Dog's Brains Are Tuned to Recognize Human Faces

Dogs have a region of the brain that reacts equally to human and canine faces.

Posted May 21, 2015

Just to see how well dogs recognize and interpret human faces and human emotional expressions a few of us played a psychological game. We placed a dog on a table and then we had a person that the dog was not very familiar with look directly at the dog. This person was then instructed to either smile or fake a scowling expression. Of course the game was to watch the dog's reactions. We next had the dog's owner do the same thing. We repeated this for an additional four dogs. Since this was simply being done as a sort of demonstration the data which we kept was quite casual. Nonetheless it was apparent that all of the dogs responded appropriately to their owners' expressions (approaching or wagging their tail to the happy smile or acting submissively, backing away, dropping their tails to the negative expression). On the other hand, while the dogs did pay careful intention to the strangers' faces they did not seem to respond much at all to their positive expression. The dogs did respond somewhat to the stranger's negative emotional expression, although nowhere near the degree that they responded to their owners.

Obviously the results of our informal game would not qualify as scientific data, however it does seem to confirm a lot of recent research which shows that dogs watch human faces, read the emotional reactions that they see on recognizable faces, and modify their behaviors on the basis of what they see (click here or here for examples). How the dogs do this is a bit of a puzzle. Some data suggests that some of the abilities that dogs have to recognize and read human faces is not based solely upon learning but may have some kind of instinctual component (click here for an example). But what kind of neural mechanism could be responsible for this?

Most neuroscientists would start their investigation by exploring the brains of dogs to get an answer. We already know that the brain mechanisms used to interpret emotions are similar in dogs and humans (see here). What about the brain mechanisms used to recognize faces? Here things get a bit more complicated. There is good scientific data which shows that there is a region in the human brain which seems to be specifically designed to recognize human faces, and if that area is damaged your ability to identify faces is lost. The face recognition area is located in the temporal lobe of the brain (a region located on the side of the brain, at the bottom middle part of the cortex which would be right behind your ears). Neurological measures and brain scans have shown that the activity in this part of the brain increases when a person is looking at a human face, or something that approximates a human face. Other primates, including monkeys, have a similar area in their brains which responds specifically to monkey faces. Two other species have been measured, namely sheep (where this area of the brain reacts to sheep-like faces) and crows (where this area responds to bird-like faces). Given such already existing data, it would not be surprising if scientists might find that this same temporal region of the brain of dogs might respond to dog-like faces. However when researchers tried to confirm this they found a bit of a surprise.

I recently received a preprint of the report of an investigation by a team of researchers headed by Daniel Dilks at Emory University. This study was specifically designed to look at how faces are processed in the dog's brain*. The method used to study the dog's brain involved functional magnetic resonance imaging (fMRI). This is a technique that indicates the level of activity in specific brain structures by detecting changes in blood flow and oxygen levels. Getting dogs to participate in the measurement of an fMRI is a difficult task, not only because it requires the dog to remain motionless in a confined space for a period of time, but also because MRI machines make a lot of noise (such as whirring gear sounds along with loud clanks and bangs) and such noises might be expected to startle a dog and cause him to move. Because of this it is not surprising to learn that 2 to 4 months of training were required for each of the six dogs measured in this study.

Daniel Dilks, Emory University
Source: Daniel Dilks, Emory University

The experimental methodology is really quite simple (once you have a well trained dog who will lie quietly, and of course the multimillion dollar MRI machine, the computers for analyses, and the highly trained scientists who will interpret the data). Each dog was shown a series of short video clips of dog faces, human faces, objects, scrambled images and random complex displays. Quite consistent with the research team's expectations was the fact that there was a region in the temporal lobe of the dog's brain that seemed to be tuned to canine faces. This area gave a very strong response to such images in comparison to when the dog was viewing videos of objects or random images. Such data indicates that the dog's brain is tuned to facial images of its own species in much the same way that humans, primates, and other social animals are tuned to recognize the faces of their own species. No surprise here, so what is the big deal? The real shocker is that this same area in the dog's brain is also tuned to recognize human faces. The response from this region of the canine brain was virtually identical when viewing the face of a person as it was when viewing the face of another dog!

How could this come about? Well it could be that during the process of domestication, over many thousands of years, we have systematically selected and nurtured those dogs which responded best to human faces. That would certainly help make communications between people and canines easier. During the course of all of that selective breeding it might well be the case that we were unknowingly fostering dogs with brains that were steadily evolving specific mechanisms that were tuned to recognize human faces in the same way that they recognized canine faces.

This current team of neuropsychological researchers summarize their results by noting that, however it arose, this particular brain specialization in dogs came about, it is a significant finding. Specifically, they point out that having a designated area of the dog's brain set aside to process information from people's faces "may help explain dogs’ exquisite sensitivity to human social cues."

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

* Data from: Daniel Dilks, Peter Cook, Samuel Weiller, Helen  Berns,  Mark Spivak, Gregory  Berns, (2015). Awake fMRI reveals a specialized region in dog temporal cortex for face processing. PeerJ PrePrints,, CC-BY 4.0, Open Access (rec: 14 May 2015).