I was working on an attention exercise with my young Nova Scotia Duck Tolling Retriever, Ranger. A colleague, who is a biopsychologist and studies animal behavior, stopped to watch me. The exercise is a very simple one and involves teaching the dog the command "Watch me." When the dog learns it, the command should cause the dog to look at my face. Once the dog is looking at me I can be sure that I have his attention and I can give him another command knowing that he is not distracted by something else.
It is an easy training procedure where you show the dog a treat, raise the treat toward your eye, then lower it toward the dog's face while you say "Watch me" and then give him the treat. Sometimes you raise the treat toward your eye and lower part way then return it toward your face and lower it again all the while repeating "Watch me" finally giving it to the dog on the second or third time.
My colleague asked me what I was doing and when I explained that it was simply a method of capturing the dog's attention and having him look at my face he smiled and said "I have a much simpler way to get your dog to look at your face — just scowl at him and make an angry facial expression. There is a lot of data which shows that when animals see an angry or aggressive expression on another animal, they will stare at that animal's face. Now I know that you are going to protest because you are positive trainer and you don't want to bring negative emotions into the training situation. However you could do it in a two-step fashion, starting by making an angry expression, which will capture the dog's attention, and then once the dog is looking at you can smile and reward him. In effect, I suppose, you are teaching the dog that the angry expression is the signal to watch you."
He paused for a moment and then added "As far as I know the research is only been done on conspecifics [the scientific word for animals of the same species], but it should still work since we already know that dogs do look at human faces and are capable of reading the emotions on them (click here for more on that)."
My colleague's idea was an interesting one, although I was much more worried about the negative effects that the angry facial expression might have on the dog's performance then he was because of some other research that I knew about (click here for more than that). However I did think that it was worth a search through the literature to see if it is true that dogs pay more attention to an angry human face than one which is showing more positive emotions.
My search was rewarded when I found a fairly recent article by a team of researchers from the University of Helsinki in Finland. The lead author of the report was Sanni Somppi and it was published in the journal PLOS ONE*. The study looked at the responses of dogs when they were presented with photographs of people or dogs with different facial expressions.
The experimental setup was rather interesting. A group of 31 dogs were shown a large set of photographs on a computer monitor. The faces could either be those of an unfamiliar person showing a threatening, pleasant, or neutral facial expression, or those of an unfamiliar dog (they used a variety of different breeds) also showing a threatening, pleasant, or neutral facial expression. These faces flipped by at a moderate rate of speed on the screen.
The really neat thing about this experiment is that close to the monitor, where the photos were appearing, was an apparatus which used infrared light bouncing off of the dog's eyes to measure the animal's eye movements. The main question which of these investigators were asking was whether dogs look at the facial expressions of people and dogs in the same way, and the general answer that they got was "Yes." When dogs are looking at the faces, regardless of whether the face is human or canine, they seem to glance first at the eyes. They next tend to look at the middle of the face where the nose is. In dog faces this is a very informative location since different emotions are associated with different patterns of wrinkles on the canine snout. They look least at the mouth (which is a bit different than people who spend a lot of time looking at the mouth to decode an emotional expressions).
Just as my colleague had said, when looking at the emotional expressions of other dogs, the dogs tested paid much more attention when the expression was threatening. They certainly looked at the canine faces with negative emotions for a longer time. The surprise was that this was not the pattern which the data showed when the dogs were looking at human facial expressions. Here the results were reversed so that the dogs actually seemed to be trying to avoid looking at the negative or threatening human faces.
The reason for this difference is not clear. We can speculate that it might have to do with evolution and the genetic wiring of animals. It makes sense for an animal to be born with a tendency to pay attention when another animal of its own species has a threatening expression. After all that threatening look can be a warning of impending aggression. When it comes to a negative expression that a dog sees on a person, however, this might also suggest that something bad may happen (or at least nothing good is about to happen). But because dogs and people are different species his understanding of the expression may be learned through interacting with humans. Anticipating something negative might occur is not pleasant, and people tend to avoid coming in contact with situations which do not seem to be set up for a happy outcome, and so it appears to be with dogs. The dogs often engage in a sort of "ostrich — head in the sand" behavior, to avoid confronting anything negative, and in this case it involves not looking at or paying attention to a human face which has an angry expression.
So based on this data my colleague's suggestion that a scowling face might help attract the dog's attention clearly seems to be wrong. This left me feeling more secure in using my food-based "Watch me" training technique.
As a postscript to this, I mentioned the fact that dogs tend to avoid looking at angry faces to another colleague of mine. She laughed and told me "When I had just graduated from high school my mother entered me in the 'Miss Kansas City' beauty pageant. There was a woman associated with the event who taught us how to walk and stand, and I remember her telling us 'If you want the judges to look at you keep smiling. One frown and they will never look at you again.' I wasn't particularly happy being primped or being stared at by everyone and so I probably didn't smile much, or maybe even frowned a bit. In any event, when I failed to win the pageant my mother accused the judges of just being a bunch of dogs. Maybe she was right!"
Stanley Coren is the author of many books including: Gods, Ghosts and Black Dogs; 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
* Data from: Somppi S, Törnqvist H, Kujala MV, Hänninen L, Krause CM, Vainio O (2016) Dogs Evaluate Threatening Facial Expressions by Their Biological Validity – Evidence from Gazing Patterns. PLoS ONE 11(1): e0143047. doi:10.1371/journal.pone.0143047