Dog's discrimination of human selfish and generous attitudes: the role of individual recognition, experience, and experimenters' gender.

Discrimination of and memory for others' generous and selfish behaviors could be adaptive abilities in social animals. Dogs have seemingly expressed such skills in both direct and indirect interactions with humans. However, recent studies suggest that their capacity may rely on cues other than people's individual characteristics, such as the place where the person stands. Thus, the conditions under which dogs recognize individual humans when solving cooperative tasks still remains unclear. With the aim of contributing to this problem, we made dogs interact with two human experimenters, one generous (pointed towards the food, gave ostensive cues, and allowed the dog to eat it) and the other selfish (pointed towards the food, but ate it before the dog could have it). Then subjects could choose between them (studies 1-3). In study 1, dogs took several training trials to learn the discrimination between the generous and the selfish experimenters when both were of the same gender. In study 2, the discrimination was learned faster when the experimenters were of different gender as evidenced both by dogs' latencies to approach the bowl in training trials as well as by their choices in preference tests. Nevertheless, dogs did not get confused by gender when the experimenters were changed in between the training and the choice phase in study 3. We conclude that dogs spontaneously used human gender as a cue to discriminate between more and less cooperative experimenters. They also relied on some other personal feature which let them avoid being confused by gender when demonstrators were changed. We discuss these results in terms of dogs' ability to recognize individuals and the potential advantage of this skill for their lives in human environments.

Individual differences in social and non-social behaviors in domestic dogs (Canis familiaris) during the acquisition, extinction and reacquisition of a problem solving task.

Dogs are able to solve different problems by trial and error learning, but it seems that they cannot understand the means-end connection. Some studies suggest that dogs' performance is influenced by their breed and by the level of familiarity with the person they interact with. In our study, we assess individual differences in both social and non-social responses in a problem-solving task during the acquisition, extinction, and reacquisition phases. In order to investigate the effect of familiarity, in the first experiment, the human present during the task was either a familiar (the dog's owner) or unfamiliar person. In the second experiment, we compared breeds (Retrievers and Shepherds) that had previously shown differences in a communicative task. The results revealed that all groups learned the task and became more efficient in the acquisition trials. These non-social responses diminished during extinction, where an increase in social responses was observed. With regard to individual differences, dogs were more persistent in searching the reward during the second extinction trial when the owner was present (in contrast with a stranger), and also looked longer at the unfamiliar person at the beginning of the acquisition trial. On the other hand, Retrievers showed greater social motivation during reacquisition and Shepherds picked up more bones during the third acquisition trial, thus suggesting a more persistent search of the reward. These findings highlight the relevance of studying different learning schedules as well as individual differences in problem-solving ability so as to improve selection and training techniques.