Summary: A new study demonstrates that some highly gifted dogs can categorize objects not just by appearance, but by how they are used. When taught words like “pull” or “fetch,” these dogs later applied the concepts to brand-new toys through natural play, without training or explicit labeling.
This shows that dogs can form mental representations of objects based on their function rather than physical features. The findings highlight the depth of canine cognition and suggest links to the evolution of human language and memory.
Key Facts
- Functional Categorization: Dogs grouped toys by use (tugging vs. fetching) instead of looks.
- No Training Needed: Skills emerged from natural play with owners, not formal instruction.
- Language Insight: Results hint at shared cognitive roots between dogs and humans in learning words.
Source: Cell Press
As infants, humans naturally learn new words and their associations—like the fact that forks are related to bowls because both are used to consume food.
In a study publishing in the Cell Press journal Current Biology on September 18, a team of animal behavior experts demonstrate that dogs can categorize objects by function, too.
In a series of playful interactions with their owners, a group of Gifted Word Learner (GWL) dogs were able to distinguish between toys used for tugging versus fetching, even when the toys in question didn’t share any obvious physical similarities—and then could remember those categorizations for long periods of time, all with no prior training.
“We discovered that these Gifted Word Learner dogs can extend labels to items that have the same function or that are used in the same way,” says author Claudia Fugazza of Eötvös Loránd University, Budapest, Hungary.
It’s like a person calling both a traditional hammer and a rock by the same name, says Fugazza.
“The rock and the hammer look physically different, but they can be used for the same function,” she says. “So now it turns out that these dogs can do the same.”
The studies took place in the dogs’ natural home environments with their human owners. At the beginning, the dogs spent time getting familiar with verbal labels for two functional groups of objects: pull and fetch. Their owners used these words with specific toys and played with them accordingly even though the toys didn’t share any similar physical features.
Next, the dogs were tested to see if they had learned to connect the functional labels to the correct group of toys before playing with more novel toys in the two distinct categories. However, this time, their owners didn’t use the “pull” and “fetch” labels for the dogs.
The team found that the dogs were able to extend the functional labels they’d learned previously to the new toys based on their experience playing with them. In the final test, the dogs showed that they could successfully apply the verbal labels to the toys by either pulling or fetching accordingly, even when their owners hadn’t named them.
“For these new toys, they’ve never heard the name, but they have played either pull or fetch, and so the dog has to choose which toy was used to play which game,” Fugazza says. “This was done in a natural setup, with no extensive training. It’s just owners playing for a week with the toys. So, it’s a natural type of interaction.”
The authors note that the dogs’ ability to connect verbal labels to objects based on their functional classifications and apart from the toys’ physical attributes suggests that they form a mental representation of the objects based on their experience with their functions, which they can later recall.
These findings provide insight into the evolution of basic skills related to language and their relationship to other cognitive abilities, including memory, the researchers say.
More research is needed to understand the scope and flexibility of dogs’ language categorization abilities. The researchers suggest future studies to explore whether dogs that don’t learn object labels may nevertheless have an ability to classify objects based on their functions.
“We have shown that dogs learn object labels really fast, and they remember them for a long period, even without rehearsing,” Fugazza says. “And I think the way they extend labels also beyond perceptual similarities gives an idea of the breadth of what these labels could be for dogs.”
Funding
This work was supported by National Brain Research Program NAP 3.0 of the Hungarian Academy of Sciences, MTA-ELTE Comparative Ethology Research Group, and TRIXIE.
About this animal cognition research news
Author: Julia Grimmett
Source: Cell Press
Contact: Julia Grimmett – Cell Press
Image: The image is credited to Neuroscience News
Original Research: Open access.
“Dogs extend verbal labels for functional classification of objects” by Claudia Fugazza et al. Current Biology
Abstract
Dogs extend verbal labels for functional classification of objects
The relationship between language and human thought is an ongoing debate, and it has been proposed that language shapes object categorization.
Words play an instrumental role in early perceptual categorization, but the capacity to preferentially utilize object function over perceptual similarity in guiding the extension of novel words emerges later, in toddlers and preschoolers. It is debated whether non-humans’ ability to categorize objects is limited to perceptual features.
Only a few studies have investigated the extension of labels to categories, and these relied on a small number of extensively trained individuals who acquired a basic use of labels. However, this differs markedly from how human infants naturally acquire language and adopt words as category markers.
We capitalized on the discovery of dogs with an outstanding ability to form a vocabulary of object verbal labels during spontaneous, natural interactions with their owners to reveal that they generalize these labels to items that share only functional properties without sharing perceptual similarities. During naturalistic-like playful interactions, the dogs classified novel objects based on functional object use rather than mere shared perceptual attributes.
This reveals functional classification beyond perceptual features connected to verbal label learning in a non-linguistic species in naturalistic setups. Such functional label generalization was only documented in a few individual animals after extensive formal training.
The results provide insights into the diverse evolution of basic language-related skills and their relationship with other cognitive abilities.
