News Release

Dog breeds differ from each other in their cognitive traits

Peer-Reviewed Publication

University of Helsinki

A study conducted at the University of Helsinki, Finland set out to determine how 13 dog breeds performed in various cognitive tests. With a sample size of over 1,000 dogs, the researchers found several breed differences in traits such as social cognition, problem-solving ability, and inhibitory control.

Previous research has shown that breed is one factor potentially affecting a dog’s personality and behaviour. However, we still know very little about the breed differences of cognitive traits, such as problem-solving ability, memory, logical reasoning, and social cognition. The aim of this study was to explore which cognitive traits may differ by breed and how each breed performs compared with others for each of the studied traits.

“The results suggest that different cognitive traits may have been favoured in different breeds”, says doctoral researcher and study author Saara Junttila.

Large-scale cognition study

The sample consisted of 1,002 Finnish dogs that took part in the smartDOG cognitive test battery. SmartDOG is a company owned by co-author Katriina Tiira, offering various cognitive and behavioural tests to dog owners who are hoping to understand their dogs’ characteristics better. The analysis only included dog breeds with a minimum of 40 tested individuals, which resulted in 13 breeds.

The dogs took part in seven cognitive tests and three tests measuring behaviour. Significant differences were found between breeds in the tests measuring inhibitory control, understanding of human gestures (such as pointing with a finger in a specific direction), spatial problem-solving ability (detouring around a V-shaped fence), activity level, greeting an unknown person, exploration in a novel environment, and behaviour during a problem-solving situation that is impossible to solve (i.e., whether the dog tries to solve the problem independently or asks the human for help instead).

“We did not find breed differences in tasks measuring logical reasoning or short-term memory”, says Junttila. “These traits may not have been selected for in different breeds or these traits are more strongly affected by environmental factors, such as training methods, previous experiences, or the test situation itself”.

How do breeds differ?

“The results provide interesting information concerning the kinds of traits that may have been favoured in different breeds”, says Junttila. “For example, the most common and important role of the modern golden retriever seems to be that of a pet dog and family member. This means that high inhibitory control and dependence on humans may be valuable traits, whereas problem-solving ability may not be as important. Other characteristics and traits may be valuable in other roles instead”, Junttila continues. “The test battery included many tests where the results could not be evaluated as “good” or “bad”. For example, low performance in the inhibitory control test may mean that the dog has a high motivation for rewards, is easily aroused in training, and reacts quickly, all of which can be advantageous in training for working roles and dog sports. This may be why breeds, such as the Malinois and German shepherd, received such low scores in the inhibitory control test”.

The results showed, for example, that the Hovawart was one of the most independent breeds, attempting to solve the unsolvable task by themselves rather than turning to a human for help. In contrast, the golden retriever was one of the most human-oriented breeds, spending a lot of time seemingly asking the human for help. The Malinois was the most successful breed at understanding human gestures, with the Labrador retriever not far behind. The Malinois and border collie were the fastest at spatial problem-solving, whereas the border collie excelled at inhibitory control.

According to Junttila, the results only tell us how these breeds perform in these specific problem situations. “Therefore, it would not be advisable to draw too many conclusions from our study results to the more general behaviour of a specific dog, as each dog is still, after all, an individual”, Junttila sums up.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.