Do Wolves Understand Cause and Effect Better Than Dogs?
The headline of an essay asserts they do, but dogs aren't dumb-downed wolves
Posted Sep 17, 2017
Media and animals; The importance of getting it right
Last evening and today I received a number of email messages and queries about an essay called "Wolves understand cause and effect better than dogs" that was published in ScienceDaily. There also is a picture of a wolf participating in the experiment about which this essay is concerned with a caption that reads, "Wolves are better in causal reasoning than dogs."
Coincidentally, as I was writing this essay, I was asked to give a talk at an upcoming meeting about the role and influence of media in representing nonhuman animals (animals) to a broad audience, and the importance of getting things right. Among the reasons I write essays for Psychology Today and other popular media outlets is to offer what we know and are learning about the cognitive, emotional, and moral lives of other animals to a broad audience who are not researchers or who conduct studies in other areas. In a previous essay called "Why Writing for Psychology Today Is a Good Idea," I wrote more about this topic as a follow up to an essay by Asit Biswas and Julian Kirchherr titled "Prof, no one is reading you." They write, "An average academic journal article is read in its entirety by about 10 people. To shape policy, professors should start penning commentaries in popular media."1 I agree, and, when they do, popular media has to be accurate, because many people will only read what they can understand or what they have time to peruse or scan. Often it's only the headlines or the first few sentences of a piece. They assume what they're reading is accurate.
Because studying dogs, wolves, and other canids, along with the process of domestication, have long been among my fields of interest, I was intrigued by the title of the ScienceDaily essay and wanted to know more about the study being discussed. It's interesting and important to compare the behavior of wolves and dogs because it's well accepted that wolves gave rise to modern day dogs.2
This ScienceDaily essay begins, "A rattle will only make noise if you shake it. Animals like the wolf also understand such connections and are better at this than their domesticated descendants. Researchers say that wolves have a better causal understanding than dogs and that they follow human-given communicative cues equally well. The study provides insight that the process of domestication can also affect an animal's causal understanding."
The ScienceDaily piece is about a research paper by Michelle Lampe and her colleagues titled "The effects of domestication and ontogeny on cognition in dogs and wolves," that is available online. In this project, the reasoning abilities of 14 dogs and 12 captive human-socialized wolves were studied. It turns out that some of the dogs and wolves were able to find hidden food using human cues such as direct eye contact and pointing gestures. Their ability to find hidden food also was tested using behavioral cues that didn't include eye contact or pointing, such as reaching out to the food, and using noise produced by an object containing the food when it was shaken.
In the ScienceDaily article we read, "Both dogs and wolves were able to follow communicative cues to find hidden food. However, without direct eye-contact, neither the dogs nor the wolves chose the correct object. In the absence of a human to show them where the food was located, only the wolves were able to make causal inferences. In this experiment, the wolves showed an understanding of cause and effect, which the dogs lacked."
What's very interesting about this study is that dogs and wolves who were living under identical conditions and who underwent the same training regime were studied. So, while I agree that the study under consideration might provide "insight that the process of domestication can also affect an animal's causal understanding," the authors themselves offer some reasons why the data they present are only suggestive of this relationship. For example, in the ScienceDaily essay we read, "It cannot be excluded however, that the differences can be explained by the fact that wolves are more persistent to explore objects than dogs. Dogs are conditioned to receive food from us, whereas wolves have to find food themselves in nature." In addition, the researchers recognize that the wolves were socialized and accustomed to human contact, and this might be why they used eye contact, whereas in other studies they did not.
The data that are presented are very interesting, but also rather preliminary. First of all, the title of both essays would have been more accurate if the word "captive" was used, especially concerning the wolves. It also would have been very interesting to control for the wolves persistence and to study non-socialized wolves or individuals who had minimal or different degrees of socialization. Further studies clearly are needed.
In the research essay we read, "Our results seem to imply that domestication impaired dogs’ ability to understand causal relationships, as in this condition, wolves outperformed pack dogs and were the only group to perform above chance level." (my emphasis) All well and good, but the phrase "seem to imply" caught my attention and that of someone who wrote to me. Of course, the results do imply, rather than seem to imply, that wolves understand this specific type of cause and effect relationship better than dogs. However, even the researchers recognize that questions still remain about just how robust and generalizable the cause and effect relationship is. Perhaps there are other cause and effect relationships on which dogs would perform better than wolves. And, it's possible that wild wolves need to be able to infer cause and effect better than dogs who have humans to help them along. There are many questions awaiting detailed comparative research.
Furthermore, how all of these data relate to the process of domestication remains open to discussion. The researchers write, "In conclusion, our results confirm that wolves can adapt their social cognitive abilities to their social environment, in this case to humans and their communication. Possibly for this reason, we found no evidence that domestication has altered how dogs use human-given cues. On the other hand, we found that domestication has left a mark on how dogs perform in a causal task."
Dogs aren't merely dumb-downed wolves
As are many experts on domestication, I too am leery of explanations of how domestication has worked when only a small number of individuals are studied in highly artificial situations. Along these lines, one very well known dog researcher wrote to me, "Wolf-dog differences cannot be explained on the basis of 1-2 experiments," while another wrote to me about their concerns about the limitations and questions the researchers themselves raised.
Variability among dogs is the rule rather than the exception
In studies of dogs, it's not at all surprising to find different results from different research groups who study different animals who have different experiences and individual histories. Research conducted in various dog labs and on free-ranging dogs uniformly shows there is an incredible amount of within-species variability among dogs. When I asked a dog expert about why there is so much variation in studies of captive dogs, they responded with the question, “Who are these dogs in all of these tests?” They went on to explain that they were referring to the fact that studies frequently treat all dogs as equivalent, but they're not. Many studies have shown that It’s extremely difficult to say without exception that all or even most or many dogs do this or that in a given situation, or that dogs and wolves are similar in this way and different in that way. Within-species variability is intriguing and highlights what is so exciting about studies of dogs and other animals.
So, while I am very interested in the results of the study under consideration, I also only cautiously accept the explanations for what the researchers discovered. Whether or not, and how, domestication has played a causal role in the appearance and perpetuation of cognitive and behavioral differences between dogs and wolves, remains a fruitful area for future research. And, I'm not necessarily expecting simple or single-factor explanations.
Captive wolves may understand cause and effect better than dogs
Let me emphasize that my questions about this study do not mean the researchers are wrong or that they don't recognize what they've actually done, but rather, it's important for readers to recognize that much more comparative research is needed. And, media has to get it right.
Perhaps I wouldn't have received a number of queries about the study if the summary essay, meant for a broad audience, had a different title, maybe something along the lines of "Captive wolves may understand cause and effect better than dogs." The researchers' qualifying statements make it clear that they're aware of the limitations of their study, and a different title would have been a more accurate introduction to what they actually learned.
What's the take-home message? What media puts out matters
The title of the ScienceDaily essay grabbed others' and my attention, and I worry about the take-home message received by the people who wrote to me saying something like I needed to know that it has been shown that wolves are better than dogs in understanding cause and effect.
Maybe this is so and maybe it's not. I look forward to further studies in this area. Dogs are fascinating beings and learning more about their cognitive capacities will be a win-win for all, regardless of how they compare with their progenitors. Ample research shows that dogs aren't merely dumb-downed wolves. Similar to numerous other animals, dogs are rather clever beings.
Please stay tuned for more exciting comparative research on our companion and other animals' cognitive abilities.
1Biswat and Kirchherr also write, "If a paper is cited, this does not imply it has actually been read. According to one estimate, only 20 per cent of papers cited have actually been read. We estimate that an average paper in a peer-reviewed journal is read completely by no more than 10 people. Hence, impacts of most peer-reviewed publications even within the scientific community are minuscule. I've always been interested in learning more about the impact of academic publications. Years ago I learned that in his book called Science as a Process: An Evolutionary Account of the Social and Conceptual Development of Science, the renowned philosopher David Hull noted that publishing a paper in an academic journal was equivalent to tossing it into a garbage pail (For more discussion please see my essay called Assessing publication impact.)
2It's also important to note that when people write or talk about a "socialized wolf" -- an individual who is comfortable around humans, for example -- as being a "domesticated wolf," this is incorrect, for a domesticated wolf is a dog. (For more on domestication please see Mark Derr's wide-ranging and detailed essays.)
Marc Bekoff’s latest books are Jasper’s Story: Saving Moon Bears (with Jill Robinson); Ignoring Nature No More: The Case for Compassionate Conservation; Why Dogs Hump and Bees Get Depressed: The Fascinating Science of Animal Intelligence, Emotions, Friendship, and Conservation; Rewilding Our Hearts: Building Pathways of Compassion and Coexistence; The Jane Effect: Celebrating Jane Goodall (edited with Dale Peterson); and The Animals’ Agenda: Freedom, Compassion, and Coexistence in the Human Age (with Jessica Pierce). Canine Confidential: Why Dogs Do What They Do will be published in early 2018. Learn more at marcbekoff.com.