Dogs Can Understand Our Intentions

If my dog isn't sleeping, the odds are most likely that he's watching me—watching me compose blog posts, watching me write computer code, watching me watch movies, watching me watch him. My dog is both a scientist and documentarian of me. With that kind of dedication, it becomes a persistent question of what is he watching for? What does he understand of what I do? Does he just remember patterns of my behavior, or does he perceive what my intentions are? In other words, what can he know of why I do the things that I do? Patterns or understanding.

According to a study out last week in the open-access journal PLOS ONE, dogs are perhaps more capable of understanding human intentions than they're typically given credit for. In experiments funded by the European Research Council and led by University of Portsmouth researcher Sarah Marshall-Pescini, dogs were found to have the capability for this sort of understanding of at least a human infant or chimp.

"The question of whether non-human species can perceive other's actions as goal-directed has been more controversial," Marshall-Pescini et al write, "however there is mounting evidence that at least some primates species do. Recently domestic dogs have been shown to be particularly sensitive to human communicative cues and more so in cooperative and intentional contexts."

Showing that human infants and-or dogs can recognize human intentions is challenging for obvious reasons, e.g. neither can very well explain themselves. The set-up used in the current study is the same as a classic set-up used to test infant awareness.

"[In a 1998 study, AL] Woodward found that when 5-month old infants repeatedly observe a person interacting with an object, they will then look longer when the actor suddenly switches to interacting with a different item in the same location than when they see the actor interacting with the usual object placed in a new location," the current paper explains. "The 'surprise' shown by infants in this paradigm has lead authors to conclude that infants understand the actor's action as being goal-directed to a particular target."

Essentially, Marshall-Pescini and her team just rewrote the experiment for dogs. "Similarly if dogs, perceive the actions of the animate agent as goal-directed," they write, "we expect a significantly higher looking time when they see the person (animate agent) interact with the novel object compared to the usual one (in the new place) but this pattern of results should not however occur with the inanimate object (i.e. the black box)." And this is more or less what the current study found.

The dogs reliably kept their gaze focused on the actions of the person as they switched objects from the familiar to the unfamiliar, but less so when the person kept the familiar object and moved to a different location. The suggestion here is that the dog is observing a relationship between person and object rather than simply remembering the object itself. This indicates some perception of intention.

Marshall-Pescini notes that it's possible that the dogs were interested in the first scenario (new object, old location) for other reasons. They might have expected to the new thing would result in some new play or food opportunity, yet the study offers that, "once the dogs were released very few investigated the objects at all, suggesting that there was no strong expectation of finding food."

The current study ultimately is, however, more proof that the infant looking-time experimental model can work in dogs than it is definitive proof that dogs understand intentions on a "mentalistic" level, e.g. understanding relationships between a person an object as representations of some internal state of the person. Instead, the dogs might be making a statistical or inferential connection just by virtue of observing the setup again and again. More research is needed.

Nonetheless, I'll continue to believe that my dog, a bafflingly patient and dedicated observer of human behavior, probably knows more about my own intentions than I do. Dogs, after all, have had 30,000 years of co-evolution to get it right.

For a further demonstration, research released last year revealed that dogs steal things (food) because they understand the different conditions (a dark room) in which they could do so and not get busted by human overseers. That study wasn't quite so precise/well-calibrated (however fascinating), but it does give further support to at least us being on the right track when it comes to understanding the ever-deepening understanding of dogs.