“But it’s unhealthy to protect your dog from everything! If you do that, it’s just like overprotecting your child. They won’t be able to cope with the real world!”
This is another one of the criticisms one often reads about force-free training. It is generally presented by someone advocating the use of aversives in training. They’ve found another reason to say training with pain is necessary. They can say it’s all for the future good of the dog!
What is Stress?
The question of stress is one that I have given some careful consideration to. Stress has many definitions, but here are a few pertinent ones:
“A physical, chemical, or emotional factor that causes bodily or mental tension and may be a factor in disease causation.”–Merriam-Webster online
“The non-specific response of the body to any demand for change.”–Hans Selye, 1936, the Hungarian endocrinologist who coined the word.
(Selye also coined the word eustress: “Stress that is healthy or gives you a sense of fulfillment.”)
Is Stress OK At All?
Clearly stress is a serious deal, but maybe not all bad. Prescott Breeden published a very nice post on Stress and Learning recently. His article deserves a careful read, in case you haven’t read it already.
Are you back now? Great! I won’t synopsize the whole article, but the two relevant points to my topic are:
- a little stress can be beneficial to learning; and
- the optimal amount of stress for learning varies inversely with the difficulty of the task.
How to Use this Information About Stress
If little bit of stress can sometimes be a good thing, how on earth would you do that in a humane way? In the first place, not with positive punishment or negative reinforcement. That means not by hurting, scaring, or pressuring your dog if they make a mistake or to get them to do the “right” thing.
Introducing aversives into training as a way of preparing the student for the real world doesn’t make sense. That’s like deciding, “I want my kids to be able to cope with the real world, so I’m going to pinch them whenever they make a mistake doing their math homework.”
Huh? I really hope that’s not the type of world you’re planning for your child or dog. As strange as our world may be, that’s not something that happens very often. (Plus, way to go! Isn’t your kid going to just LOVE math now!) Much more often, if your daughter grows up to use math in her job, the difficulty will come if she has to do it while 1) the person in the next cube in her office is playing music; 2) her boss just fussed at her; or 3) she has a sick kid at home.
Now we’re getting somewhere. Those are real life stressors.
In psych experiments, stress is induced in humans in a variety of ways, but generally by exposing the humans to something stressful before or during the actual work of the experiment that is not contingent on their performance of a task. It may be social pressure, a self control exercise, or something like a cold pressor test, where a person is asked to leave his or her arm immersed in extremely cold water as long as they can stand it. (There are certainly experiments with all sorts of animals and humans where various unpleasant or painful things are used for punishment, and we would do well to heed the outcome of those experiments. My point here is that you don’t need contingent aversives to raise a person or animal’s stress level.)
So if a little stress could be OK, can we get any ideas about how to fit it into a humane training structure? You bet!
Enter Mr. Skinner and his cohorts. Contrary to his popular image as someone overly cerebral in a white lab coat who trained animals in a completely sterile environment, Skinner was also passionately interested in human education. Preparation for real world situations was a cornerstone of his methods.
Skinner believed that an animal or human did not have to make mistakes to learn the correct behavior or answer. I haven’t gotten to the original Skinnerian source yet, but this has been borne out in subsequent research. (Here’s one of several articles: “The Implicit Benefits of Training Without Errors“)
This brings up so called “errorless learning” again, since making errors is generally considered to cause stress, even when teaching with positive reinforcement. During Skinner’s time there was much discussion about whether stress was necessary or beneficial to learning.
Here is a very interesting quote from “B.F. Skinner: A Reappraisal,” by Marc Richelle:
Skinner […] insisted on building the required behavior with as few errors as possible, the ideal being errorless learning. This was subject to debate among the first generation of specialists, some of them arguing, on the contrary, that errors have some virtues and that in any case they can never be completely eliminated in practice. One must admit that a natural life environment does not provide many occasions for errorless learning, and that education should prepare for real life, which implies some tolerance of failure and frustration. (bold added)
Skinner didn’t even consider the use of punishment in teaching. He said, in “The Technology of Teaching, A Review Lecture” (1965):
“Teaching is the arrangement of contingencies of reinforcement under which students learn.”
“…aversive control is the most shameful of irrelevancies.”
Punishment is not considered. So when he says that education should prepare students for tolerance of failure and frustration, how did he mean to do that? By making the learning process more challenging to the student. And there are at least two ways to do that:
- By deliberately making the presentation of the material itself faster or more challenging; and
- By allowing the learning to take place in progressively more difficult and distracting environments.
I submit that in training our dogs, it takes no special effort to expose them to environmental stressors. They happen all the time! It does take an effort to make sure that the challenges we give our dogs are gradual and fair. Noise, activity from other dogs, riding in the car, being left alone, meeting strangers, uncertainty in a routine, extremes of temperature, being boxed in or confined, being on leash, being in a new environment, and dozens of other things can all cause stress in your dog, especially if she has not been prepared for it. People ask their dogs to perform in these situations frequently.
Don’t you think there is much more danger of overdoing it than underdoing it?
Real Life Example
Here’s an example video of something that sounds pretty benign: I was practicing leash, Zen (self control), and attention exercises with my dogs Zani and Summer on my front porch. There was quite a bit going on. In the video you can see Zani give a little flurry of stress signals when a dog barks loudly. She works at paying attention and recovers nicely. Summer was also working hard to pay attention but the situation was more difficult for her. They both got practice dealing with a stressful distraction, but I think Zani’s was just at the right level, and Summer’s was a little too tough for her. See what you think.
By the way, we practiced the next day, and Summer handled some similar distractions with less apparent stress. The video on my post about my dogs working for kibble has some of the footage of Summer from the next day. (The footage of Zani in that video is mostly the same as that in the video above in this post.)
Most of us do need to teach our dogs to be able to pay attention in challenging environments, just like we ourselves need to learn that. Think of all the times you had to learn something or perform a task where perhaps there was lots of noise. Or you had a sore toe. Or you just had a fight with a loved one. We ask our dogs to the the doggie equivalents of those frequently and we often don’t even know it!
The argument that you need to use aversives in training in order to teach your dog coping skills is a completely empty one.
Are you conscious of what makes your dogs stress out? Have you been able to teach them to cope? I’d love to hear about it.
Thanks for reading!
- The Ex-Pen Garden
- What’s in a Name?
- Arguing with Logic and Grace
- When Management Succeeds
- Level 1 Breakfast (quick behavior drills)