What's Your Dog’s IQ?

Measuring canine intelligence

Posted Oct 14, 2016

Most dog owners will agree that their pets are pretty smart. Dogs not only learn tricks and useful tasks, they also have a high degree of social intelligence. But is there any way to measure the canine intellect? Can we come up with a doggie IQ test? British psychologist Rosalind Arden and her colleagues are searching for ways to measure the intelligence of dogs.

Intelligence testing goes back more than a century. But your approach to testing it depends on how you view intelligence. Many psychologists today buy into the arguments of Charles Spearman, an early twentieth-century researcher. Spearman argued that intelligence was a singular quality that each person has to a greater or lesser degree. He called this general intelligence, or g for short.

According to Spearman’s theory, g is mainly determined by genetic factors. Some people have high levels of g because their nervous systems process information faster and retain it better. However, Spearman also recognized that experience was important. Even people with high levels of g may not perform well the first time they try a new task. Still, they’ll learn that task much faster than a person with low g.

In high school, you probably knew someone who did well in all subjects—math, science, language arts, social studies. That person also was good at athletics, played a musical instrument, and was socially active in various clubs. Perhaps the person I’m describing was the valedictorian at your high school graduation. Such a person exemplifies Spearman’s concept of g.

To put this another way, we can think of g as the tide that raises or lowers all boats. People with high g will perform well on all tasks that they apply themselves to, whereas people with low g may never perform well no matter how much effort they expend.

It’s important to note here that not all psychologists argue that intelligence is a single construct. These are the “multiple intelligence” psychologists. They argue that people can be highly intelligent in some areas but not very intelligent at all in other areas. You probably also knew people like this in high school. The computer nerd with zero social skills. The football jock who failed every class.

My experience in reading the literature on intelligence has been that the single-intelligence psychologists pretend that the multiple-intelligence psychologists don’t exist, and vice versa. So I should note here that Arden and her associates are single-intelligence psychologists. In fact, the very idea of IQ—an intelligence quotient—is based on the idea that there is a single intelligence, g, that can be measured on an intelligence test.

Which brings us back to the furry question of how to measure doggie IQ. Obviously, human intelligence tests are no good. This is because, as Arden and her colleagues point out, g is only meaningful within a species and can't be used to compare average intelligence across species. This is because “intelligence” means different things for different species.

It may not surprise you that psychologists can’t agree on a definition of intelligence. But generally speaking they agree on some of the characteristics that make up intelligence. These include the abilities to solve problems relevant to one’s life, to adapt to changing circumstances, and to benefit from experience. Thus, we must consider what it means to be intelligent within the context of a particular species struggling to survive within a given context.

For example, we may all agree that chimpanzees aren’t as smart as humans, but largely this is because we’re holding chimpanzees to human standards. Chimpanzees are faced with particular problems in their struggle to survive and reproduce, and some chimpanzees are better at it than others. Extending Spearman’s theory, some chimpanzees have more chimpanzee-g than other chimpanzees. Most humans would not be able to survive in natural chimpanzee environments, and so we could assume they’re low in chimpanzee-g.

This is not to say that we can’t make cross-species distinctions in intelligence. Humans live in technologically-complex societies, whereas chimpanzee technology consists of finding suitable rocks to crack nuts and stripping the leaves from twigs to fish for termites. And dogs have cast their lot as servants to humans, as opposed to their wild cousins the wolves, who prefer the freedom to fend for themselves. So obviously there are cross-species differences in intelligence, but the concept of general intelligence can only be applied within a particular species.

Thus, intelligence testing for canines is intended to find individual differences between dogs, not differences between dogs as a species and humans (or any other animal). So to measure doggie IQ, you have to set up tasks that are relevant to the canine form of intelligence. For instance, dogs are well known for their spatial navigation abilities, and so several tests of doggie intelligence involve a “detour task” in which the animals need to find a way to navigate around an obstacle to reach a target.

The research of Arden and her colleagues investigates the range of doggie intelligence. That is, it helps us understand what it means to be intelligent as a dog, which can be quite different from what it means to be intelligent as a human.

In the end, though, all intelligence tests are about finding individual differences among members of a group. We generally agree that the person with an IQ of 130 is more intelligent than the person with an IQ of 70. Likewise, the tests that Arden and her colleagues are developing are designed to separate the smart dogs from the not-so-smart. Which reminds me of the old Ken-L-Ration commercial: “My dog’s smarter than your dog…”


Arden, R., Bensky, M. K., & Adams, M. J. (2016). A review of cognitive abilities in dogs, 1911 through 2016: More individual differences, please! Current Directions in Psychological Science, 307-312.

David Ludden is the author of The Psychology of Language: An Integrated Approach (SAGE Publications).