How Puppies Learn From Dogs and People

Source: Africa Studio/Shutterstock

To survive, young individuals, including dog or wolf puppies and human toddlers, must learn where to go, what is safe, and what they are expected to do. If these youngsters had to learn only by trial and error, by interacting with the world and making mistakes, many would suffer injuries, and many more would not live to tell the tale. For this reason, animals which live in social groupings benefit from something called "social learning."

Simply put, social learning refers to the fact that young individuals watch the behavior of more experienced individuals (usually their mother or other older individuals in the family or social group), and thus they learn which behaviors are most likely to bring them rewards and which behaviors to avoid. It is a reasonable assumption that evolution predisposed young animals to observe the behaviors of other animals of their own species in order to safely learn new behaviors. Thus we know that young children observe human family members in order to learn more about their environments, while young wolf pups observe members of their packs behaving. However, things are a lot more complex for domestic dogs. They have evolved in a human environment, and it is extremely complex, with lots of important information that must be learned. Furthermore, there are two different species actively behaving around domestic dog puppies, not just other canines, but also humans. Both of these types of individuals can provide important information which is valuable for the puppy's safety and happiness. Given the complex social world that puppies grow up in, a team of researchers, headed by Claudia Fugazza of the Department of Ethology at Eötvös Loránd University, asked the question, "Do young puppies learn from observing human behavior as well as from observing the behaviors of their mother or other dogs?"

To answer this question they used a group of eight-week-old puppies of a number of different breeds, including Labrador Retrievers, Border Collies, and Belgian Tervurens. They would be asked to solve two problems: Both involved opening a box in order to get a sausage treat inside. One of the boxes could be opened by lifting its hinged lid, while the other could be opened by sliding the lid aside in its grooved track.

The puppies were first shown that each of the boxes contained sausage treats and were allowed to nibble some of them from the open box. This focused their interest on these objects. The research showed that if puppies were simply allowed to explore the closed box and to manipulate it in a trial-and-error fashion, only about 50 percent of them would actually manage to open the box to eat the treat within the two-minute test period allowed.

In the first test condition, the puppies were placed in a dog kennel about a meter and a half away from the test box. From here, they could observe it quite clearly. Now they were allowed to see other dogs solving the problem. This gave the pups an opportunity to learn what to do without actually having to fool around with the boxes in a trial-and-error manner themselves. The demonstrator dogs could either be the puppy's mother or an unfamiliar dog. The demonstrator dogs had been trained over a number of trials on how to efficiently open either box. The puppies then got a chance to observe one of the demonstrator dogs open one of the boxes for six different trials. At the end of this observation period, the pup was released and allowed to go to the box to try to open it himself.

Having the chance to observe other dogs solving the problem did provide a benefit for the observing puppies. There was a small (5 percent) improvement in the likelihood that the pups would solve the problem in the two minutes given when they had observed their mothers opening the box. Surprisingly, however, there was a large (29 percent) improvement in the success rate when the demonstrator was a stranger. It turns out that this result is probably due to the fact that the puppies were more likely to carefully observe the stranger's behavior, while they spent less time looking at the behavior of their mothers. (I wonder if this has implications for the home-schooling of human children.)

It is important to note that the puppies are not simply imitating the behavior of the adult dogs, but rather extracting information about how these puzzle boxes worked. We know this, because some of the adult dogs opened the boxes by manipulating them with their muzzles, while others opened them by using their paws. However, regardless of the specific action performed by the adult demonstrator, all puppies used their muzzles to open the boxes. This means they had learned how the boxes work through observation and were not simply mimicking specific movements that they had seen.

This demonstrates that young pups can learn from observing other dogs behaving in the environment. It would be sensible for evolution to have programmed young individuals to watch the actions of animals of their own species to improve the likelihood of their survival. But now the crucial question remains: Since dogs have evolved in the human environment, have they also been wired in such a way so that from early puppyhood they are predisposed to observe human beings behaving and to extract information which they can later use?

As before, the puppies got to observe six separate trials, only now with the human being opening a specific box. When released, the puppies showed that they had learned from watching the people behave. Now there was a 42 percent improvement in the likelihood that the puppies would solve the problem in the allotted time period. This means that, at least in this study, the puppies benefited more from watching a person's behavior than from watching the behavior of another dog.

Environment Essential Reads

Can We Stop Sabotaging Our Climate? Yes, With These Mindsets

How Nature Makes Us Human

The research also showed that these benefits persisted, and are likely to be permanently remembered, since when retested an hour later, the puppies still remembered how to solve the puzzle boxes.

Dr Fugazza summarized the implications of this study by saying, "We can train puppies from a young age by showing them what to do, like their mothers. If we want them to fetch a stick, we should fetch it first, and if a person wants them to use a new bed, they should lie down in it first."

For more about social learning in dogs, click here.

Copyright SC Psychological Enterprises Ltd. May not be reprinted or reposted without permission