Think you understand your dog? Think again.

Dogs can’t talk, but their body language speaks volumes. Many dogs will bow when they want to play, for instance, or lick their lips and avert their gaze when nervous or afraid.

But people aren’t always good at interpreting such cues — or even noticing them, a new study suggests.

In the study, the researchers presented people with videos of a dog reacting to positive and negative stimuli, including a leash, treat, vacuum cleaner and scolding.

Asked to assess the dog’s emotions, viewers seemed to pay more attention to the situational cues than the dog’s actual behavior, even when the videos had been edited to be deliberately misleading. (In one video, for instance, a dog that appeared to be reacting to the sight of its leash actually had been shown a vacuum cleane.)

“When it comes to just perceiving dog emotions, we think we know what’s happening, but we’re actually subconsciously relying on a lot of other factors,” said Holly Molinaro, a doctoral student at Arizona State University and the first author of the new paper, which was published in the journal Anthrozoös.

That bias could mislead owners about their dogs’ well-being, Molinaro said. People who want to be attentive to their dog’s experiences and emotions need to “take a second or two to actually focus on the dog rather than everything else that’s going on,” she said.

The idea for the study was born in 2021, when Molinaro was just beginning her doctoral work in canine emotions but the COVID-19 pandemic had sharply limited her ability to do in-person research.

She was inspired by studies that explore how context clues affect people’s perceptions of others’ emotions. She also was inspired by a distinctly pandemic-era technology: Zoom. The video conferencing software has a feature that blurs out workers’ backgrounds.

Molinaro and her adviser, Clive Wynne, a canine-behavior expert at Arizona State, began to wonder if they could do something similar, creating videos that allowed people to see a dog’s behavior without seeing what was unfolding around it.

And so, while visiting her parents, Molinaro began recording videos of her family dog, Oliver, a 14-year-old pointer-beagle mix, interacting with Molinaro’s father.

In some of the videos, Molinaro’s father did things that Oliver was likely to respond to positively, such as show him his leash or a toy. In others, he did things that were likely to elicit more negative reactions, such as gently scold Oliver or present him with Molinaro’s cat, Saffron. (“He was not a fan,” she said.)

Then, after a crash course in video editing, Molinaro made versions of each video that removed all of the situational context, leaving footage of Oliver, alone, on a black background.

The researchers asked hundreds of undergraduates to watch both sets of videos and assess Oliver’s emotional state in each clip. When the subjects evaluated the original videos, they rated Oliver’s emotions as more positive in the positive scenarios than in the negative ones. But when the context was removed, they rated Oliver’s emotions as equally positive in both types of situations.

Then, the scientists took things a step further by splicing together footage from different situations — showing, for instance, Molinaro’s father presenting a vacuum alongside footage of Oliver’s response to seeing his leash.

Viewers seemed to be swayed more by the context than by Oliver’s behavior. When Molinaro’s father was depicted doing something positive, subjects judged Oliver’s emotions to be positive, even if he had been filmed reacting to something negative.

“There’s no evidence at all that people actually see the dog,” Wynne said. “They seem to have a sort of a big blind spot around the dog.”

The study has limitations, including that it was based on the behavior of just a single dog. People might also perform better when asked to evaluate the emotions of their own dogs, Wynne said, and probably would have noticed signs of intense terror or trauma. (The scientists did not subject Oliver to any extremely negative experiences.)

Still, he hoped that the study will be a wake-up call for pet owners.

“I’m taking it to heart in my own life,” said Wynne, who recently adopted a retired racing greyhound. “I’m making it a project to learn how she expresses herself. Because if I know what makes her happy and unhappy, well, then I can guide her life toward greater happiness.”

Alas, Oliver did not live long enough to see the study published. “But it’s sweet that he’s memorialized in this research,” Molinaro said.