It was at an advanced dog obedience class when a cluster of students had gathered together to chat about a dog obedience competition that was coming up on the weekend. Ellen* was quite concerned because she had entered her dog Chucky who only needed one more qualifying score to earn his open obedience title. However Ellen was scheduled for some surgery on Thursday and it was certain that she would not be able to attend the competition. Sue had offered to take Chucky to the show and handle him in the ring. Sue knew the dog reasonably well although she had never trained him or worked with him in obedience exercises. The conversation was centering on whether the dog would work for someone who was not his owner, and some opinions were that having the dog compete with a different handler might not only be ineffective but might even do some psychological damage and impair his future performance and sense of security.
Most of us feel that our dogs respond best to us, their owners, and some of us may not be sure that our dogs can be as reliably controlled by other people — not just strangers but even people who that are familiar to him. Actually, although such feelings are probably universal among dog owners, there have been very few systematic studies that have tested this notion. However, a recent study directly bearing on this issue has been accepted for publication in the journal Behavioral Processes**. It is authored by a team of researchers headed by Andrea Kerepesi at Eötvös Loránd University in Budapest, Hungary. Specifically this research tested various interactions between dogs and humans changed depending upon the dog’s familiarity with the person. In this case the levels of familiarity used involved the dog’s owner, someone who the dog was familiar with, and a complete stranger who had only interacted with the dog briefly prior to experimental testing. The researchers whether familiarity played a role in a variety of different situations, some involving simple social interaction, and some involving obedience and control.
It should not be surprising to find that the data revealed that in a strange situation the dogs will tend to seek out their owners and spend more time with them. The familiar person and the unfamiliar one were relatively ignored in that situation. Similarly it is not surprising to find that in a play setting the dogs preferred interacting with their owners, treating the familiar and unfamiliar people indifferently. In a situation where the dogs are threatened (being directly approached by a person wearing a dark raincoat who is moving stiffly and staring at them) 85% of the dogs ran directly to their owner. In a situation where the dog was to be physically manipulated (rolled on its back) the owner was most successful and the familiar person was no more successful than the stranger. All of these tests might be viewed as involving social interactions rather than obedience and control, and one would expect that the bond between the owner and the dog would show itself in these. However what about situations involving obedience to commands and control?
This study employed three tests which looked at the ability of familiar and unfamiliar people to control the dog’s behavior. In the first one a piece of food was placed in a wire kennel crate where the dog could see and smell it but could not reach it. The task of each individual was simply to call the dog to come away from the location of the food and to approach them. The result was unambiguous with the dogs responding best to the owner and making no distinction between the familiar and the unfamiliar person. The second test of control was a little bit more tempting. In this one each of the people had a piece of sausage strapped to their leg. The dog was moved around to allow it to see that their owner, the familiar person, and the unfamiliar person each had some of the mouth-watering bait. When the dog approached each person and nosed at the bait they were given one forceful “No!” command to inhibit the dog from taking it. The dog was then released and there was no further verbal interaction. The dog was allowed to do as it wanted with no further restrictions or commands. Overall the dogs seem to recognize that taking the food was prohibited and 60% of them did not take the food from anyone. For the dogs who disobeyed, they were much more likely to take the food from their owner or the familiar person and least likely to take the food from a stranger.
What about the influence of a person’s familiarity on the dog’s response to formal obedience commands? After all that was the issue which started this discussion. This test was carried out an open field and the dog was allowed to move freely. The owner, the familiar and the unfamiliar person were each allowed in turn to see how the dog responded to common obedience commands. Specifically they were told to call the dog, command it to sit, and lastly to make it lie down and stay in the down position for 15 seconds. Only gestural signals and verbal commands were allowed — food reward or touching the dog were prohibited. Each participant had one minute to complete each task. Two measures were taken: whether the dog obeyed the command, and how long it took for the dog to respond.
Since these were pet dogs that were being tested rather than obedience competitors, perfect performance was not expected from all of them. In the recall test, 100% of the dogs came to their owner, 90% came to the familiar person and 70% came to the unfamiliar person. In the sit command 95% of the dogs responded correctly to their owner, 60% to the familiar person and only 45% to the unfamiliar person. For the down and stay command 90% of the dogs responded correctly to their owner, 55% to the familiar person, and only 35% to the unfamiliar person. What is clear from these results is that the dogs respond best to their owner, they respond acceptably (more than half of the time) to the familiar person and their obedience performance is poorest for the unfamiliar person. However that was not the full story.
It is when we measure the time that the dogs take to respond to obedience commands that we find the largest effects. The dogs respond much more quickly to their owner then to either the familiar or the unfamiliar person. For example in the recall command the dogs respond in an average of 3 seconds to their owner, but take 11 seconds for the familiar person and 21 seconds for the unfamiliar person. The sit command is responded to in 8 seconds when it is given by the owner, 23 seconds when given by the familiar person and 27 seconds when given by the unfamiliar person. Finally, dogs respond to the down command in 12 seconds when the order is given by the owner, 29 seconds when given by the familiar person and 34 seconds when given by the unfamiliar person.
The overall conclusion to be drawn from this research is that dogs respond to obedience commands best to their owner, less reliably but still mostly adequately to a familiar person, and the poorest to an unfamiliar person. However even when they are responding well to either the familiar or the unfamiliar person they are certainly going to respond much more slowly than they do for their owner.
Stanley Coren is the author of many books including: The Wisdom of Dogs; Do Dogs Dream? Born to Bark; The Modern Dog; Why Do Dogs Have Wet Noses? The Pawprints of History; How Dogs Think; How To Speak Dog; Why We Love the Dogs We Do; What Do Dogs Know? The Intelligence of Dogs; Why Does My Dog Act That Way? Understanding Dogs for Dummies; Sleep Thieves; The Left-hander Syndrome
Copyright SC Psychological Enterprises Ltd. May not be reprinted or reposted without permission
* The names of the people and dogs have been changed lest they stop talking in my vicinity.
** Data from: Kerepesi, A., Dóka, A., Miklósi, Á. (2014), Dogs and their human companions: The effect of familiarity on dog-human interactions, Behavioural Processes, http://dx.doi.org/10.1016/j.beproc.2014.02.005