Dogs Might Really Understand Your Intentions, Study Finds

In first-of-its-kind experiment, dogs behaved differently depending on whether the actions of a human experimenter were intentional or unintentional.
September 1, 2021, 2:07pm
In first-of-its-kind experiment, dogs behaved differently depending on whether the actions of a human experimenter were intentional or unintentional.
A good dog. Image: Emanuel Maiberg
210329_MOTHERBOARD_ABSTRACT_LOGO
ABSTRACT breaks down mind-bending scientific research, future tech, new discoveries, and major breakthroughs.

The bond between dogs and humans is so strong that people often feel as if their furry friends can understand their thoughts and intentions. However, it’s one thing to sense that your dog might be aware of the motives behind your actions, and quite another to scientifically test this facet of canine cognition.

Now, researchers have done just that by running a common experimental setup, known as the “unable vs. unwilling paradigm,” for the first time with dogs. Probing the capacity of dogs to distinguish human intentions could help establish to what extent they possess “theory of mind,” or the ability to ascribe mental states to others, not to mention that this skill “would have been of immense value for dogs’ common history with humans,” according to a study published on Wednesday in Scientific Reports.

Advertisement

The study addresses “a huge question about whether animals have a theory of mind,” said  Juliane Bräuer, head of the DogStudies Lab at the Max Planck Institute for the Science of Human History and senior author of the new study, in a call. “This overall question is very important in comparative psychology. This is what we do: we compare animals with humans and look at what is unique.” 

The “unable vs. unwilling paradigm” is a setup that involves the denial of rewards in unintentional (unable) or intentional (unwilling) contexts. In other words, it tasks the animal with intuiting whether an action is an accident, or on purpose. This framework has been applied in previous studies to human infants, chimpanzees, African grey parrots, capuchins, Tonkean macaques, and horses, but the new research represents the first time that the paradigm has been adapted for dogs.

This canine version included three scenarios, all of which placed a transparent partition between a female experimenter and a dog. A total of 51 dogs (27 female and 24 male) of various breeds and ages were tested in the study. 

Screen Shot 2021-09-01 at 9.28.58 AM.png

Dogs were fed through the gap before the experimenter started to withhold the reward intentionally or unintentionally. Image: Josepha Erlacher

In the “unwilling” scenario, the experimenter offered a food reward to the dog through a gap in the partition, then withdrew the treat back and exclaimed “ha ha!.” She then held it in front of her where it could be seen, but not reached, by the dog. 

The experiment also featured two “unable” scenarios. In the “unable-clumsy” test, the woman tried to feed the dog through the partition, but dropped the treat on the wrong side and said “oops!” In the “unable-blocked” condition, the woman tried to feed the dog, but failed because the partition gap was blocked. In this case, the verbal cue was “oh!”

Advertisement

The idea was to study the dogs’ reactions to different implied motives of the human treat-dispenser. In the unwilling scenario, the reward seemed to be intentionally denied to them, whereas in the two unable scenarios, the treat was unintentionally withheld.  

The results revealed that dogs “clearly behaved differently depending on whether the actions of a human experimenter were intentional or unintentional,” though there could be many explanations for this “surprise” outcome, according to the study.

Most significantly, the dogs waited for longer periods in the unwilling scenario before they decided to walk around the partition to get closer to the withheld reward. The dogs were also more likely to sit, lie down, or wag their tails in this scenario, perhaps because they thought these appeasing gestures might earn them back the treat.

“They maybe understand: ‘I am not getting this food now, but maybe when I do something right, when I sit down and I’m a nice dog, she will give me the food,’” Bräuer explained. “They expect that she will change her mind.”

The dogs also waited slightly longer in the “unable-clumsy” condition compared to the “unable-blocked” scenario, perhaps because they understood that there was no easy way to get the reward from the human when the gap was blocked, spurring them to walk around the partition.

Bräuer’s team emphasized that their study does not confirm that dogs possess “theory of mind,” nor do their results necessarily imply that these pets can perceive intentions in humans, even though they display distinct responses to them, as there are other explanations for the behavioral differences. 

“Future research needs to address whether dogs’ distinguishing reaction really reflect a capacity to read human intentions or only some form of behavior reading based on learned associations,” the researchers said in the study. “Nevertheless, our findings provide important initial evidence that dogs may have at least one aspect of Theory of Mind: the capacity to recognize intention-in-action.”

The experiment also confirmed what any dog owner already knows: these domestic animals are profoundly attuned to human actions and behaviors.

 “They are extremely sensitive in that they can distinguish between these three situations that are quite similar,” Bräuer said. “Their sensitivity makes them very special and maybe they can even understand intention.”