I have no doubt that if you asked the majority of people to tell you the characteristics that they most want in their dog, trainability would be very high on that list. Obviously for working dogs, their ability to learn and to respond to commands is a vital portion of their function. For family dogs, basic training, at least to the level where the dog reacts reliably and does what you tell him to do, is often the difference between whether your dog is a comfortable companion, or whether it is an intrusion into your life. For this reason many canine behavioral researchers have been looking for ways to determine whether any given puppy will grow up to be an easily trained and responsive adult or not.
There are a set of canine intelligence tests which can be administered individually to dogs, and these do a reasonable job at predicting just how trainable a given animal will be. An example of one such test appears in my book The Intelligence of Dogs. The problem with the these tests is that they require the dog to be at least one year old (and preferably a bit older than that) in order to be valid. Of course, they need to wait until the dog is a year or year and a half in age makes such tests not particularly useful as a tool for selecting puppies while they are still in their litter. For that reason many researchers decided to see if there were differences in trainability among the various dog breeds. It turns out that there are systematic differences with certain breeds being much more trainable than others (see for example here or here to get more information on this). The advantage of using breed as an indication of trainability is that it can be used as a technique to help select a puppy. However, there are problems with this method since, although the differences between breeds are generally reliable, within any breed there can be individual variations. Consider for instance, Labrador Retrievers (which are the most popular dogs in the world), which are ranked as the seventh most trainable dog breed. Some Labs are so brilliant that it seems to take only moments, perhaps a half a dozen repetitions, for them to learn many tasks. But of course, as any dog trainer can tell you, there are some occasional Labs who respond to training no more quickly than river rocks. So using breed alone only gives you a general estimate of trainability.
Many scientists continue to look at the nuances of breed differences in behavior, therefore I was interested when I had the opportunity to read a research report by a team of researchers headed by Helena Eken Asp from the Swedish University of Agricultural Sciences, in Uppsala, Sweden. This report was recently accepted for publication by the journal Applied Animal Behaviour Science 1. The aim of this research was to look at breed differences in the everyday behavior of dogs, specifically grouping dogs into working dogs versus other breeds using the classification criteria from the Swedish Working Dog Association. This study involved analyzing an extensive set of questionnaires filled out by 3591 dog owners. The data from these questionnaires could be broken down into 18 separate subscales, each of which represented a behavioral predisposition (a typical behavior pattern which characterizes the dog) or a personality characteristic. Included in these personality characteristics were things like various types of fearfulness, aggressiveness, excitability, energy level, playfulness, and trainability. Although the researchers did find systematic differences in the behaviors of the 20 dog breeds represented in their study, what I found particularly interesting was the relationship that they found between playfulness and trainability. Basically they report that the dogs who showed the most interest in playing with humans also turned out to be the dogs which were most trainable.
There are some practical reasons why I was caught by this particular finding. Personality tests can be given a lot of earlier then intelligence or trainability measures. In fact a number of popular dog personality or temperament tests are typically given when puppies are around seven or eight weeks of age. So if the personality characteristic that we call playfulness is a good predictor of trainability then this might be a useful measure to take when assessing puppies in a litter.
Nonetheless, one reported result from a single laboratory, especially on a set of dimensions which were not the major focus of the research, is not convincing enough for most scientists to use to base recommendations upon. However it turns out that this is not an isolated report since more than a decade ago Kenth Svartberg, who was then at Sweden's Stockholm University, published a series of papers2, 3 in which he also looked at personality differences among various breeds of dogs, and their relationship to trainability. His research was impressive since it used data from a standardized behavioral test from 13,097 dogs of 31 breeds. His data could also be analyzed in terms of personality dimensions. What he found was that the best predictors of trainability in dogs involved a combination of personality factors which included the dog's sociability (friendliness), lack of fearfulness, and playfulness. The reason that playfulness was of particular interest is because it is so easily measured. Basically all you have to do is to toss a rag in front of the dog and observe how interested he is in chasing it, and then playing a bit of tug-of-war with the person giving the test. This is the kind of simple measure that one could use to test puppies. Because we have more than one research report that confirms the relationship between playfulness and trainability it now seems to be more sensible to suggest that we might use a simple test of playfulness as an additional method if we are trying to select the most trainable puppies in a litter.
Svartberg was quite impressed by the strength of the relationship between playfulness and trainability. He would later go on to speculate that if breeders concentrated more on producing and selecting for more playful dogs, it could eventually lead to an increase in the overall trainability of the specific dog breeds, or breeding lines, which adopt this focus.
Why should trainability be related to playfulness? Svartberg suggests that playfulness is really part of a complex of behaviors that include sociability and a lack of fearfulness around people. If the dog is more playful then it is also more interested in human interactions in general and therefore is more likely to be paying attention to his handler's communications and behaviors. A dog which is not paying attention is unlikely to be easily trainable. The data from Svartberg and others provide a hopeful note for how we can shape the nature of our dogs in the future since playfulness systematically varies among dog breeds and it appears that it is passed on genetically. Therefore if breeders use playfulness as one of the criteria when they are selecting mating pairs, the amount of playfulness (and therefore trainability) can be increased in any line of dogs. In any event, if you are interested in selecting a trainable puppy from a litter, on the basis of this research, I think that it might be worthwhile to bring along a rag or a soft pull toy and use it to see how playful each of the puppies is.
Stanley Coren is the author of many books including: The Wisdom of Dogs; Do Dogs Dream? Born to Bark; The Modern Dog; Why Do Dogs Have Wet Noses? The Pawprints of History; How Dogs Think; How To Speak Dog; Why We Love the Dogs We Do; What Do Dogs Know? The Intelligence of Dogs; Why Does My Dog Act That Way? Understanding Dogs for Dummies; Sleep Thieves; The Left-hander Syndrome
Copyright SC Psychological Enterprises Ltd. May not be reprinted or reposted without permission
1. Helena Eken Asp, Willem Freddy Fikse, Katja Nilsson, Erling Strandberg (2015). Breed differences in everyday behaviour of dogs. Applied Animal Behaviour Science, http://dx.doi.org/10.1016/j.applanim.2015.04.010
2. Kenth Svartberg, (2002). Shyness-boldness predicts performance in working dogs. Applied Animal Behaviour Science, 79, 157–174.
3. Kenth Svartberg, (2006). Breed-typical behaviour in dogs – historical remnants or recent constructs? Applied Animal Behaviour Science, 96, 293–313.