How Do Dogs Interpret Human Facial Expressions?

What ever is a dog to make of a human smile? Or a frown for that matter? On the one hand, it seems to me to be trivial for a dog to distinguish between obviously different expressions on the face of a human. Dogs, after all, are highly visual and the preponderence of their social communication is based on visual signals. But here’s the question I’ve always wondered about: What signals from our faces are salient to dogs?

T W Smile smallMy experience has suggested that dogs are exceptionally good at noticing (and interpreting) the following, whether done by a person or another dog: a still body versus a relaxed one, a hard, direct stare versus a soft or indirect gaze, and a loose, relaxed, open-mouth face versus one that has a tightly closed mouth. These are, obviously, signals that appear to be highly salient in canine communication, and my impression is that they transfer from one species to another. However, what of the signals that we humans consciously focus on, like smiles and frowns? Do dogs pay as much attention to them as we do? If so, what aspects of those expressions are salient? Members of both species may be aware of the difference between a look of mild irritation versus extreme anger, but we might be cuing on different things. Could we be focusing on the position of one’s eyebrows while dogs are primarily focused on the stiffness of the head and neck?

We know that people all over the world both express emotions on their faces in similar ways and also interpret them in the same way, no matter what their culture, native country or language group. (See the work of Paul Ekman, who has done over 30 years of research on the universality of human facial expressions, and helped me with my book For the Love of the Dog when I was writing about that issue.)

But we don’t know that much yet about how dogs interpret them. I’ve thought about this issue for years, (see my blog of 2010 on a related study) and so was especially interested in a study making the rounds late last week in Current Biology. Titled “Dogs Can Discriminate Emotional Expressions of Human Faces,” the study by Müller et. al. asked not just if the dogs could discriminate between “happy” faces and “angry” faces, but also if they could generalize what they’d seen from one part of the face to the other.

 Here’s a summary of their study, taken from the study’s abstract:

“After learning to discriminate between happy and angry human faces in 15 picture pairs, whereby for one group only the upper halves of the faces were shown and for the other group only the lower halves of the faces were shown, dogs were tested with four types of probe trials: (1) the same half of the faces as in the training but of novel faces, (2) the other half of the faces used in training, (3) the other half of novel faces, and (4) the left half of the faces used in training. We found that dogs for which the happy faces were rewarded learned the discrimination more quickly than dogs for which the angry faces were rewarded. This would be predicted if the dogs recognized an angry face as an aversive stimulus. Furthermore, the dogs performed significantly above chance level in all four probe conditions and thus transferred the training contingency to novel stimuli that shared with the training set only the emotional expression as a distinguishing feature. We conclude that the dogs used their memories of real emotional human faces to accomplish the discrimination task.”

There’s lots of interesting information here, which makes the study far more interesting than one that just shows dogs can tell the difference between an angry face and a happy face.  Note that the dogs were initially only shown the lower or upper half of the face (from photos on a computer screen) and could 1) generalize its visual features to the face of another person and 2) could generalize from the upper half of the face to the lower half. This is especially important, because it implies that dogs were learning more than a simple visual cue, and matching the emotion expressed by the bottom half of the face (with a big smile for example) to the upper half of the face, with open eyes and relaxed eyebrows. Especially interesting was the result that dogs learned the discrimination more quickly if they were rewarded for cuing on the “happy” face. We have to be careful about interpreting that result, but the author’s suggestion that the dogs recognized the angry faces as aversive is reasonable.

The author’s conclude by asking whether the dog’s abilities shown in the study were based on the dog’s experience as individuals, on selection pressures over time to select for dogs able to better interpret the expressions of humans, whether this ability is simply hard wired into many species of mammals and dogs happen to be one of them. (I would add that all three could potentially  be occurring simultaneously.) I’m happy to say that the last few years have seen a flurry of studies related to visual communication between people and dogs. The research ranges from a study in Argentina by Jakovcevic et. al. that looked at whether breed affected how long dogs would look at their owner’s face without reinforcement (Retrievers did longer than Poodles or German shepherds), to one by Turcsán and colleagues at the Miklosi lab in Hungary, which found that if an owner had a happy expression while handling an object their dog was more likely to retrieve it.

Food for thought, yes? I anticipate that some will read this about this study and think “Well, what a waste of money! Of course we know that!” But actually, we don’t. I’d argue that this is exactly the kind of study we need to do, examining what we think might be true with what really is true. So kudos to Müller et. al. for doing this well-designed study. Can you see me smiling?

MEANWHILE, back on the farm: Brrrrrr. It’s really, really cold out here. (May I mention, however, that I’m grateful we haven’t had 6 feet of snow in a few weeks like New England? I’d love to hear from any of you slammed by snow. How are you doing?!) Saturday’s high here in Wisconsin was about 6 Farenheit at the farm, but the problem was the wind. It was strong enough that you just couldn’t stay warm, no matter how bundled up you were. At least, I couldn’t, so much of the day was spent inside. All the dogs are learning new tricks, including “Find and nose-touch a white square any where in the room” (Willie), “Roll over” (Maggie), “Spin on a verbal cue only” (Tootsie). Nothing especially creative, but still fun. I’d say after a few sessions that the only thing warm on the farm was the clicker, which always sees a lot of use from me when it’s crazy cold outside.

No wind today, so even though it’s cold you just need the right clothes to be comfortable outside. (You know that saying? There’s no bad weather, just bad clothes? I’m in agreement, except when it’s windy.) The BCs and I took a nice long walk in the woods just now, me armed with my camera and the lesson from my Contemplative  Photography class to see the world in a new way. Here’s what I came up with:

bark 2-2015

Oak leaves 2-2015

Authors: Trisha

Read more

Powered by C2C Consulting, LLC