New research out of Hungary points out that puppies and human toddlers face similar problems when it comes to interacting with the environment. There are many things in the world which are simply not safe for young individuals. Hot surfaces, open flames, machines with moving parts, and a myriad of other dangerous things populate the world. These can damage inquisitive little fingers or puppy paws and even produce injuries that may be serious or fatal. How is a puppy or a child supposed to find out about such perilous things without going through the hazards of trial and error learning?
Dogs and people share an advantage that helps in such situations because they are social animals. Both can learn by gathering information from other individuals that they associate with. For example, dogs are sensitive to emotional expressions in other dogs and also in humans. If a puppy sees another dog acting frightened or uncomfortable when they encounter an unfamiliar object, then it is probably a good idea for the pup to stay away from that object. Similarly if the object or situation produces a positive response, then it is probably safe for the pup to approach and interact with it. This is a process called "social referencing".
Social referencing is the procedure in which an individual relies on emotional cues from others when they are faced with an unfamiliar situation. Using social referencing is of particular value to young individuals, mainly because, with their lack of experience, nearly everything which a young thing encounters is novel to them. The youngster's very survival can depend on his ability to read the emotional responses of individuals in his environment so that he knows how to react properly.
There are two easily seen behavioral patterns associated with social referencing. These behaviors are seen in both puppies and human kids. The first is what is called "referential looking". This is where the toddler or pup looks at the unfamiliar object or situation and then looks back at their social partner to pick up any possible emotional cues. This may be repeated a number of times. When their social partner looks at the suspect situation and then back at the youngster while expressing some emotional response, then the puppy or child should be able to pick up some valuable information. The second part of social referencing involves acting on that information. If the information which he received was negative (for example, the other individual shows fear or discomfort), then it is best to stay away from that object or place. However if the emotional response was positive then it is probably safe to go over and investigate this novel thing.
One important question is "Who is a puppy supposed to trust as a source for such information?" It would make sense for evolution to have wired puppies to monitor the responses of other dogs, particularly familiar dogs, such as the pup's mother. But is it possible that evolution has also modified the responses of young puppies so that they can gather information from humans about the safety of things, even at a young age? This was the question which a team of researchers headed by Claudia Fugazza from the Department of Ethology, at Eötvös Loránd University in Budapest, Hungary, set out to answer.
Since the investigators assumed that social referencing would be especially valuable for young inexperienced individuals, they decided to see if it was already detectable in eight week old puppies who were still living with their mother and litter-mates.
First they needed some ambiguous objects which a puppy might consider to be suspicious. They used two different stimulus objects. One was a plastic bucket hanging from a frame which contained a speaker which could broadcast loud unfamiliar sounds such as creaking and sirens. The other was an oscillating fan with plastic streamers that flapped in the air stream. For each test trial one of these unfamiliar objects would be placed in a corner of the test area.
The first condition of interest involved social referencing using information from other dogs. One third of the group of 48 puppies tested was placed in the test area along with their mother (who had previously been exposed to the test object and had been acclimated so that she felt comfortable with it nearby). Another group of puppies had an unknown adult dog placed in the room with them. This dog was similar in breed and size to their mother but was completely unfamiliar to the puppies. The final sample of puppies was placed in the room without any dog present.
The puppies acted as expected when the mother was present. They looked back and forth between the ambiguous object and their mother. When their mother simply remained calmly sitting in the room, the vast majority the puppies clearly decided that the situation was safe and wandered over to explore the novel object. When the dog in the room with them was unfamiliar the puppies still looked back and forth between their canine companion and the test object, but they were less likely to get up and investigate that strange thing, and if they did so it took a lot longer than when their mother was around. When there was no dog in the room, the puppies acted very suspicious toward this odd thing, and were not very likely to approach and interact with it.
That result makes sense from an evolutionary viewpoint. It indicates that puppies gather information from observing other dogs that they know and trust, and use the emotional responses that they see to determine the safety of objects in their environment. However, next these scientists asked the question "Have dogs co-evolved with humans to the degree that these young puppies would be willing to trust information from a human about the safety of objects in the world, even at the tender age of eight weeks?"
To answer this part of the question the investigators substituted the dogs used in the previous experimental conditions with a female human. Over a period of an hour or so this woman had interacted with the puppies so they were comfortable around her. Again three conditions were used. The first involved the female experimenter in the room with the puppy and the strange object. In this scenario the human expressed positive emotions toward the unfamiliar stimulus. She did this by looking back and forth between the test object and the puppy, while smiling and making appropriately positive and happy vocal sounds. The second group of puppies had the same woman in the testing area except that she acted neutrally, expressing no emotion toward the strange object. The third group was placed in the room with the suspicious object, but no person or dog was present. [A negative emotional group was not included for ethical reasons, especially given the age of the puppies that were being tested.]
The interesting thing here is that the puppies showed referential looking when the human was present. They gazed back and forth between the person and the unfamiliar object, just as they did when their social companion was another dog. It was quite clear that the puppies were trying to pick up the emotional reaction of the person in order to learn how to respond to the strange thing in the room. When the experimenter showed positive emotions toward the unfamiliar object, the puppies were most likely to approach and interact with it, and they did so relatively quickly. When the human showed a neutral expression, the puppies were less likely to approach and explore the unusual stimulus and if they did they approached it more slowly. With no one in the room, the puppies treated the object as untrustworthy and were much less likely to attempt to approach or interact with it.
The conclusion seems to be clear. One of the ways in which puppies manage to learn about their world in a safe way is by extracting information from the emotional responses of other individuals when they are faced with unfamiliar and ambiguous situations. The puppies respond appropriately if that individual is a trusted and familiar dog. However the truly interesting finding from this research is that the puppies are also quite comfortable using a human being as their source of information. This suggests that there is something in the nature of the evolution or domestication of canines which has predisposed dogs toward paying attention to human emotional responses. Once they have interpreted those human emotions and determine which aspects of the environment triggered them, they use that information to guide their own behaviors. Thus they don't need to actually get hurt by something dangerous in order to know that it should be avoided. The interesting additional finding is that it seems that from puppyhood, dogs are willing to rely upon the validity of the emotional responses of humans in matters that can affect their own safety.
To my mind the idea that a young puppy will look at a newly met human being, and have the faith that this person's emotional responses can be trusted to the same extent as those of its own mother, gives me a warm feeling.