Workpackage 2: Consequences of being imitated


Huber, A., Windsteig, V. & Barber, A.L.A. (2014) The Chameleon-Effect in Dogs – Do dogs recognize whether being imitated? Project report, Into Science course, Vetmeduni Vienna.

Following Paukner et al. (Paukner et al., 2005; Paukner et al., 2009) we started to examine dogs’ change in behaviour during/after being imitated by a human person. Although the abilities of dogs to imitate movements shown by a demonstrator were already proven by Huber et al. in the intraspecific context (Range et al. 2007) as well as in the interspecific context with humans (Huber et al. 2009), the ability of dogs to recognize whether they are imitated by a human has not been studied so far. For this reason the affiliation of the dog to humans who were or were not imitating them was investigated. The project was conducted in the Clever Dog Lab by two Master students, Annika Huber and Viola Windsteig. Twelve dogs of different breeds and ages and both sexes participated in this study. A prerequisite for the dogs was the ability to perform four to five different commands (i.e.: down, sit up and beg, turn-around, make a roll) in the distance of about 1.5 m to the owner. The study was conducted by three experimenters, the handler (H) of the dog, the imitator (I) and the non-imitator (NI). After the baseline phase to check whether the dog had a preference for either the I or NI, the imitation/non- imitation-phase started. While the imitator copied the dog's action, the non-imitator showed one of the other actions. Afterwards we repeated the preference tests to check whether the imitation of the dog had an effect on the relationship between the dog and the I or NI. Preference scores did not differ from chance level in any of the four preference tests. Although the study did not reveal any changes in the dog´s behaviour after being imitated by a human person, this project resulted in an advanced set up and a repetition of the research question. 

photo: Clever Dog Lab/Vetmeduni Vienna 


Berczik, J., Fürdös, I. & Barber, A. (2014) Do dogs prefer humans, imitating humans?  Project report, Into Science course, Vetmeduni Vienna.

This study investigated whether dogs prefer a person they had previously observed in a video clip imitating to a person they had observed not imitating another human. Our hypothesis was that dogs prefer an imitator of another person's actions, shown by higher preference for the imitator in a two-choice task (in terms of first choice and time spent with the person). However, during the video presentation, the dogs were expected to pay more attention to the non-imitation video as following different movements of two persons might require more concentration than imitation movements. The imitation situations were videotaped in advance: A model performed a set of four different movements, each movement was imitated by one of two experimenters. All actions were neutral movements and were carried out while sitting on a chair. For the non-imitation situation, the model performed the same movements as in the imitation video, while the experimenter performed them in a different order than the model. Each imitation/non-imitation video sequence was approximately 25 seconds long. In total, 41 adult dogs of various breeds (20 male, 21 female, ages 2-13 years) were tested. Regarding the two-choice task, more dogs chose the non-imitator than the imitator (21 versus 16), however, this difference was not significant. There was also no significant difference regarding the time spent in proximity to each experimenter in total or on a group level. Finally, there was no relationship between attention duration, first choice and time spent in proximity. In summary, we could not find any evidence for dog preferring imitators over non-imitators. A possible reason for this negative result may be the presentation of videos instead of live presentations or the mix of video presentation and live choice. Still, the study provides a first step and suggests a number of changes in the experimental design.

Photo: Anjuli Barber/Vetmeduni Vienna