychology have challenged this idea and believe that animal emotions can and should be empirically studied. Unfortunately, there is still no way to tell how an d if animals can feel and reflect on their emotional states. In his pioneering work, Darwin (1872/1979) argued that the expressions of emotion are evolved and adapted to serve an important communicative function. He called these expressions "the language of emotions" and believed that they were outward manifestations of an inner state. Since he thought that expressions were hereditary and evolved, he also drew parallels to animal emotion and suggested that emotional expressions in animals may be antecedent s to human expression. He proposed three principles: the principle of serviceable habits, the principle of antithesis, and the principle of direct action of the excited nervous system on the body. The principle of serviceable habits states that offspring will inherit useful expressive habits that were acquired during experience. Darwin described opening one's eyes in surprise as a useful habit because it increases the field of vision and allows for the eyes to dart about quickly. The principle of antithesi s states that some actions are performed merely because they are opposite of a serviceable habit. The principle of direct action of the excited nervous system on the body states that some reactions are physiological reactions and thus cannot be controlled by will. An example is trembling out of fear. This action seemingly serves no use and actually may hinder the animal's chance to escape, but it is

PAGE 23

! 16 uncontrollable because it is a physiological reaction. Darwin also posited that emotional differences among a nimals were in degree and not in kind, and that similar emotions could be experienced by many species. Many researchers think that it is possible to discriminate the different emotional states of animals. Changes in body posture, muscle tone, and vocaliza tions are just some of the different emotional responses (Bekoff, 2000). Many of the emotional expressions of animals can be described as primary emotions. Primary emotions such as anger and fear are considered to be inborn and somewhat automatic. They are wired to the so called "emotional part of the brain," or the limbic system, especially the amygdala (MacLean, 1970). Many species share similar structures in the limbic system so it is possible that they also share similar emotional reactions to events. I t is pretty widely accepted that at least some of the primary emotions (e.g. fear and anger) are evident in a variety of species (Ekman, 1992; Izard, 1992). Secondary emotions are ones that can be experienced and reflected on. These emotions require highe r brain functions and seem to be reserved for humans and some of the great apes. They include emotions such as jealousy, guilt, and shame. Some of the secondary emotions have been referred to as "self conscious emotions" and thus require the ability to dis tinguish oneself as a distinct entity (Tangney et al., 1996). If this is truly the case, then animals that are not self aware should not have the capacity for secondary emotions. However this idea is hard for some animal enthusiasts to completely accept. A fter spending so much time around animals and pets, even ones that the scientific community does

PAGE 24

! 17 not accept as having self consciousness, it is hard not to notice certain behaviors that may suggest the capacity for secondary emotion. Anecdotal evidence fr om pet owners suggests that secondary emotions may exist in non primates (Morris, Doe, & Godsell, 2008). The researchers surveyed 907 animal owners about emotions that they had observed in their pets. They found that owners reported primary emotions more o ften than secondary emotions but that their animals exhibited jealousy, a secondary emotion, at high levels (especially in horses and dogs). The researchers then followed up and interviewed 40 dog owners about the contexts and behaviors associated with the ir perceptions of their dogs' jealousy. All of the owners were able to describe specific instances where their pet appeared to be jealous. Descriptions of jealous dogs were consistent across the participants showing that owners have a common idea of what i t means for a dog to exhibit jealousy. Jealousy requires an evaluation of one's own situation as well as a comparison to another individual's. Since a dog cannot describe how it is feeling, it is incredibly hard to demonstrate if it is actually feeling jea lous. What can be studied are the reactions of dogs in situations that may induce jealousy. While these studies will not provide conclusive evidence of secondary emotions, they can potentially identify precursors to higher thought processes. Range, Horn, V iranyi and Huber (2008) found an inequity aversion in dogs. When in the presence of a partner that was consistently given a more attractive reward for the same effort, the dog would decrease its participation. The dogs that were in the presence of the rewa rded partner refused earlier, hesitated to

PAGE 25

! 18 obey commands, and showed more stress behaviors than dogs in a condition where there was no reward and no partner. The results suggest that dogs may be sensitive to unequal reward distribution. Again, this is not conclusive evidence of secondary emotions. The reaction of the dog cannot be explained as jealousy, but it could be an indication of a precursor to the understanding of equality and fairness. Many owners express the idea that their dogs exhibit a look of guilt when they have disobeyed because the dog understands that it has done something wrong (Horowitz, 2009). To look at this Horowitz (2009) recorded the behavior of 14 dogs. Dogs were le d into a room where the owners instructed them not to eat a trea t. Afterwards the owners left the room. Upon returning, they were informed that the dog either obeyed or disobeyed and were instructed to respond accordingly. Some of the owners were misled and were told that their dogs disobeyed when they had not. It wa s found that b oth "guilty" and "non guilty" dogs were equally likely to exhibit behaviors associated with guilt. But there was a n effect of owner response. Dogs exhibited more behaviors associated with guilt if the owner scolded them. It was also found tha t more behaviors were exhibited when the owner scolded a dog that did not eat the treat. The results show that the "guilty look" can appear even when no transgression has occurred suggesting that dogs do not necessarily possess an understanding of disobedi ence and are instead merely reacting to the owner.

PAGE 26

! 19 Human Guilt Guilt is a relatively common form of emotional distress in everyday human life and possibly plays a critical role in self regulation. Lewis (1971) proposed that with guilt, the primary con cern of the afflicted person lies with the transgression and not with the self, as is the case with shame. Therefore, guilt is generally a less painful experience and involves a feeling of remorse or tension over a bad act. Research suggests that guilt is best described as an interpersonal phenomenon proposing that the most common cause of guilt is the infliction of harm, loss, or distress to a partner (Baumeister, Stillwell, & Heatherton, 1994). It would also predict that the person would show stronger gui lty tendencies in close relationships than distant ones. Also, it should be noted that even though guilt is seemingly a social emotion, guilty feelings can occur when alone (Tangney et al., 1996). In their study, the researchers investigated the differenc es between the emotions of guilt, shame, and embarrassment and whether they required an evaluative audience or could be experienced while alone. A total of 182 participants were asked to fill out questionnaires as well as write detailed accounts of when th ey felt particular emotions. They found that around 10% of the participants described guilty experiences when they were alone and were not in the presence of an evaluative party. Even though guilt can be felt when alone, it is still seemingly predominant i n social contexts. The interpersonal approach also suggests that guilt serves three functions (Baumeister et al., 1994). First, it motivates relationship enhancing behaviors,

PAGE 27

! 20 perhaps by serving as a punishment for communal transgressions. This ever looming idea of punishment can keep a person's behaviors in line with the moral standards set by the community. Guilt could also be used for manipulative purposes. A person in a relationship could try to induce guilty feeling into the partner in order to get the partner to act a certain way. The third function is to redistribute emotional distress between the parties. If a transgression is committed, the victim may feel better if the transgressor shows guilt because it would suggest that the transgressor was exper iencing diminished enjoyment. This could show the victim that the transgressor cares about the relationship and/or may serve as an implication that the transgression will likely not be committed again. Amodio, Devine, and Harmon Jones (2007) investigated guilt as a dynamic model. They proposed that at first guilt provides the person with a negative reinforcement cue. Upon receiving this cue, the person would experience an initial reduction in approach motivation. However, the person would experience an inc rease in approach motivation if presented with an opportunity to make amends for the transgression. To see this phenomenon, 47 white American females viewed a series of multiracial faces while they were fitted for EEG readings. All participants filled out a baseline state affect measure before the experiment. After receiving fake feedback from the experimenters suggesting moderately negative responses to black faces, the subjects took the state affect test again. Participants reported elevated levels of gui lt, which were associated with changes in frontal cortical asymmetry. This change indicated a reduction in

PAGE 28

! 21 approach motivation. The participants were then given an opportunity to engage in prejudice reducing behavior. It was found that guilt was a predicto r for greater interest in prejudice reduction and that this was associated with an approach related shift in frontal asymmetry. The results suggest that the expression of guilt may consist of multiple stages. The feelings of guilt can affect motivational behavior even after a week (Ketelaar & Tung Au, 2003). Participants played two social bargaining games that included the prisoner's dilemma and an ultimatum game. In the prisoner's dilemma, participants had to choose between cooperating, and thus upholdin g social norms of fairness, or defecting, thus violating social norms. Guilt was assessed through manipulation and self report. They found that participants who experienced guilty feelings from defecting in early stages of the game were more likely to coop erate in subsequent rounds. They also found that this behavior lasted even after a week. Using Amodio's (2007) model, it is possible that participants with guilt were more motivated to make amends for their transgressions and that this motivation can last over a week. One issue surrounding theories of guilt is whether or not young children are capable of experiencing guilt. Under his definition of guilt, Freud (1930/1961) suggested that guilt was a result of impulses from the id that conflict with the supe rego. He claimed that children therefore could not experience guilt until the superego was established. Another theory suggested that guilt requires children to know that they are in control of their behaviors (Kagan, 1984). The generally accepted sequence of emotional development in humans is that the primary

PAGE 29

! 22 emotions emerge within the first year of life whereas the secondary emotions do not emerge until the second year or later (Draghi Lorenz, Reddy, & Costall, 2001). However it should be noted that there is some suggestion of early forms of secondary emotions that may not necessarily be delayed until the second year of life. But even if young children can experience guilt, it is still not at the same level of adult guilt. It seems as though a certain level of cognitive ability is needed in order to self evaluate and adhere to moral guidelines. Thompson and Hoffman (1980) had children from first, third, and fifth grade listen to stories describing acts of wrongdoing. The children were interviewed afterw ards and asked questions about how they would feel if they were the wrongdoers and why. They found that the younger children were more likely to describe concern over detection and punishment whereas older children talked more about internal principles of right and wrong. The results from this study suggest that younger children have a more simplified experience of guilt that may be more of a fear of punishment than true guilt. Williams and Bybee (1994) investigated descriptions of guilt from 5 th 8 th and 11 th graders. Most of the children from all groups described guilt over a transgression. Descriptions of guilt being caused by inaction, neglect of responsibilities, and failure to obtain ideals more than doubles from the 5 th grade to the 11 th grade. This marks a shift from guilty feelings over externalizing to internalizing events. They also found that females were more likely to feel guilty over violating norms of compassion and trust, whereas males were more

PAGE 30

! 23 preoccupied with external events. It seems as though the experience of guilt develops and becomes more complicated and internalized with age in humans. Anthropomorphism Anthropomorphism refers to the tendency to use human characteristics to describe non human animals. The problem with anthropomorph ism is that the descriptions cannot be determined empirically. For example, when a pet tilts its head to a side this may look like puzzlement, but there is no way to truly know how the animal is feeling inside. Since there is a void of this information, ma ny pet owners tend to fill it in with known human actions and feelings. It has also been suggested that there may be an evolutionary component to anthropomorphizing. It is possible that the phenomenon developed once humans possessed the capacity for "refl exive consciousness," or the ability to use knowledge of what it is like to be a person to understand and anticipate other's behaviors (Humphrey, 1983). This ability may have helped early hunters to predict the behavior of prey and thus create better hunti ng strategies (Horowitz & Bekoff, 2007). It may also have played a role in the evolution of the human dog relationship. Humans' drives for affection may be the cause for the shift from dogs as tools to companions. Anthropomorphizing may allow the human to think of an animal companion's behaviors in human terms, which may make it easier for the pet to provide beneficial social support for the human (Serpell, 2003). The social role of some dogs is so important to their owners that the death of the dog can be taken as seriously as the death of a human family member (Keddie, 1977). It is somewhat common for dogs to be regarded as part of the family, with a

PAGE 31

! 24 tendency for the owners to see them more as infants rather than adult humans (Belk, 1996). Dotson and Hya tt (2008) found that women were more likely than men to consider their dogs as part of the family. The researchers sent surveys to 749 dog owners focusing on owner interactions with their dogs. The researchers identified seven main dimensions that comprise the human dog relationship: symbiotic relationship, dog oriented self concept, anthropomorphism, activity/youth, boundaries, specialty purchases, and willingness to adapt. The researchers found that overall younger people experience the dimensions more st rongly than do older people. The gender of the owner also had an effect on all of the dimensions with women scoring higher than men. Women are more likely to view their dogs as part of the family and see them as being more like people than wild animals. P rato Previde, Fallani, and Valsecchi (2006) also looked at the differences in the interactions of men and women with their dogs. A group of 25 dog owners, 15 women and 10 men, participated in a modified version of the Strange Situation test with their pet dogs. Participants were given a questionnaire looking at their attachments to their pets. They found that women tended to talk more to their dogs and engaged in more mother infant behaviors including infant directed speech. Both genders engaged in play and there were no gender differences in reported attachment. The results suggested that the human dog relationship might have similarities to the mother infant relationship.

PAGE 32

! 25 Current Study It is clear that humans have a tendency to anthropomorphize their do gs, even more so than other pets (Morris, 2008; Szasz, 1968), and that this tendency may lead to incorrect attributions of emotion. However, it is also clear that dogs are able to appropriately react to emotional cues, which could suggest that the coevolut ion of dogs and humans may have led to a unique sensitivity to emotion by dogs. The current study investigates the relationship between anthropomorphic tendencies and beliefs about emotions in pet dogs. If anthropomorphic tendencies correlate with high bel iefs of emotion, it could show that the dog's close proximity to the daily life of the human makes it too hard to see its actions only in animal terms. However, if the two are unrelated, it could show that attributing human emotions to dogs is not necessar ily caused by anthropomorphic tendencies. There are also likely to be gender differences. It was hypothesized that women would be more likely to report higher anthropomorphic tendencies than men. The current study also further investigates the claims made by Horowitz (2009) by recreating the potentially guilt producing paradigm. Adult humans are able to experience guilt when not in a social context (Tangney et al., 1996) so it is possible that the dogs will show behaviors associated with guilt while the ow ner is out of the room. It is also possible that the guilty experience of dogs is closer to that of a human child and therefore the dogs will have more concern over the possibility of detection than of the morality of the transgression itself. If guilty be haviors really are only reactions to or anticipation of scolding, then the dogs

PAGE 33

! 26 should show no guilty behaviors if the owner shows no emotion. If they do show guilty behaviors at this time it could show that they understood that they committed a transgres sion and guilty behaviors are a way of making amends of some sort. Using owners' perceptions of their dog emotions may allow for the detection of subtle emotions that someone that is inexperienced with the individual dogs may not see. Method Participants A total of 51 participants (27 males, 24 females) from two age groups were recruited for the study. The two age groups consisted of "student adults" with an average age of 21.1 years ( SD = 1.72) and "family adults" with an average age of 44.3 years ( SD = 7. 78). Participants were categorized as student adults if they were over 18 and presently attending a school. All of the student adults were recruited from a small, liberal arts college in southwest Florida. The family adults were categorized as such if they were over the age of 18 (although all were over 28) and had children of any age. The family adults were recruited in Tallahassee, Florida with the majority having children that were involved in a soccer league. Participants answered surveys about their dogs and were allowed to take multiple surveys if they had multiple dogs (participants noted that their dogs had different personalities). Multiple people were also allowed to take a survey for the same dog if they lived in the same house and had interacti ons with that dog. Including multiples, there was a total of 70 surveys about 54 dogs (23 males, 31

PAGE 34

! 27 females) with the average age of the dogs being 5.7 years ( SD = 4.1). A total of 25 dogs participated in a filmed interaction with their owners. Materials P articipants were asked to complete a questionnaire consisting of three surveys investigating owners' beliefs about their dogs' behaviors in a test situation, their dogs' emotions, and interactions that they have with their dogs. The aim of the questionnair e was to get numerical scores about owner perceptions of emotions (both primary and secondary) as well as anthropomorphic tendencies. The survey began with questions about the age and sex/gender of both the dog and the owner. Refer to the Appendix for a co py of the full questionnaire. The first survey looked at pre test beliefs (Appendix A). The test consisted of an interaction where the owner instructed the dog not to eat a treat and then left the dog in the room with the treat while a camera caught the do g's behaviors. Pre test beliefs were measured using six Likert scale questions. These questions included "will your dog eat the treat after you tell it not to?" and "will your dog show guilty behaviors even if you do not scold it?" Responses ranged from 1 Definitely Not to 7 Definitely Will. The second survey investigated the occurrence of a range of emotions and was modified from Morris, Doe, and Godsell (2008) (Appendix B). In this study, questions about emotions of different species of pets were asked using two forms: a yes/no followed by a Likert scale asking about the confidence in the answer with 1 being not at all confident and 5 being very confident An example would be: "Is your animal ever surprised? If yes, how confident are you about your

PAGE 35

! 28 deci sion?" The present study combined the questions into one, but kept the initial question the same. An example would be "Is your dog ever surprised?" with responses from 1 Strongly Disagree to 7 Strongly Agree. This survey was used to ask about the occurren ce of 16 emotions, which were based on emotions in Morris (2008). These included anger, fear, surprise, joy/happiness, sadness, disgust, guilt, jealousy, embarrassment, love/affection, interest/curiosity, shame, pride, empathy, anxiety, and grief. Out of t he 16 emotions, 8 were primary (anger, fear, sadness, joy, anxiety, love, curiosity, and surprise) and 8 were secondary (guilt, jealousy, disgust, empathy, grief, embarrassment, shame, and pride). The maximum total emotion (all 16 emotions) score was 112. The maximum scores for primary emotions and secondary emotions were both at 56. The third and final survey came from Topal, Miklosi, and Csanyi (1997) and looked at anthropomorphic tendencies of the owners towards their dogs (Appendix C). Questions inclu ded "to what extent does your dog identify with your emotions?" and "what kind of cognitive abilities does your dog have?" Responses were scored using the scoring system presented in Topal (1997) with the maximum score being 54. A higher score indicated a stronger anthropomorphic attitude. A total of 34 owners and 25 dogs also participated in a filmed interaction. The interaction took place after the questionnaire in a room where the dog could be left alone with a treat on a plate. The entire interaction w as recorded using a camera set up on a tripod.

PAGE 36

! 29 Procedure The filmed interaction portion of the study took place at the participants' residences so that the dog would be in its most natural setting. Participants filled out the survey while a camera was se t up in an approved room. After the survey was finished the owner was instructed to walk into the room with the dog while carrying a treat on a plate. When he/she placed the treat down she instructed the dog not to eat the food and then walked out of the r oom. The participant then waited outside of the room for no more than one minute. Owners were instructed to enter the room, return to the place where the treat was, and with a completely emotionless face look down at the plate and then at the dog. Afterwar ds the video was shown and the owners were asked to narrate what was happening. A note was made if the participant mentioned that the dog was showing guilty behaviors. Participants who did not participate in the filmed portion just filled out the questionn aire. Results Although 70 surveys were collected, it was decided to only analyze one survey per person so that an individual would not be represented twice. If a person had filled out multiple surveys for different dogs, only data from one dog were select ed. This was done randomly with a flip of a coin. A total of 51 surveys (N male = 27, N female = 24) from 51 participants between the ages of 19 56 (N family = 26, N student = 25) were analyzed.

PAGE 37

! 30 Anthropomorphism Analyses were conducted to see if anthropomorph ism scores were related to emotion scores, age, and gender/sex. Since data generated from Likert scales do not necessarily possess a normal probability distribution, comparisons were calculated using Spearman correlations. A Spearman correlation is a non p arametric test that does not require the data to be normally distributed. Correlations were calculated using SAS9.2. Using conventional criteria ( = .05), there was a weak positive correlation between the total emotion score (primary + secondary) and anth ropomorphism score r s (49) =0.3135, p =0.025. As emotion scores increased, anthropomorphism scores also increased. There was no correlation between the secondary emotion score and anthropomorphism although it approached significance, r s (49)=.2534, p = .0 73 Th ere was not a significant correlation between primary emotion score and anthropomorphism r s (49)=.230, p = .105. Anthropomorphism scores were also analyzed for differences in owner gender and age as well as dog sex and age. Age was analyzed using the Spe arman correlation while gender/sex were analyzed using t tests. There was no correlation between the age of the owner and anthropomorphism r s (49) = .036, p =0.803. There was however a significant weak correlation between the age of the dog and the anthropom orphism score r s (49) = 0.290, p =0.039. As the age of the dog increased, the anthropomorphism score decreased. There was a significant difference in anthropomorphism scores between women ( M = 38.1, SD = 5.6) and men ( M =32.9, SD = 7.5) t (49)=2.82, p = .007. Wome n were more likely to

PAGE 38

! 31 anthropomorphize their dogs than men. There was no difference in the anthropomorphism score between male dogs ( M = 34.3, SD = 8.4) and female dogs ( M = 36.1, SD = 5.9) t (49)=.92, p = .36. Emotions Emotion scores were also analyzed for dif ferences in age and gender/sex. There was no difference in total emotion scores between male dogs ( M =80.9, SD = 6.8) and female dogs ( M =80.6, SD = 10.7) t (49)= .09, p = .93. Both sexes were equally likely to have emotions perceived by their owners. Analyses were conducted to look at differences between the owner age groups (Table 1). There was no difference when it came to total emotion t (49)= .10, p = .925, primary emotion t (49)= .06, p = .950, and secondary emotion t (49)= .18, p = .855. Both age groups were eq ually likely to perceive both primary and secondary emotions in their dogs. Table 1 Mean (SD) Emotion Scores for Student Adults and Family Adults Owner Age Group Emotion Student Family Primary 45.4 ( 4.56 ) 45.3 ( 5.44) Secondary 35.2 ( 6.29 ) 35.6 ( 6.93 ) Total 80.6 (8.10 ) 80.9 ( 10.27 )

PAGE 39

! 32 There were differences seen between owner genders (Table 2). There was a difference in the total emotion scores between men and women t (49)= 2.94, p= .005. There was also a significant difference in secondar y emotion scores between men and women t (49)= 4.10, p< .001. Women were more likely to perceive secondary emotions than men. There were no differences in the primary emotion scores between the genders t (49)= .361, p= .719. Table 2. Mean (SD) Emotion Scores f or Female and Male Owners Gender Emotion Female Male Primary 45.6 ( 5.10 ) 45.1 ( 4.97) Secondary** 38.9 ( 4.60 ) 32.3 ( 6.50 ) Total** 84.5 (7.99 ) 77.41 ( 9.01 ) ** = p < .005 Participants were much more likely to perceive primary emotions th an secondary emotions t (100)= 8.62, p < .001. The Likert scores from the survey were then transformed into yes and no (scores 1 4 were marked no, scores 5 7 were marked yes ) in order to better visualize differences in the perceptions of emotions from all pa rticipants (Figure 1). This also served to compare the findings to those in Morris (2008). The figure shows that participants were more likely to agree that their dogs had primary emotions than secondary emotions and

PAGE 40

! 33 that jealousy and guilt were the most c ommon secondary emotions. These percentages were not analyzed further. Analyses were conducted to see if there were gender differences between individual primary emotions (Table 3). A score of 5 or higher suggested the owners believed their dogs had that emotion and lower than that suggested they did not perceive the emotion. Table 3 Mean (SD) of Individual Primary Emotion Scores between Men and Women Gender Emotion Male Female Sadness 5.26 ( 1.48 ) 5.08 ( 1.77) Anger 3.44 ( 2.19 ) 3.21 ( 1.96 ) Surprise 6.04 (0.81 ) 5.50 ( 1.29 ) Joy 6.44 ( 1.15 ) 6.75 (0 .44 ) Fear 5.63 ( 1.50 ) 5.25 ( 1.67 ) Love 6.52 (0.70) 6.63 (0 .65 ) Curiosity 6.37 (0.69) 6.63 (0 .49 ) Anxiety 5.74 (1.32) 5.71 ( 1.46 )

PAGE 41

! 34 There were no significant differences betwee n men and women with any of the primary emotions (Figure 2). Both men and women were equally likely to perceive that their dogs had primary emotions. Individual secondary emotions were also compared between men and women (Table 4). Table 4 Mean (SD) of I ndividual Secondary Emotion Scores between Men and Women Gender Emotion Male Female Guilt* 4.37 ( 1.92 ) 5.38 ( 1.69) Jealousy 5.41 ( 1.76 ) 5.92 ( 1.21 ) Embarrassed* 3.30 (1.79 ) 4.17 ( 1.40 ) Disgust 3.33 ( 2.02 ) 2.83 ( 1.46 ) Empathy 4.7 4 ( 1.46 ) 5.33 ( 1.24 ) Grief 3.74 ( 1.68 ) 4.17 ( 1.40 ) Pride 4.48 ( 1.53 ) 4.63 ( 1.47 ) Shame** 3.89 ( 1.67 ) 5.29 ( 0.69 ) = p < .061 ** = p < .001

PAGE 42

! 35 There was a significant difference between men and women in perceptions of shame t (49)= 3.82, p= .0 004. Women were much more likely than men to say that their dogs felt shame. Gender differences in the perceptions of guilt t (49)= 1.97, p= .055 and embarrassment t (49)= 1.91, p= .061 approached significance, but were not significant by conventional criter ia. Women possibly perceived guilt and embarrassment more than men. Filmed Interaction There were a total of 25 dogs that were involved in the filmed portion of the study. Out of the 25 dogs, 24 (96%) ate the food. Only 1 (4%) avoided the food altogether This dog showed submissive behaviors even before the owner had instructed it not to eat the food, however the owner did not perceive any of the behaviors as guilt. Of all the dogs, 19 (76%) waited to eat the food until the owner was no longer focused on the dog or the food. This was seen in two ways: waiting for the owner's back to be turned or waiting for the owner to leave the room completely. Five (20%) of the dogs waited for the owner's back to be turned while 14 (56%) waited for the owner to complete ly leave the room (Figure 4). The remaining 5 (20%) dogs ate the food while their owners were still instructing them not to. Only 6 (24%) of the dogs were described by their owners as showing behaviors associated with guilt. None of the owners perceived guilty behaviors right after the dog ate the treat or while the owner was out of the room. All of the described guilty behaviors were seen when the owner entered the room, even though there was no scolding. Most of the behaviors were subtle and would likel y

PAGE 43

! 36 be completely ignored by an independent coder. Many of the owners that identified guilt were describing anticipatory behaviors. While viewing the tape of her dog, one owner commented: "she's slightly guilty, she realized I wasn't saying anything and rela xed." Another owner said "you can see how her tail tucks a little bit, she kind of knows, like I think I may be in trouble', but then she's like ok' and starts wagging her tail." Most of the owners laughed as their dogs ran out of the room looking for mo re treats, thus showing no guilt at all. An analysis was conducted to see if there were differences in owners' perceptions of their dogs mental capacities based on the dogs' behaviors. The first analysis compared owner perceptions of mental capacity betwe en dogs that waited for the owner to completely leave the room and dogs that did not wait. A higher score denotes a perception of a higher mental capacity. There was a significant difference in the perceptions of the mental capacities of the dogs that did not wait for the owner to leave ( M = 2.93. SD = 1.1) and dogs that waited for the owner to leave the room completely ( M = 4.05, SD = .92), t (34)=3.30, p = .002. Dogs that waited for their owners to completely leave the room had higher perceived mental capacitie s than dogs that ate the food with the owner still in the room. However, there was no difference in the perceptions of secondary emotions between dogs that waited ( M = 36.9, SD = 6.1) and those that did not wait for the owner to leave ( M = 36.2, SD = 6.8), t( 3 5)= .32, p = .74. There was a significant difference in the anthropomorphism score between dogs that waited ( M =38.2, SD = 5.4) and dogs that did not wait ( M = 33.8, SD = 5.5), t (34)=2.44, p = .0202.

PAGE 44

! 37 Dogs that waited for their owners to leave were also more like ly to be anthropomorphized. Pre Test Beliefs There were no significant differences between men and women when it came to any of the pre test beliefs. It was found that there were no differences in beliefs that the dog would eat the treat after being instr ucted not to between women ( M =5.58, SD =1.69) and men ( M =5.04, SD =2.05). Both genders agreed that the dogs would probably eat the treat. Men and women were also equally likely to think that their dogs would show guilty behaviors in all three situations ( p > .05): right after eating the treat (men: M =3.26, SD =2.0; women: M =3.79, SD = 1.6), when the owner was out of the room (men: M =2.30, SD =1.27; women: M =2.83, SD = 1.63), and if the owner did not scold the dog at all (men: M =3.19, SD =1.73; women: M =3.21, SD = 1.69). It should be noted that the results show that both men and women think that the dog will most likely not show guilty behaviors in all three situations. Discussion Emotions This study investigated dog owners' beliefs about dog emotions, anthropomor phic attitudes, and interactions potentially involving guilt. Most owners reported that their dogs had the primary emotions of sadness, fear, anxiety, surprise, curiosity, joy, and affection. Anger was the least reported of the primary emotions and the ave rage answer of owners was that they did not believe their dogs showed anger. This is consistent with findings from Morris (2008). In

PAGE 45

! 38 their study, only 65% of dog owners reported that their dogs were ever angry. This also was the lowest reporting from all o f the primary emotions. In their study as well as the current study curiosity, joy, and affection were the top three reported emotions. The secondary emotions looked a lot different than the primary emotions (this is especially evident in Figure 1). Owne rs were much more likely to report that their dogs did not have secondary emotions. Both men and women said that their dogs did not express embarrassment, disgust, and grief. These were also the lowest when it came to the percentage of participants. Less t han 30% of participants reported these three emotions. Participants were more confident in reporting that their dogs had shame, empathy, and guilt than the other three emotions, but these were still lower than most of the primary emotions. The highest repo rted secondary emotion was jealousy, which was also the case in Morris (2008). The results from the current study successfully replicated the findings in Morris (2008). Participants answered similarly, and guilt and jealousy were the two highest reported o f the secondary emotions for dogs. The current study also looked at gender differences in beliefs about emotions. No gender differences were found for beliefs in primary emotions, both men and women were equally likely to report all 8 of the primary emot ions. There were, however, gender differences in beliefs about secondary emotions. Women were more likely than men to report secondary emotions (Figure 3). Specifically they were more likely to report embarrassment, guilt, and shame. This is interesting be cause these three emotions are somewhat similar and all require self

PAGE 46

! 39 reflection (Lewis, 2011). The line that separates these into distinct emotions has not been fully standardized (Tangney et al. 1996). Lewis (1971) suggested that shame was characterized b y painful feelings about oneself while guilt had more of a focus on the thing that was done as opposed to the self. Embarrassment was considered to be a lesser form of shame that occurred after some sort of public humiliation, whereas shame could be felt w hile alone. It seems as though what separates these emotions are the internal thoughts that accompany them and not necessarily the physical behaviors. It is possible that women picked up on a specific behavior from their dogs, such as distress, in differen t contexts and attributed complex human emotions to explain the dogs' behaviors. It is also possible that women have a unique sensitivity to emotional expression. The differences in perceptions of dog emotion between men and women may have something to do with differences in attitude towards their dogs. Anthropomorphism Women were more likely than men to anthropomorphize their dogs. It is not surprising that there were gender differences in both anthropomorphic attitudes as well as perceptions of e motion. Research shows that men and women interact differently with their dogs (Prato Previde, 2006). Women are more likely to talk to their dogs and engage in mother infant behaviors including infant directed speech. They are also more likely to think of their dogs as children (Dotson, 2008). If emotion scores and anthropomorphic tendencies were not related, it would show that attributing emotions to dogs is not necessarily caused by

PAGE 47

! 40 anthropomorphic tendencies. Therefore it would be possible that those em otions are not just reserved for humans and could also be seen in dogs and possibly other species. However, the correlation between anthropomorphic tendencies and perceptions of emotions was significant. As anthropomorphic tendencies increased it was likel y that the perception of emotions also increased. It is possible that attributing human qualities onto the dog is what causes the perception of emotion. This would suggest that owners pick up on random or learned behaviors from their dogs and explain the m using human terms. This in turn leads to the idea that their dogs are capable of feeling and expressing complex emotions such as jealousy and guilt. On the other side, it is possible that owners detect emotional expression in their dogs and this leads th em to think of the dogs more as humans. Therefore a dog that is more emotive may be more likely to be anthropomorphized. It is also possible that there is another variable that affects both emotion and anthropomorphism scores. More research needs to be don e to see what the direction is. Dogs and Guilt In the filmed interaction all but one of the dogs ate the food. The owner of the exception said that the dog had already been trained not to eat food while the owner was away. A couple of the other owners m entioned that they also had trained their dogs not to eat treats when instructed, but not if the owner left the room. It is possible that the dogs did not know that the owners were instructing them not to eat the treat ever. These dogs had mostly been trai ned to wait for the owner to give permission to eat the treat. Perhaps the owner's absence made it

PAGE 48

! 41 seem like indirect permission was given. If this was the case then this paradigm may not be enough to elicit guilt. If the target food was something that the dog was not used to eating it is possible that it could have an effect on the dog's behaviors. A future study should be conducted to see the differences in behaviors if there was no treat, a familiar treat, or an unfamiliar treat such as a plate of human food. An unfamiliar food may elicit more avoidance behaviors than a dog treat. It is also possible that placing the plate on the ground made it less of a transgression for the dog to eat the food. Most dogs are fed in bowls on the ground so it could be t hat they associate the ground with food that the owner is giving them. If the food was somehow raised off the ground, this could also affect behavior. More dogs may wait for the owner to completely leave the room because the act of taking food off of a tab le is more punishable than taking it off of the floor. Changing the severity of the crime, so to speak, may elicit more "guilty" behaviors upon the return of the human. Research shows that dog owners tend to believe that their dogs understand the rules o f the household and actually show and feel guilt after committing a transgression (Sanders, 1993). However, Horowitz (2009) suggested that the guilty look was really only in response to scolding. The look of guilt shares the characteristics of a fearful re sponse. These behaviors include lowering the ears and body, tucking the tail between the legs, and sometimes rolling over on the back to expose the underside (Darwin 1872/1979). These are all submissive behaviors that suggest that the dog may be preoccupie d with the expectation of

PAGE 49

! 42 punishment than the act of a wrongdoing. If this is the case, it seems similar to the behaviors of young children. Some researchers suggest that children cannot experience guilt until they are able to recognize themselves as disti nct entities. Any early forms of guilt are better described as distress (Draghi Lorenz, 2001). Even when they develop this ability, guilt in young children is still simpler than that of adults. Younger children are more likely to describe concern over det ection and punishment and as they develop they begin to experience more guilt over internalized events such as violations in principles of right and wrong (Thompson, 1980). Dog owners have a tendency to view their dogs as infants or children (Belk, 1996). So if dogs are more similar to human children than human adults, then it would make sense if dogs expressed guilty behaviors more similar to those of young children. Some of the dogs in the current study were described as showing guilty behaviors. These b ehaviors looked more like anticipatory behaviors as though the dog was expecting punishment. The owners explained that their dogs expressed relief that they were not punished upon the return of the owner. These behaviors suggest that the dogs had learned t hat the act of eating forbidden food might lead to punishment. These results suggest the possibility that the expression of guilt in dogs has similarities to "guilt" in very young children. That is to say that dogs do not have an internalized sense of rig ht and wrong and that a violation of this would not lead to self reflection. Instead, dogs learn punishable events and can show submissive behaviors possibly due to a fear of detection.

PAGE 50

! 43 Some dogs that have thunderstorm phobias begin to show fear responses before the fearful event actually begins (Cottam, Dodman, & Critzer, 2005). It is possible that they recognize changes in the environment that are indicative of an oncoming storm such as changes in barometric pressure or levels of static electricity (Over all, Dunham, & Frank, 2001). In any case it suggests that dogs can have an anticipatory fear response. So if dogs have a fear of detection/punishment then it is possible that they show anticipatory submissive behaviors. In the current study none of the dog s showed any guilty behaviors while the owner was out of the room. Since some showed slightly submissive behaviors upon the return of the human, it could be that the guilty behaviors are for the benefit of the human instead of being a fear response to the threat of punishment. It is possible that dogs have been trained by their owners to show guilty behaviors. Upon seeing a submissive dog the owner may not be as harsh in his punishment. This would reinforce the behavior so that the next time the dog commit ted the transgression it would be more likely to show submissive behaviors before punishment. This would suggest that the dog may not necessarily have a fear of detection but instead a learned behavior of submission. Horowitz (2009) found that dogs that h ad received prior obedience training showed more guilty behaviors when the owner scolded them (even if they did not eat the treat) than dogs that did not receive the training. She posited that these dogs might have learned to express submission in the pres ence of certain owner behaviors. Future research could be done to find out what behaviors of the

PAGE 51

! 44 owner, if any, are the causes of the submissive "guilty" look. For instance, it could be the body language, the facial expression, the stern voice, or even a c ombination that leads to submissive behaviors in dogs. In the current study, the one dog that did not eat the treat showed submissive behaviors upon the return of the owner, even though the owner did not scold him and it did not commit a transgression. It is possible that the owner's body language or some other cue set off the dog's learned submissive response. It is also possible that the submissive behaviors are a conditional emotional response. Conditional emotional responses (CERs) are learned emotiona l reactions to predictive cues. An example is the conditioned fear response in the classic Little Albert experiment (Watson & Rayner, 1920). In this experiment Baby Albert was allowed to play with a white rat to which he initially showed no fear. Then in s ubsequent trials, the researchers made a loud noise with a piece of steel and a hammer every time the baby touched the rat. By the end of the experiment, Little Albert would start crying out of fear when he saw the rat, even before the sound was made. He e ven showed fear responses towards other cues including a dog, a Santa beard, and a rabbit. The dogs in the current study did not show a prolonged fear response as some do with thunderstorms, but this could be because of a lack of cues. The change in the e nvironment before storms gives the dogs cues for an approaching storm, which can then elicit the conditional fear response. Leaving a dog alone in a room with just the knowledge of the wrongful act may not be enough of a cue to elicit fear of punishment. P erhaps if the owner messed with the door handle for a

PAGE 52

! 45 bit or talked right outside of the door before entering it could provide a cue that the owner was returning. If the dog feared punishment then it is possible that it would show guilty behaviors before t he owner entered the room upon receiving the cue. It is also possible that the dog only associates punishment with an angry person. Therefore the dog would require angry cues from the owner in order to elicit the fear response. Again, more research can be done to see what cues dogs pick up on before they show submissive behaviors. Conclusion The discussion of guilt and other complex emotions in dogs is incredibly difficult because so far there are no methods that successfully provide empirical evidence abo ut subjective states in animals. Therefore it should be noted that the results from this study could only provide suggestions about the emotional expression of dogs. The results indicated that the perception of secondary emotions in dogs was less likely th an primary emotions. It is possible that certain behaviors of dogs are misconstrued and given human attributes, which seems to be the case with guilt. It seems unlikely that the dogs in the study really felt "guilt" over eating the treat. This could be bec ause the paradigm was not appropriate for eliciting guilt, but it seems more probable that the owners saw submissive behaviors and gave them human attributes. Future research is needed to look more in depth at specific emotional behaviors and the events th at lead to them. Fully understanding the emotional expression s of dogs is important because it could lead to a better relationship between dog and owner.

PAGE 53

! 46 References Amodio, D.M., Devine, P.G., & Harmon Jones, E. (2007). A dynamic model of guilt: Implica tions for motivation and self regulation in the context of prejudice. Psychological Science, 18 (6) 524 530 Baumeister, R.F., Stillwell, A.M., & Heatherton, T.F. (1994). Guilt: An interpersonal approach. Psychological Bulletin, 115, 243 267 Bekoff, M. (2 000). Animal emotions: Exploring passionate natures. BioScience, 50, 861 870 Belk, R.W. (1996). Metaphoric relationships with pets. Society and Animals, 4 (2), 121 145 Benecke, N. (1987). Studies on early dog remains from Northern Europe. Journal of Arch aeological Science, 14, 31 49 Call, J., Brauer, J., Kaminski, J., & Tomasello, M. (2003). Domestic dogs (Canis familiaris) are sensitive to the attentional state of humans. Journal of Comparative Psychology, 117 (3), 257 263 Coppinger, R. & Coppinger, L. (2001). Dogs: a startling new understanding of canine origin, behavior and evolution. Chicago, IL: Chicago University Press Cottam, N., N. Dodman, and T. Critzer. 2005. Use of a cape (the Storm Defender¨) in the treatment of canine ( Canis familiaris ) th understorm phobia. In: D. Mills et al. (eds.), Current issues and research in veterinary behavioral medicine: Papers presented at the 5th International

PAGE 54

! 47 Veterinary Behavior Meeting (pp 165 167). West Lafayette, IN: Purdue University Press Darwin, C. (1872/ 1979). The expression of emotions in man and animals. St. Martin's Press, New York, NY. Descartes, R. (1646 1649/1976). Animals are machines. In T. Regan & P. Singer (Eds.), Animal rights and human obligations (pp. 60 66). Englewood Cliffs, NJ: Prentice Hall Draghi Lorenz, R., Vasudevi, R., & Costall, A. (2001). Rethinking the development of "nonbasic" emotions: A critical review of existing theories. Developmental Review, 21, 263 304 Ekman, P. & Friesen, W.V. (1971). Constants across cultures in the face and emotion. Journal of Personality and Social Psychology, 17, 124 129 Ekman, P. (1992). Are there basic emotions? Psychological Review, 99 (3), 550 553 Freud, S. (1961). Civilization and its discontents. (J. Strachey, Ed. & Trans.). Oxford, Englan d: Hogarth. (Original work published 1930) Gacsi, M., Miklosi, A., Varga, O., Topal, J., & Csanyi, V. (2004). Are readers of our face readers of our minds? Dogs (Canis familiaris) show situation dependent recognition of human's attention. Animal Cognitio n, 7, 144 153 Groves, C.P. (1999). The advantages and disadvantages of being domesticated. Perspectives in Human Biology, 4, 1 12

PAGE 55

! 48 Guo, K., Meints, K., Hall, C., Hall, S., & Mills, D. (2009). Left gaze bias in humans, rhesus monkeys, and domestic dogs. Animal Cognition, 12, 409 418 Hare, B., Brown, M., Williamson, C., & Tomasello, M. (2002). The domestication of social cognition in dogs. Science, 298, 1634 1636 Horowitz, A. & Bekoff, M. (2007). Naturalizing anthropomorphism: Behavioral prompts to o ur humanizing of animals. Anthrozoos, 20, 23 35 Horowitz, A. (2009). Disambiguating the "guilty look": Salient prompts to a familiar dog behavior. Behavioral Processes, 81, 447 452 Humphrey, N. (1983). Consciousness regained. Oxford: OUP Itakura, S. & Tan aka, M. (1998). Use of experimenter given cues during object choice tasks by chimpanzee (Pan troglodytes), and orangutan (Pongo pygmaeus) and human infants (Homo sapiens). J Comp Psychol, 112 119 126 Izard, C.E. (1992). Basic emotions, relations among emotions, and emotion cognition relations. Psychological Review, 99 (3), 561 565 Kagan, J. (1984). The nature of the child. New York: Basic Books. Keddie, K. (1977). Pathological mourning after the death of a domestic pet. British Journal of Psychiatry, 131 21 25 Leonard, J.A., Wayne, R.K., Wheeler, J., Valadez, R., Guillen, S., & Vila, C. (2002). Ancient DNA evidence for Old World origin of New World dogs. Science, 298, 1613 1616

PAGE 56

! 49 Lewis, H.B. (1971). Shame and guilt in neurosis. Psychoanalytic Review 58, 419 438 Lewis M. (2011). The self conscious emotions. In: Tremblay RE, Boivin M, Peters RDeV, eds. Encyclopedia on Early Childhood Development [online] (pp. 1 6) Montreal, Quebec: Centre of Excellence for Early Childhood Development and Strategic Knowledge Cluster on Early Child Development MacLean, P. (1970 ). The triune brain in evolution: Role in Paleocerebral functions. New York: Plenum Press. Maros, K., Pongracz, P., Bardos, G., Molnar, C., Farago, T., & Miklosi, A. (2008). Dogs can discrim inate barks from different situations. Applied Animal Behaviour Science, 114, 159 167 Masson, J. & McCarthy, S. (1995). When elephants weep: the emotional lives of animals. New York: Plenum Press McKinley, J. & Sambrook, T.D. (2000). Use of human given cues by domestic dogs ( Canis familiaris ) and horses (Equus caballus). Animal Cognition, 3, 13 22 Miklosi, A., Kubinyi, E., Topal, J., Gacsi, M., Viranyi, Z., & Csanyi, V. (2003). A simple reason for a big difference: Wolves do not look back at hum ans, but dogs do. Current Biology, 13, 763 76 Miklosi, A., Polgardi, R., Topal, J., & Csanyi, V. (1998). Use of experimenter given cues in dogs. Animal Cognition, 1, 113 121 Morris, P.H., Doe, C., and Godsell, E. (2008). Secondary emotions in non primate

PAGE 57

! 50 species? Behavioral reports and subjective claims by animal owners. Cognition and Emotion, 22, 3 20 Nagasawa, M., Murai, K., Mogi, K., & Kikusui, T. (2011). Dogs can discriminate human smiling faces from blank expressions. Animal Cognition, 14, 525 533 Overall, K.L., Dunham, A.E., & Frank, D. (2001). Frequency of nonspecific clinical signs in dogs with separation anxiety, thunderstorm phobia, and noise phobia, alone or in combination. J. Am. Vet. Med. Assoc. 219, 467 473 Pang, J.F., Kluetsch, C., Zou X.J., Zhang, A., Luo, L.Y., Angleby, H., Savolainen, P. (2009). mtDNA data indicate a single origin for dogs south of Yangtze River, less than16,300 years ago, from numerous wolves. Mol. Biol. Evol. 26 (12), 2849 2864 Pfungst, O. (1911). Clever Hans. The horse of Mr. von Osten. Henry Holt, New York Pongracz, P., Molnar, C., Miklosi, A., & Csanyi, V. (2005). Human listeners are able to classify dog (Canis familiaris) barks recorded in different situations. Journal of Comparative Psychology, 119 (2) 136 144 Prato Previde, E., Fallani, G., & Valsecchi, P. (2006). Gender differences in owners interacting with pet dogs: An observational study. Ethology, 112, 64 73. Range, F., Horn, L., Viranyi, Z., & Huber, L. (2009). The absence of reward induces i nequity aversion in dogs. PNAS, 106, 340 345

PAGE 58

! 51 Sackheim, H.A. Gur, R.C., & Saucy, M.C. (1978). Emotions are expressed more intensely on the left side of the face. Science, 202, 434 436 Sanders, C.R. (1993). Understand dogs: caretakers' attributes of minde dness in canine human relationships. Journal of Contemporary Ethnography, 22, 205 226 Savolainen, P., Zhang, Y., Luo, J., Lundeberg, J., & Leitner, T. (2002). Genetic evidence for an east Asian origin of domestic dogs. Science, 298, 1610 1613 Schleidt, W.M. (1998). Is humaneness canine? Human Ethology Bulletin, 13 (4), 14 Schleidt W. M. & Shalter M. D. (2003). Co evolution of humans and canids. An alternative view of dog domestication: Homo Homini Lupus? Evolution and Cognition, 9, 57 72 Serpell, J. A. (2003). Anthropomorphism and anthropomorphic selection Beyond the "cute response." Society & Animals, 11, 83 100 Staats, S., Wallace, H., & Anderson, T. (2008). Reasons for companion animal guardianship (pet ownership) from two populations. Society o f Animals, 16, 279 291 Szasz, K. (1968). Petishism: Pets and their people in the Western World. New York, NY: Holt, Rhinehart and Winston. Tangney, J.P., Miller, R.S., Flicker, L., Barlow, D.H. (1996). Are shame, guilt, and embarrassment distinct emot ions? Journal of Personality and Social Psychology, 70 (6), 1256 1269

PAGE 59

! 52 Thompson, R.A. & Hoffman, M.L. (1980). Empathy and the development of guilt in children. Developmental Psychology, 16 (2), 155 156 Topal, J., Miklosi, A., & Csanyi, V. (1997). Dog human relationship affects problem solving behavior in the dog. Anthrozoos, 10, 214 224 Trut, L.N., Plyusnina, I.Z., & Oksina, I.N. (2004). An experiment on fox domestication and debatable issues of evolution of the dog. Russian Journal of Genetics, 40 (6), 64 4 655 Vila, C., Savolainen, P., Maldonado, J.E., Amorim, I.R., Rice, J.E., Honeycutt, R.L., Crandall. K.A., Lundeberg, J., & Wayne, R.K. (1997). Multiple and ancient origins of the domestic dog. Science, 276, 1687 1689 Viranyi, Z., Topal, J., Gacsi, M., Miklosi, A., & Csanyi, V. (2004). Dogs respond appropriately to cues of humans' attentional focus. Behavioural Processes, 66, 161 172 Watson, J.B. & Rayner, R. (1920). Conditioned emotional reactions. Journal of Experimental Psychology, 3, 1 14 Williams, C. & Bybee, J. (1994). What do children feel guilty about? Developmental and gender differences. Developmental Psychology. 30 (5), 617 623

PAGE 60

! 53 Figure 1. Percentage of participants reporting primary and secondary emotions in their pet dogs.

PAGE 61

! 54 Figure 2. Average scores for men and women on beliefs of primary emotions. Scores over 4 suggest a belief that their dogs do express the particular emotion. = p< .08 !

PAGE 62

! 55 !"#$ %&'()' Average scores for men and women on beliefs of secondary emotions. Scores over 4 suggest a belief that dogs do express the particular emotion. ** = p < .05; = p < .08

PAGE 63

! 56 Figure 4. A common behavior during the filmed interaction. The dogs were instructed not to eat the food (A), waited for the owner to leave the room (B), and then proceeded to eat the food (C).

PAGE 64

! 57 Appendix A Age:_________________ Gender:_________________ Dog's Age: ___________________ Dog's Sex: _________ ______ Instructions: Circle the best answer Pre Test Beliefs 1. Will your dog eat the treat after you tell it not to? 1 2 3 4 5 6 7 (definitely not) (no) (probably not) (don't know) (maybe) (yes) (definitely will) 2. Will your dog eat the tre at while you are still in the room? 1 2 3 4 5 6 7 (definitely not) (no) (probably not) (don't know) (maybe) (yes) (definitely will) 3. Wil l your dog wait till you are out of the room? 1 2 3 4 5 6 7 (definitely not) (no) (probably not) (don't kn ow) (maybe) (yes) (definitely will) 4. Will your dog show guilty behaviors right after he/she eats the treat? 1 2 3 4 5 6 7 (definitely not) (no) (probably not) (don't know) (maybe) (yes) (definitely will) 5. Will your dog show guilty behaviors while y ou are out of the room? 1 2 3 4 5 6 7 (definitely not) (no) (probably not) (don't know) (maybe) (yes) (definitely will) 6. Will your dog show guilty behaviors even if you do not scold it? 1 2 3 4 5 6 7 (definitely not) (no) (probably not) (don't know) (maybe) (yes) (definitely will) Other expectations:

PAGE 65

! 58 Appendix B From Morris, Doe, & Godsell (2008) Instructions: Circle the answer that best describes how much you agree with the occurrence of different emotions in your dog. 1. Is your dog ever angry? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewhat disagree) (Neutral) (somewhat agree) (agree) (strongly agree) 2. Is your dog ever fearful? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewhat disagree) (Neutral) (somewhat agree) (agre e) (strongly agree) 3. Is your dog ever surprised? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewhat disagree) (Neutral) (somewhat agree) (agree) (strongly agree) 4. Is your dog ever joyful/happy? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (s omewhat disagree) (Neutral) (somewhat agree) (agree) (strongly agree) 5 Is your dog ever sad? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewhat disagree) (Neutral) (somewhat agree) (agree) (strongly agree) 6. Is your dog ever guilty? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewhat disagree) (Neutral) (somewhat agree) (agree) (strongly agree) 7. Is your dog ever jealous? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewhat disagree) (Neutral) (somewhat agree) (agree) (strongly agree) 8. Is your dog ever embarrassed? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewhat disagree) (Neutral) (somewhat agree) (agree) (strongly agree)

PAGE 66

! 59 9 Does your dog ever show love/affection? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewh at disagree) (Neutral) (somewhat agree) (agree) (strongly agree) 10. Does your dog ever show disgust? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewhat disagree) (Neutral) (somewhat agree) (agree) (strongly agree) 11. Does your dog ever show int erest/curiosity? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewhat disagree) (Neutral) (somewhat agree) (agree) (strongly agree) 12. Does your dog ever seem anxious? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewhat disagree) (Neutral) (som ewhat agree) (agree) (strongly agree) 13. Does your dog ever show empathy? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewhat disagree) (Neutral) (somewhat agree) (agree) (strongly agree) 14. Does your dog ever show grief? 1 2 3 4 5 6 7 (strong ly disagree) (disagree) (somewhat disagree) (Neutral) (somewhat agree) (agree) (strongly agree) 15. Does your dog ever show pride? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewhat disagree) (Neutral) (somewhat agree) (agree) (strongly agree) 16 D oes your dog ever show shame? 1 2 3 4 5 6 7 (strongly disagree) (disagree) (somewhat disagree) (Neutral) (somewhat agree) (agree) (strongly agree)

PAGE 67

! 60 Appendix C From Topal, Miklosi, & Csanyi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

PAGE 68

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