Social Referencing in Dog-Owner Dyads. Do dogs get information from their owners before reacting toward a potentially scary object?

Do Dogs Get Information From Their Owners Before Reacting Toward a Potentially Scary Object?




Naci Berkoz and Contact Points Will share a moment together. Photo courtesy of


Social referencing is the seeking of information from another individual to form one’s own understanding and guide action. In a study, adult dogs were tested in a social referencing paradigm involving their owner and a potentially scary object. Dogs received either a positive or a negative message from their owner. The aim of this research was to evaluate the presence of referential looking to the owner and behavioral regulation based on the owner’s actions towards the object (a decorated fan).

Most dogs (83%) looked referentially to the owner after looking at the strange object, appearing to seek information about the environment from the human, but few differences were found between dogs in the positive and negative groups as regards to behavioral regulation. Finally, a strong effect of observational conditioning was found with dogs in the positive group moving closer to the fan and dogs in the negative group moving away, both mirroring their owners behavior. These results are interesting in relation to studies on human-dog communication, attachment, and social learning.

The Details

Until now, little was known about the presence of referential looking to the owner, behavioral regulation based on the owner’s (vocal and facial) emotional message, and observational conditioning following the dog owner’s actions towards an unfamiliar object. The behavioral regulation theory is a relatively new development that answers two key questions about instrumental conditioning. What makes something effective as a reinforcer? And how does the reinforcer produce its effect?

To address this knowledge gap, Dr. Isabella Merola, Dr. Emanuela Prato-Previde, and Dr. Sarah Marshall-Pescini from the University of Milan, Department of Biomedical Sciences, studied 90 dogs (37 males, 53 females; 61 pure breed, 29 mixed breed). Dog-owner dyads were semi-randomly assigned to one of four groups, balancing for age and sex. All in all, 44 dogs participated in the study with their owners as the informant (the informant being the person sharing information with the dog about the object, a fan). Of these, 26 were tested with the owner conveying a positive emotional message about the object (owner-positive group) and 18 with the owner giving a negative emotional message (owner-negative group). Forty-six dogs were tested with the same female stranger acting as the informant. Of these, 21 witnessed the stranger giving a positive message (stranger-positive group) and 25 a negative message (stranger-negative group). All the dogs that participated in the research lived at home with their owners.

The experimental stimulus was the same for all dogs in all groups: a 50 centimeter tall and 34 centimeter wide electric fan, with plastic green ribbons attached to it. This object was selected because it evokes a cautious reaction in most dogs, neither very positive (approaching directly and touching) nor very negative (running in the opposite direction or strong stress such as trembling, or hiding.)

In order to assess how much of an influence the informant’s identity had on a dog’s referential looking, the researchers made either the dog’s owner or the stranger that was acting as the informant, sit reading quietly in the testing room. The dogs’ behavior was then measured when the informant delivered the message (positive or negative to the dog about the ambiguous stimulus).

Through observations, Dr. Merola, Dr. Prato-Previde, and Dr. Marshall observed that the results suggest that the dogs were probably more sensitive to the emotional expression of the stranger, yet the manner in which they changed their behavior was dependant on their relationship with the informant. Thus, when a positive message was being conveyed, significantly more dogs interacted with the fan if the owner, rather than the stranger was the informant. These results are similar to those in infant studies.

Interestingly enough, there are two possible explanations for dogs’ not approaching the object. First, the object was slightly intimidating and the motivation to explore it may have been low and only activated by the owner’s encouragement. Second, the researchers also thought that perhaps the owner reading a magazine and facing away from the dogs could have lowered their motivation.

The results from this study indicate that dogs are able to distinguish the informant’s emotional message. Moods and facial expressions by the informants were also considered to be possible factors influencing the differential behavior of dogs in the different groups.

Results also indicate that dogs were affected by the general mood of the informant (more specifically, the owner), rather than understanding that the emotional message referred to a specific object. Mood modification is a process by which the observer is affected by the emotions of the actor and then mirrors those same emotions. It was noted that dogs’ behavioral changes were specifically directed to the fan and the area around it.

Dr. Merola reports that the motivation behind the dogs’ gaze in this research was “gaze alternation behavior.” This is behavior that is manifested by the subject between the object and the sharer of attention. This shows joint attention. For the dogs participating in this research, Dr. Merola states that the dogs need to have motivation to share attention and interests with others with no other goal. Secondly, the dogs and owners need to know that they are sharing joint attention.

Meanwhile, the results of the gaze alternation study indicated that the motivation behind the dogs’ gaze alternation behavior in general could not be considered a desire to obtain the object since dogs were somewhat intimidated by it. There was also an active search on the dogs’ parts to involve the owner when he or she was inactive by the gaze alternating between him/her and the fan. If dogs wanted owners to attend to them, they did not need to gaze alternate towards the object. Other attention-gaining behaviors to the owner alone would have been sufficient. The results seem to suggest that dogs wanted their owners to attend to the same object that they were attending to. This may be because a stranger’s feedback was not sufficient enough or relevant for them.

The current studies by Dr. Merola, Dr. Prato-Previde, and Dr. Marshall indicate that dogs look back not just to request a desired object or food, but to also check their owner’s (but not a stranger’s) reaction to an ambiguous object. It is the first study that shows dogs will modify their behavior towards an object depending on the informant’s positive vs. negative message. Thus, dogs do use social referencing in their interactions with humans, and when confronted with a potentially scary object, their behavior towards the scary object will be selective and dependent on the relationship with the informant.

If you are introducing your dog to a new or potentially scary object, your reaction matters! Appearing positive and happy may make a big difference in how your dog reacts immediately and in the future when approaching scary things. For more information visit USDAA.

Screen Shot 2016-01-03 at 6.03.39 PM


Thanks for visiting my blog.

Woofs & Wags!

Copyright © 2016 Claudia Bensimoun


Leave a Reply

%d bloggers like this: