11.1; P , 0.001). The study dogs exhibited 6 vomiting eventsin comparison to 1399 grass-eating events.The results of the current study suggest that grass-eatingbehavior in domestic dogs is innate and that the mother’seating habits further facilitate the puppies’ grass eating. Asthe puppies matured and were weaned, they spent more timeeating grass, further supporting the notion that grass is seen asa food source (Bjone et al., 2007). Some researchers contendthat dogs use grass as an emetic. However, the low prevalenceof vomiting in the current study does not support this claim.

Key words: dog; feeding behavior; grass eating; learning;maternal influence

Reference:

Bjone, S.J., Brown, W.Y., Price, I.R., 2007 Grass eating patterns in the do-

mestic dog, Canis familiaris. Recent Advances in Animal Nutrition in

Australia 15, 45–49.

91

A DOG’S GAZE TOWARD ITS OWNER FUNCTIONS ASSOCIAL ATTACHMENT AND INCREASES THE OWNER’SURINE OXYTOCIN LEVELMiho Nagasawaa,*, Takefumi Kikusuia, Tatsushi Onakab,Mitsuaki Ohtaa

aDepartment of Animal Science and Biotechnology, AzabuUniversity, JapanbDepartment of Physiology, Jichi Medical University, Japan*Corresponding author: mihnag@gmail.com

Oxytocin (OT) has been shown to play an important role insocial bonding in animals (Kikusui et al., 2006). However,it is unclear whether OT is related to interspecies socialbonding. In the process of domestication, dogs have devel-oped humanlike visual cognitive abilities (Hare et al.,2002). In this study, we measured urinary OT concentra-tions of owners before and after interaction with theirdogs to examine the possibility that urinary OT concentra-tions of owners were increased by their dog’s gaze, repre-senting social attachment to their owners.Fifty-five pairs of dogs and their owners participated in 2experiments. Prior to the experiments, the owners wereasked to complete questionnaires concerning their owner–dog relationship. Dog owners interacted with their dogs in amanner typical for them for 30 minutes (interaction exper-iment) or were instructed not to look at their dogs directly(control experiment). We observed the behaviors of ownersand their dogs during the experiments and measuredOT concentrations by radioimmunoassay in owners’ urinesamples collected just before and 20 minutes after interac-tion with their dogs (Amico et al., 1987).Using cluster analysis, owners could be divided into 2groups: one experienced longer gaze durations from theirdogs and reported a higher degree of relationship with theirdogs (LG); the other received a shorter duration of gaze and

reported a lower degree of relationship (SG). There was astatistically significant interaction between times andgroups in urinary OT concentrations (F[1.84] 5 9.03,P , 0.01, 3-way analysis of variance [ANOVA] with re-peated-measures, factors: groups [LG and SG], times col-lecting urine [pre and post], and experiments [interactionexperiment and control experiment]). In post hoc analysis,urinary OT was higher in LG than SG after typical interac-tion with their dogs (P , 0.01), but not in the controlexperiment. In the interaction experiment, a high correla-tion was found in LG between the frequency of behavioralexchanges initiated by the dog’s gaze and the increase inurinary OT (rs 5 0.73, P , 0.01), but not in SG.These results suggested that the dog’s gaze induced activa-tion of the OT neuroendocrine system. We concluded thatinteractions with dogs, especially those initiated by the dogsgazing at their owners, can increase the urinary OT concen-trations of their owners as a manifestation of attachmentbehavior.

Key words: dog; gaze; attachment behavior; urinaryoxytocin

References:

Amico, J.A., Ulbrecht, J.S., Robinson, A.G., 1987. Clearance studies of ox-

ytocin in humans using radioimmunoassay measurements of the hor-

mone in plasma and urine. J. Clin. Endocrinol. Metab. 64, 340–345.

Hare, B., Brown, M., Williamson, C., Tomasello, M., 2002. The domesti-

cation of social cognition in dogs. Science 298, 1634–1636.

Kikusui, T., Winslow, J.T., Mori, Y., 2006. Social buffering: relief from

stress and anxiety. Philo. Trans. R. Soc. Lond. B. Biol. Sci. 361,

2215–2228.

98 Journal of Veterinary Behavior, Vol 4, No 2, March/April 2009

92

HUMAN-DIRECTED GAZING BEHAVIOR IN DOMESTICDOGS (CANIS FAMILIARIS)Lisa Horn1,*, Peter Pongracz2, Zsofia Viranyi3,Ludwig Huber1, Adam Miklosi2, Friederike Range1

1Department of Neurobiology and Cognition Research,University of Vienna, Austria2Department of Ethology, Eotvos Lorand University,Hungary3Konrad Lorenz Institute for Evolution and Cognition,Austria*Corresponding author: lisa.horn@univie.ac.at

Previous studies have shown that domestic dogs (Canis fa-miliaris) initialize communication with humans (e.g., look-ing, approaching) when they are faced with an insolvableproblem. In the current study, we wanted to explore thepossibility that this behavior can be influenced by formerinteraction between dog and owner. We asked whetherverbal encouragement by the owner increases the gazingbehavior of the dog when it faces an insolvable problemcompared to dogs that were not encouraged previously.

reported that individuals take complementary roles whenhunting for the same prey. However, to test whether or notcooperative behavior reflects understanding of the need forpartners and of each partner’s role, controlled experimentalstudies are needed.In our pilot study, we tested 2 of 3 timber wolves (bothmales, 12 and 15 years of age), kept in the ZooSchonbrunn/Vienna in a cooperative string-pulling task toinvestigate whether individuals would coordinate theirbehavior for getting access to a food reward. After thewolves quickly learned individually to pull on a ropeconnected to 1 of 2 baited platforms for themselves, weconnected the 2 platforms so 2 wolves had to pull theirropes synchronously to move the platforms and reach thefood. We randomly varied the following conditions: (1)1 wolf could solve the task alone (separated platforms); or(2) both wolves had to cooperate to pull the platformssufficiently close to reach the food reward (connectedplatforms). In both conditions we baited either both plat-forms or just one.In the condition with the separated platforms, the wolvesalways reached the food. In the connected condition thesuccess rate dropped only slightly to 95% of the trials.Furthermore, after an initial refusal to cooperate by thesubordinate animal when only 1 platform was baited, thewolves reengaged in their cooperation (80% of the trials)by changing position relative to the baited platform so thatoften the dominant wolf was on the same side as the food.However, in those cases when the subordinate wolf pulledthe baited platform, the dominant one took the food fromhim in only 35% of the successful trials.Our preliminary results show that the wolves were willingto cooperate in most of the trials. They were able tomaintain their cooperation even when only a single piece ofmeat was provided. We saw quite some tolerance, e.g., not

Interactions between dogs and humans 99

During training sessions, all dogs had the opportunity tolearn how to manipulate an apparatus to get food rewardefficiently. The owners of 1 group of dogs (n 5 13) wereasked to reinforce the looking behavior of their dogs by en-couraging them verbally whereas owners of the other ex-perimental group (n 5 12) did not encourage their dogsthroughout the training session. After the training session,2 tests were carried out. In Test 1 the apparatus was empty,whereas in Test 2 the apparatus stopped working after thedogs had obtained the first pieces of food reward. To inves-tigate the gazing behavior of the dogs we measured the du-ration and direction of gazes.Although the latency to look at the owner was the same forboth groups (ANOVA, F 5 2.72, P 5 0.106), we found thatin both Test 1 (apparatus empty) and Test 2 (apparatusblocked) dogs that were encouraged during the training ses-sion tended to look for longer periods at their owners com-pared to dogs that were not encouraged (ANOVA, F 5

14.13, P 5 0.001). In Test 1 (apparatus empty), dogs ofboth experimental groups increased their looking time to-ward the door through which the experimenter had usuallyentered to refill the apparatus during the training sessioncompared to Test 2 (apparatus blocked) (ANOVA, F 5

7.99, P 5 0.007).The results indicate that the former encouragement in-creases human-directed gazing behavior in dogs. However,even dogs that did not receive encouragement increasedtheir looking time toward the owner to some extent whenfaced with a problem. Further, the results of Test 1 (appa-ratus empty) indicated that dogs associated the experi-menter (i.e., the door where she usually entered) with therefilling of the empty apparatus.

Key words: problem solving; dog–human interaction;gazing; dependency; association

93

COOPERATIVE STRING-PULLING IN WOLVESHelene Moslinger1,*, Kurt Kotrschal1,2, Ludwig Huber3,Friederike Range3, Zsofia Viranyi41Department of Behavioral Biology, University of Vienna,Austria2Konrad Lorenz Forschungsstelle Grunau, Austria3Department for Neurobiology and Cognition Research,University of Vienna, Austria4Konrad Lorenz Institute for Evolution & Cognition,Altenberg, Austria*Corresponding author: helene.moeslinger@gmx.at

Cooperation is a widely discussed topic in the context ofhuman evolution and animal behavior. Cooperation, in thesense that 2 or more individuals synchronize their behavior toreach a common goal, probably depends on mutual toleranceand thus on social relationships and may rely on certaincognitive abilities as well. For investigating cooperativeinteractions, during hunting for instance, field observationsalone may not be conclusive. In lions and wolves it has been

just the dominant wolf got the meat when just 1 piece wasavailable and the dominant wolf apparently did not forcethe subordinate one to cooperate. The success of the 2wolves in this cooperative task was seemingly based ontheir behavioral flexibility and their quick adaptation to thedifferent experimental conditions.

Key words: wolf; cooperation; string-pulling

94

CHANGES IN EMOTIONAL COMPETENCES OF DRUGOFFENDERS DURING DOG-ASSISTED GROUP TRAINING(MTI)Birgit U. Stetinaa,*, Barbara Kuchtaa, Barbara Gindla,Tamara Lederman Mamana, Ursula Handlosb,Wolfgang Werdenichc, Ilse Kryspin-Exnera

aInstitute for Clinical, Biological and DifferentialPsychology, Faculty of Psychology, University of ViennabSonderpadagogisches Zentrum 9, Vienna, AustriacJustizanstalt Wien-Favoriten, Austria*Corresponding author: birgit.stetina@univie.ac.at