One of the biggest criticisms of “positive-only” training (side note: there is no such thing as “only” positive training, this is a misnomer based on a misunderstanding of the quadrants of operant conditioning. But I digress…) is that we never ‘correct’ our dogs for doing the wrong thing, so how could they possibly learn what is ‘wrong’? Many ‘balanced’ trainers (meaning trainers who use all quadrants, including positive punishment and negative reinforcement)say that without this information, dogs will be unruly and out of control. However, the idea that our dogs do not have consequences to behaviors that we don’t like is simply untrue. Force-free trainers will make use of negative punishment, which means removing something the dog likes, to teach the dog that their behavior can make good things go away. The true difference isn’t that we don’t provide consequences for undesired behaviors, it’s that we take an ethical stance against using force, pain, and intimidation as consequences.
The following is a five step process that can help you modify many of your dog’s undesired behaviors:
Figure out EXACTLY what it is you don’t like about what your dog is doing. When we’re working with another species, we can’t explain to them in our language what we’re working towards, so we need to make sure that the criteria is crystal clear for all involved.
Figure out EXACTLY what it is you’d like your dog to do instead.
Generously reinforce your dog for what you WOULD like them to do.
Remove reinforcement for the undesired behavior.
Manage the dog during his training to prevent slip ups.
The following is this process put into the specific context of ‘loose-leash walking’:
I DON’T like that my dog pulls forward like a freight train on leash.
“Not pulling on leash”, while self explanatory for us, could mean any number of things to our dogs. Would we like them to be behind us? At our side? On one side? Walking on their hind legs? (I wouldn’t suggest that last one for long durations, but a fun trick to teach!) We must be specific in our criteria. For this example, we’ll say the “dog can be anywhere in front, to the side, or slightly behind us, but the leash MUST be loose.” So, we have one criteria we’re looking for: a relaxed, loose leash.
When you’re out walking your dog, bring your clicker and treats, and allow the dog to move forward as long as the leash is slack. For an extra layer of training, we can click/treat our dog every time they walk nicely by our side (I like to think of the dog’s head lining up with my knee as the ‘clickable’ moment) and click/treat them when they give us eye contact.
When your dog DOES pull, and the leash goes tight, say “Oops, too bad” in a light, relaxed tone as your no-reward marker, and turn around to walk in the opposite direction. This is a consequence, but no pain or fear was needed. We taught our dog “Oops, pulling loses you the opportunity to sniff that tree”, and you remove the desired thing (the opportunity to sniff). The minute your dog catches up with you, you can click/treat them for position, and then raise the rate of reinforcement as you approach the area your dog was pulling toward, teaching him “yes, you can absolutely sniff that cool thing, but you have to walk with me to get there first!”
Management for this behavior would include a front clip harness to decrease pulling, as well as to keep you safe from being pulled over. Additionally, you can precede each walk with an exhausting game of fetch or tug to attempt to tire your pup out before the excitement of the walk.
You can apply this same formula to almost any other behavior that you don’t like, and you never have to use pain or fear to make your dog behave the way you’d like! Will it take some time and consistency? Absolutely, and the more practice you and your dog get, the faster the desired behavior will develop. But what a wonderful thing it is to know that you never have to scare or hurt your dog again to modify their behavior!