Ever wondered if cats or dogs have the edge on intelligence tests? You're not the only one: scientists have been interested in evaluating the smarts of domestic pets for quite a long time. And new studies released recently have found that, on one measure — the neural construction of the brain — dogs may have the edge. But when it comes to accurately assessing the intelligence of cats in actual tests, where they're asked to count or make logical deductions, there's a hilarious problem that hampers our communal understanding and scientific research in general: cats just cannot stop being cats. And that means being seriously unhelpful.
A study published in Frontiers of Neuroanatomy at the very end of 2017 compared the levels of neural connections in the brains of eight species: the ferrets, banded mongooses, racoons, cats, dogs, hyenas, lions, and brown bears. All are mammals and have reputations for relative intelligence. Lions, for instance, can learn complicated techniques like managing food traps by watching their compatriots just once, a new study filmed for the BBC program Big Cats found in early 2018 — and in that they're unique among all big cat species.
When the scientists behind the study compared neuron levels in the species of the animals, though, lions and bears had fewer neurons than golden retrievers (which was the type of dog tested — usually relatively bright and capable of being highly trained). Cats also featured less neurons than dogs, and that got some people thinking that we had, at last, reached definitive proof that dogs are in fact smarter than our feline friends. But it's not that simple.
Measuring animal intelligence isn't all about neurons — and while a lot of neurons do appear to be necessary for complicated cognitive decisions, it's not 100 percent clear that there's a direct relationship between more of them and a more intelligent animal. For one, cognitive abilities are shaped by where in the brain neurons concentrate; elephants have many more neurons than humans in their massive brains, according to a 2016 study, but 98 percent of them were focussed on the cerebellum at the base of the brain, while a much higher proportion of human neurons live in the prefrontal cortex, where a lot of higher thought happens.
And there's another issue: just because an animal is capable of doing something doesn't mean it's actually going to do it, particularly under laboratory conditions. Enter the cat — and the huge difficulty of trying to test an animal that has so much free will it won't do anything you say.
David Grimm, who has written extensively on animal intelligence, told Slate in 2014 that one behavioral scientist told him, “We did one study on cats—and that was enough!” Christian Agrillo, an expert on numerical ability in animals who specializes in fish and dogs (he proved last year that dogs aren't susceptible to an illusion about differently sized bowls), has also done a lot of work with cats, and reveals that it's incredibly difficult to get them to do anything. He told Grimm that when cats are put in experimental conditions to try and make them show a cognitive ability, like counting, "they just walk away."
Yep, your house cat is a terrible test subject. Which means that assessing feline intelligence often means looking at wild cats in their natural environments — like a study that came out in early 2018 revealing that wild pumas, who live alone in the wild, will share their feeding grounds with other pumas as long as the strange pumas share their feeding grounds too, revealing a pretty sophisticated bit of thinking.
So the next time there's a headline that claims that dogs or cats have finally "won" the intelligence wars, take it with a grain of salt — because chances are that they've had some serious trouble lining up cats cooperative enough to do their tests, rather than walking off and lying in the sun aloofly.