Does Activity Level Predict Trainability in Dogs?
Data shows that higher activity level improves trainability up to a point.
Posted August 10, 2022 | Reviewed by Lybi Ma
- A dog low in energy level will not test many behaviors and may be slow in the responses that trainers want him to learn.
- A dog high in activity level will act in a berserk manner and miss cues and behaviors that are important for proper learning.
- Energy and activity levels are related to dog intelligence and trainability. Energy too low, or too high, impairs performance.
The intelligence and trainability of dogs is a major concern to many people. When it comes to service dogs, such as those used to assist the blind or handicapped, or those working with the military and police, it is important to have an intelligent animal that can successfully complete a training program and work in the field. A dog that flunks out of a training program in midstream can cost many thousands of dollars because of the wasted instructional resources and training time.
For the average person who simply desires a dog that can participate in sporting activities, hunting, obedience, agility, and rally trials, or for someone desiring a pet that understands what is wanted of him at home and responds appropriately, there is also a need to know which dogs will prove to be intelligent and trainable. For these reasons, it is important to find variables that can predict a dog's cognitive abilities and ability to learn.
Energy Level and Trainability
Anecdotally dog trainers often feel that an important factor in the trainability of dogs is their activity and energy level. According to many trainers, dogs with a lot of energy, who move rapidly, going from one point in the environment to another, are most likely to learn quickly. This is consistent with some basic psychological theorizing. From the beginning of his research on operant conditioning, the psychologist BF Skinner pointed out that if an animal performs an action and it is rewarded, then that action is more likely to be performed in the future. It is from such rewards and reinforcements that our dogs learn tasks and solve problems. A dog that has a low activity level will not explore many behavioral options and therefore is less likely to find out which behaviors are correct. Such a behavior predisposition will certainly reduce a dog's ability to learn things.
I had a dog who illustrated this point. Banshee was a Cavalier King Charles Spaniel and probably the most laid-back and relaxed animal that I have ever had. This made him a fine companion around the house, but very difficult to train for obedience competition. He never did learn to do scent discrimination simply because if he made a choice and it was not immediately rewarded, he typically would just stand there looking around and not checking out any other options. That meant that he never went on to find the appropriately scented object for which he could earn a reward and also learn what was wanted of him.
Maybe Not Too Much Energy
There is a contrary opinion that is held by many dog trainers and it is also supported by some psychological evidence. Here the argument grants that dogs too low in energy will not learn very efficiently, but on the other hand, it contends that dogs with very high levels of energy do not learn as well either. The hypothesis: If a dog's energy level is high enough the dog essentially acts a bit berserk, going from one action to another without waiting long enough to even see if it is rewarded. This means that the overactive dog is not processing all of the information available in the environment either, which can also result in slowed learning.
The effects of energy level on behavior, in general, were demonstrated around the beginning of the 20th century and became known as the Yerkes Dodson Law. It was originally couched in terms of arousal level, referring to levels of generalized neural action, but it is usually behaviorally measured in terms of overall activity and energy levels.
A New Data Analysis
As a psychologist, it seems to me that it would be useful to know if a dog's habitual activity level is actually associated with its intelligence and trainability, and if so what is the nature of that relationship. I was surprised to find no scientific work specifically addressing that question in canines. Therefore, I turned to a database that I already have in my possession. It will not allow me to make predictions on the basis of individual dogs, but since different dog breeds characteristically have different activity levels, it still would allow me to test the relationship between activity level and trainability in dogs at the breed level.
The data set assessed the personality traits of 133 individual dog breeds. It was collected for one of my books, Why Does My Dog Act That Way? The data was based on ratings by 96 dog experts, each of whom was willing to contribute several hours to rank dog breeds on five different behavioral dimensions. Two of these dimensions are useful to answer the current question. The first was called Energy Level and it is a composite measure that looks at the dog's activity level overall. The second was labeled Intelligence-Learning Ability, which is a measure of how easily a dog learns and solves problems.
The dog breeds were ranked for energy and trainability and then the composite rankings were divided into quartiles and assigned a score from 1 through 4 for each of the two behavioral dimensions separately. Dog breeds that fell in the bottom 25 percent were assigned a score of 1, and dog breeds that fell in the top 25 percent for that dimension were assigned a score of 4.
The scores on energy and trainability were then cross-tabulated. An analysis showed that there is a statistically significant* relationship between a dog's energy level and its intelligence and trainability. To make these results clearer let's look at the likelihood that a dog breed will be above average (in the top 50 percent) in terms of trainability for each given energy level. When we do that kind of breakdown we get the results shown below.
First, looking at the dogs with the lowest activity level, just as operant conditioning principles predict, these dogs show the lowest trainability with only 29 percent of these slow-moving dogs scoring above average. However, the results do not show a linear relationship between activity level and trainability.
The pattern of results is actually similar to those expected on the basis of the Yerkes Dodson Law in view of the fact that trainability increases with increasing energy level up to a peak, and then declines for the highest category of energy level. The peak performance in trainability is achieved for dogs that are in the medium-high level of energy (the quartile running from 50 to 75 percent from the bottom). These clearly seem to be the most trainable dogs since the probability that a dog with this energy level will have higher than average intelligence and trainability is 70 percent. This decreases to only 47 percent for the highest level of activity.
These results clearly show that activity and energy level in a dog does make a difference when it comes to its trainability. All other things being equal, it appears that laid-back, easy-going dogs are not going to be readily trainable, however, neither are their hyperactive cousins. It is the dog with a medium-high level of energy who is the most likely to prove to be a quick learner.
It is important to note that these results are based on a breed-wide level of analysis. I would hope that some canine cognition laboratory would repeat my observations using the activity levels of individual dogs.
Copyright SC Psychological Enterprises Ltd. May not be reprinted or reposted without permission
* Statistical Note: The chi-square statistic for these data is 17.66 (9 df), p < 0.05.
Coren, S. (2006). Why does my dog act that way? A complete guide to your dog’s personality. New York: Free Press.
Ferster CB, Skinner BF (1957). Schedules of Reinforcement. New York: Appleton-Century-Crofts. ISBN 0-13-792309-0.
Yerkes RM, Dodson JD (1908). "The relation of strength of stimulus to rapidity of habit-formation". Journal of Comparative Neurology and Psychology. 18 (5): 459–482. doi:10.1002/cne.920180503