In order for our dogs to be able to learn, the environment can’t feel distressing. Key word is feel, not if you think it isn’t. Feeling anxious, fearful, frustrated, hungry, thirsty, hot, tired or sore, having to poop, and being overstimulated, all inhibit learning because the dog has meeting her basic needs on her mind. So, don’t withhold food before training, and don’t run her on the treadmill until her tongue hangs out in hopes she’ll behave better in obedience class. Rather, stay attuned to your dog’s needs and deliberately orchestrate conditions that increase the likelihood that she succeeds.
A dog in conflict or when confused falls back into a behavior that has a high reinforcement history. These actions very solid. Pay attention which actions you want, and rehearse and reinforce these. It’s more functional that a dog doesn’t move on cue than rolls over when you say “bang”.
This photo was taken in Peggy's Cove, Nova Scotia, Canada. We asked Bowie to "wait", to not move, and it was one of the very few times he chose to sit. His default was to lie. It was also Will's. And Davie's. But all of them didn't move when we asked them not to, and that was what mattered. Don't sweat the unimportant stuff.
Your dog decides what he finds reinforcing. Knowing what really turns your dog’s crank deepens your relationship and raises training and behavior modification success. Conversely, if a dog shies away or refuses what you think is a reinforcement, often a pat on the head or side, it isn’t one. Don’t offer it again. That can include food. Don’t shove a treat into your dog’s mouth if he wants something else at the moment.
Reinforce when your dog does something you’d like to see again. Dogs offer behaviors we like all the time —don’t ignore, but capture, name, and reinforce them. That is effortless training. Once your dog associates a specific word with a particular action, you can prompt the action it with your word.
Reinforce when your dog checks in with you. I do not teach “watch me” because I want eye contact to be authentic, not a trick. Reinforce with what your dog wants at the moment which is, uh, despite popular belief, typically not a treat. A dog who checks in does so to seek information or support. Give him that—ideally via a taught informational cue.
When your dog acts in a way you don’t want, interrupt with a positively taught cue like “leave”, and reinforce the moment he changes his behavior. Ignoring won’t work despite what the science says. The hiccup is that we don’t live in a science lab. With our dogs an action almost always evokes consequence that matter to her: the passersby passes by when she barks out the window; the action is self-reinforcing because for some dogs barking and jumping feels good in it’s own right. Nothing extrinsic needs to happen for an innate behavior to continue. So interrupt, and reinforce.
And interrupt and redirect to a wanted behavior that also feels good for your dog. If you remove one action, don’t keep your dog in a “now what?” mental vacuum, but replace it.
Here is a nugget from the wonderful Eileen Anderson: https://eileenanddogs.com/blog/2016/06/25/ignoring-bad-dog-behavior/
Reward each time when your dog learns a new cue/action combination, and right away. Actions that are rewarded consistently are learned fast, and I want my dog to link my signal and his action quickly. Make sure that you are reinforcing the exact action. For fleeting moments it helps to use a marker signal such as a clicker or a specific word like yes. Reinforce the behavior you are working on. For example, if it’s come, reinforce when your dog reaches you and don’t request a sit or down. If it is not moving, reinforce when she doesn’t move. What I often see people do when they teach stay is walking away and then calling their dog to come. That’s a great way to practice the recall command but a lousy one to practice the stay. Dog sport folk often teach two cues: one that means “don’t move but I can tell you to do a variety of things” and the other “stay where you’re to ‘til I comes where you’re at”—a Newfoundland phrase I love. For this, I use the word “wait”. For a companion dog that’s all you need, but if you are ambitious teach both.
Once the behavior is learned, operant condition law tells us that a variable rate of reinforcement, which means the reinforcement is not produced each time or after a predictable amount of time or set of repetitions, creates better responses. In real-speak, a dog who only gets rewarded sometimes for a specific action and can’t predict when, becomes addicted to the action—like a gambler who is addicted to the slot machine. However, that may not be so clear-cut. Behavior expert Eileen Anderson said, and I quote, “that a variable schedule of reinforcement makes a behavior resistant to extinction (which means it is harder to get rid of like any addiction), which is entirely different from reliable, consistent and enthusiastic performance” (which is what we want with our dogs). So it seems that specifically actions that are important, like coming when called or leaving something alone, ought to be reinforced each time. But don’t be troubled by that. Reinforcing doesn’t only mean food. Social attention, play, are reinforcing as well, and there is no reason to not lovingly acknowledge your dog for a lifetime when he cooperates.
Be real with your approval. Don’t make a big deal of every sit, but be big deal happy when your dog has a lightbulb moment and when she has a choice and chooses the behavior you want. For example, she is about to jump on you but then self-corrects and doesn’t. Jackpot that moment and don’t wait for the sit or you are spoiling the moment. When I am genuinely happy my dog knows it and not just with treats. I become animated and yippy.
Don’t always make things more difficult but surprise your dog every so often with a super reward for little effort. Surprise rewards raise neural activity in the brain, with us too. I remember a bed and breakfast I stayed in many years ago who served a different breakfast each day. I couldn’t wait to get out of bed to see what delicious grub I’d find on my plate.
Along that line, what about the kind of rewards? Always the same? A variety? That depends on your dog. Some are bored with the same old thing and others do better with a predictable consistency. I could have fed our Will rotisserie chicken for breakfast, lunch, and dinner and she still would have worked for it. She never satiated, which means it didn’t have to be a limited resource. Same with Davie and her ball. Our Newf on the other paw responded better when I had a bunch of stuff up my sleeve. Know your dog.
Once the dog knows an action/cue combo, behavioral science tells us to only reinforce when the action is cued, not when the dog offers it. It’s fancily called putting the behavior under stimulus control. Most of my clients don’t do that, with the result that some dogs offer this action or the other all the living-long day in an attempt to score. So putting a cue/action under your control is technically true, and in real life too, and yet I still periodically reinforce offered behaviors—the ones that are important to me and I want my dog to offer a lot. For example, checking in with me and waiting for info when he is confused or conflicted. But even if you reinforce a bunch of other actions your dog offers because you are a dog owner and not a behavioral scientist, you have a copout: teach an off switch cue like all-done.
Enforce a cue once you gave it. That is super important. Don’t teach your dog to second-guess you. If you can’t enforce it, don’t give it. The person who is working with the dog is the one who enforces it. The leash (or long line outside) is your best friend until you have solid responses and habits.
If you made a mistake and your dog doesn’t respond as asked, don’t punish her. Reset her and try again, this time managing better, becoming more engaging or upping your reinforcement, or becoming clearer in what you want.
If your dog still doesn’t play along, end the interaction and make the reinforcements you have unavailable, but give her another chance right away to prevent frustration—that’d likely occur if the resource would become completely unavailable.
Don’t raise the bar too quickly. Also super important. Rushing things can tag along avoidance and anxiety, regressions and plateaus—frustration in person and dog that stifle motivation in both. Repeated failure paralyzes, and achieving the end goal will ultimately take longer. You wouldn’t advance a child from grade 1 to grade 5 and expect success, would you?
Dogs need about 20-50 repetitions before they make the link between cue and action. I know 20-50 is rather vague, but it really depends on the individual. If your dog needs 50 repetitions he is not dumber than other dogs, but might have an aptitude for different things. Just like people. My spacial intelligence is below ground level, but I can write decently. So I tell myself. Please be patient with your dog, and if that isn’t your strong suit, sign up for a course you have no aptitude for.
How do you know that your dog knows what you taught him? If he responds 100% of the time when you say the cue in the context you practiced it. Not 8 out of 10, not 9 out of 10, 10 out of 10. But only in that context. For example: if your dog sits when you say sit 10 out of 10 times in the living room with his body turned to you, you can assume that he knows what to do when you say your cue word. But that doesn’t mean he knows it when he is beside you, you sitting on the sofa, or both of you out on a walk. Or he may know it. Some dogs generalize better than others. But don’t expect it.
I also like world-renowned dog trainer Jean Donaldson’s 3 out of 3 concept: 1 out of 2 correct responses = backtrack to the last successful level, 2 out of 2 = continue to practice, 3 out of 3 = now you can raise the bar.
By how much? There isn’t a template how small the increments must be, but my general guideline is 10-20%. For example, if you work on reactivity to dogs and you are reliably successful being 50 feet away, you shouldn’t decrease the distance by more than 10 feet—if my math is correct. That though, too, depends on the individual. Key is that each new step is small enough that your dog is able to respond as desired. If he is too overwhelmed or not challenged enough, adjust accordingly.
One general rule you can put in your back pocket is that whenever you add a new component, make the others easier. Continuing with the above scenario, animation will grab your dog’s attention, so increase the distance to the other dogs. Once your dog acclimated to the activity, you can close in to your original distance.
You must practice if you want reliable responses. Repetitions don’t only build brain, but also muscle, memory. However, keep your sessions short and sweet: 5-10 minutes 2-3 times a week is enough. If you have a busy dog, do more, but always remember that there is a fine line between challenging and pressuring. Challenging fosters confidence; pressuring leads to distress and avoidance. Learn to comprehend “doggish”, and watch your dog’s communication signals.
Ultimately, rather than seeing compliance as training, aim for cooperation as a fluid part of your life together. Sprinkle what you taught your dog, your cues, into day-to-day’s situations. Then you don’t have to reserve any extra practice time, but more importantly, your dog learns that mental connection and shared attention, to the environment but also to you, happens anytime, anywhere, in any context.
Comments