I used an experimental design called object choice. First I hid a piece of food from the dog in one of two containers. In the first condition I tried to inform the dog of where the food was hidden by giving an informative cue, e.g. a pointing gesture. In the second condition I instead prohibited the dog from approaching one of the containers (the one containing food) by using the same gestures, but now combining them with a strict No! Before this condition I had established a competitive relationship with the dog by taking away food from it. This condition tested if the dog could draw a conclusion from the fact that I did not want it to find the food.

After the cue was given the dog was allowed to chose one container and eat the food if it chose the correct one. The dogs choices were recorded and the results showed that they were able to use the informative cues to find the food, but they did not understand the forbiddance, which means that the dogs could not draw any conclusions from the competitive motive. Nor could they make an inference from the forbiddance itself.

The experimental set up can be seen as an illustration to the right.