Social Referencing in Dogs

Do dogs look to their person when something stressful happens, like infants do their parents? This study says yes.

By Zazie Todd, PhD

When human infants see something they are unsure of, they look to their caregiver to see what their reaction is. This is called social referencing. It has two components: first of all, a look from the object to the caregiver; and second, a reaction to the object (approach or avoidance) that is influenced by the caregiver’s response. This is well established in infants at twelve months of age. Do dogs do the same thing?

Two recent papers by Merola and colleagues set out to investigate, using a method similar to infant studies. They needed a slightly-scary object; something that will make a dog feel cautious, but not so scary that it will turn and run away. They decided to use an electric fan with streamers attached. 

In the first paper, published in Animal Cognition, the owner brought their dog into the room with the fan. The fan was at the far end of the room, and as soon as the owner closed the door, the fan was turned on by remote control. The owner stopped at a specific mark on the floor and released the dog.  The owner then stood still and looked at the fan with a neutral expression (phase 1).

What happened next depended on whether the dog was assigned to a positive or negative condition. There were three further phases in which the owner gave a positive or negative response to the fan using speech and facial expression (phase 2), and movement towards or away from the fan (phases 3 and 4). These reactions were later rated by people who were blind to the aims of the experiment, to confirm that they were indeed positive or negative as intended. The dog’s behaviour was observed throughout.

A pretty biewer yorkie with an autumnal display of fruits and foliage
Photo: Liliyana Kulianionak / Shutterstock

Seventy-five dogs took part. Seventeen were excluded from analysis because of mistakes by the owners, and some were excluded from most of the analyses because they confidently approached the fan right at the start. Of the remaining dogs, 83% looked at their owner at least once after seeing the fan. Once the owners started reacting to the fan, dogs in the negative group were significantly more likely to remain still than those in the positive group. Also, in the last two phases of the study (once the owner had moved), dogs in the negative condition spent significantly more time interacting with the owner. This shows they have responded to their owner’s negative message by avoiding the fan.

This study shows that most of the dogs engaged in social referencing to see what their owner thought of the fan. However, the experiment had an initial phase in which the owner was neutral, which is different from studies with infants. This disrupted normal social referencing behaviour, since the owner was artificially silent in response to the dog’s look, and only began to respond after a pre-set time had elapsed. Also, it would be interesting to know whether dogs do social referencing with a stranger as well as their owner.

The second paper, published in PloS One, involved a new set of ninety dogs. This time, both the owner and a stranger were in the room with the dog. Either the owner or stranger acted as the informant, while the other person sat on a chair and read a book throughout, paying no attention to the dog. The experimental procedure was similar, except this time the informant began responding to the fan as soon as the dog looked back at them (phase 1). Then, the fan was turned off (phase 2), and the informant stayed in the same place, but continued to give the positive or negative message every time the dog looked at them. This design is much closer to that used in the infant cognition literature. 

As before, some results had to be discarded either because of errors made by the owners or because the dogs were very confident about the fan. Results from the remaining dogs showed that 76% of dogs with the owner as informant, and 60% of dogs with the stranger, showed referential looking at the informant.  There were significant differences in behaviour depending on whether the message was positive or negative. Interestingly, there is also evidence that dogs responded more to their owner than to a stranger. When the informant was the owner, dogs in the positive group reached the space near the fan more quickly, and dogs in the negative group took longer to reach the fan, than when the informant was a stranger. When the message was negative, dogs looked to the seated person more if it was the owner rather than a stranger, suggesting that they wanted information from their owner too. 

At the end of both studies, the fan was turned off and the experimenter sat next to it and gave the dogs treats. This was to make sure the dogs would not be frightened by fans in future.

These two studies show that dogs use gaze to look at a person for information when they are faced with something they are unsure about, and their subsequent behaviour is based on the person’s reaction. With some small differences, these results are very similar to those found in studies of infants.

Have you ever noticed your dog looking to you for information about something unfamiliar?

This page contains affiliate links.

Zazie Todd, PhD, is the author of Wag: The Science of Making Your Dog Happy. She is the founder of the popular blog Companion Animal Psychology, where she writes about everything from training methods to the human-canine relationship. She also writes a column for Psychology Today and has received the prestigious Captain Haggerty Award for Best Training Article in 2017. Todd lives in Maple Ridge, BC, with her husband and two cats.

Useful links:
References
Merola, I., Prato-Previde, E., & Marshall-Pescini, S. (2011). Social referencing in dog-owner dyads? Animal Cognition, 15 (2), 175-185 DOI: 10.1007/s10071-011-0443-0
Merola, I., Prato-Previde, E., & Marshall-Pescini, S. (2012). Dogs' Social Referencing towards Owners and Strangers PLoS ONE, 7 (10) DOI: 10.1371/journal.pone.0047653

As an Amazon Associate I earn from qualifying purchases. As an Etsy affiliate, I earn from qualifying Etsy purchases.