Interactions between shelter dogs: some new research

When dogs at a shelter are housed in groups, it helps them engage in normal canine behaviours--and dominance does not explain them.

By Zazie Todd, PhD

Some animal shelters house dogs in pairs or small groups. This can enrich their lives, but it could also potentially be a source of stress if the dogs are not well-matched. A new paper by Irena Petak, of the University of Zagreb, Croatia, examines the communication patterns between dogs housed in groups.

At the Dogs Trust in Salisbury, England, there is a sanctuary for long-term residents.  There is a ‘mountain area’ with an artificial mountain and three kennels, and a tree area with grass and trees. There is also a small introductory pen for new dogs who are coming in to the sanctuary. The sanctuary is enriched with a sand box, tunnels through the mountain, ramps and toys for the dogs to play with. During the day, the dogs are allowed to run free in the enclosure, and at night the dogs can choose one of the three kennels to go into. 

At the time of the study, there were twelve dogs in the sanctuary and two in the introductory pen. All of these dogs were neutered males who had been re-homed several times. The time they had spent at the sanctuary varied from one week to seven years, and their estimated age varies from seven to twelve years.

The dogs were observed for 162 hours, during the day-time, over a period of seven weeks, always at times when the care-takers were not there so the dogs could interact as they wished. Interactions during this time were recorded, although a few were missed because it was a large area (for example, sometimes trees were in the way).

The results showed that dogs interact with the other dogs in different ways. This is as expected, but it emphasizes the need to be careful in selecting dogs to share housing, as different dogs have different preferences. 

The interactions between pairs of dogs were grouped into proactive neutral, proactive aggressive, and reactive scent-marking (reactions to the scent-marking of other dogs). Aggression was characterized by things like growling, snarling, mounting, attacking and chin resting.  Group interactions were classified as vocal, visual or olfactory, according to the behaviour of the dog that initiated the interaction.

The most common type of interaction was proactive neutral, followed by reactive scent-marking. Aggression was the least common type of behaviour. Individual differences included one dog that tried to initiate many proactive neutral interactions, a couple of dogs that were responsible for most of the aggression, and one dog that didn’t really interact with the other dogs. There were also two dogs who barked the most, and often other dogs barked in response to hearing them. Some dogs were more active than others in exploring the environment and participating in social behaviours.

There was an interesting pattern in that the dogs who started olfactory interactions with other dogs tended also to receive this kind of behaviour at other times. On the other hand, dogs who initiated visual or vocal group interactions were not also the recipients of such behaviour.

There was a typical pattern to interactions between two dogs. It was “usually initiated by one dog approaching another dog and was followed by sniffing body parts. Recipients frequently did not try to stop the initiator from sniffing them and did not try to sniff the initiator.” 

There was also a lot of scent-marking and sniffing of places where other dogs had urinated. The frequency of scent-marking and sniffing was such that Petak suggests that olfactory communication should be considered as enrichment activity.

The patterns of interactions between the dogs are very complex, and do not support the idea of a dominance hierarchy in dogs. The scent-marking that was observed also cannot be linked to dominance or aggression. 

It is interesting that the two dogs that most often initiated aggression were also most likely the recipients of it. This suggests that some dogs can have relationship problems over a long period of time. However, it should be remembered that these encounters were rare, and not serious.

The results suggest it is important to match dogs carefully, and this would apply to group-housed shelter dogs and to people adopting a second dog as a friend for one they already have. Matching activity levels is important, as if one dog is too boisterous it may make the other dog unhappy. Similarly, if one dog vocalizes a lot, this may stress another dog.
This study shows that housing dogs in groups enables them to engage in normal, social behaviours.  Many shelters already house dogs in pairs or groups, and other shelters may wish to follow suit.

For anyone interested in reading the full paper, it is freely available for a limited time (along with some other papers about animal shelters) thanks to a collaboration between the ASPCA, the Animals and Society Institute, and the Journal of Applied Animal Welfare Science.  Just click on the link below.

Do you have more than one dog? If so, how do they get along?

Reference
Petak, I. (2013). Communication patterns within a group of shelter dogs and implications for their welfare. Journal of Applied Animal Welfare Science, 16(2), 118-139.