For the first time, researchers have shown that not only can humans differentiate between different languages: the brains of dogs also show different activity patterns, depending on whether the four-legged friend is familiar with that language. Yes or No. This is especially evident in older dogs that have had a long time in their lives to learn the sounds of a language. In addition, four-legged friends recognize whether spoken words are actually a language. Non-linguistic sound fragments are processed in a different brain region.
Before human babies learn to speak, they can recognize and differentiate different languages. Since communication through language plays an important role for humans, our brain has evolved in the course of evolution in such a way that it can recognize, organize and reproduce linguistic patterns. But is this ability uniquely human? How do animals understand our language? Are there differences in the different human languages for our four-legged companions?
Spain to Hungary
Laura Kuaya of Iotvos Lorand University in Hungary asked herself this question when she and her dog Kun-Kun moved from Mexico to Hungary. “Before, I only spoke to him in Spanish,” Kuya says. “So I asked myself if Kun-kun noticed that people in Budapest speak a different language, namely Hungarian.” To find out, he and his colleague Kun-kun and 17 other dogs trained on it, all in a recumbent brain-photographer. In this way, the researchers were able to follow the animals’ brain activity while they listened through headphones to excerpts from the book “The Little Prince” by Antoine de Saint-Exupery – sometimes in Spanish, sometimes in Hungarian.
“All the dogs had heard only one of two languages from their owners, so that we could compare a very familiar language with a completely unknown one,” Kuya reports. “We also played dogs with destroyed versions of these passages that sound completely unnatural. So we can test whether they even recognize the difference between language and non-language.”
different brain activity
The result: different activity patterns were found in their primary auditory cortex, depending on whether the dogs heard a non-linguistic scrap of sound or one of the two languages. In this area of the brain, sounds are processed and classified according to volume and pitch. Based on brain activity in the primary auditory cortex, the researchers could not tell whether the dogs heard Spanish or Hungarian, but they could tell whether it was speech or non-speech. However, it’s not clear whether this makes a qualitative difference for animals: “While human brains are specifically adjusted to language, dog brains can more readily recognize the naturalness of sounds,” says Kuya’s colleague Raul Hernandez- Perez explains.
A better distinction was found in the dogs’ secondary auditory cortex. This area is responsible for the detailed processing of sounds—in humans, for recognizing words and melodies, among other things. The dogs also showed different activity patterns depending on whether they heard a known or unfamiliar language. The older the dog, the more obvious the differences become. “Every language has multiple auditory regularities. Our results suggest that over the course of their lives with humans, dogs experience auditory regularities of the language they encounter,” says Hernández-Pérez.
Unique to man’s best friend?
The study shows for the first time that non-human brains are also able to recognize and recognize different human languages. “This is exciting because it shows that the ability of a human to learn the regularity of a language is not just a human’s,” says Attila Andics, Kuya’s colleague. It is not yet clear whether this is characteristic of dogs or whether other animals have this ability. “Dogs have been with humans for thousands of years,” Andics says. “It’s possible that this has resulted in changes in their brains that make them better speech listeners. Future studies will need to find out if this is indeed the case.”
Source: Laura Kuaya (Iotvos Lorand University, Hungary) et al., NeuroImage, DOI: 10.1016/j.neuroimage.2021.118811
