“Clicker training” has become something of a buzzword, and people all over the world are using clickers in their training. While this is an encouraging trend, many people don’t realize that there is more to clicker training than just using a clicker.
Clicker training encompasses both a technique and a philosophy. The technique of clicker training involves using an event marker (often the “click” sound of a clicker) paired with reinforcement (a desirable reward, such as a food treat or toy).
The philosophy of clicker training involves looking for behaviors you like and reinforcing them, rather than punishing behaviors you don’t like.
Part 1 of this article examines the technique of clicker training. Part 2 of this article discusses the philosophy of clicker training.
Clicker training is a system of training that uses an event marker followed by positive reinforcement to increase the likelihood that a desired behavior will be repeated. There’s a lot of jargon in that sentence, so let’s look at some of the words in it a little more closely.
An “event marker” is anything — a sound, a flash of light, or a touch, for example — that acts as a signal pinpointing a certain instant in time (and, by association, the behavior taking place at that moment) for the learner. Clickers make good event markers because they create a brief, distinct sound that is the same every time the clicker is used. The sound clickers produce is also quite unique, and unlikely to occur in other contexts.
The clicker isn’t always an ideal event marker, though. Some animals, such as fish, don’t hear sounds produced in the air very well. Others, such as deaf animals, simply can’t hear at all. The sound of the clicker might also be drowned out in a noisy environment. In cases like these, a flash of light, gentle touch, or other non-sound-based event marker is a better choice. (Note that the term clicker training is commonly used to refer to any type of training where you are using an event marker and following it with positive reinforcement, even if the event marker isn’t a clicker.)
“Positive reinforcement” is something added to the learner’s environment that increases the likelihood that a behavior will be repeated. “Positive” refers to the fact that something has been added to (rather than removed from) the environment. In other words, the term “positive” is used as it applies in mathematics. (Although the removal of something bad is sometimes referred to as a “positive” thing in common parlance, that common use of the word “positive” does not apply in this case.) “Reinforcement” (or “reinforcer”) is shorthand for anything the learner likes and will work for. Common reinforcers for pets include food (most often in the form of treats) and toys. “Positive reinforcement” can therefore refer, for example, to the giving of food to the learner.
Behaviors that are positively reinforced are more likely to occur again in the future. In clicker training, the trainer marks certain behaviors and then reinforces them (in other words, makes the behavior more likely to happen again) by adding something desirable to the environment.
Note that the learner, not the trainer, decides whether something is reinforcing. Put another way, just because you would do anything for a Klondike bar doesn’t mean every other person on the planet would (with all due respect to Klondike bars, of course). If you insist on giving Klondike bars to a learner who doesn’t like Klondike bars (in hopes of making them repeat a certain behavior), you won’t get very far.
When training using marker-based positive reinforcement training, a.k.a. “clicker training,” trainers click as the learner is performing the desired behavior. In most situations, the goal is to click for movement, rather than for completed actions.
Let’s look at teaching your pet to tap your hand with his or her nose (a fairly simple targeting behavior) as an example. One way to do this is to hold your hand right in front of your pet’s nose and then click as your pet begins to reach his or her nose towards your hand.
Most learners will complete the action they have started, as long as your hand is close enough, so your pet is likely to finish tapping your hand even though you marked the reach with your click, rather than the tap. Even if your pet actually stops without quite tapping your hand, your pet has now learned that reaching for your hand earns a click, so if you wait a millisecond longer before clicking the next time, your pet will complete the action.
Clicking for movement towards your hand generally works better than clicking while (or after) your pet taps your hand, for the following reason: Most pets will move their noses away from your hand after the tap, so if you attempt to click as the nose taps your hand but your timing is slow, you wind up teaching your pet to move away from your hand — the exact reverse of what you intended. When in doubt, click a little too soon. You can always delay the click next time.
After clicking your pet for reaching towards your hand, you would follow the click as quickly as possible with a treat, toy, or other reward your pet enjoys. Your pet would soon learn that tapping your hand with his or her nose results in a click, followed by a reward he or she likes, so your pet would repeat the behavior to earn the click and the reinforcement again.
Clicking at the right time obviously requires perceiving the behavior as it happens. That involves observing the learner (which can be a lot of fun – there’s a reason people tend to have great stories about the things their pets do). In the example above (nose to hand targeting), the trainer would watch and wait for the instant the pet’s nose begins to move towards the trainer’s hand. It can take a little practice to see subtle behavior as it is happening, but as with all other things, the more you do it, the easier it gets.
To recap, clicker training involves the following three steps:
- Observe (look for the behavior you want)
- Mark (click your clicker — or activate another event marker — as the behavior is happening)
- Reinforce (quickly follow the event marker with a reward the learner enjoys)
As you can see, the technique of clicker training is quite straightforward, although it can be challenging in practice. A later article will describe games you can use to hone your observation skills and ensure you time the event marker properly and deliver the reinforcer promptly.
To learn more about clicker training and get in-person coaching, seek out a clicker trainer in your area (feel free to contact me even if you are not in my area as I can provide referrals, or use the links on the right side of the page). If you are interested in meeting other clicker trainers and learning about clicker training in depth from a variety of experts in the field, consider attending ClickerExpo.