Although I am of two minds about whether Internet discussions convince many people, I have written the posts listed here with those very discussions in mind.
Trainers who do not use force and choose humane, reward-based methods for their dogs get the same arguments thrown at them over and over again. I think in most cases the people hurling the arguments are not apt to change their minds. I mean no insult by that. Science tells us more and more that changing one’s ideas about something, even in the face of overwhelming evidence, is very difficult and practically goes against our very wiring.
But these arguments are truly based on misconceptions. I do think in some cases a calm, polite, persuasive presentation of a reasonable position can explain the misunderstanding to some readers, if not the main combatants.
Each piece is a full post if you link on it, but since there are synopses you can also use the page as a cheat sheet. But if you are having a discussion with someone and you think a post would help, I recommend that you link to the individual post and not this page.
- Being Open-Minded About Training Synopsis: Trainers who base their methods on positive reinforcement are often called close-minded. This post is about why this is not necessarily true, and how to respond when someone says that to you.
- 6 Myths about Positive Reinforcement-Based Training Synopsis: From the idea that we just ignore bad behavior to the claim that we seek to prevent every aversive in our dogs’ lives, this post covers some of the common myths and misunderstandings about R+ based training.
- But We Don’t Give Our Kids a Cookie Every Time they Tie their Shoes! Synopsis: 1) Positive reinforcement for desirable behavior isn’t really such a bad idea for kids. It is not the same thing as indulgent or permissive parenting. For a behavior that is not initially reinforcing to a child, your other options are coercion or punishment. Are they better? 2) The job of any teacher or trainer is to use whatever type of reinforcers are necessary to get the behavior, then transition the child or animal to naturally occurring reinforcers. In the case of the child, these are generally social reinforcers or a product of interests that change with maturity. For animals, they are life rewards that are directly related to performing the desired behavior.
- But It’s Unhealthy to Protect Your Dog From All Stress! Synopsis: I prepare my dog for life in the real world by gradually raising the challenges in her training sessions and asking for performance in the face of large distractions. This is the real world she needs to prepare for. Hurting her in the name of training has nothing to do with preparation for the real world.
- But Every Dog is Different! Synopsis: That’s absolutely right! To use positive reinforcement effectively I have to truly study and get to know the animal and what floats their boat. There are no limitations with this method. An animal that can’t be reinforced is probably dead. On the other hand, using force or pain (which the proponents of this argument are usually suggesting), is very easy. That’s the true cookie cutter tool.
- But What if Your Dog Runs Out Into Traffic? Synopsis: First, because positive reinforcement training is so powerful and I have trained my dog to stay with me, it’s very unlikely to occur. Second, if it were to occur, I have trained several safety behaviors including barrier training, recall, and drop in position, and one of these likely could save my dog. Third, if my dog is on leash and this were to happen, I would pull him out of traffic by the leash. Just because I do not train using leash pressure doesn’t mean I wouldn’t pull my dog out of danger to save his life!
- But I Want to Use All the Tools in the Toolbox Synopsis: First, someone who relies on positive reinforcement generally develops a giant toolbox, because it is up to them to find what turns each individual dog on. It’s not hard to figure out how to hurt a dog. Second, almost everyone rejects some methods as inhumane. So even the people who say they want to use “all the tools” draw a line. They just draw the line in a different place.
- But Purely Positive is a LIE! Synopsis: This is almost always a straw man argument, since almost no one is going around claiming to be able to train exclusively with positive reinforcement. At the same time, this fact shouldn’t discourage us from aiming for that paradigm and using absolutely the most humane methods within our capabilities.
- Leaving the Scene: Clarifying the Science of Negative Reinforcement Synopsis: This is about the claim that adding distance from something aversive is positive reinforcement because something is being added. This is a semantic trick. “Adding distance” could be stated just as easily as “reducing proximity.” Both erase the aversive, which is that actual thing that is being added or removed by its movement or movement of the learner.
- But Isn’t it Punishment to Withhold the Treat? (Extinction Part 1) Synopsis: No. It’s extinction. Extinction isn’t necessarily a picnic, but when combined with reinforcement of another behavior, it is generally considered to be more humane than positive or negative punishment.
- How to Make Extinction Not Stink (Extinction Part 2) Synopsis: A review of the difference between the misery of extinction by itself, and the comparative comfort of extinction as a part of a differential reinforcement protocol.
- Only if the Behavior Increases: Synopsis: A review of the definition of reinforcement. We can give treats for a behavior all day long, but if the behavior doesn’t maintain or increase, reinforcement has not occurred. We tend to focus on our actions, whereas we need to look also at the outcome.
- Only if the Behavior Decreases: Synopsis: A review of the definition of punishment. Again, punishment has not occurred if the targeted behavior doesn’t decrease. The classic example is yelling at your animal. If it stops, but that behavior is just as frequent and strong in the future, all you did was interrupt it. You didn’t punish it.
- Can We Determine Whether Training with Food is Positive or Negative Reinforcement? Synopsis: Yes. Some people claim that training with food is negative reinforcement because the animal is escaping the unpleasant state of hunger. However, the common sense response is that everyone knows that food tastes good and is a big motivator even when one isn’t hungry. And the scientific response is that some very clever experiments have shown that there is a strong, independent positive reinforcement effect from the use of food. For instance, one experiment used substances that tasted good but had no nutritional value. Even though there was no satiation of hunger, the food was a strong positive reinforcer.
- But if You Use Negative Reinforcement, Aren’t You Also Using Positive Punishment? Synopsis: No, not necessarily. If no behavior decreased because of the onset of the aversive (see Only if the Behavior Decreases above), then positive punishment didn’t occur. It is quite possible to initiate the aversive at relatively random times so that no one behavior is affected. However, this shouldn’t be used to claim that negative reinforcement is OK.
- Force Free Training and the Continuum Fallacy: Defining Ourselves An explanation of the deliberate obfuscation of the difference between trainers who avoid the use of aversives, and those who use them. Discussion of open and closed concepts and the continuum fallacy. This one is too hard to synopsize; I just hope you’ll read it.
- Why Scratching an Itch is not the Same as Performing a Force Fetch Synopsis: An explanation of the difference between automatic and socially mediated negative reinforcement. An answer to those who trot examples of benign sounding negative reinforcement such as scratching an itch and washing one’s hands, when the topic is actually a situation where one person is creating a contingency on the behavior of another, or on an animal, through control of an aversive.
- We Don’t Need to Stop Discussing “The Quadrants” Synopsis: It’s all well and good to say that we need to concentrate on the individual animal to determine whether a technique is humane, but this omits the fact that most people are even worse at reading dogs than they are at learning some scientific principles. Argues that study of the quadrants and observation of the animal are both important and can inform each other.
- But I’ve Seen Stressed-Out Dogs During Positive Reinforcement Training Too!: Synopsis: The post includes an analysis of the common errors in R+ training that can create stress in the dog. It also points out that much stress in R+ training sessions is from an accidentally added aversive (therefore not due to the R+ training itself). This can happen in any kind of training, and will have the least fallout if the training itself were R+. Points out the unfairness of comparing bad R+ training with aversive based training. If the problem in the R+ training was an added aversive, why then should we use it to excuse training that is based on aversives?
A nice poster about logical fallacies. Too detailed to link as a graphic, but a handy reference. Take a look!
© Eileen Anderson 2015 eileenanddogs.com
25 thoughts on “R+ Misconceptions”
Are there unnatural reinforcers?
Good question! What comes to my mind immediately was the experiment where rats were conditioned for electric shock to be a secondary positive reinforcer. That’s pretty “unnatural” in the casual sense of the word. But perhaps my use of language wasn’t the best in the post. When I said “natural” or “naturally occurring” I meant either a primary reinforcer or a secondary that didn’t take much conditioning. Something that was already reinforcing without training intervention. But “natural” and “unnatural” can be very misleading words. Thanks for a good point.
Fantastic blog… As usual! I foresee myself linking to it a lot!
Why thank you!
Positive only training is a hoax. [Remainder redacted because I am not required to provide a platform for people who don’t argue reasonably–Eileen]
Jerry, if you’d done even the tiniest bit of reading before posting, you’d know that I am not a “positive only” trainer and in fact I have posts that criticize that term.
In the meantime, enjoy this video of a hippo, one of the most dangerous animals on earth, in unprotected contact with a zookeeper/trainer. It holds its mouth open for 2+ minutes for a mouth exam and tooth brushing for four pieces of fruit.
I’ve seen probably too many balanced trainers say that corrections are fine because mother dogs nip and push around their pups when they do something wrong. Your thoughts, if you will?
Good question. Two things about that. One is that it appears that mother dogs actually rarely nip their pups. There are perhaps 2–5 YouTube videos that show mother dogs scruffing and intimidating their pups. At least one of them (cane corso) looks pathological. Something bad wrong going on there. If it were as common as people say, and it’s such a strong belief, wouldn’t there be more? Those few videos get lots of hits precisely because they are so rare. I’ve heard from many breeders who say that those videos mostly show abnormal behavior and they would never breed a bitch that acted like that. It is far more common to see a bitch walk away from her pups when they bother her. (Interesting that in one of the videos, the mother can’t get away, which could be part of the problem.)
Even in the worst videos, the mother dog rarely actually nips.
The second thing is kind of related, but it’s that being rough with puppies can scare them during a very impressionable period. In one of the videos, you can tell that the pups are terrified of the mother (if she is their mother; she is a lactating female but that’s all we know). And we certainly don’t want puppies terrified of us. That’s the risk of using corrections with puppies.
i’m a balanced trainer but most of the time instead of correcting my dog i will guide her to do the right thing . yes i do support the the use of tools. but before i ever got my dog a prong i tested it out on my skin and YANKED it ( guess what it did not hurt )and before you say that a dog have a thinner skin on their neck that is incorrect the thickest skin on a dog is the neck .
Ryly,it may surprise you to hear that the well-respected balanced trainers are very clear that prong collars hurt and/or are very uncomfortable when used in a correction. At least they are honest about it.
Your “test” was pretty feeble. To do a fair test, you first must fasten the prong on your neck tightly, so that the prongs are already pressing into your skin. Take a look at Leerburg’s instructions about this. I’ don’t want to link them here but I’m sure you can find them. The individual prongs should be pressing into your skin before anyone administers a correction. Once the prong is very tight and fastened, attach a leash to the live ring, so the collar can tighten further. Hand the leash to someone else. Have them control you with the leash for, say, 10–30 minutes, including steady pulling and sharp corrections.
I would like to change my previous statement, I can not seem to delete it. I always learning, I will openly admit my personal dog Mercy was very shut down on the prong. I did not see it at the time, Since I have switched over to a flat and clicker and she is thriving. Her engagement is through the roof. Im not against any tool, but I overused the prong collar, and shut her down. Sometimes I will use a gentle leader in a busy environment. I would like to give an example of a time that a harness was given with a very negative association(im not bashing R+). Our other dog, a ridge-back named Noca has a Julius K-9 harness. This harness is never used because my dog is so scared of it. We used this harness at the vet’s (rookie mistake), and a few off leash dog have attacked her while she was wearing the harness (stupid irresponsible dog owners. Im amazed that Noca is not dog reactive). If she sees this harness, she will run to the couch and burrow her head in the pillows, trying to hide. When Noca sees her everyday leash, she gets super exited and wants to go to the car to go somewhere. I would just like to give this update of my dog training process.
– ryly m
Thanks so much for the update! It takes a brave person to embrace that kind of change. And it’s true, shut down behavior is hard to see at first.
I’m sorry Noca got scared of her harness. I’m glad she can enjoy going somewhere on a leash and collar. Have you tried a completely different kind of harness for her? the Julius harnesses have a unique look (I use one). Maybe a Hurtta or a Balance harness? Just a thought.
Congrats to you!
Nope. Humans have thicker, less sensitive skin than dogs.
“Canine skin has several layers, including an outer epidermis that is constantly being replaced and an inner dermis that contains nerves and blood vessels. Canine skin is thinner and much more sensitive than human skin.”
“The epidermis of a dog is 3-5 cells thick however in humans it is at least 10-15 cells thick.”