Do pets understand our language?

Posted by Susan Lubejko on January 14, 2021 in Animals, Behavioral Neuroscience, Language, Neuroscience | 2 Comments

Humans are unique in our use of sophisticated spoken language to communicate. While other animals use communicative calls and sounds, human language features complex grammar and structure that allows us to convey a nearly infinite number of ideas. Homo sapiens, the species name for modern humans, appears to have been far superior to other hominid species like the Neanderthals at using symbols, such as art, which archaeologists and anthropologists often associate with the use of language. Language is often credited as a major factor in the success of our species [1]. 

However, for a species that has created such an advanced system designed to communicate with other humans, we spend a whole lot of time talking to our pets. You probably don’t know a single pet owner who doesn’t hold full conversations with their pet, whether in baby talk about the extreme fluffiness of their cat’s tail or serious prose regarding their dog’s opinion on societal issues. But how much does Fluffy or Rover understand of these interactions? Does your pet know the name of their favorite toy, or even their name? What can neuroscience tell us about how pets process language?

Dog days of neuroscience

Luckily for us, neuroscientists have begun to use modern brain imaging techniques to answer just this question about man’s best friend. In 2018, neuroscientists from Emory University [2] asked whether dogs could differentiate between words they know and fake words with a similar structure using a common neuroimaging technique called functional magnetic resonance imaging (fMRI), which measures blood flow to different parts of the brain as a signal that those areas are more active during a certain task. First, the dogs were trained at home to discriminate between two toys (e.g. Libby learned to associate “hedgehog” and “duck” with her toys in the picture below). Right before their brain scan, the dogs were tested by their owners in the laboratory to make sure they could retrieve the right toy when the owner said its name. Then, the dogs were placed in the fMRI machine, and their brains were scanned while their owner said either their expected words (“hedgehog” or “duck”) or pseudowords that are similar to these short English words, but aren’t words at all (like “cloft” and “bobbu”). I must point out that fMRI scans require patients to stay completely still for long periods of time, proving this group of dogs to be the goodest boys and girls to remain stationary in the machine without the use of restraints or anesthesia. Check out this Tedx talk from Dr. Gregory Berns, the leader of this research group, to see how the dogs were trained to have their brains scanned.

When the scientists analyzed the fMRI data, they found that the dogs had increased activity in a part of the brain called the parietotemporal cortex, which is known to be involved in sensory processing, when they heard the pseudowords compared to known words. Interestingly, this finding is opposite from human brain imaging, which indicates that we have greater brain activity in the parietal and temporal cortices when we hear words that we know. The authors posit that this increased processing in dogs to new words might help them learn the new object that follows. Also, behaviorally, dogs tend toward new stimuli, so it makes sense that their brain is wired for novelty. 

Additionally, the researchers wanted to know how the dogs’ brains discriminated between the two words that they were trained on. How did Libby know the difference between “hedgehog” and “duck”? Further analysis of the brain scans revealed that some of the dogs had distinct activity patterns for the different words in brain areas such as the temporal cortex (which is involved in sound processing) and the amygdala (which is involved in emotional processing), which suggests that each word was being processed, and perhaps perceived, differently. Taken together, the researchers concluded that the ability for dogs to “understand” our language involves brain activity for detecting novel words, and, in some dogs, brain activity involving auditory and emotional processing.

What we can’t tell from this study is if these dogs actually understand language, or if they are merely trained to use it in this narrow context. Instead of understanding the learned words as words, the dogs may simply be associating a specific noise with getting a specific toy. In 2016, a separate group of researchers [3] used fMRI to determine the similarities between how dogs and humans process speech information. They specifically wanted to know how word information (positive vs. neutral words) and intonation (positive, which tend to be higher in pitch, vs. neutral tones) are processed in the dog brain, and how these components come together to provide a rewarding experience for the dog. These scientists found that word and tone processing in dogs looks a whole lot like that of humans in that the processing of the word components themselves was mainly achieved on just one side of the brain. Does this mean that dogs understand language the same way humans do? So far, it isn’t clear whether dogs actually have specific areas of their brain delegated for our language and communication with humans like we do. These authors hypothesize instead that these pathways for processing some type of information specific to one side of the brain may be shared by many animals throughout evolution, but humans were the first to dedicate that processing power for use in language.

Another finding from this study that is probably of interest to dog owners is that the parts of the dogs’ brains responsible for processing rewarding experiences were most active when positive words known to the dog through training were paired with a positive tone. This implies that you might not always be able to trick your dog by saying negative words your dog has heard before in a positive tone. 

Cat got your tongue?

What about our other furry friend, the cat? While cats have always been important in studying neuroscience, especially for understanding how the brain processes vision, the research surrounding how their brains process our language is a bit behind dogs. This is probably because, as most cat owners will tell you, cats are notoriously difficult to train and sometimes not all that interested in what you want them to do anyway. Therefore, while there aren’t studies like the one above that correlate brain activity with language recognition, there are a small number of scientific papers that look at cat behavior to determine how they might be interacting with our language.

First, you may be wondering whether your cat recognizes your voice when you talk to them. A 2013 study tested this question in 20 cats by playing the voices of 3 strangers, then the cat’s owner, saying the cat’s name. The scientists found that most of the cats responded to the voices by turning their heads or ears toward the sound; the cats almost never actually moved toward or communicated back to the voices. The cats responded pretty strongly to the voice of the first stranger, but mostly lost interest by the time the third stranger called their name. However, once the recording of their owner was played to each cat, their head and ear movements perked up again, and their response looked like that of the response to the first voice. Unfortunately for cat owners, the cats were not more likely to approach or communicate with their owner’s voice than that of a stranger. However, the fact that they perked up to the sound of their owner after starting to lose interest in the task indicated to the scientists that these cats did indeed recognize the voice of their owners [4].

In the above study, the researchers used each cat’s name to get their attention, but are cats familiar with their name? Are they able to discriminate their name from other spoken words? In 2019, the same group of researchers investigated a group of cats’ ability to distinguish between their name and nouns with the same length and phonetic structure or the names of other cats in the same household. The names and the words were also spoken in the same tone and manner to make sure the cats couldn’t use other phonetic clues to determine the difference. Similarly to the previous study, the researchers relied on cats getting bored of the unfamiliar nouns or other cat’s names over time before presenting the cat’s own name to see if they would perk up. Household cats oriented their heads and ears toward their own name over other words and other cat names, regardless of whether they were spoken by their owner or a stranger. Interestingly, when the researchers tested cats that live in cat cafes, those cats could distinguish their name from nouns, but didn’t respond to their name versus the name of other known cats [5].

The future of the field

While language has an important place in our world almost from birth, our communication is much different from that of animals. It is clear, however, that pet owners can interact verbally with the animals they live with, even if we don’t quite understand how our pets process that information. Recent research on language processing in dogs marries adorable dog behavior with advanced neuroscience techniques to try to address this question and opens doors for future research on our furry friends. According to Dr. Berns, this technology may even be useful in determining the propensity for individual dogs to become good service animals, a process that involves intense training that many dogs aren’t able to complete. While the work continues, I suspect the researchers involved may tell you that performing their experiments is quite literally like herding cats.

References

1. Pagel M (2017) Q&A: What is human language, when did it evolve and why should we care? BMC Biology, 15:64
2. Prichard A, Cook PF, Spivak M, Chhibber R, Berns GS (2018) Awake fMRI reveals brain regions for novel word detection in dogs. Frontiers in Neuroscience, 12:737
3. Andies A, Gabor A, Gaesi M, Farago T, Szabo D, Miklosi A (2016) Neural mechanisms for lexical processing in dogs. Science, 353(6303): 1030-1032
4. Saito A, Shinozuka K (2013) Vocal recognition of owners by domestic cats (Felis catus). Animal Cognition, 16:685-690
5. Saito A, Shinozuka K, Ito Y, Hasegawa T (2019) Domestic cats (Felis catus) discriminate their names from other words. Scientific Reports, 9:5394

Cover Image: practicalfmri.blogspot.com