Behavior Genetics, Breed Bans, and Getting the Facts Straight

We read an article in the Denver Post last winter, an Op-Ed piece by Krista Kafer, with a title of “End bans on specific dog breeds”.  In the article, she admonishes the town of Breckenridge, which was considering a breed ban at the time, to “keep a few facts in mind”.  She then goes on to quote dog trainer and “behaviorist” Lisa Lucero that “30% of a dog’s temperament is the result of genetic coding while 70% comes from environmental factors”.

Really?  It is just flat-out not possible to know how much of any given behavior is due to genetics or the environment.  This claim reflects an utter lack of understanding of behavioral genetics and the complexity of how environmental and genetic factors work together to influence behavior.    Let’s start by talking about how behaviorists evaluate the variability of a particular trait among individuals and the relative contribution of genetic and environmental factors.

Because this article was about the potential for aggression in a particular breed-type of dog, let’s use that behavior in our example.  We first have to define a specific measure of aggressiveness, so let’s use how long (in seconds) it takes a dog to snap at a fake hand that is placed in his food bowl when he is eating.  We could also use a different type of behavioral measure, such as a ranking of various behaviors. We might score a dog a 1 if he wags his tail in response to the hand, all the way up to a 5 if he bites and holds onto the hand. 

We’d next have to test a population, or group of dogs, whose genetic relationships to each other were known.  There are different ways to do this but suppose we use the scores from twenty litters of Labrador puppies and those of their parents.  After testing, we’d have a score for each dog, either in latency to snap or its behavioral ranking.  Clearly, there will be variability among the scores – dogs would score differently. 

Then, using some sophisticated (well beyond the scope of this article) statistical procedures, that evaluated the degree of genetic relatedness with the behavioral score, we would come up with a number between 0 and 1 that would be a measure of heritability.  If the result was a 0.4, we could then conclude that for this specific group of Labradors, tested at this time in this specific way, 40% of the variability in our measure of aggression was due to differences in the genetics of the dogs.  This heritability measure would NOT be generalizable to ANY other group of dogs, or even the same dogs tested at a different time or tested in a different way.

This is quite different from saying that 40% of an individual dog’s tendency toward aggression was due to genetic factors.  Why? Because measures of heritability refer to the VARIATION in a behavior trait in a population.  We CANNOT parcel out the relative influence of genetic or environmental factors in a specific individual because behavior is always the product of complex interactions among genes and environments.

But wait, there’s more!  A review of studies of heritabilities of different behavioral traits by Silvia Ruefenacht and colleages found wild variation in the results, depending on the behavior examined, the dogs tested and the specifics of the study.  For example, they found that heritability for “temperament” varied between .10 and .51 in different studies.  And all this doesn’t address how aggression and temperament are defined and measured and how the two are related to each other.   Temperament gets defined in a variety of ways, and in some studies, temperament is viewed as different from aggression.

To over-simplify these complex interactions into an unsubstantiated claim of percentages as was done in the Denver Post article is a total misrepresentation of what behaviorists can and cannot say about the genetics of behavior. This is something that no scientifically trained behaviorist would ever do.

Why is this an important distinction to make?  In the case of the article about breed bans, the fact that the genetic information was so wrong, and so off-base, makes any other claims or arguments suspect.  If the facts are wrong about the genetics, how can we expect readers to trust any information or recommendations contained in the article?

In fact, there are other problems with the article.  Kafer’s major thesis and concluding sentence “Educating owners – not euthanizing dogs – is the answer to public safety concerns”  is itself incomplete and an over simplification. The causes of dog bites are complex.  It’s not just the breed of the dog, nor is it just the education level of the owners. There are a number of factors, including behavior of the victim, elements of the context  where the bite occurred, and more, that influence bite likelihood and must be considered if we are to design programs to reduce dog bites.  But that’s a topic for another article.  

Regardless of one’s stance on breed bans, we do agree with one point in the article however – it’s a good idea to get one’s facts straight if one wants to be taken seriously and have credibility with readers.

If you want to get the science right and be taken seriously, join us on March 27th for our upcoming CAABChat on "Social Roles and Relationships in Dogs” with two Certified Applied Animal Behaviorists who have spent the majority of the careers studying dog behavior.   Click on the link above or visit www.CAABChats.com

And if you really want to delve into the science of animal behavior as it applies to dogs, take our four session, On Demand Certificate course “Shining the Light of Science on Canine Behavior”  available at PetProWebinars.com.