Old dogs CAN learn new tricks, but the way that they learn may be different than when they were younger. So concludes a recent study called “Aging effects on discrimination learning, logical reasoning and memory in pet dogs”. The study was conducted on 95 pet Border Collies who ranged in age from 5 months to over 13 years old. Researchers purposely chose dogs of the same breed in order to minimize any differences in performance that were unrelated to age.
Starting out with a spoiler, older dogs did worse than younger dogs on one of the tests, they did better on a second test, and there was no effect of age in the third test.
For the first test, dogs were trained to associate four images on a touch screen with a positive experience—receiving a treat. So, if the dogs touched these images on the screen, they received a treat. Another four images were associated with a time out, meaning that touching any one of them resulted in no opportunities to touch images for a brief period of time. After being taught these associations, dogs were tested with a pair of images that always included one randomly selected “treat” image and one randomly selected “time out” image. Sessions consisted of 30 tests with a pair of images. Dogs were considered to have mastered this task when they chose the right image 20 out of 30 times for four out of five sessions in a row. There was a linear relationship between age and the number of sessions it took dogs to learn this task, meaning that younger dogs learned it faster than older dogs.
In the second test, dogs were again shown a pair of images on the touchscreen, but only one of those images was one that the dog had seen before. In each case, the familiar image was one that the dog had learned had a negative association because it led to a time out if touched. The dogs could therefore make an inference that the unfamiliar image was the correct choice and would lead to a treat if touched. (These trials were interspersed in sessions that included pairs of images just as in the first test in which both images were familiar to the dogs.) In this experiment, the older dogs were, the more likely they were to choose the correct image, meaning that older dogs were better at solving this puzzle than younger dogs.
The final test in this experiment looked at long-term memory. Dogs were tested at least 6 months after the other parts of the study were completed to determine how well they retained what they had learned. When presented with pairs of images just as they had been in test one, over 90 percent of the dogs performed better than chance level (at least 22 correct out of 32) and there was no effect of age on the success rate.
This study shows that there are differences in cognitive abilities between older and younger dogs, but not that dogs of certain ages have better abilities than dogs of other ages. The way that age affects performance depends on the specific task dogs are asked to do.