Some trainers claim that dogs, trained with gobs of positive reinforcement are undisciplined and… crazy. They give examples of happy dogs, vibrating, barely able to think. Dogs that pull into events, demand bark, whine and yip. Over aroused. Quivering with anticipation. I can’t disagree with their observations. I have seen many such dogs. Where I have an issue with this line of reasoning is that it is used to justify punishment or vilify the use of positive reinforcement. The conclusion is wrong. The fix is ridiculously easy and involves neither punishment nor the absence of reinforcement.
We know that reinforcers such as food, play and toys have the potential to create a willing, happy worker. With this happiness, the dog gets a bounce in their steps. “Sit? Yes please! I would love to.” Nothing novel thus far. We like to see dogs working happily. We like to see happy responses to our cues.
As we train more, we go to more classes, more events, more outings, more stores with cookie wielding sales clerks. Positive associations form to these people, places and things. It’s ring a bell and salivate. Pavlovian conditioning.
It starts innocently enough. People start to smile and say, “It’s so cute. He gets excited when we turn on the street where our trainer, daycare, friend, pet store is. He’s so happy!”
The learning does not stop there. Before too long the dog is tap dancing as soon as they hit the highway. As more time passes, the dog is whining and pacing during all car trips. Cars lead to fun places. Woo hoo hoo!
The Pavlovian associations continue to reach back through the predictable sequence of events. It isn’t long before the sight of a leash, shoes or keys gets the dog worked up.
Each step in the series of events compounds and adds more anticipation of the outcome.
Shoes might mean an outing. Looking good.
Leashes definitely mean one. Woo hoo!
Doors opening mean off to a fun place. Woo hoo hoo!!!
The car means it’s super awesome. OMG Losing my mind!
Agility equipment? Dog park? Daycare? Over aroused and out of control.
By the time the dog arrives anywhere it’s so out of control they may even snub food. They can barely execute skills. Each step along the way, from leash to outing jacks the dog up a little more.
It’s not a problem created by using positive reinforcement. It’s a problem of failing to address Pavlov. He’s always on your shoulder.
Adding more training under the guise of calling it control building adds more woo hoo to the dog’s anticipation. It’s why people struggle with getting calm and focus. They keep working with Skinner when Pavlov is the problem. More training isn’t bad. It just doesn’t address this issue.
Sadly, when more training fails, the cry for punishment starts. Humans create an over aroused dog via a predictable sequences of events. The dog gets punished. Aversives will take the happy out of the dog’s training experience. It takes the happy anticipation out of the dog because, let’s face it, punishment sucks. It works if you’re happy with sacrificing enthusiastic happy responses in favour of sluggish procrastination. It works if you want to sacrifice things like speed or latency. It works if you want to risk tainting motivation. You don’t have to.
There is a fix that does not risk tanking motivation. It’s ridiculously easy.
Remember, this is a problem that Pavlov is creating. Positive associations are forming. The places, the vehicles, landmarks, shoes and even the leash have developed a positive association. Associations bled backwards through a predictable series of events. The problem bled all the way back to something like the leash or a pair of shoes. The dog starts getting jacked up before you leave the house.
Dogs tap dancing at the sight of the leash is cute but also the first sign of impending arousal. Each step along the way, shoes, door, car, turns, seeing the facility then cumulatively add additional fuel to the fire. It builds up to over aroused.
Positive associations are super easy to break. Think cats that love can openers. They get excited to hear them run. Except you know which cats don’t get excited? Those cats who live with people who eat loads of canned soup and ravioli. Those cats don’t think can openers predict food because frankly, can openers are not reliable at predicting cat food. They just don’t get as jazzed up by them. Meh…feh…mildly interesting.
That’s essentially the fix. Break the associations in the steps leading up to the activities.
You can do things like:
Pick up the leash. Put it down. Do this until you get meh feh.
Put the leash on the dog. Take it off. Work to meh feh.
Leash on. Take the dog to the living room. Leash off. Again to meh feh.
Leash on. Shoes on. Shoes off. Leash off. Always to meh feh.
Leash, shoes, door. Meh feh.
Leash, shoes, door car. Back inside. Meh feh.
Leash, shoes, door, car, trip around the block. Meh feh.
Longer trips until you get to the park or agility. Then go home. Meh feh.
I tell some of my clients to drive out, turn around and go home. Not on class night, but at other times. Manage the dog’s expectations. I do this for regulars when I see Pavlov is getting a little too problematic.
Not all trips should lead to doggie Disneyland. Go to boring positive dog friendly places. Many farm supply stores, garden centres and auto shops allow dogs. Positive but rather boring. Balance out expectations. The habits we have shouldn’t create predictable, conditioned signs that clearly tell the dog to get excited. I want them thinking, focused happy, not spun out of control.
What should get the dog jazzed up?
For my dogs, it’s certain words. Their commands (cues) have a happy association. Down? Yes I’d love to. I might say, “wanna work?” They happily run over. Not bat shit crazy. Happy, focused, on point. It’s reasonably happy because until I said those words, they were not already teetering on the edge. It’s possible to have enthusiastic and focused responses, using loads of positive reinforcement AND to have calm without using aversives.
Enthusiasm isn’t bad. Gobs of positive reinforcement, food, toys are all fantastic. You do need to give Pavlov a nod. If you address Pavlov and drive the associations to the right place you can increase the reinforcers with little uptick in arousal. Really think, “Where do I want positive associations forming?” and “Where do I not want them?” So the fun training question becomes, “What little habits do you notice your dog amping up over?”
Also rewarding calming behavior by giving the thing they were initially excited about, Premack it!! Let the impulse pass and they might not even want it anymore. Find this a lot with sniffing or saying hi, will walk right by sometimes.
I have a similar problem with one of my border collies getting excited with people. He gets excited when he sees a dog. He will either loudly whine or scream. He is very friendly with dogs, however it’s too much, and it frightens the other human being. If I say not to worry he is just excited and will not hurt your dog, well that never works (nor should it).
I do not want to punish my dog at all, let alone for being a happy, friendly dog. Any suggestions, please?
Hi Dan,
You can do a similar exercise with the leash. That can sometimes help a bit.
The trick to leash reactive – where the dog is frustrated not getting to what they want but is super friendly is in recognizing that the skill of walking nicely is weak. The past history of getting to fun things is strong. In the moment, when there is a distraction, you are in a losing position.
To get around that, you need to build the skill of walking nicely separately. Then you need to drill OTHER distractions. Each rep you do is putting more power in your corner. So when you get to “face the people/dogs on the street” you have enough history on what you want to be successful.
It’s important that when you do this, dog should be 80% right. That way if you do need to take a few steps back if you lose your dog’s focus, it’s just a mistake the dog learns from. “This no longer works.” Not a frustrating thing. Just to be really technically honest, taking 3 steps back is considered punishment. It’s like a football penalty.
But put “mass” and “Power” behind the skill before trying to beat something that already has way too big of a reinforcement history.
Great article by the way!
I have a puppy and am really struggling with people out in public getting her overstimulated. I know it’s an age old problem but I’m at my wits end.
Oy…yeah. Tough one that most of us struggle with. It’s a bit of a balance between telling people to lower their excitement to what the dog can handle. But you can also focus on drills at home where you teach the dog to focus on you despite distractions. Build the focus up in a controlled environment. Build up the “mass” of the behaviour so you have a fighting chance out in the real world.
Thanks! Good advice! We are going to start some greeting practice with neighbors (the ones I can control haha)
Thanks! Glad you liked it.
Neighbours you can control are always good. hahaha