[In the first experiment] The results showed that the dogs in the compatible group were able to ga
in the reward much more easily and with fewer trials than those who had to counterimitate their owners.
This strongly suggests that dogs, like humans, learn by automatic imitation.
In a second experiment all the dogs were rewarded if
they imitated their owners, and in this case the dogs who had been in the incompatible group fared worse than those in the compatible group, making more errors through counterimitation.
The researchers say this suggests the imitation depends on “sensorimotor experience and phylogenetically general mechanisms of associative learning” and that imitation in dogs is shaped more by their interactions with people than by their evolutionary history of domestication.