Repetitive behaviour in kennelled domestic dog: stereotypical or not?

Repetitive behaviour is common in kennelled dogs, yet its motivational basis remains relatively unexplored. We examine the repetitive behaviour of 30 kennelled working dogs in ten contexts both coinciding with, and in the absence of, commonly occurring arousing stimuli, such as care staff, other dogs and food preparation. A large proportion (93%) of subjects performed some repetitive behaviour, most commonly bouncing, but only 17% in the absence of the arousing stimuli. Subjects could be divided into four groups according to the stimuli eliciting, and the duration, of their repetitive behaviour, and these groups were compared on the basis of their cortisol response to an acute psychogenic stressor--a veterinary examination. Urinary cortisol/creatinine response curves differed significantly between the groups. In particular, those dogs which performed repetitive behaviour at times of minimal stimulation, showed a distinctly different pattern of response, with cortisol levels decreasing, as compared to increasing, after the veterinary examination. We conclude that dogs showing repetitive behaviours at times of high arousal are motivationally distinct from those "stereotyping" in the absence of stimulation. We suggest that those dogs showing spontaneous repetitive behaviours may have past experiences and/or temperaments that affect both their reactions to a veterinary examination and to long-term kennelling. For example, some dogs may find isolation from humans particularly aversive, hence affecting their reactions both to being left in a kennel and to being taken to the veterinary surgeon. Alternatively, such dogs may have atypical responsiveness of their hypothalamic-pituitary-adrenal (HPA) axis, possibly brought about through chronic stress. High levels of repetitive behaviours in response to inaccessible husbandry events may be explained if such behaviour has inadvertently been reinforced by attention from staff, and therefore may not always be indicative of aversion to kennelling or compromised welfare.