How good are people at interpreting the behavior of dogs? Dogs and people have been around together for perhaps 100,000 years, so you would think that people would be pretty good at it. However, it turns out that people aren't very good at all.
A recent article in the journal Applied Animal Behaviour Science (Tami, G. and A. Gallagher, 2009, Descriptions of the behavior of domestic dog (Canis familiaris) by experienced and inexperienced people, Applied Animal Behaviour Science 120: 159-169) tested 60 people in the vicinity of Edinburgh, Scotland, on their ability to interpret dog behavior.
The 60 people were divided into four groups: dog owners, veterinarians, dog trainers, and non-dog owners. The non-dog owners had no experience with dogs, while the others had considerable experience. Dog owners had a median of 30 years of experience, dog trainers had a median of 20 years of experience, and veterinarians had a median of 11 years of experience.
The authors of the study first walked a border collie past 8 other dogs on separate occasions and recorded video clips of the responses of the other dogs to the border collie. They then categorized the responses of the dogs as aggressive, confident, fearful, friendly, submissive,, play solicitation, actual play, defensive, or indifferent.
The 60 people were shown the video clips and asked to describe the behaviors. As we can predict from the long association between dogs and humans, both dog-oriented people and non-dog people had similar descriptions of the dogs' behaviors. Except for two categories of behavior (confidence and play solicitation), there were no significant differences in the responses of any of the groups.
The amazing thing is that both the dog-oriented people and the non-dog people got many of the behaviors wrong.
Both the dog-oriented people and the non-dog people could mostly identify indifferent (90% of observers scored correctly), and friendly behavior (73 percent scored correctly). Fearful behavior was correctly identified by 67% of the observers, and play solicitation behavior was correctly identified by 62%.
However, for the other behaviors, both the dog-oriented people and the non-dog people got things mostly wrong. Aggression was correctly identified by only 38% of the observers, confident behavior was also identified by only 38%, and actual play (as opposed to play solicitation) was correctly identified by only 30% of the observers.
Most of the observers seemed to focus primarily on the movement of the tail, rather than on all of the other signals that a dog was displaying.
This study points out some interesting things. One is the possibility that neither dog owners nor veterinarians necessarily know enough about dog behavior to always correctly identify the behavior of a dog. Another is that being around dogs does not necessarily give people an edge in interpreting the behavior of dogs. A third thing, on a positive note, is that most people can identify a friendly dog, even if they can't identify an aggressive one.
--Con Slobodchikoff
As commented before, it would be great to see the videos. I probably incorrectly identify my own dogs' behavior even though I know them well. It is best for any "dog people" to research animal behavior to help them understand their dogs. A little knowledge is better than none.
Posted by: Shaun | October 22, 2009 at 09:45 AM
It would be a lot more interesting and valuable had the videos themselves been posted so we could come to our own conclusions regarding the 'correct' answers, along with a breakdown of what each respondent said regarding each video.
Posted by: calmassertiv | October 15, 2009 at 10:59 AM