Training a dog to help the disabled or to sniff out bombs is a long and involved process, so if you're going to do it, you certainly want to end up with a "usable" animal at the end. A new cognitive-ability test could help determine in advance which canines are going to make the cut, so time isn't wasted on those that won't.

Currently, trainers often do assess a dog's physical characteristics and temperament in order to decide if they're likely to make a good service animal. Even so, according to the University of Arizona, only about half of the pups that are specifically bred to be assistance dogs will successfully complete their training.

Not only does this mean that a lot of time and effort gets wasted, but it also results in a relatively small number of trained dogs being available to the people who need them – it's not unusual for individuals to have to wait up to two years before getting one.

This situation inspired Evan MacLean, director of the university's Arizona Canine Cognition Center, to see if an assessment of dogs' cognitive abilities could better determine which animals were best suited to certain jobs. The resulting study started with 164 dogs from Canine Companions for Independence, which trains assistance dogs for the disabled, and 222 animals that were being trained by the Navy to sniff out explosives.

Beginning when the dogs were 18 months old and about to enter a training program, MacLean and colleagues subjected them to the Dog Cognition Test Battery, which consists of a variety of game-based exercises designed to assess 25 different cognitive skills. After the training program was over, the scientists looked at which of those skills were shared by the dogs that successfully graduated from the assistance program, or that performed best at explosives detection (the Navy program has no official graduation).

In a second part of the study which involved 180 dogs, the scientists performed an abbreviated form of the test. This was designed to assess only the key cognitive skills that were previously identified as being indicators of potentially good assistance dogs, or of good bomb-sniffing dogs. Based on this revised test, the team made predictions as to which animals would graduate the assistance program, or be best-suited to work with the Navy.

It turned out that when it came to predicting the 25 percent of the dogs that had the best shot at graduating the assistance program, the scientists were 86 percent accurate. Although the figures weren't as neat and tidy for the Navy dogs, there was nonetheless a strong correlation between the animals identified in the prediction, and those that ultimately excelled at their jobs.

"We study these abstract questions about how animals think about the world and how they solve problems, but there aren't always a lot of situations where you can say, 'Why does that matter? What does it allow an animal to actually do?'" says MacLean. "This is some of the first evidence that suggests that these processes that we measure, which differ between individual dogs, have some real consequences related to something that's quite worthy in society."

A paper on the research was recently published in the journal Frontiers in Veterinary Science.