Hey DOG, WHAT are you looking at? Science starts to unlock the secrets of the dog’s brain.
This has been the holy grail of dog research for years.
Every owner swears that their dog knows more about them, than the dog necessarily lets on. That their dog understands their every move, their intention, and emotion.
And much of this might be true. But one lingering question is, are dogs smart enough to understand all of our intentions ? To anticipate our actions IN THE SAME WAY as we process information?
We look at three recent dog studies over the last decade and see the picture of what a dog SEES when it looks at humans.
This blog is based on four recent research papers. And aims to answer:
- Do Dog brains prefer faces of humans or dogs, or the back of their head?
- Can they can ‘read’ some emotions from owners
- can dogs follow a human gaze or finger point when instructed
- Do dogs prefer processing action information from owners faces or from their owners actions ?
These conclusions might shatter or reinforce what you think about your dog – but understanding what the research is suggesting, might help you communicate with your dog even better.
Research is always ongoing, and often limited by funding and technology – but at last, things seem to be moving on quantifying what is going on in your beloved dogs brains, when they look at us.
Dog brains do NOT prefer faces
From a vanity point of view, this is a blow to all human kind. Surely, we are at the centre of our dog’s universe? Or so most of us would like to muse.
But it turns out that “Unlike human brains, dog brains do not contain face-sensitive areas“ ref 3.
And while dogs can gaze into our eyes, it seems that canines do not process information like most mammals do. So even more remarkable that dogs are the oldest and most loved domesticated creatures alive today, and that they seem to be able to do a work around to communicate so well with us.
How did the researches find this out? In 2020 they already knew using fMRI technology (specialised medical scanning technology) that in humans we have a special face network area in our brain, the ‘fusiform face area.’ On the fMRI this area lights up when humans look at human faces. The standard test to see how important the face is to a human, seems to be if the brain area lights up more when a human looks at a face, rather than the back of a human head.
“In contrast, dog brains (50 dogs tested) only showed species preference (ie for dogs). Visual areas had greater activity in response to seeing a dog over a human, and no activity difference between seeing a face vs. the back of the head. “ ref 3
In tech speak “In dogs, the bilateral mid suprasylvian gyrus showed conspecific-preference, no regions exhibited face-preference, and the majority of the visually-responsive cortex showed greater conspecific-preference than face-preference. In humans, conspecific-preferring regions (the right amygdala/hippocampus and the posterior superior temporal sulcus) also showed face-preference, and much of the visually-responsive cortex showed greater face-preference than conspecific-preference.” Ref 3
Dogs can INFER IMPLICIT information from human emotions
It is curious that the above information says that dogs don’t have a space in their brain to particularly process human faces, and that a they don’t care anymore about the front of a head or the back – when watching videos of humans doing tasks for having fun.
Yet another 2022 paper says that a DOG’S “ability to infer emotional states and their wider consequences requires the establishment of relationships between the emotional display (OF A HUMAN) and subsequent actions. These DOG abilities, together with the use of emotional information from others in social decision making, are cognitively demanding and require inferential skills that extend beyond the immediate perception of the current behaviour of another individual. They may include predictions of the significance of the emotional states being expressed. “ REF 4
The experiments were purely based on setting up Silent interactions between humans (giving or receiving objects) – and displaying facial emotions (rather than a lot of physical movement of the human body.
Researchers believe that these behaviour experiments show that “dogs actively acquire information from affective cues produced by people and make functional use of this to solve ecologically relevant problems. “ ref 4
The experimenters then infer that this means that the dogs were “able to infer the emotional states of people from representations they have generated and stored in their memory, probably based on their previous experience with each type of stimulus.” Ref 4
There is a lot more analysis that the authors of the experiment describe in their ‘discussion’ section.
However at this point given the other information I show in this blog I am not convinced of how much dogs read every nuance of facial human expression OR EMOTION, or are using some other clue about what is happening and what they should do.
There is no doubt that dogs can be very smart – but it seems that the experiments continue to be explorative and results extrapolative rather than definitive.
Do we know if dogs know what we are looking at?
It was also an important question, because it was found that in ‘Gaze following’ to distant space that a human stared at, was found in “many species such as primates, domesticated goats, several bird species, dolphins, fur seals, the red-footed tortoise and wolves.”
Unfortunately, the results for dogs were found to be mixed.
Dogs were able to “follow human gaze to objects such as food or toys, but not for the comparatively simpler task of following gaze into distant space.” Ref 2
The experiment that the results were drawn from had “The experimenter obtained the dogs’ attention using its name and the command “watch” after which the experimenter turned her head swiftly to look at the door of the testing room in the test condition, or looked down to the floor next to her feet in the control condition. If the dogs responded by looking at the door within two seconds in the test condition but did not look at the door in the control condition, a gaze following response was recorded.” Ref 3
The good news is (if you are after your dog seeing where you have thrown your ball in the park) is that “Dogs’ tendency to follow human gaze is influenced by training for eye contact“. And it was independent of age – ie older dog age didn’t decrease the dogs ability any more than at any younger aged dogs.
ALSO “Dogs which had a higher amount of formal training over their lifespan showed a lower gaze following response compared to dogs with little or no training. Similarly, short-term training also decreased dogs’ gaze following response and increased gaze to the human face.“
As you can see – unravelling exactly how a dogs brain works, and how it will respond is MUCH MORE complex than first thought.
The final piece of the current dog brain puzzle is actually understanding how they think in general. Or in fact …
How a dog’s brain represents what it sees
This research was only released in September 2022 so it might be the most relevant of all.
Using fMRI technology again, “Scientists have decoded visual images from a dog’s brain, offering a first look at how the canine mind reconstructs what it sees.” Ref 1
The found that “The results suggest that dogs are more attuned to actions in their environment rather than to who or what is doing the action.”
And yes, this kind of dovetails into not all dogs being about to accurately read an owner’s face or their specific emotions, from the face alone.
The scientists used “fMRI neural data for two awake, unrestrained dogs as they watched videos in three 30-minute sessions, for a total of 90 minutes. They then used a machine-learning algorithm to analyse the patterns in the neural data.” This data was compared to mapping that resulted from two humans watching the same videos.
Imagine this, in 2022 to “a limited degree, (researchers) were able to reconstruct what the dogs were looking at !
“Activities included dogs being petted by people and receiving treats from people. Scenes with dogs also showed them sniffing, playing, eating or walking on a leash. Activity scenes showed cars, bikes or a scooter going by on a road; a cat walking in a house; a deer crossing a path; people sitting; people hugging or kissing; people offering a rubber bone or a ball to the camera; and people eating.“
The models worked perfectly for the humans BUT for the dogs “ the model did not work for the object classifiers. It was 75% to 88% accurate, however, at decoding the ACTION classifications for the dogs.”
The overall analysis concluded that “Dogs appear to be less concerned with who or what they are seeing and more concerned with the action itself.””
While the study used only 2 dogs (funding and difficulty of getting unrestrained dogs to watch 3 30 minute videos), it paves the way for refining the understanding further.
CONCLUSION
The recent dog research has show that dogs can seem to read some human emotions, but are not particularly fascinated with the front of our head any more than the back.
Dogs can follow our gaze or pointing gestures – but this varies dependent on training etc.
Overall, it appears that dogs are more ACTIOIN orientated, and might use hearing, and smell where possible as much as visual clues to get the best understanding of a situation.
NONE of this research is conclusive, but it does give us glimpses into the wonderful world of dogs.
References
Ref 1 Machine learning gives glimpse of how a dog’s brain represents what it sees – Emory University September 15, 2022
Ref 2 What are you looking at? Dogs are able to follow human gaze Source: University of Veterinary Medicine – Vienna Date: June 12, 2015
Ref 3 Dog brains do not prefer faces JNeurosci October 5, 2020
Ref 4 Dogs can infer implicit information from human emotional expressions. Natalia Albuquerque, et al. Animal Cognition volume 25, pages 231–240 (2022)