Dogs take the human perspective into account when deciding whether to take food that they have been told not to take. This is the conclusion of a recent series of experiments by Juliane Kaminski, Andrea Pitsch and Michael Tomasello described in a research paper called Dogs Steal in the Dark.
The experiments all had a similar set up, in which a dog was in a room with a food item on the floor. Dogs were pre-tested to ensure that they understood the cue not to take the food. They were told not to take the food and then the experimenters recorded whether the dog took the food or not and if so, how long it took for them to do so. In the first experiment, a human was present in the room and there were four conditions: 1) the food and the human were both in darkness, 2) the food was illuminated but the human was in the dark, 3) the food was in the dark but the human was illuminated, and 4) the food and the human were both illuminated.
The dogs took the food most often when both the food and the human were in darkness, and least often when they were both illuminated. The dogs took the food faster when it was not illuminated, but whether or not the human was illuminated had no effect on the time until the food was taken. These results suggest that the level of light in the room had an influence on the dogs’ behavior.
In the second experiment, the food was either illuminated or in the dark, but the human left the room after giving the cue not to take the food. The goal of this experiment was to determine if dogs were avoiding food when it was lit. The results were that the dogs almost always took the food no matter what light situation it was in, but they took it faster when it was illuminated than when it was in the dark. This indicates that the dogs are not simply avoiding food that is illuminated.
In the third experiment, the researchers investigated whether the overall amount of light in the room was important to dogs or whether they were responding to the specific location of the illumination. There were two different situations tested. For one group of dogs, the human in the room was always illuminated. Half the time, the food was illuminated, and half the time the light was directed at another spot in the room. For the other group of dogs, the human in the room was always in the dark and half the time, the food was illuminated, and half the time the light was directed at another spot in the room.
In this study, whether or not the human or the food was illuminated had no influence on the likelihood that the dogs took the food, but they waited longer to take the food when it was illuminated. This suggests that the location of the lit area matters, as opposed to dogs just reacting to the overall amount of illumination in the room. This shows that the visibility of the human is not the factor that causes the difference in behavior.
Overall, the conclusions drawn from these studies are that dogs do take into account illumination when taking food that they have been told not to take. Dogs were more likely to take the food when it was dark compared to when it was light. They took the food faster when the food was in darkness when a human was present, but took the food faster when it was illuminated when they were alone. Whether or not a human in the room was illuminated did not affect their behavior.
This research supports the idea that dogs are aware of and consider the human point of view when deciding whether to take food or not. To put that into a larger context, this relates to the ever-increasing body of evidence that dogs have a theory of mind, meaning they have an understanding that other individuals have different perspectives, knowledge and emotions.
This is very well done research that teases apart a number of variables to add to what we know about the canine mind. Unfortunately, most of the articles in the popular press only say, “Dogs steal more food in a dark room than in a light room,” which is a huge oversimplification of the complexity of this interesting work.