A new study indicates that the body language of a human can trick a dog and influence its actions. Published in the journal PloS ONE, it deals with the ability of a dog to want to follow to cues projected by humans. The research involved 149 dog owners who were asked to bring their dogs to University of Milan’s psychology laboratory.
The head of the study, Sarah Marshall-Pescini, and other colleagues designed a number of experiments that consisted of giving dogs choices between two bowls of food. In some scenarios, one bowl had only one piece of food while the other had six pieces. Other experiments involved dogs having to choose between two bowls with the same quantity of food. In some instances, dogs were allowed to choose which bowl they preferred.
There would also be times when a person would be present before the dog was given the opportunity to choose the bowl. Sometimes, the person would favor one of the bowls by staring at it. Other times, the person would talk to the dog while showing a piece of food from one of the bowls. These interactions of a person between dog and food were varied.
The study showed that in majority of the trials where no human was present, 73% of dogs made the obvious choice of favoring the bowl with more servings. But in scenarios where a person was involved, the dogs would generally pick the bowl favored by the human even if this had the smaller portion of food. The study showed that the most convincing gesture that would influence the dog’s decision was the hand to mouth action by the person.
In one situation, dogs favored a smaller plate when a person approached the food, held a piece over her mouth, and then put it down again before leaving. In another scenario, a person did the same thing but she also talked to the dog while holding up the food. In the third situation where the food was held up and placed back in the bowl by utensils behind the curtain which hid the person, the dogs were not more inclined to choose that bowl. This suggested that the dogs took cues from human actions.
According to the researchers, “the current study adds to a small but growing literature showing that social learning is not necessarily always the best strategy and provides an experimental paradigm which may potentially be used to explore when an animal will rely on private vs. social information.”