We wrote THIS post some time ago and it got more traction than any other. But it’s been a while, we’ve grown and learned from our clients, their dogs, our own new dog experiences, and many other trainers. So it’s about time we did an updated version of our 10 rules for training your dog.
1. Force the dog to be right instead of waiting to address them for being wrong.
We do not shy away from the use of punishment in our training when it makes sense: to stop dangerous and antisocial behaviors. That said, we find it much more productive to force the dog to do the right thing (use negative reinforcement) than to punish them for doing the wrong thing.
Punishment is EFFECTIVE (and doesn’t have all the nasty side effects that people claim… if it’s done correctly) but not EFFICIENT. The functional purpose of punishment is to suppress a targeted behavior and create an inhibition. But it doesn’t teach the dog what they should do instead. So it makes sense that you would have to rebuild other behaviors using reinforcement after the use of punishment to suppress some part in a behavior chain.
But when we use negative reinforcement we apply pressure from the moment the dog chooses to do something wrong, until they are back on track. In doing so, we teach them to avoid the incorrect behavior while also teaching them to complete the correct behavior. In this way, negative reinforcement is both EFFECTIVE AND EFFICIENT. We force them to do the right thing, which implies we are forcing them to avoid doing the wrong thing.
As a side note… you can’t support speed or style in a behavior with pressure. Use positive reinforcement for that. But you can create and support commitment. And that is what we’re talking about here… using pressure to create commitment to the correct behavior.
2. It’s your job to teach your dog the answer to two questions: How do I access reward? How do I turn off pressure?
If we only use rewards (positive reinforcement) in our training, our dogs choose when they want to work, what they are willing to work for, and what behaviors they are willing to perform. If we only use pressure (negative reinforcement, force, coercion, correction, whatever you want to call it) in our training, our dogs will only work when they know we can make them.
Before we can start “proofing” any training, meaning work to show the dog that the behaviors should be reliably performed in all contexts, we need to make sure we have prepared them for the information we’ll provide by making sure they understand for any action we ask them to perform: How do I turn off pressure? How do I access reward? What this means is that we have taught our dogs the appropriate thing to do in response to both types of reinforcement.
Whenever you give your dog a command, you are asking them those two questions. Please be certain that you really took the time to assure that they know the answer to both.
3. Your dog’s TRAINED response to pressure is your responsibility.
Their untrained response to pressure is not your responsibility. I’m including this one because I think a lot of people have experiences where their dog is visibly uncomfortable and that is off putting enough that they quit doing whatever they were doing. This usually happens when they were right on the cusp of a breakthrough. And because they quit, or go try to find a more palatable option for resolving their behavior, they end up at a vet’s office getting a medical solution for a behavioral problem (hint… that won’t work) or they pay a bunch of money to have someone make them feel better about having a problem instead of actually fixing it.
If you have a dog who is fearful of people, and you hire a trainer to help your dog work through that problem, the trainer’s simple presence is going to stress (put pressure on) your dog. There is no way around that, nor should you worry about trying to avoid it. Whatever the dog’s initial response to pressure is: fight, flight, freeze, whatever… is not our concern and we don’t have any control over it. This in no way excuses poor handling, ham handed training, or anything else of the sort. Shouldn’t have to say that, but nowadays there’s always someone who takes a good idea and twists it into something it was never intended to be. Abusus no tollit usum: the abuse of a tool is not an argument against its proper use.
Same with leashes and collars, as well as electronic collars. Some dogs have a short lived adverse reaction to these things. That doesn’t mean they are hurt or even all that uncomfortable. Frankly what it means is that they have been taught to turn on the hysterics when they want something to stop. Congrats on raising a drama queen.
We work with plenty of dogs who object to being handled at the vet. Most of them learned (were trained, intentionally or not) that acting out makes the pressure (the handling of their body) turn off. So when we teach them that doens’t work anymore, many of them have trouble with that at first. Once they learn that calm, quiet behavior makes that pressure turn off, the overwhelming majority of those dogs change quickly. They are happy to do whatever they need to do in order to get what they want: turn off pressure, and access reward. In this case, for most dogs, the biggest reward IS to turn off pressure.
What is, 100%, our concern and responsibility, is the dog’s TRAINED response to pressure. See rule #2 for how that works. And if you’ve been working hard for a while, and your dog still responds poorly to pressure, it’s a problem with implementation of the training at the thumb-loop-end of the leash.
4. Leave something in it for the dog.
When I got Anja, my first hunting dog puppy, I was amazed at how much work ethic she had and how willing she was to learn and try new things. And I realized it was because there was always a reward she found valuable available to her during training. It was inherent to the activity. She is genetically wired to hunt, to desire to find game. So she is always highly motivated to do whatever she can to access those rewards.
If you don’t have a built in reward, that your dog very obviously values greatly, you’ll need to spend more time finding and using some outside reward(s). It’s nice to think that our dogs should just do what we want all the time out of the goodness of their hearts. But we can’t apply our altruistic outlook on things to an animal who doesn’t think like we do. Aside from that, using (correctly taught) play as a reward is a good way to show your dog more appropriate ways to interact with you and others, and is a good bonding experience.
Rule 4.1 (or 4a… take your pick) in here is get a dog that wants to do the things you want to do with them. This makes training much easier because you feel like you are working together towards a common purpose rather than trying to motivate a dog that doesn’t deeply value what you want them to do.
5. In training, handle to show the approved route. In real life, handle to avoid failure.
At a Pat Nolan seminar we went to last year, he said, “In training we handle to show the approved route. In trial we handle to avoid failure.” In our last round of rules, one of them was “Always train, never test.” But I like this updated, more clear version now. With retrievers, Pat is trying to get them to go fast, go straight, and go long, taking a line to a bird. So in training, he’s going to use directionals to keep the dog going on that straight line path to the reward. In trial, he’ll handle the dog less, but will work hard to use handling to avoid places where failure is likely.
We should do the same thing with our dogs. In training, we’ll push the envelope and really push their impulse control and discrimination skills. But when we are out in public we should require them to do their job while ALSO working hard not to set them up for failure. This is called good “handling,” as in how you handle your dog through situations. So if you see the wild dog dragging its owner down the street, I don’t have any issues with you moving to the opposite side of the road. Yes, your dog should be trained and respond to you appropriately, no matter what. That doesn’t mean you make a habit of leaning on it in situations where you are subjecting your dog to potentially unfair treatment from others, or anything that could be dangerous when the risk isn’t necessary.
6. Reward the weakest behaviors more.
Most people want to reward the dog when they do something well. And I’m not really suggesting that you reward less than full effort. But by rewarding the behaviors that need more work, instead of the ones that are good already, you’ll find that those behaviors come around quickly instead of turning into a point of contention in your training. If your dog sit reliably, but won’t lay down or stay put, skip rewarding the sit, and reward those other behaviors when they are done correctly more often.
Dogs tend to perform behaviors in direct proportion to how frequently that behavior was rewarded. Most dogs sit as a default behavior because that got rewarded more than anything else. Start rewarding down more than sit… you get more downs more reliably because you draw more attention to them.
We often see clients have issues with stay and recall because they just get their dog staying nicely and then start calling the dog out of a stay to come to them. What ensues is a confusing event for the dog where they are corrected for breaking a stay a few times, and then immediately after that, asked to leave the stay to come to the owner. A better option is usually to reward the recalls until they are solid, and then reward the stays to make them competitive with the recall. But particularly when you are working two opposing actions together, try to put emphasis on the one that needs help by rewarding it more during that session, rather than using correction. When the dog has a decent idea of both, then we can add correction when needed.
7. The four quadrants of a good dog. (Not operant conditioning)
Your dog’s behavior is a function of four things:
Genetics are what they got from mom and dad. If you got your dog from a breeder, you picked that. Maybe you picked well, and maybe you didn’t. If you didn’t meet parents or see pedigrees that confirmed important behavioral traits, you defeated the whole (intelligent) purpose of going to a breeder. If you got your dog from a rescue, you most likely decided, actively or not, that you didn’t care too much about genetic predictability. That might come back to bite you in the butt, and since it’s your decision, it’s your “fault”.
Socialization is the early exposure to things in everyday life that determines how your dog views the world for the rest of their life. We’re talking birth to four months old. So again… if you got your dog from a breeder, some of this is on them and the rest is on you. If you got your dog from a rescue, unless they were a puppy, you made a decision that you were fine with having no control over this. This is the nurturing, pruning, and conditioning of the raw materials provided by the genetics.
The environment is everything around the dog. Every person, place, and thing your dog interacts with. Your dog is always responding to pressures from the environment. So the better you are at controlling them in ways that help your dog be successful, the better your dog will be able to behave. And if you properly socialized your dog, the amount of time you’ll have to spend controlling the environment will be significantly less.
Training is the way you teach your dog to reference environmental elements around them. Whether they should interact with them or not, how they should interact with them, if they should signal some default trained behavior, and so on.
The point of this rule is to recognize that decisions you made prior to actually obtaining your dog play a role in their behavior for their entire life. Only you can prioritize your values as they relate the the choices you make when selecting a dog. But only you can be to blame for those choices as well. Anybody can be a one hit wonder. Ignore genetics and socialization at your own peril and understand that if you don’t think those things ever matter, your expectations are probably so low that they… well… don’t. It’s probably not wise to subscribe to an “any dog will do” mantra unless you really mean it. Put as much effort into selection of a dog as you do any other major life decision that will affect you for the next 10 - 20 years. So the next time you decide to slag on somebody for getting a healthy, purpose bred dog from a reputable breeder… consider how very responsible it was of that person for being really thoughtful about what they wanted in a dog, how well it will fit into their lifestyle, and how all of that sets them up for success.
8. The worst thing about dogs is they do what you train them to do.
You are inconsistent. You behave differently in class than you do out on walks. The way you allow your neighbors to interact with your dog and your unwillingness to hold your dog accountable for improper behavior in certain situations… all of these things TRAIN your dog to behave differently in different situations.
You passively endorse disobedience. You tell your dog to sit but don’t have a leash on them and don’t feel like getting up after a long day of work. So you just brush it off. But you just weakened the classically conditioned connection between the command “sit” and the act of sitting. You are literally un-training your dog.
Your dog gets fussy, or maybe downright violent when they are handled in ways they dislike at the vet. And when they do, the handling stops, no matter how briefly. You are training your dog that fussy or violent behavior is how they are able to get people to stop doing things they object to.
Other ways this comes out that folks don’t think about? We get calls about:
mouthy dogs whose owners roughhouse with them
dogs that kill small animals who get to play with squeaky toys
dogs that jump but are allowed to do it when the person says “it’s ok”
resource guarding dogs whose owners constantly yank things out of their mouth
In each case, there is a behavior perpetuated by the owner that is contributing to the “problem” behavior.
So before you say “the training isn’t working…” make sure you are actually implementing the training correctly and thoughtfully.
9. Watch the PERSON, not the dog.
Folks have a tendency to focus on the dog. The dog’s behavior, or their natural ability, their temperament. These are the things that get focused on. But what made them that way is the result of the person.
If a dog has good genetics, look at the breeder and find out what they are doing. Where they source, their criteria for putting dogs together.
If a dog shows great natural talent for some activity, look at the handler and figure out what they did to imprint, condition, and nurture that talent. What type of exposure did they provide to allow those things to come out?
And if you are looking at how obedient or well trained a dog is, look at the trainer. Watch what they do and how they do it. I think clients are often watching their dog during training sessions instead of watching us, the trainers. But when you go home and have to do homework, it’s what WE were doing that you’ll need to do, not your dog. After all, your dog was responding to what WE were doing.
Your job is to do the play the part of the person. You can’t be a good understudy if you were reading the lines for the wrong character. You cannot repeat what the person does if you are not watching the person. So focus on the person. Not the dog.
10. Measure dog training time in “good reps” not minutes, hours, days, or years.
We teach a six week introductory curriculum. And we advertise that we can get ANY DOG from untrained to meeting the requirements to pass the UKC SPOT in that time frame. But here’s what is implied with that… A MINIMUM of
36 laps of long line work.
42 sessions of tethering
450 reps of sit
300 reps of stand
300 reps of down
18 physical handling/grooming/vetting sessions
13,500 steps of heeling
And remember these are GOOD REPS which means you did your job correctly AND your dog did their job correctly. The latter usually happens easily when the former does. But if you perform the work incorrectly, those reps don’t count. In fact, depending on what we’re talking about, bad reps do worse than simply not count, they put you in the negative. For habituated undesirable behaviors, incidents where your dog is overwhelmed with fear or defensiveness, and so one, each occurrence probably adds a minimum of 10 repetitions that need to be done correctly just to make up for that one bad rep.