Back in March I wrote this post looking at which species are better models for the evolution of human behavior and cognition. A new paper on the subject has come out in the journal Behavioral and Brain Functions.
The research compared several groups of dogs. The first comparison involved breeds selected for their ability to work while separated from humans, breeds that work in close proximity to humans, and mongrels. The second compared dogs with a high cephalic index and more centrally placed eyes to dogs with low cephalic indexes and more laterally placed eyes. The comparisons involved the ability of dogs in each group to follow the human pointing gesture.
In the first case, dogs that worked in close proximity (defined as within visual range) to humans were more successful at following the pointing gesture than the independent workers and the mongrels. In the second case, the high cephalic index, centrally placed eye group were more adept at following the pointing gesture. They conclude:
Regardless of our categorization, all groups of dogs showed an ability to rely on the human momentary distal pointing. However, the current study is the first, to our knowledge, to reveal striking difference in the performance of breed groups selected for different characteristics.
In accordance with our hypothesis, the performance of working dog breeds selected for intense visual contact with the owner proved to be better in the utilization of the human distal momentary pointing gesture than those selected for working independently of or visually separated from the owner. We suggest that this might not be attributable to differences in the cognitive abilities per se, but rather reflects a genetic tendency to be responsive to social stimuli in a cooperative context. Frank … proposed that one important difference between dogs and wolves is that dogs can more easily be brought under the control of artificial stimuli. For example, dogs can be trained also to respond to verbal or non-verbal human cues whilst this is usually difficult with wolves. However, in the current study, a general sensitivity gained during the process of domestication does not account for differences in performance of independent and cooperative working dogs. It is more likely that direct selection for utilizing human visual signals endowed cooperative breeds with an ability to inhibit their own spontaneous behaviour and benefit from human social cues.
For the second case the authors conclude:
In Study 2, as predicted, we found that the brachycephalic breed group was significantly better in using the human pointing gesture than the dolichocephalic group. The superior performance of these dogs can be explained by their much more focused attention on the signaller, and strongly indicate that a difference in a morphological characteristic can also influence the performance in communicative task.
The authors argue that this means that two independent phenotypes impact the ability of dogs to understand visual cues.
Update 1: I have fixed the link to the paper.