The Rise and Fall of the Clicker Fad

This was going to be a post about the appropriate use of clickers in agility training, but I got sidetracked. Instead, here’s a potted history of twenty years of dog training techniques in New Zealand!

Shaping: A Training Revolution

Clickers first came to the NZ agility scene in the late 1990s, when a few clubs teamed up to bring an Australian trainer called Steve Drinkwater out to run some seminars. Sadly I missed his Wellington seminar because my parents made me go to school, so I gathered it second

At that time most agility handlers had a rather limited understanding of the science of how dogs learn. A lot of training involved pushing, shoving and pulling on the dog – drag him up an Aframe on lead, push him into a tunnel, use your leg to guide him into the weave poles, smack him on the bum if he won’t sit. At more enlightened clubs, they tried dangling a cookie in front of the dog’s nose for him to follow around instead.

Steve’s seminars brought a radically different way of training. “Big” behaviours like weaving 12 poles or running a course were broken down into little pieces, like just walking through the gap between two poles. The dog was encouraged to learn these little pieces by experimenting and earning rewards for getting it right. Then the pieces were combined together until ultimately the dog could run a full course with all the obstacles.

This process is known as shaping.

What’s That Got To Do with Clickers?

You don’t actually need a clicker to do shaping, but it helps a lot. When you’re working on small precise behaviours, the reward timing is very important. The clicker is a way to indicate to the dog that you are going to reward him for the thing that he is doing at that precise moment, but you don’t quite have it ready yet.

Clickers are better than words like “good dog” or “yes” for a couple of reasons. The sound is very short and sharp, much shorter than even a one-syllable word. This reduces confusion about exactly which moment of behaviour you are rewarding. It’s also a sound that the dog won’t hear in his everyday life, whereas many people are prone to verbal diarrhoea and use words like “good dog” outside of a training context.

Some handlers (and some clubs) went through a phase in the early 2000s of trying to use a clicker for everything, and discouraged beginner agility students from using verbal praise at all. Overseas agility experts like Susan Garrett also participated in this fad, and provided a lot of the inspiration for kiwi trainers.

Clicker Training Without the Clicker

Back in the 1990s we had dogs that were used to being pushed and pulled around, and understandably wary to volunteer behaviour. Then we all went home from the Steve Drinkwater seminar, sat down with our brand new clickers, and rewarded our dogs copiously for every little thing that they did. The dogs cottoned on fairly quickly – when the clicker was in the hand, it was time to experiment until they got a click.

The dogs still had years of baggage from the older way of training. The people still had habits from years of training with force and yelling. The dogs tended not to offer random behaviours outside of the very specific “clicker and treats in hand” scenario that they knew was safe. The people began to rely on the clicker as a crutch to inspire spontaneity in their dogs.

The next generation of dogs grew up differently. We were so excited to see that spark of creativity in our puppies, and we tried not to squash it out of them with too much old-style training. During the 2000s we learned how to control our pups’ environments so that we didn’t have to correct them so much, and our dogs matured into eager agility partners who loved learning. We discovered that we didn’t need a clicker in our hands to do what we called “clicker training”.

You Don’t Butter Bread with a Scalpel

The clicker is a fantastic tool for isolating a very small part of a dog’s behaviour that you want to reward. A lot of agility training is not that – remember my example before of the handler who clicked after they’d finished running a sequence?

Clickers can be handy for training the small steps of foundation skills – things like targeting, wrapping around a cone, or exploring a wobble board. They’re particularly useful for things that require a high degree of precision, like the fitness skills that many people are now teaching their dogs.

Even in these situations, they’re really not necessary. We’ve improved our training mechanics a lot since the 1990s. We’re good at coming up with really easy starting points to build from, and at controlling the environment and placing our rewards to help our dogs succeed. Check out this video from Susan Garrett comparing her 1999 training to her 2012 training – what a difference!

Personally I don’t use a clicker much any more. My last dog Spring (not the brightest crayon in the box, bless her heart) would freeze in position when I clicked, standing still for many seconds while she pondered what she’d done to earn the click. At least half of these real-life freeze-frames showed me that I’d clicked too early, too late, or for the wrong thing altogether. This feedback on my awful timing encouraged me to rely more on verbal praise, where I don’t tend to get dozens of repetitions of the Wrong Thing because I accidentally clicked it once.

Once the individual behaviours have been built up into chains and sequences, the clicker is definitely no longer needed. The dog’s reinforcement is that the sequence continues and he gets closer to earning his reward. If there is a problem with a particular skill, it needs to be refined in isolation. Stay tuned for more on this next week!