How dogs get the point what enables canines to interpret human gestures

Guest post by: Lucia Lazarowski, PhD candidate. Her research is available via free promotional access in the journal Behavioural Processes until February, 2016.

Hi Mia and Julie,

As a long-time fan of the blog, it is an honor to be a guest contributor! I am especially excited to tell DYBID readers about this research because it was somewhat of a pet project (pun intended). I am now a PhD student at Auburn University, but this study was done while I was working at North Carolina State University’s College of Veterinary Medicine. At NCSU, I worked with a team of veterinarians and animal behaviorists on a several projects aimed at improving selection and training of military working dogs, and I was primarily involved with studies related to explosives detection. 

Meanwhile in the canine cognition world, a hot topic was that of dogs’ ability to follow human gestures. Several studies have demonstrated that dogs are able to use human gestures, like pointing, to find hidden treats. An interesting finding that fueled a lot of the research in this area is that dogs perform better on these tasks than chimpanzees, our closest relatives, and wolves, dogs’ closest relatives. Is it possible that dogs are able to read and use human gestures because they co-evolved with humans, endowing them with a specialized human-like type of social cognition that their ancestors missed out on? Or, is it that dogs are such an integrated part of our lives that through our daily interactions they learn that paying attention to our body language pays off?

These two viewpoints have sparked a heated debate among canine scientists. In order to tease apart the roles of domestication and experience (or the nature/nurture debate, as your high school psychology teacher would call it), researchers have tested canines of different species (domesticated and wild-type) and different life histories (human-reared and feral). The domestication hypothesis, which suggests that point-following is an innate skill that dogs have acquired in a case of convergent evolution with humans, predicts that domestication alone is sufficient for point-following. The learning hypothesis, on the other hand, contends that dogs must learn through experience to follow human gestures, regardless of domestication status. 

The fact that chimps and wolves do not appear to utilize human pointing as dogs do seems to support domestication as an explanation. But, (plot twist!) if wolves are raised with humans from an early age and are tested in appropriate conditions, they can perform as well or even better than dogs.  To recap, groups that have succeeded at human pointing tasks include canines that are domesticated and socialized (pet dogs), non-domesticated and un-socialized (wolves), and non-domesticated and socialized (hand-reared wolves).  Hopefully at this point the missing piece of the puzzle is obvious: what about domestic dogs that have not been heavily exposed to humans? This vital yet untested sub-group of canines would help tip the scales in the domestication vs. experience debate.

At NCSU, we were gearing up to begin a new study investigating factors related to olfactory learning in canine explosives detection. The dogs acquired for this study were mixed-breed males around 1 year old, and unlike our previous studies which used trained military working dogs, these were laboratory-reared dogs. It occurred to me that this would be the perfect opportunity to test a group of dogs that met all of the proposed criteria for the “missing link”: laboratory dogs lack the same experiences that pet dogs living in human homes have (including the possibly critical opportunity to learn about human gestures), but they are socialized to humans at an early age and thus not fearful like feral dogs may be. Another bonus is that their life histories are known and documented, unlike dogs found in a shelter that at some point may have lived with people. If the opportunity to learn about human gestures is critical for point-following behavior to develop and not just domestication alone, these dogs would be expected to perform worse than pet dogs on point-following tasks. 

We tested 11 laboratory dogs and 9 pet dogs using methods established in previous studies in which dogs watched as humans performed two types of point (“easy” and “hard”, for simplicity’s sake).  What we found was that while pet dogs followed the harder point to the correct container significantly higher than chance, the laboratory dogs did not. Both groups of dogs were able to locate the correct container using the easier point, demonstrating that any failures were not due methodological flaws or to an inability to perform the demands of the task (note that success on these easier point trials can be explained by simpler mechanisms like physical proximity to the container).
 
Our results seem to suggest that exposure to humans and the opportunity to learn about the meanings of gestures plays an important role in dogs’ ability to follow pointing. 

Interestingly, a few dogs in the pet group performed just as poorly as the laboratory dogs, which would lend further support to the idea that individual experiences shape these abilities. Further, failures by the laboratory dogs are not likely caused by cognitive deficits due to an impoverished environment; the dogs received environmental enrichment including daily interactions with kennel and research staff, play-time with conspecifics, outdoor exercise, and a variety of toys (and after completing this experiment, participated in daily socialization and reward-based training sessions to facilitate future adoptions). Though domestication may likely contribute to dogs’ gesture-reading skills, specific life experiences may also be critical for their manifestation. 

P.S.: A happily-ever-after to this story: one of the subjects from this cohort, ‘Captain’, was adopted upon completion of the studies... by me! 

Lucia and Captain - all smiles!

Author
Lucia Lazarowski is a PhD student at Auburn University in the Comparative Cognition Laboratory. They collaborate with the Canine Performance Sciences program at Auburn University (Facebook).

If you found this interesting, you may also enjoy our guest post by Dr Bradley Smith: Take a walk on the wild side: Dingo science, or see all of our guest contributors.

Images: Copyright Lucia Lazarowski. 

References:
Lazarowski L. (2015). A comparison of pet and purpose-bred research dog (Canis familiaris) performance on human-guided object-choice tasks, Behavioural Processes, 110 60-67. DOI: http://dx.doi.org/10.1016/j.beproc.2014.09.021 [OPEN ACCESS until Feb 2016]

Kaminski, J., Nitzschner, M., 2013. Do dogs get the point? A review of dog-human communication ability. Learn Motiv. 44 (4), 294–302.

Udell, M.A.R., Dorey, N.R., Wynne, C.D.L., 2010. What did domestication do to dogs? A new account of dogs’ sensitivity to human actions. Biol Rev. 85 (2), 327–345.

Reid, P., 2009. Adapting to the human world: Dogs’ responsiveness to our social cues. Behav Process. 80 (3), 325–333. 

© 2015 Lucia Lazarowski | Do You Believe in Dog?