Investigating the reliability of a separation : and reunion test performed on pet dogs (Canis familiaris) and their owners

Sammanfattning: Our long mutual history together with the dog (Canis familiaris) has influenced health, physique, wellbeing, as well as behaviour – on both parts. Today, we refer to dogs as family members rather than working tools. Previous research has shown that a dog and its owner show similar attachment behaviour towards each other, as would a child and its parent. The urge to express attachment behaviour is said to be activated by stress, e.g. during and after a separation from each other. Therefore, separation and reunion are commonly used when evaluating the relationship quality. However, this method had not yet been evaluated for its reliability (or trustworthiness), which this study aimed to correct. Nor has its validity been studied (whether the test actually measures what it is supposed to measure). This, however, was only briefly investigated due to a small sample size. Fifteen dogs and their owners participated in a separation- and reunion test at two separate occasions each, consisting of a departure phase (the owner leaves the dog), separation phase (the dog is alone for three minutes), and a reunion phase (the dog and owner reunites for three minutes). As an attempt to initiate a validation of the test, the owners answered four different questionnaires regarding their adult attachment style (how they bond and interact with other people), their relationship to their dogs, and their dogs’ personalities. The adult attachment style scores of the owners were then correlated to their owner-dog relationship scores, the dog temperament scores, and owner and dog behaviour in the separation and reunion test. The results showed no difference in behaviour between the two tests, indicating that daily conditions or mood did not affect the behaviour of the dog nor the owner during the test scenarios. There was a correlation between owners being more anxious and dogs paying more attention to and staying close to them when the owners were leaving the room during the test. These dogs also explored less. When alone, owners that were more anxious had dogs that whined or barked more often during separation. During the reunion phase of the test, more avoidant owners did not talk much to their dogs and in return the dogs paid less attention to them and did not initiate much physical contact, but instead explored the room. Regardless of owner adult attachment style, dogs that scored high on attention seeking in the C-BARQ paid more attention towards the door during separation. In addition, owners who considered themselves emotionally close to their dogs in the MDORS had to a larger extent dogs who explored the room both during departure and reunion in the test. Emotional closeness also correlated positively to owner and dogs spending a lot of time together on a daily basis (MDORS). We can only speculate, but the results may indicate that dogs owned by more anxious owners are more sensitive to external challenges that many dogs are exposed to, such as meeting unfamiliar people, being in a novel environment, being left alone, etc. Dogs having more avoidant owners may be less dependent of their owners or having different strategies to cope with stressors as compared to if the dog instead had a more anxious owner. In conclusion, we can consider this method as being reliable for use in the study of dog and owner interaction and relationship quality. Moreover, results indicate that the test is valid, based on the associations between questionnaires and the behavioural test, but these results should be interpreted with caution as the sample size was small and because it was difficult to identify a proper measure to check validity against (i.e. a comparable template).