A study by Rooney, Bradshaw and Robinson (2001. Do dogs respond to play signals given by humans? Animal Behaviour, 61:715-722.) investigated dogs’ responses to human play signals. These researchers found that humans do communicate a playful intent to their dogs and that their various behaviors when doing so can be considered interspecific play signals.
Additionally, they found that the success of human signals at instigating play was unrelated to the frequency of use. For example, patting the floor as well as whispering were both often used by people attempting to initiate play with their dogs, but dogs showed a low rate of playful responses to these signals.
In contrast, running towards or away from the dog as well as tapping their own chest were two human signals that were highly effective at initiating play with dogs but neither was used frequently by participants in the study.
The least effective human behaviors in their study for eliciting play in dogs were kissing the dog, picking up the dog, and barking at the dog, none of which ever resulted in play. Stamping their feet and pulling the dog’s tail had very low rates of playful responses by the dogs.
The human signals that most frequently elicited play were the forward lunge (the person makes a sudden quick movement toward the dog), the vertical bow (the person bends at the waist until the torso is horizontal), chasing the dog or running away from the dog, the play bow, and grabbing the dog’s paws.
Part of their study involved looking at the effects of vocalizations on play solicitation by humans of dogs. They used two composite play signals and asked whether they were more effective at soliciting play with or without vocalizations. The two composite signals were a play bow combined with patting the floor with their hands and foot shuffling combined with a forward lunge in the vertical bow position. The vocalization was saying “come on” in a whisper followed by “come on little dog” in a high-pitched voice.
The two composite signals used to initiate play with their dogs were more successful at eliciting play when accompanied by play vocalizations. It could be that the more components a play signal has, the more effective the signal is.
Interestingly, their data suggest that different play signals may invite different kinds of play. The composite bow signal with or without vocalization increased physical contact between the person and the dog and resulted in more licks of the person by the dog, but the composite lunge signal did not, either with or without vocalizations. Perhaps the bow acts as an invitation to approach and make contact, but the composite lunge signal does not. Perhaps one signal acts to solicit close contact play but the other instigates chase games.
Their study, though it was published eight years ago and is widely known in academic circles, does not yet seem to have reached wide audiences in the dog world. This is a shame because people and dogs could benefit from its findings. This study indicates that we should pay attention to whether or not the way we try to entice our dogs to play is effective, what the effects are in terms of the type of play our dogs’ do, and that we should consider adding vocalizations to our visual play signals.