Canine stress responses : a study on the impact of petting on behaviour, saliva cortisol and heart rate of dogs

Abstract: From previous researches it has been shown that dogs being petted by familiar people show more calming gestures and social approach. The aim of this experiment was to investigate whether a 5-minute tactile stimulation reduces or buffers the stress reaction in dogs. The participants were 20 privately owned dogs and their owners. The experiment was carried out in a room which contained a chair, a blanket for the dog to sit/lie on and a water bowl. There was one person present in the room to collect the saliva cortisol samples and put on the heart rate measuring device on the dog. The experiment had 2 different treatments. The first treatment was that the owner petted the dog for 5 minutes and then the dog was exposed to a stressor which was a scary person entering the room with some initial loud noises like bagging the doors. The second treatment included the same scary person making some noise but without the owner petting the dog before. All dogs went through both treatments, and half of the dogs were randomly selected to start with the first treatment and the other half to start with the second treatment. Both treatments were recorded for 60 minutes. The saliva cortisol samples were collected at 0 minutes, 20 minutes and at 60 minutes, and additionally two saliva samples were collected by the owners at home between the two treatments. The heart rate values were collected manually from the monitor of the heart rate device attached on the dogs at the start after some minutes when the heart rate was stabilized, after the scary person (stressor) entered the room and some minutes after the stressor had left the room. The dogs were video recorded and with the program Boris the number of recordings of different behaviours during 20 minutes before during and after the scary person entered the room were analysed. All data was statistically analysed by a Logistic Regression with a generalized mixed model testing the effect of treatment, personality of the dogs and gender with age as a covariate. When the dogs had been petted before the stressor arrived, they showed significantly more sniffing on the floor (p<0.05) and less whining (p<0.05), and tendencies for less growl and bark at strangers (p<0.01). The saliva cortisol levels showed a significant difference with time and treatment. Cortisol tended to be higher when the dogs had been petted before the stressor compared to when they were not petted, but cortisol did not change over time for petted dogs. When the dogs were not petted before the stressor saliva cortisol showed a trend to increase from 0-20 min. Heart rate was not significantly affected by petting in these dogs. In conclusion, petting before a scary person arrived was related to more exploratory dogs that made fewer vocalizations and that did not show an increase in saliva cortisol or heart rate. The stressor is assumed to not have been stressful enough to trigger stress responses in the majority of dogs participating in the experiment. On the other hand, vocalizations towards strangers seemed to be reduced by petting.