Since childhood, we’ve heard “Everything in moderation.”
Or “eat a balanced diet.”We’ve been told how beneficial it is to have a “good balance” between work and home. Dog training hasn’t been overlooked – balanced training has become a buzzword in the industry. It sounds good, it makes intuitive sense. But does a balanced trainer get better results?
Generally, a balanced dog trainer uses both corrections (punishment) and reinforcement (rewards) to teach dogs. As a society, we encounter this every day in our school systems, work environment, and everyday lives and we rely more heavily on corrections than on rewards in all those scenarios. For instance, at school kids lose points if they don’t turn in their homework (punishment). At work, if you’re late, your paycheck will be docked (punishment). And if you speed, you’ll get a ticket
(punishment). Rare is the teacher who gives points to on-time assignments; a boss who pats you on the back when you arrive on-time for a week; the police officer who writes an “atta boy” citation for obeying the law.
So while it makes perfect sense to us to use both punishment and reward in training, it doesn’t actually translate to stellar results in your dog. Why? Let’s take a look at balanced training and see why what makes sense doesn’t actually work…
Old-fashioned trainers relied heavily on punishment in the early years of dog-training.
Either the dog learned what he needed to learn, or he suffered the consequences. Consequences included physically punishing the dog (hitting, kicking, kneeing, shocking, hanging, etc.), as well as implied punishment (yelling, raising a hand or fist, etc.). Dogs learned quickly to do what they were told. Or else. So when push came to shove, the training actually worked – dogs did what people wanted. At least most of the time. Slowly, praise made its way into training. Dogs got verbal “atta boys” when they did something right. They still got punished if they did something wrong, however.
Let’s look at this in a human example: your child’s teacher uses both reward and punishment in class. A wrong answer gets your child a punishment (he loses a star for the day) and a correct answer gets a reward (he earns a star for the day). How often do you think your child will raise his (or her) hand in class? How involved in discussion will your child be? Do you think your child will volunteer guesses? Or will be willing to take a chance on a creative answer? Probably not.
Why won’t your child want to participate? The teacher is offering opportunity for reward, right? Yes, but there’s also the opportunity for punishment. And unless your child is 100% confident that his answer is correct, it’s risky for him to participate and volunteer that answer. It’s safer to do nothing than be wrong.
Now let’s take a look at the dog training world. If you put your dog in the same scenario – the correct behavior earns a reward and incorrect behavior earns punishment – he’s likely to do nothing, which according us is wrong and earns punishment.
It looks like this:
You ask your dog to roll over. He’s not 100% sure what “rollover” means and he also knows that guessing is not the best option. So he doesn’t move. And what happens? If you believe in balanced training, you’ll punish the dog for his incorrect response.
A positive reinforcement trainer doesn’t blame or punish the dog for the wrong behavior (or for the absence of a behavior). Instead, a positive reinforcement trainer takes the responsibility for teaching the dog. If the dog isn’t doing what we like, it’s not the fault of the dog, it’s the fault of the teacher. Positive reinforcement trainers try to find a different way to teach the behavior.
Behavior science tells us that behaviors that aren’t reinforced will go away. This is called extinction. Punishment merely suppresses behavior, but it doesn’t extinguish it. Think of this in terms of a fire at your house: do you want to suppress the fire or extinguish it? I want that fire gone, I want it extinguished! If it’s merely suppressed, the fire can (and probably) will come back later. Definitely not what I want!
The following is from a Purdue University website focused on childhood education. These principles also apply directly to our work with our dogs (although I’ve left the human scenarios in because I think it underscores just how much punishment affects us as well as our dogs).
Punishment merely teaches what NOT to do. There are an awful lot of things we DON’T want our dog to do. Teaching your dog in this manner could take a long, long time. It’s more productive to focus on what you want your dog TO DO.
Punishment often causes avoidance behaviors. The recipient of punishment is likely to avoid both (a) the person who administered the punishment, and (b) the situation in which the punishment was administered. Thus, in schools where the assistant principal is the one whose main job seems to be to administer swats to unruly students, students often avoid the assistant principal. Since parents and teachers (and even assistant principals) are interested in teaching children appropriate behaviors, they make their job considerably more difficult if they must first make the children stop avoiding them before they can initiate positive contacts. Likewise, if children are frightened of school or of staying around the house out of fear of being punished, it will be difficult to help them develop adaptive skills.
Punishment often results in a mere suppression of the undesirable behavior. Punishment merely teaches what not to do and therefore the punished person may stop the behavior until it appears that the aversive situation has been successfully avoided, or until a time when the pleasant results outweigh the aversive results. In any event, the behavior is likely to resurface with the right conditions.
Punishment often results in a sort of behavioral constriction. The person who is punished may discover that the safest way to avoid punishment in the future is to avoid doing anything that even remotely resembles the punished action.
These are only a few of the side-effects associated with punishment. While balanced training may look appealing at first glance, deeper investigation shows there isn’t any good reason to be a balanced trainer. You can teach your dog all the things he needs to know without adding punishment.
If it can be done, then why are people still using punishment to teach their dogs?
Unfortunately, I believe it’s because we can. We’re physically larger than our dogs and we’re able to administer punishment. However, just because we can administer punishment doesn’t mean we should. Especially since behavior can be taught more efficiently without it. In this instance, I say throw balance out the window – you’ll get better behavior (and a better attitude) from your dog without it!
If you need more help, please contact a Karen Pryor Certified Training Partner or a certified professional dog trainer that uses positive reinforcement, and dog-friendly training techniques in your area. If you need help locating a trainer, drop me a line, I’d be happy to help you find a qualified trainer in your area.
Thank you for this post! It is so clear and simple. I often find myself trying to get dog owners to read books on clicker training, but it seems they never have the time or don’t see the point. Now I have something to send to them that includes great examples and an easy read.
Much appreciated!
Found you on twitter BTW
Awesome article!! Worth printing and handing out or forwarding. Well written, clear and really awesome!
Very nice article. Very informative and a wonderful introduction to why R+ trainers might not recommend a “balanced trainers.”
I do have a comment about your school child example. It makes sense about little Johnny getting a star for a right answer and getting a star removed for a wrong answer, but what if little Johnny starts beating up another child? In that situation, if you can’t use punishment (anything from corporal punishment to time outs to getting a star removed to a strong “No!”) it seems like you would have two problems:
1) The only option you would have would be to engage in a lion’s share of management. Little Johnny would have to never be placed in a situation where there is the remotest possibilty of him making the mistake again because if he gets to rehearse that behavior you have just taken two steps back in trying to eliminate the behavior.
2) If the behavior occurs learning stops. If your doing true R+ only training your only option is to reset the situation and try again, but this time reinforce the good behavior in an attempt to keep the bad behavior from happening again. There is no feedback about the consequences of that behavior on Johnny or the kid he beat up.
Also, do you think of the research on how different parenting styles (authoritative, authoritarian, etc) affect the child’s perception of punishment could have any application to dog behavior?
Since there is no industry standard definition as to what “balanced training” is or isn’t, it’s a lot like “alpha training,” in that the meaning is contextual, and thus has little value without explanation. Everyone’s entitled to an opinion, but it would be a lot more helpful to struggling dog owners if you made it clear that that is what you’re promoting in pieces like this: a personal opinion based almost exclusively on what YOUR definition of “balanced training” is.
Thanks for your comment! I love Twitter and spend quite a bit of time there. I’m following you and am looking forward to your tweets!
Thanks Donna. I appreciate your kind words. 🙂
Alexandria,
Thanks for your thoughtful comment. In the article, I’m talking specifically about teaching behavior. Of course you’re going to jerk the arm of a child backward if she’s about to step out in traffic. Or you’re going to break up a fight that’s occurring in front of you. That’s not a teaching/learning situation – it’s about keeping someone safe. I wouldn’t use those tools to *teach* behavior.
In your example, I’d work hard to determine what the antecedent(s) are for the behavior. I’d also try to figure out what the consequences are for little Johnny — what is he getting out of doing the behavior. I can probably change the antecedents and/or the consequences in order to teach little Johnny the appropriate behavior.
One other thing: it would definitely be necessary to identify specifically what behavior I want from little Johnny *instead of* the fighting. If I can reinforce Johnny for THAT behavior, I have a much better chance of getting him to do that. It’s impossible to get rid of a behavior without replacing it with something else. So it’d be my job to (1) figure out the antecedent, (2) the consequence, (3) what behavior I want Johnny TO DO, (4) set the situation up so Johnny can succeed and (5) heavily reinforce the appropriate behavior (or approximation of that behavior).
Hope this helps clarify my post. Thanks very much for your comment!
Hi Dee,
Thanks for your comment. In the first sentence of the second paragraph, I describe my definition of balanced training: “Generally, a balanced dog trainer uses both corrections (punishment) and reinforcement (rewards) to teach dogs.”
Research has been done on this topic at the University of North Texas. Dr. Jesus Rosalez-Ruiz presented the data at a dog behavior conference. Indeed, dogs that were taught with both treats AND tugging on the leash to respond to a cue had much slower response times. The group that was trained with just treats had a much quicker response time.
I hope this additional information helps clarify that the scientific research supports the information in my post.
Thanks again for your comment, I appreciate hearing other points of view.
Well put Dee.
I would consider myself a “balance trainer”. No, I don’t punish my dogs with leash pops, hitting etc. However, I do “punish” my dogs if they jump or don’t follow a command they know 100%. This “punishment” is me turning away, ignoring them or an “uh-oh”. For one of my dogs, he is responds with “ok, not right try again” and the other (a much softer dog, she might think the world is ending.
My male dog has developed a terrible problem. When anyone comes into our house, he runs up to them barking. It is a very insecure bark and once he jumped up while barking. This to me is completely unacceptable because a dog shouldn’t do, he is 26″ at the shoulder and 53 lbs (making too big and scary) and he should feel secure enough with me as his person not to have to do that. For this, I’m fixing the problem with the people giving him treats when he’s friendly and quiet. However, I do live with 3 roommates and sometimes people come in when I don’t have him under control. When that happens, I usually issue a “leave it/eh-eh” and remove him. I have used a can of air before sprayed by him (not at him) which he hates and it redirects him then I remove him from the situation. Would I consider that punishment? Yes, is going to cause him not to think when we are training? Absolutely not, it is simply an aversive technique an appropriate situation.
Hi Kaitlyn,
Thanks for your comment and details. Punishment is defined as anything that decreases a behavior. So if you’re decreasing the jumping, you’re using punishment. Taking him away from the situation is technically a punishment ONLY if it results in less jumping.
Many dog owners knee their dogs in the chest, but complain that their punishment isn’t working. If it were truly punishment, it would be working — as that’s the definition of punishment! In the case of many dog owners, what they *think* is punishing is really reinforcing! :0
I’ve trained service dogs for nine years and we don’t use a conditioned punisher (your eh-eh) or any positive punishment.
One more note: aversives aren’t necessarily punishers. My alarm clock ringing at 5:30a is really aversive. But I keep setting it! So it’s aversive, but it’s not a punisher. My behavior of setting the alarm does not decrease.
The science of operant conditioning is wonderfully fascinating to me. I could talk about behavior, the quadrants, operant and classical conditioning all day. I’m glad you posted – and look forward to continued discussions. I love this stuff! 🙂 Thanks again for your comment.
Laurie – dogs aren’t children, children aren’t dogs. Putting the two, and how they learn, and all learning, into the limited space allotted in a blog post leads naturally to oversimplification of key points and absolute definitions, and nothing good can come of that in the long run. 😉
I take issue with the characterization of balance in dog training as somehow equal helpings of both correction/punishment and reward. Balance, in my world, is about what works most effectively for a specific dog in a given situation – and that doesn’t automatically mean that for every potential click or cookie a dog can earn by offering the behavior I need, a leash pop is the automatic alternative for offering a different behavior.
I recently read an article where a trainer who is well-versed in operant conditioning characterized stepping on a leash to prevent a dog from lunging as ‘negative reinforcement.’ IMO, that is P+, the application of a condition (stepping on the leash) to a situation (dog lunging) with the goal of stopping the dog’s lunging behavior. It’s an aversive (the dog gets a leash correction in the process of the lunge. It’s positive (applied action) punishment (correction/aversive.) But hey, we can’t say that in our out-loud dog trainer voices, or we won’t be perceived as being ‘good’ trainers. 😉
My job as a trainer is multi-focal: to teach the dog acceptable behaviors; to teach the owner effective techniques to get and sustain those behaviors; to teach the owner how to problem-solve and, finally, to teach the owner that appropriately naming positive punishment when that’s the technique in use doesn’t make them bad people…as current dog training fashion would have them believe. 😉
In the category of teaching the dog (and the human student) acceptable behaviors – my name is Pat, and I’m a balanced trainer. “Balance’ applied to my training methods doesn’t mean that I use equal parts of correction/punishment and praise/reward. It means I approach each dog’s learning plan with equal parts of logic, analysis and sanity. And I try to do the same with my human dog-owner students, who can often be more trying and more complicated learners than the dogs.
The feedback I provide can take many forms, and is appropriate to getting the behavior goals on my training agenda. Some days and some dogs, it’s far more praise, coaching and rewards and relatively little of anything that the dog perceives as aversive. Some days and some dogs, I need to use more aversion. Teaching a dog scent discrimination isn’t the same training agenda as say – snake-proofing – and the feedback tools I use are chosen accordingly. And yes – aversion is used heavily in snake proofing and I’d dispute that just because it’s a safety situation, it can’t also be a learning situation for the dog.
There is far too much characterization of ‘balanced training’ as a level seesaw with equal parts of every type of feedback. The whole point of balance is to maintain equilibrium – but that doesn’t necessarily mean that all trainers who use both correction and rewards use equal parts of both, or can’t be using more rewards than correction (or vice-versa) depending on the dog or the individual training situation.
And there’s way too much denigration involved in what shoudl be reasonable discussions of feedback. Positive means applying something, negative means removing something, and punishment doesn’t have to be physically or emotionally devastating – it was only ever intended to describe something the subject found aversive. It’s people who’ve emotionally charged and mis-used the words. I’d hope that when trainers use the words, we could demystify them and set a better example of how to use them without the emotional baggage.
Hi Pat,
Thanks for your comments. I appreciate your taking the time to write and give your opinion.
All animals with a brain learn the same way: operant conditioning isn’t different depending on the species you’re talking about. We all learn the same way: we do what works, we quit doing what doesn’t work. If the behavior doesn’t pay off, we stop doing it. Period. It’s not different between species.
Check out TAGteach (www.tagteach.com). TAGteach is the application of operant conditioning to people. Check out http://www.clickertraining.com and you’ll see the principles of operant conditioning and the laws of behavior applied to horses, guinea pigs, crabs (yes, crabs), rabbits — you name it, it’s been taught with operant conditioning.
It’s kinda like gravity — you don’t have to believe in it, but it definitely exists! Operant conditioning happens to us all the time. And it’s how we know what behaviors to keep doing (behaviors that are reinforced), what behaviors aren’t paying off (extinction) and what behaviors aren’t safe (punishment).
All aversives are NOT punishment. I commented in an earlier post. Just because we don’t like something doesn’t mean it’s punishment. Most of us hate traffic, but if it’s the only way to get to work, we do it. I hate my alarm clock, but I keep setting it every night before I go to bed. It’s aversive, but it doesn’t decrease my behavior. Therefore, the aversive isn’t punishing.
The scientific research is clear: mixing corrections with positive reinforcement slows learning.
Do I sometimes use punishment? Yes. Often by mistake. Meaning: I don’t realize the consequence I’m giving is punishing. For instance, one of my dogs hates to be pet on the head (the Great Dane service dog in training, Talos). Every now and then, I forget that and put both my hands on his head. That IS both punishing and aversive to him. He hates it AND he comes up to me less frequently.
But do I use punishment or negative reinforcement to teach Talos the behaviors he needs to learn? No – I use positive reinforcement.
Thank you again for taking the time to write your comments and join the discussion. It’s through communication like these that progress can be made. The more time I spend teaching both dogs and people, the more I rely on scientific data to bear out my thoughts and discoveries. Thank goodness for the science! Otherwise we’d be left with opinion and conjecture!
Dr. Ruiz-Rosalez is hard at work at University of North Texas and he and his graduate students are revealing very interesting research into the field of behavior modification and learning. It’s an exciting time to be in the field!
By the way, have you read Karen Pryor’s “Reaching the Animal Mind.” There is a ton more research listed in her book. It’s an easy, fun read – with lots of good data.
Laurie – while all animals with a brain learn the same way (I don’t dispute that), they don’t communicate what they know in the same way. More to the point, people aren not as good at deciphering the intended communication of, say, dogs (or crabs ) as some claim to be in deciphering the intended communication of say, kids. And FWIW, teenagers likely qualify as a communication sub-group all to themselves … but I digress. In any case, to directly compare an evaluation of how kids learn to how dogs learn skips the part about kids and dogs communicating differently, and people understanding their efforts to communicate incompletely. And I still contend that no good can come of that. 😉
But more important, your characterization of ‘balanced dog training’ is an attempt to characterize something that means different things to different people. You’ve fallen into a common trap – that balance is comprised of equal helpings of punishment and praise with no moderating circumstances. Not so, in my experience, but many people persist in this thinking. I’m a balanced trainer – but because I have the guts to admit that now and then I use R- or P+ in training, the assumption is that those are the bulk of the techniques I use. Yep, let’s disregard any news from that quarter because this person admits that removing a dog from the course for disregarding a command is (to the dog) P+. Or R-, depending on your perspective and how you describe the action. In any case, I didn’t let the dog continue the R+ of the agility run because he didn’t hold the first exercise (stay.) And I named my training behavior for what it is. Oh my! 😉
I hear all the time trainers describing techniques as anything but what they are – negative reinforcement, or positive punishment. I hear all the time the self-description that “I am a 100% positive trainer.” Your own admission that while you sometimes ‘accidentally’ use positive punishment because you forget or don’t realize that an action is a punisher in the dog’s eyes, you never use P+ or R- to train a behavior is common, too.
I’d submit that most people use a lot more P+ or R- techniques with their dogs’ training than they will categorize as such…but that in any case, using those techniques isn’t wrong in and of itself, and as trainers, we need to stop denigrating the correct use of those techniques. Using any technique incorrectly is unfair to the dog, and won’t necessarily help teach him what you want to teach. But using P+ or R- correctly can be very powerful, and judicious (balanced) use in combination with R+ will accomplish training goals. The ‘balance’ is different for each dog, and in my experience it’s not usually an equal split.
I’m aware of and up on the science – I’ve worked in drug safety research for 24+ years with my own ever-changing multi-species colony of animals to observe. I’ve read Karen Pryor (more than once) and been using operant conditioning techniques since the early ’80s. I simply find it foolish to disregard an entire range of techniques because people have tied emotional baggage to their names and convinced themselves that it’s too hard to learn how to use them correctly…
Train well.
Hi Pat,
Thanks for the dialogue. I don’t want to confuse teaching with communication skills. Thankfully using marker style training (clicker training is but one example) allows us to teach even when we don’t share a language. Or when we aren’t even the same species.
I used to be an entrenched choke-collar trainer. It’s taken me years to change. Education, experience, and working with clients and dogs have led to my awareness of the science and data.
So while I don’t use positive punishment or negative reinforcement in my training today, I relied heavily on it “back in the day.” I got results, yes. But as I switched from punishment- to reward-based training, I noticed lots of really nice side effects in my dogs. And then when I got involved with service dog training and have had nine puppies in as many years — all of them very different in temperament and learning styles — what really struck me is how successful I’ve been since relying on science to guide my training.
Thanks again for the discussions. I think we’ll leave it at agreeing to disagree.
Forgot to add — take a look at the studies by Dr. Ruiz-Rosales. The researchers didn’t use heavy punishment in their training. They actually used negative reinforcement (tugging on the leash until the dog did the correct behavior). They found that even the slight tug (not a jerk or a “collar correction”), slowed the dogs’ performance and learning. Pretty powerful stuff.
Thank you for this article Laurie. I will be sharing it with others.
Laurie, Loved your reference to Dr. Ruiz-Rosales. He once told a story about how he used to love going grocery shopping. One day his wife gave him a list of things she needed for dinner, and he cheerfully went off to the store. Not finding the exact brand she wanted, he substituted what he thought might work instead. When he got home with his dozen purchases, his wife focused only on that one incorrect purchase. The next time she asked him to run to the store, he found his mind wandering, his desire to do something else increasing–the fallout from that last trip took away the joy of going out and exploring the store, for fear of coming back with a wrong item again.
I did a similar thing with my husband–sent him to the hardware store with a list of a dozen things to pick up. WHen he got home, he had picked up the wrong object to purchase. It happened to be the most important thing on my list, so I ignored all the CORRECT objects he did get, and focused on the one he screwed up on . I realized what I had done immediately, so I stopped, paused, and (thankfully something that can be understood by humans) apologized for just ‘poisoning the cue” for him. Then I suggested we both take the wrong item back to the store, and stop for frozen yoghurt on the way, and watch airplanes (his favorite past time). By recognizing I’d nearly ruined his joy for running errands for me, I took steps to quickly remedy the situation.
In the classroom, my husband is a teacher. The accepted response to classroom disruption and tardiness was to write the kids up, give ’em a slip to take down to the principal’s office. This was not a punisher to gang kids–they didni’t want to be in the class, and they carried their slips around on their notebooks as badges of honor.
So what my husband did was simply this: When the bell rang, the door was closed. THose in their seats were issued “Geo Dollars” (geometry class tokens to be turned in for things they did want–like choice of music for the day, pencils, erasers, paper, etc.) NOthing was said to those who were late, and finally sauntered in. They just missed out on an chance to get all the goodies. Took two days before they realized they were getting rooked by staying out in the hallway, and they chose to be in class sitting down before the bell rang.
Once, a child began kicking out at my service dog in a mall while his mother was busy talking to a cashier. Rather than reprimand him (I could see he was simply tired, hungry and cranky), I blocked him (management to keep dog safe at that moment), then said, “Wow, that’s quite a strong kicking leg you have. Would you be interested in helping me train my service dog? I started ripping pages out of my notebook, balling them up. Then I gave the boy a paper bag, showed him how to hold it open. I asked him if he would kick each paper ball as far across the floor as he could, then I’d send my dog to go fetch them and put them in the paper bag the boy was holding. The frustrated attitude the boy was displaying disappeared immediately. He was now re-directed into doing something fun and positive.
Beautiful description (and execution) of turning problem behavior into reinforcable behavior — without punishing!
I wish I put R+ into practice long ago. I love your husband’s reinforcement of the on-time students. And even better, the way he extinguished the undesirable behavior of being late. It’s sheer brilliance. Our culture slips toward punitive measures far too quickly – it’s exciting to see it being used in real life, especially with kids in school!
I love Jesus’ “can of beans” story. My R+ training friends refer to that story at least once a month.
Laurie,
Thanks for the post and replying to the various comments.
I needed that reminder about the ‘can of beans’ story. My husband was grateful when I read “What Shamu Taught Me About Life, Love, and Marriage: Lessons for People from Animals and Their Trainers”.
I will be in your KPA Georgia class this fall and am looking forward to it not only for dog training, but for all aspects of working with people.
Lois
The author couldn’t be more wrong. Her “impression” of how a balanced trainer works, is not reality.
I agree with Pat and j1boss, the author is not describing balanced training, she is describing what idiots do. Who would punch a dog? Hang a dog? An idiot seriously in need of being bitten by said dog. This blog (I refuse to call this an article) implies that this is how balance trainers train, and she could not be more wrong.
There is absolutely nothing wrong with learning consequences for behavior, good or bad. Finding a balance between the two produces a reliable dog (or child)(or adult employee) who understands what is expected and delivers. Not out of fear, out of understanding rules and boundaries.
I will direct you to an old blog I wrote years ago: http://dogsbeerandchocolate.blogspot.com/2008/05/what-is-wrong-with-world.html
Hi Janet,
Thanks for your comment on the article. I’d like to start a dialogue here — how would you describe a balanced trainer?
Thanks for contributing to the discussion.
Laurie
Hi Lois,
Ah, yes, the “can of beans” is really taking on a life of it’s own — and for that I’m glad! It’s a good reminder, I think. Looking forward to seeing you in Georgia for Karen Pryor Academy! It’s a fun time, and a lot of good, hard work.
Thanks,
Laurie
I’m a balanced trainer. I would not more correct a dog for not understanding a behavior, than I would reward them for it. I TEACH. Finding what works for each dog (AND handler, as they are a TEAM) is balanced.
I don’t know where you get this:
“It looks like this:
You ask your dog to roll over. He’s not 100% sure what “rollover” means and he also knows that guessing is not the best option. So he doesn’t move. And what happens? If you believe in balanced training, you’ll punish the dog for his incorrect response. ”
In your attempts to making balanced trainers be the “bad guys”, you’re fabricating scenarios.
Hi Cyn,
First, I love your blog name! Dogs, beer, and chocolate make my top 10 list, for sure! Can you tell me what parts of my article you think are inaccurate, and offer up your vision of a balanced trainer?
For clarity, I’m not talking about consequences being good or bad. I’m specifically talking about teaching. I suffer the consequences of my actions everyday — both good and bad. Everyone (and every animal) does.
My article is about how to teach behaviors. Not about the day-to-day consequences we all face. Have you read the studies by Jesús Rosales-Ruiz, Ph.D.? What do you think of his findings?
Thanks for the dialogue, I’m always interested in discussing the science of behavior!
Thanks,
Laurie
P.S. I’m so glad this article is generating so much interest and comments! I love nothing more than “behavior talk!”
I’m going to be honest…how do you expect someone to take a blog post seriously when it’s full of such vitriol and sarcasm and buzz words like “purely positive” or buys into the claim that positive reinforcement training = dogs being euthanized?
Your post shows a distinct lack of understanding of what +R/-P training actually is.
In this case I completely agree with Janet. The core of this “opinion statement” is a misrepresentation of the true concept of Skinnerian “punishment” in behavior modification and training. Those who react emotionally to the term P+ seem to need to present extreme distortions and silly examples in their effort to promote their “cause”. If “pure positive” or exclusively R+/P- were actually better than “balanced” training, it would stand on its own and truly become the method of choice for all trainers and behaviorists. But the fact is that proper administration of aversives (or P+) in the case of a dog (or person) not performing a requested behavior that is KNOWN and UNDERSTOOD is effective and necessary. The examples given are always based on an idiot’s attempted use of P+, and not at all what an intelligent and rational trainer or even novice would do. Such ridiculous examples are only one of many fallacies of R+ extremists, and anyone who thinks logically will see the flaws in such rhetoric and totally reject the paradigm. Balanced trainers always use R+ for teaching new behaviors.
Wow, I’m so happy we’re getting so many comments. Dialogue is great! Thanks, everyone, for putting in your opinions. With dialogue, we can all learn.
Hi Paul,
Thanks for your comments. I like the science because it takes emotion out of the conversation. Punishment, as a scientific term, is defined it’s effect on behavior, not whether I (or anyone) thinks it’s a correction, bad, or anything else. So if a behavior decreases as a result of the consequence, it is technically a punisher.
Aversives, however, aren’t always punishers. Interesting, I think! I find my alarm clock wickedly aversive. I *hate* waking up to it! However, I set that sucker every night during the week. It’s aversive, yes, but it’s not punishing.
I would like help from you — and all the readers. Do you think this sentence is inaccurate:
“Generally, a balanced dog trainer uses both corrections (punishment) and reinforcement (rewards) to teach dogs.” If so, let me know how you would define a balanced dog trainer’s use of the four quadrants.
Again, thanks for participating in the dialogue, I love this stuff!
Thanks for chiming in! I think we can find agreement in some of what you wrote: “But using P+ or R- correctly can be very powerful.” Yes, indeed! Very, very true! That’s part of my point. It’s very powerful.
I hesitate to throw out the findings of scientific research just because it’s done in a lab. That’s pure science. It’s interesting how we, as a culture, rely so heavily on scientific evidence in courtrooms and in hospitals, but kind of throw it away when it comes to behavior. I’m not sure why that happens, but it is curious.
For instance, in a courtroom, scientific evidence is the pinnacle – it’s what every prosecutor and defense attorney — and juror! — are after. When we find a drug that cures a disease or significantly improves quality of life, we’re all about the scientific evidence. But the same isn’t true in our society, for the most part, when it comes to behavior. We tend to disbelieve the research. Or qualify it in some way when it doesn’t match our world view.
However, I find myself relying more on science now than ever before. It can be replicated. It can be defended. It is pure fact, not opinion. I like facts. Sometimes the research doesn’t match my opinion. Then I’ve got to sit and do a lot of thinking. And wait for more research to back up the initial findings. That confirmation usually comes and over time, I integrate the new facts into my repertoire. I’m always learning, always listening, always searching.
Again, I’m enjoying this conversation immensely. Let’s keep the discussion going!
A cut and paste of your entire blog would be redundant, but these paragraphs are particularly offensive:
(you said)
*****Generally, a balanced dog trainer uses both corrections (punishment) and reinforcement (rewards) to teach dogs. As a society, we encounter this every day in our school systems, work environment, and everyday lives and we rely more heavily on corrections than on rewards in all those scenarios. For instance, at school kids lose points if they don’t turn in their homework (punishment). At work, if you’re late, your paycheck will be docked (punishment). And if you speed, you’ll get a ticket (punishment). Rare is the teacher who gives points to on-time assignments; a boss who pats you on the back when you arrive on-time for a week; the police officer who writes an “atta boy” citation for obeying the law.
So while it makes perfect sense to us to use both punishment and reward in training, it doesn’t actually translate to stellar results in your dog. Why? Let’s take a look at balanced training and see why what makes sense doesn’t actually work…
Old-fashioned trainers relied heavily on punishment in the early years of dog-training. Either the dog learned what he needed to learn, or he suffered the consequences. Consequences included physically punishing the dog (hitting, kicking, kneeing, shocking, hanging, etc.), as well as implied punishment (yelling, raising a hand or fist, etc.). Dogs learned quickly to do what they were told. Or else. So when push came to shove, the training actually worked – dogs did what people wanted. At least most of the time. Slowly, praise made its way into training. Dogs got verbal “atta boys” when they did something right. They still got punished if they did something wrong, however.
Let’s look at this in a human example: your child’s teacher uses both reward and punishment in class. A wrong answer gets your child a punishment (he loses a star for the day) and a correct answer gets a reward (he earns a star for the day). How often do you think your child will raise his (or her) hand in class? How involved in discussion will your child be? Do you think your child will volunteer guesses? Or will be willing to take a chance on a creative answer? Probably not.
Why won’t your child want to participate? The teacher is offering opportunity for reward, right? Yes, but there’s also the opportunity for punishment. And unless your child is 100% confident that his answer is correct, it’s risky for him to participate and volunteer that answer. It’s safer to do nothing than be wrong.
Now let’s take a look at the dog training world. If you put your dog in the same scenario – the correct behavior earns a reward and incorrect behavior earns punishment – he’s likely to do nothing, which according us is wrong and earns punishment.
It looks like this:
You ask your dog to roll over. He’s not 100% sure what “rollover” means and he also knows that guessing is not the best option. So he doesn’t move. And what happens? If you believe in balanced training, you’ll punish the dog for his incorrect response.*****
I would say your entire blog reflects a totally inaccurate view of balance training and is simply propaganda.And yes, teaching any animal anything also reflects the consequences of behaviors. I mean seriously, who cares if the dog understands roll over if it doesn’t have a solid recall? Who cares if the dog understands roll over if it won’t do a down stay for it’s own safety? Or walk calmly at your side in a crowd? Or any behavior at all without the constant need for direct reward?
Verbal and physical praise didn’t slowly leak into the dog training world, it’s always been there. Tell you what, I know a really good balance trainer in your area. How about a little challenge? You both go to a local shelter, get an unknown dog and train it to well mannered, off leash reliability in say 6 weeks time (I’m stretching out my normal training time by double, just in case of anything unforeseen happening). The dogs are expected to complete say Novice level work with a relaxed and happy attitude. Now that is not an unreasonable period in which to accomplish this task. This has the making of a decent documentary, both trainers being taped and monitored, expected to keep training logs and produce a trained, well mannered dog at the end
Theory is all well and good. Let’s see how it plays out in practice.
Michelle, I have no idea where you got the idea that I believe dogs are being euthanized because of purely positive training unless you are a mind reader. Actually the purely positive trainers are excellent for my balanced training business, at least for those dogs who get a second chance at training.
And truly, I don’t care if you take my blog post seriously.
Hi Cyn,
I couldn’t find where you wrote what your definition is of balanced training in your comments. I am very interested in learning what you consider to be the definition of balanced training.
I’d appreciate your reasonable dialogue on this topic — it’s something of great interest to me as a professional in my field. I welcome the opportunity to hear your definition.
Thank you for the post and thank you everyone for the dialogue. It’s refreshing to see that we can all discuss such a delicate topic with respect for one another’s practices.
I believe this will all come down to personal opinion: whether or not you feel comfortable as a trainer to introduce punishers. The more I am able to train without +P or aversives, the less I rely on them in every day life. In training, I’ve found I can accomplish anything as long as I manage the environment properly. I like the sense of empowerment when I take responsibility for the learning that’s going on. 🙂
Could you please give some better citations for Dr. Ruiz-Rosalez’s studies so we can find them. I have read a lot of literature on training and operant conditioning and I haven’t seen anything that approaches what I would consider how P+ is used in a real training situation by a good trainer. The problem is that good scientific experiments try to isolate a variable and be consistent across all trials, which is not the way a good trainer operates.
I also agree your article doesn’t give examples that describe how a good trainer uses corrections. In fact it bears no resemblance at all to the way a balanced trainer works. I use the term correction because if I do use an aversive it is not just punishment in order to suppress a behavior, it also shows the dog what he is supposed to do. Corrections are not normally used for teaching. The come in during the proofing phase and even then you try to set the dog up to succeed so few corrections are needed. Sometimes a correction helps clarify what is correct and what is not correct in the dog’s understanding. Sort of along the line of the title story in “Please Don’t Eat the Daisies :-)”
If you say you used to use punishment-based training it sounds like you started with a bad version of old-style training. Even when I first trained a dog almost 50 years ago I would not have called it punishment-based because punishment was only a minor part of the process.
The dialogue going on here has been fascinating to follow 🙂
One question I have for Laurie: in speaking about punishment suppressing behavior, not extinguishing it, doesn’t the definition of punishment itself mean that the behavior is on it’s way to extinction? I never really thought about it before because I often counter-condition with a DRI or DRO in addition to any negative punishment I may be using.
I believe that balanced training, as with all others, is on a continuum with varying levels throughout. The very first trainer I met and learned from calls her way of training balanced, using praise/petting and the occasional treat in addition to collar corrections. At the time, it seemed to me to work well and it was the way I trained for a couple of years also.
At the time, I had no knowledge of learning theory but once I began to learn more about it and clicker training, I decided to change the way I trained based on the logic, the science, and that it made me ‘feel’ better to train this way.
I find solace in the science of my profession and rely on that to guide my decisions in training. I don’t know a whole lot on the mixing of the quadrants but what common sense tells me from looking at it is it has the potential to be detrimental to the learning process and should be approached knowledgeably and carefully, if at all.
Jesus Ruiz-Rosales’ DVD the Poisoned Cue is on my “to buy” list and I look forward to learning more on the subject from him.
Thanks for writing a blog that has sparked so much discussion, Laurie 🙂
-Nick Hof, CPDT-KA
I’m not Paul, but I’d like to comment on this topic.
Your sentence is not exactly inaccurate, but it is incomplete and misleading. First of all, by listing punishment before reward you imply that a balanced trainer uses punishment more than reinforcement, or at least to an equal degree, which is not at all true in my experience. You then go on to say “we encounter this every day in our school systems, work environment, and everyday lives and we rely more heavily on corrections than on rewards in all those scenarios” with the implication that the scenarios include the methods used by balanced trainers. Again, that is wrong.
I would describe a balanced trainer as one who uses all four quadrants with the intention of producing a dog that works reliably and enthusiastically.
I think it is unfortunate psychology uses the term punishment, as it is a loaded word and it is the rare person and the even rarer dog trainer who can use it dispassionately. From the way you write, I doubt you are one of them.
Another example of a loaded word is “blame” as in “a positive reinforcement trainer doesn’t blame… the dog for the wrong behavior” implying that a balanced trainer does. The word blame carried a moral judgement “the act of censuring, holding responsible, making negative statements about an individual or group that their action or actions are socially or morally irresponsible, the opposite of praise” [wikipedia: blame]. A balanced trainer doesn’t blame a dog for the wrong behavior, especially while the dog is in the learning phase and even after the dog knows the behavior, I wouldn’t use the word blame for my reaction to nonperformance.
I am a scientifically grounded person. Years ago I worked in ethology (animal behavior) including field work on canids. I still read a lot of science and I prefer to read the actual papers. Unfortunately oftentimes it isn’t easy to get more than the abstract if you aren’t an academic, which is not adequate to understand an experiment.
My problem with the science as applied to dog training is that even when an experiment is well-formulated and conducted (many are not) the results are extrapolated by most people, and often in the discussion section, far beyond what is supported by the research. These days even the best experiments raise many more questions than they answer.
As applied to dog training, I think the science is in its infancy. Remember how 20 or 30 years ago research in nutrition blamed fats for pretty much everything and recommended extremely low-fat diets? Now we know there are different kinds of fats and some are essential for good health. There are many many examples of this and I’m sure many more to come. I think the “science” of dog training is at the stage the science of nutrition was a generation or two ago.
You and Dee are both correct that there is no standard definition of balanced training, and it is true there is a lot of variation. Unfortunately your definition doesn’t have any resemblance to how any balanced trainer I know actually trains.
Great post. I see a lot of disgruntled trainers/teachers commenting about exact definitions, etc., but for the layperson, I think it’s great.
I routinely get frustrated with people punishing their dogs and claiming that punishment is the best way to train. I use the classroom/star example a LOT when I explain to people why I positively train my dog and use a clicker. Don’t get me wrong, I do get frustrated occasionally and yell–but I realize it’s because I need a time out and learning is no longer fun for either of us. People just don’t get that we use the wrong training methods because it makes us feel better, and that there is some sort of “justification”.
I know some people who only use shock collars to train their dogs. Yes, they are well behaved. Shoot! I would be too if someone shocked me all the time.
These people also tried to “teach” me, on how to get my dog to sit. He was an older dog from rescue. Had some health issues. I had only had him for two weeks.
She grabbed him by the collar, pulled up and then pushed down on his butt. He bit her. She was about ready to hit him. I grabbed her hand. Told her not to hit him and let him go. I then told her he has a spine disorder, and that it hurts him to be manipulated like that. I also told her that he was “teaching” her not to treat him like that.
I told her that I only use postive training techniques. And that I find her way of “teaching” a dog was rather brutal.
He knew how to sit, but in the foster home he came from, he also had a visual cue along with the sit command. …Which by the way came in handy, because, he got loose once and I used the visual command. He went into a sit and waited for me.
I am not a professional trainer. But working with a rescue, I’ve had a lot of dogs (huskies and mixes) in my home.
I’ve had perfectly normal dogs. I’ve had dogs who are fear biters, OCD, separation anxiety, kennel anxiety, guarding issues, etc….
I’ve had dogs, where Daddy, alpha rolled the dog, to teach him to be submissive, and when Mom did it, she got bit. (wooHOO, now we have a dog in rescue with a bite history) I’ve had dogs that ate feces, because the bigger dogs got the food first; and had no contact with more than one human ever! And OCD dogs…. Oy vey.
I’m not a strong person. I’m not an alpha (because I’m not a dog or wolf). I figured out a long time ago, that dogs have big teeth. I want all my dogs to have a relationaship with me. I want my dogs to know they can trust me. I am not going to hit, or alpha roll them, as they have been treated in the past. Even if I don’t know their past, I still want them to trust me.
Things I’ve learned along the way: Baby gates are my friend. Distraction works. If I leave food on the counter top unattended, and the dog gets it; I am a stupid human.
I’ve had much success with using the methods laid out by Karen Pryor since 2002.
Some ppl call it a gimmick. Whatever! I know it works. It has worked for some of the worst behavior out there. Yes, some behaviors can never be remedied, just modified. But that’s okay; we try to find a home for our dogs, where hopefully the ppl will continue to use our methods of positive re-enforcement, even if it’s not with the clicker.
Thanks for your thoughts. It’s definitely a divisive topic. I’m glad there are researchers out there who are now finally interested in this topic. The next 10 or 20 years should produce even more interesting information, if the last 5 years are any indication of the depth of interest.
I’m also one who feels empowered when I take responsibility for the learning that’s happening. If I take the credit when things look great, I’ve also got to take the responsibility when things *don’t* go so great!
Hi Nick,
Thanks for your thoughts on this article. About your question of extinction: extinction happens when there isn’t any consequence to the behavior. Karen Pryor writes: “Behavior that produces no results — not good results or bad results, just no results — will probably extinguish.” (Don’t Shoot the Dog).
I, too, have taken the science and run with it. It’s nice to have the science to guide my training, not my emotion.
Thanks again for your thoughts — this is indeed an interesting discussion!
Daya,
Thanks for taking the time to comment. I love to rely on the scientific definitions — free of emotion or “baggage.” Just a definition we can all agree on. If something follows a behavior, and the frequency of that behavior decreases, it’s punishment. Simple, judgement-free, emotionless.
Thank goodness for the scientific research — it gives us all a common dictionary with clear definitions.
Susan,
I never used a shock collar, but I was pretty good at using a choke collar back in the day. It did suppress behavior, but it didn’t get rid of it. And in some cases, it escalated the behavior. I had a dog who was on the fast track to euthanasia for his aggression. The more corrections he received, the more aggressive he got. Makes sense now, but I was a slow learner back then. 😉
A veterinary behaviorist introduced me to positive reinforcement and very kindly, but firmly told me to lay off ALL punishment. And Lucky lived out his entire life here in our home.
I think most people who use different techniques truly believe they’re working in the dog’s best interest. I don’t think they are trying to be mean or brutal. It’s just that the more you know, the wiser your decisions. I’m glad science is there to continue to illuminate my path. (My dogs are glad, too!)
Bingo! Trust! Yes, yes, yes! love it! I’m with you, call it a gimmick, call it a trick, call it whatever you want. It works and has very little fallout.
Linda, thanks for posting what you think a balanced trainer is. I don’t think our definitions are too far apart. You say a balanced trainer “uses all four quadrants.” I wrote a balanced trainer “uses both corrections (punishment) and reinforcement (rewards).” I think that’s pretty close! 🙂
I love that science has defined punishment for us! I have written before, in this blog, about how I have accidentally punished one of my dogs by petting him on his head. He will turn around and go the other direction when I do that, but I sometimes forget (we sometimes have 5 dogs here), and would absent-mindedly reach down and give him a pat. Oof! Was he punished for coming over to see me? You bet! And I did it! I gave out punishment! Did I intend to do it? No. Was it something horrible? No. It was a pat on the head. But to that dog in that situation, it was punishment.
However. I do not use punishment as part of my teaching repertoire. Haven’t used P+ in a long, long time to teach something to a dog.
So while I do agree that some people (usually non-dog trainers) have a loaded definition of punishment, I rely on the scientific, un-emotional definition: does the consequence decrease the probability that the behavior will happen again. If it does, it’s punishment.
Here’s an example I hear at least once a week. “I punish my dog whenever he barks by spraying water in his face, but he keeps barking!” Clearly, that’s not punishment! The dog may not like water sprayed in his face, but because the behavior of barking doesn’t decrease, it can’t be called punishment.
That’s why I love the science! 🙂
Where I think you got it really wrong wasn’t in the basic definition, it was in the way you (strongly) implied that balanced trainers rely heavily on P+.
I also did not object to science defining terms. What I feel is unfortunate is that it uses the word punishment, which has a different, though related, meaning to the general public (a meaning I think also colors the emotions of dog trainers when they hear or use the word). But even if you have overcome that association, you talk to people who haven’t.
Of course you have unintentionally punished your dogs. There is no way you could avoid it if you live closely with another being. One of the things I find remarkable about dogs is that they seem to have some understanding of an apology.
I would like to know what you think about my earlier comments on the state of the science in dog training.
I went looking for articles by Dr. Rosales-Ruiz, https://faculty.unt.edu/editprofile.php?pid=2033&onlyview=1#6. Only one of them looks like it deals with dogs. It says it is peer reviewed but Teaching Dogs Magazine doesn’t look like a scientific journal to me.
Weird…this comment was just now delivered to my in-box, but it looks like it came in on Tuesday. I’m just now seeing it for the first time. Sorry if it looked like I was ignoring it.
I agree, I think the research coming in is really exciting — and to think, it IS just in it’s infancy! The information will continue to be refined each year. It like we’re starting with something very large and general (just like the nutrition example), and as we learn more, we’ll refine it even further. It’s amazing what the researchers are producing, I’m exited to think what the future holds.
I love that the research produces questions at the same time it’s providing us answers. I think it bodes well — if there are more questions, there will be more research! I think we’re just now seeing the tip of the iceberg when it comes to cognition and behavior. I’m excited to hear Jesús Rosales-Ruiz, Ph.D. speak at ClickerExpo every year. The research he’s doing is amazing!
I did a Google Scholar search for UNT’s Dr. Rosales-Ruiz. Here is the URL for that search. There are over 117 listings, though, be forewarned! :0
http://scholar.google.com/scholar?hl=en&q=Jesus+Rosales-Ruiz&btnG=Search&lr=lang_en&as_sdt=0%2C21&as_ylo=&as_vis=0
In addition to the many scientific conferences, he presents his findings at ClickerExpo — much to the delight of the dog trainers who attend! His sessions are usually packed, we’re all so happy to have researchers working on what we do as a profession! 🙂
It is not easy to design good experiments about dog training. Experiments try to isolate variables and conduct trials consistently, which is the opposite of good dog training. I would really like to see the paper described in which reinforcement was combined with leash tugs. Were the tugs applied with good timing and appropriately to each individual dog or were they just nagging? Was the reinforcement given in such a way as to counter the effect of the tugs or just doled out mechanically? To be honest, the use of the word ‘tug’ in itself sounds like nagging. But I am curious.
Just like we now know there are fats and then there are other fats, I expect we will find there is punishment and then there is punishment. One of the reasons I say this is experience, my own and that of much better trainers, that the proper use of P+ does not have the negatives you list. I think there are times when it is appropriate and even necessary to get the results needed.
I also think it is difficult to learn how to train effectively with R+ (P-). How many of your students are average dog owners who just want a dog with good manners that will come reliably when called? How many of them can call their dogs off high distractions? And how many of those are sighthounds and terriers?
I speak not as a close-minded P+ trainer, but as someone who tried hard to learn how to do it (2 1/1 dogs worth – seminars and much reading) and was unable to get the results I needed (such as a reliable recall).
As an aside, my crossover dog was just as quick to offer behaviors as the ones I started from puppyhood and probably the most creative of them. He had no fear of not doing the right thing. But I have never been a harsh trainer and I am pretty even tempered. I can only recall a few times when I got mad at a dog (we are talking decades of dogs), and never in a training situation.
I would have to argue your alarm clock is punishing (common English) but not punishment (operant science terminology). 🙂
But it is interesting observation because it gives an opening for the use of aversives in training that are intended to increase a behavior, as in the use of the opposition reflex, or a light switching on a Schutzhund dog. I even see something like that when my dogs play with each other.
If you take a look at the later comments, you’ll see the citations for the articles. Or, better yet, do a Google Scholar search and they (lots of them) should pop right up. Just didn’t want you to think I was ignoring your request. 😉
Hi Linda,
Thanks for continuing this dialogue. There isn’t just a description of the experiment, there is video, of course, of the experiment. We saw it at Clicker Expo, and the research panel, naturally, used that video as well when evaluating the research.
Thankfully, behavior is behavior, whether it’s in a rat or a cat or a giraffe. The results of the research can be applied to all behavior, not just dog behavior or cat behavior. Check out the article on how targeting improved the trailering of horses. That data can be extrapolated to targeting in general, not just targeting in horses.
Ken Ramirez, VP of Training at Shedd Aquarium is also another valuable resource when looking at the research and those “in the trenches” actually carrying out the day-to-day training. He’s got an amazing book — Google him for more information. It’s a great read.
I’m tickled that we’re in agreement again! My alarm clock is indeed aversive. Very much so! It isn’t, however, punishment. It does not decrease the behavior of my setting it every weeknight. We agree more than you think, I believe. Which is good — we’re closer in our thinking than it originally appeared, perhaps.
So the research you described is not published? As a person interested in the science I am looking for research published in peer-reviewed journals. Will it be published in the future, or is at least the video available on the Internet?
So what did you think about my second paragraph?
It’s been published! I’ve seen the data collected — the research and the data was presented at a conference. I’m running out of time, but check Google Scholar — tons of stuff there. Or contact the researcher directly — can’t go wrong when getting it from the horses mouth. Sorry so short, really running this p.m.! Too much to do, never enough time. Story of my life…
I will try contacting him. I wonder why the list of publications is so sparse on his University bio? I did look at the link you supplied. There was an awful lot of irrelevant and/or stuff mixed in and I got tired of slogging thru it and checking everything out. It was time consuming with little of value in results. Not your fault, of course.
Discipline (correcting) is NOT the same thing as punishment. If what you are saying is true then technically a puppy should be afraid of his/her mom for CORRECTING him/her right from day one.
Get the facts.
Thanks for stopping by! I appreciate your time. Punishment, thankfully, is defined by science as reducing the frequency of a behavior based on the consequence of doing said behavior. So, technically, by definition, anything that reduces a behavior is punishment. Easy peasy! I love it when science takes the emotion out of discussions. Makes it so much easier! Thanks again for stopping by!
What scores are you getting in competition?
http://www.dogshowscores.com/dogs/search?utf8=%E2%9C%93&search=l+luck&searchtype=owner
If your training is working well, you need to show the world!
Hi Robin, I’m so glad to get so many visitors, especially who are interested in talking about this topic. I don’t spend much time in competiton venues. I’m a service dog trainer, so all of the work I do is “real world” stuff, stuff that makes someone’s life easier, safer, more fun. I realize that competition has ribbons and awards, but my rewards don’t come in neat packages like that. This stuff is “all the time” behaviors, not just necessary in the ring. We use clicker training exclusively for this very big job. From puppy raisers, to trainers, to recipients — positive training is a reality and a requirement. No need for balanced training when we get behaviors that are very reliable, very fluent, and under complete stimulus control. Maybe one of these days, when I’m done with helping people live more independent lives through the partnership with their dog, I’ll have the chance to experience the thrill of competition. It sounds lovely.
Citation needed.
Thank you, Pat. Couldn’t have said it better!!!
Beautiful reply 🙂 I love this stuff too!!! Yay for scientifically backed responses!
Ines, thanks for the positive reinforcement! I really appreciate both your time and your thoughts. I get lots of comments on the blog, even long after the article’s been written. It’s really nice to get well-reasoned ones. And I would have said that even if you’d disagreed with the post (in a well-reasoned way)! Thanks again!
I just wrote this same blog! You definitely did it better! 🙂 Awesome overview! love it!
Laurie,
I have been a big fan of yours since I decided to start my life path in the direction of becoming a dog trainer. I was at a crossroads of what school to attend and what type of training methods I wanted to pursue. I decided that I wanted to attend a school that taught the benefits of both reward based and balanced training because having an open mind and a complete knowledge of both types of would help me to decide for myself what path I would choose and we can never have too much knowledge. Why not learn and absorb every bit of information available to me? I wish that most people who are so against balanced training and talk up extreme measures of this training that are not actually used today could have walked a mile in my shoes to see what I saw at this school. I saw happy, healthy, changed dogs for the better come from both balanced and reward based training. I would like to come to the defense of balanced training and point out some notes you made about how the methods are used. First, you mention that if a dog is asked to “roll over” and doesn’t know what it means and doesn’t do it, he will be given a correction. I would like to say that balanced trainers DO NOT issue a correction if the dog doesn’t know what we want. That would cause confusion in communication and the dog REALLY wouldn’t know what he was getting corrected for other than the current behavior he or she is doing at the moment the correction is given. Balanced trainers work with the dog through molding, luring and clicks to ensure the dog understands the behavior FIRST, then when he or she is doing the command with little to no prompts from us, we add the cue. Once the dog has learned the behavior on cue and we know he understands the reaction that accompanies the cue after lots and lots of practice, we can then go into correcting when the dog is not performing the cue when asked because the dog really does know what behavior is expected. However, with trick training corrections are not necessary or really wanted because it is taught by positive reinforcement and most dogs are willing to comply because there is a treat or reward at the end of the behavior. I cannot think of one time I corrected any of my dogs for not doing a trick. Next you say that punishment causes avoidance behaviors. Well, in balanced training that means the dog is “avoiding” the correction. Just like I mentioned above, the dog is taught the behavior, then the cue, then when we don’t get the desired result from the cue we administer a correction. It lets the dog make a choice. Sit or get a correction. Hence avoidance of correction, not the cue. Now with heeling, we do the exact same method. Behavior, cue, correction. If the dog gets too excited and makes the choice to pull ahead of me, I give a correction, let the dog readjust, praise for getting back into position and continued praise and guidance that I like the behavior. Now precise timing and practice the dog is heeling politely to avoid the correction and in hopes of getting the treat. It’s not out of fear or intimidation. I have never once seen a dog cower or cry or submit to a correction unless it is a timid, fearful dog, then corrections of that nature should not be used in the first place. Sometimes a correction can be a very slight tug on a slip lead, flat collar or martingale. Every dog takes a correction differently. Some like my dog hardly notice a prong collar, some we wouldn’t even think about using a prong collar on ever because a simple firm toned “no” works. We work to find a balance for the dog. Not one that destroys the spirit and breaks the animal down. I would like to clarify that punishment (the infliction or imposition of a penalty as retribution for an offense) can range from a small tug on a collar to a quick pop on a prong collar to a time out for an unwanted behavior or even ignoring the dog for an undesired behavior. We ALL use punishment. Reward based trainers ignore the unwanted behavior, take the dog to another room for unruly behavior ect. Punishment is what it is. A consequence for unwanted behavior. However, when describing how balanced trainers use punishment you make it some like a dirty word. Something that is beat into the dog but we also use methods like the ones you use or a correction. You also said that behaviors that are not reinforced will go away. That we are only teaching what not to do but that is not true. We teach what to do, how to do it and encourage the dog to make the right choices. You also said the behavior is likely to resurface with the right conditions. In ALL aspects, even positive training I believe this to be true. Dogs are never going to be perfect, for any trainer. No human is perfect either. We make mistakes, we learn and we strive to do better. I’m not here saying this to you to cause controversy or try to change your mind because your mind is already made up. Balanced training isn’t for you. It’s not for everybody. Just like reward based training isn’t for everybody. I only ask that you try to see that we are not barbaric trainers who are intimidating, breaking down and hurting dogs. We are not always running around choking the dogs, yanking on their neck, shocking them at extreme high levels or beating them, we incorporate the exact same amount of praise, love, affection and teaching as reward based trainers. Yes, there are bad “balanced” trainers out there who use extreme force, yelling harsh corrections ect, but that’s not balanced and its people like that who give balanced training a bad name. But I for one have met a lot of balanced trainers and I have never seen one use those methods. I just ask that your open your mind and see that we are not all bad. That the methods you say we use and the way we use them are in fact incorrect. You still may not agree with them, like them or think people should use them and that’s your opinion but I just wanted to set the record straight. From a balanced trainers perspective.
This article is incredibly inaccurate. A balanced trainer doesn’t punish the dog for not rolling over when he doesn’t understand the command. A balanced trainer uses motivation and praise to teach the command. Once the dog understands the command but then chooses to disobey, the dog is corrected. To put this in human terms, it’s the equivalent of punishing a child for not cleaning their room when asked. Not punishing a child for not cleaning their room when they didn’t know they were supposed to, which is what the article implies.
Once the lines of communication are open between dog and handler, the dog will want to work for the praise of the handler without the need of a clicker or treats. The dog will learn how to learn, so the more you teach the dog the less you will even need to administer corrections.
Jasmine – here’s the problem. How do you determine that the dog knows the cue but is choosing to disobey? If the dog rolls over once when you ask him to roll over, does that mean he knows it? What about 5/10 times, does he know it then? (Hint: he doesn’t). And if he doesn’t respond the next time you give the cue, how do you determine that he is deliberately choosing to “disobey”? If appropriate motivation is being used, as you describe, what incentive does the dog have to “disobey?” Your wording implies that you might explain this using Dominance Theory, which has been misapplied to dog training and behavior for decades.
I think the article does a fantastic job of explaining the emotional impact that is often seen with balanced training. There are many reasons why a dog may not respond to a cue in a given situation, and the idea that the dog is deliberately choosing to “disobey” is extremely low on the list. Rather, the dog may not know the cue in the first place (and as the article explains, once he is punished for not responding correctly, he is much less motivated to guess), the cue may not be generalized to the given scenario, there may be a competing motivator in the envionment, etc.
Additionally, “corrections” are euphamisms for punishment. In operant conditioning terms, the correction you are referring to is positive punishment. It’s animal behavior science – using a different word doesn’t change that.
Also on the topic of operant conditioning, clickers are used during training to mark correct behaviors. They can be followed by anything that is reinforcing – treats, play, praise. They are simply a communication tool to tell a dog that a given behavior earned them a reward, and they make training more efficient. They are not relied upon to get a dog to do a given behavior.
Lastly, dogs do not have an innate “desire to please.” Praise can be semi-motivating for some dogs, and not at all for others. For a dog who receives both praise and punishment during training – people like to interpret the dog’s correct responses as a dog who is happy to work for praise, when often times, the dog is simply working to avoid punishment.
“Once the lines of communication are open between dog and handler, the
dog will want to work for the praise of the handler without the need of a
clicker or treats. The dog will learn how to learn, so the more you
teach the dog the less you will even need to administer corrections.”
Just curious as to how you ” open the lines of communication ” with a dog that is withdrawn, or fearful? In my experience the best way is using food. I’ve yet to meet a dog I do not know ” want to work for praise” UNLESS it has been conditioned by food first, also using a clicker and food means you do not have to rely upon any equipment, such as collars, leash pops or e collars etc to get the dog to ‘obey’ . No force means the dog really does want to learn.
“Once the dog understands the command but then chooses to disobey, the dog is corrected.” How do you know if the dog actually chooses to ” disobey” ? What if your training is not clear enough and the dog does not understand? So you punish, excuse me, ” correct ” a dog for ” disobeying” when maybe s/he had no idea what you were trying to train them? Test it: take a loose dog ( that is not yours), no collar, leash, nothing! Train it out in the open field somewhere and use your method; “motivation and praise to teach the command.” Let me know your results.