The begging test has been used to test the ability of dogs to interprete human attention. Facial cues, such as gaze and head position are the main signs in the communication between humans and dogs. Yet, the structure of the begging test used in research did not allow checking whether there are graduations of attention. Thus, it's not possible to say if the head and look directly driven (much attention) to the dog increases the performance in relation only to head position, but without looking directly (little attention) or head plus eyes diverted (inattention). Hence, a test using 58 mixed and pure bred dogs (Golden Retrievers) was performed, in three attention conditions, with men and women strangers to these animals, which served as experimental subjects. Results showed that regardless of race, dogs' gender of or people's gender, dogs have fewer errors on inattention condition Conversely, the performance of the dogs did not differ between the conditions of much attention and little attention, but was significantly higher than in the condition of inattention. In conclusion, dogs are good interpreters of human attention, but they make their choices based on cues related more to the head position than the look. Therefore, it is suggested that there isn't a interpretation degree of human attention in dogs. The emotional stimulus generated by the direct gaze, sustained among social animals, may explain why the gaze is averted as a cue of human attention in a test with strangers to the dog.
domestic dog; human being; attention; intention; communication
