Medical Moment: Teaching dogs to communicate

Published: Nov. 10, 2022 at 5:39 PM EST
Email This Link
Share on Pinterest
Share on LinkedIn

(WNDU) - If you’ve ever tried to teach your dog commands like sit, stay, fetch, or rollover, you know it can be a tedious job.

Now, imagine teaching your dog to use words and phrases to ask for what they want and share with you how they feel. One of the largest studies of its kind is underway to prove dogs can communicate.

Meet Stella, a spunky Catahoula Blue Heeler mix. Her dog mom, Christina Hunger, a speech-language pathologist, is making headlines around the world after teaching Stella to use augmentative and alternative communication devices to express herself.

Stella’s learned 50 words and can create phrases up to five words long but Stella is not the only dog finding their voice.

UCSD Cognitive Scientist Federico Rossano is leading the largest animal communication citizen science study ever done. It’s called, “How They Can Talk,” and involves six thousand animals in 47 countries.

“What we are trying to do is trying to understand to what degree they can communicate more complex thoughts,” Dr. Rossano explained. “The first time I saw them putting together three, four buttons to sound like a sentence, I was pretty impressed and shocked. What amazed me more was when they would follow up on that because it then really felt like they were actually engaging in a conversation. The other thing that I thought was remarkable was really to see them communicating about the needs of somebody else.”

The time it takes for a dog to learn to talk depends on how much time their human puts into teaching them. But Rossano believes the time spent could lead us all to a better understanding of our furry friends.

Cats are also included in the study, although Professor Rossano said they are much harder to train. We also asked if one breed was a better learner than others. He said their study involved a lot of mixed breeds, but border collies, poodles, and terriers seem to all be quick learners. Poodles and terriers seem to be quick learners, with some learning not just dozens, but hundreds of words. But again, it all depends on how much time their owners put in to teaching them.

Researchers are looking at using machine-learning algorithms to analyze the calls of rodents, lemurs, whales, chickens, pigs, bats, cats, and more.

Analyzing animal language is very different from analyzing human language.

Computer scientists must instruct software programs on what to look for and how to organize the data. This process depends not only on accruing a good number of vocal recordings, but also on matching these vocal recordings with the visual social behaviors of animals.

Making a Google Translate for animals has been an aspirational project that’s been in the works for the the last decade. Another factor that researchers are considering is the fact that there might be unique elements to animal language due to physiological and behavioral differences. There are suggestions for using self-supervised learning algorithms to analyze audio data in which the computer tells the researchers what patterns it’s seeing in the data.

These patterns may unveil connections that are missed by the human eye.