Clicker Training
written by Kerry Bradbury
Home
Introduction
Why Clicker
Getting Started
Using the Clicker
Teaching the Basics
Targeting
Some tricks
Links
Getting Started
Clicker training is quite a relaxed and flexible form of training. It is not for the impatient, as it allows the dog to use his brain and figure out what is required as opposed to being shown. The best way to use a clicker is to have some rewards on hand and then just watch your dog. When he does something that you like, click & treat and then wait to see if he does it again.
It is easiest to start with, if you have a few things in mind that you would like to teach your dog but simply click whatever happens first and then go from there. Inexperienced dogs will take more time to figure out what is happening, as they will not yet have learned to connect the correct behaviour with hearing the click. Dogs that have learned a lot of behaviours with clicker training tend to begin experimenting as soon as they realise that rewards are on offer and quickly repeat what you reward.
The important thing to remember about rewards is if the dog doesn’t want it, it’s not a reward. It would be like trying to reward your children with treats of cabbage and peas, when they prefer chocolate and crisps.
Which rewards?
What you use to reward your dog is up to you and should depend entirely on what your dog likes. Small pieces of food are best when first shaping a behaviour as many repetitions are required. If you reward your dog with large treats he will soon become full up and will not want to play anymore.
You may need to vary your rewards according to where you are. You may find that the treats your dog likes to eat at home won’t quite hold his interest out in the park where there are many other exciting things to do.
Most dogs being trained with a clicker however see the clicker itself as a signal that great fun is about to be had and so want to stick around for that reason. Toys can also be used but may not be as useful in the early stages of training as you would need to keep getting the toy back. Having the same reward again and again can for some dogs loose its appeal after a while. Chase or retrieve motivated dogs (such as Border Collies or Springer Spaniels) however will probably work very happily for a toy once the behaviour has been learned, or even in the learning stages.
Introducing the clicker
Opinion varies on how the clicker should be introduced to a dog. If your dog is at all sound sensitive, it may be wise to muffle the sound initially by holding the clicker in your pocket. Once the dog has heard the sound followed by a reward numerous times, the sound will not have any bad associations.
You can introduce training with a clicker in two ways.
-
Simply teach the dog that a click is always followed by a treat, without aiming to shape any behaviour. Do this by just randomly clicking and then treating your dog. Don’t always hand the treat to your dog, sometimes throw it away from you so that the dog associates the click with food, not the click with being near you. Once you can see that the dog is responding to the sound by looking for a treat, you can then move on to training a behaviour.
-
Go straight ahead and begin to shape a behaviour. Progress will probably be a little slow for the first few clicks as your dog will not know what the click means. Once you have rewarded your dog several times he will soon get the idea.
The disadvantage of no. 2 is that you may not have as many opportunities to click as your dog may not be offering the behaviour that you want. Always go for a very simple task to start with so that your dog can achieve success. How you introduce the clicker really depends on your dog. If you have a dog that is highly motivated and active then it is probably best to get them thinking right away. With a distracted or less motivated dog, the first way will work best.
To start with go for very simple tasks and click & reward these. You can change your mind once you have started reinforcing one thing if your dog then does something you like better. Be careful not to do this too often however as you run the risk of confusing your dog and ending up with him learning nothing much at all.
When you are teaching your dog a new behaviour it is important to have some level of success early on in order that your dog may be rewarded and so will continue to want to learn. The reason many people have trouble in teaching new things to their dog is that they aim too high initially. If you ask too much of a dog in the beginning he will have no chance to be successful and so will not see the point of continuing to try hard.
Whenever you teach any animal (humans included) something new you should break the task down in to small stages, progressing through each stage after achieving success. Of course, some things are relatively easy for dogs to learn and so only require one or two stages. Others will need far more.
Rewarding
It is very important to reward your dog every time when you first begin to teach him something new. The reason for this is that the beginning of training is when your dog is trying to figure out what it is you want. We can’t tell our dogs verbally when they are doing things right as they don’t understand English and as far as they are concerned we may be telling them they are doing it wrong. The most effective way to let them know we like what they are doing is to reward them.
Lets look at this from the dog’s point of view.
1 | Fido is jumping about - nothing happens |
2 | Fido is staring at you standing up - nothing happens |
3 | Fido sits and looks at you - click and treat. "Wow! Why did that happen" thinks Fido - "lets experiment." |
4 | Fido walks around - nothing |
5 | Fido sits - click and treat. "Ahah" - thinks Fido -could it be when I stick my bottom on the floor she likes it?" |
6 | Fido sits - nothing happens. Confused Fido - "can’t be that, what does she want?" |
So we have gone from a dog going in the right direction to a dog that no longer knows what is required. This is why, when we develop a new behaviour, it is important to reward every time.
Once a dog has learned a behaviour and understands what is required we can improve on the performance by not rewarding every time. This will also help to prepare the dog for times when you are not able or do not want to reward and will teach him not to give up. Again lets go back to Fido to get his view of the situation.
Fido has now learned the behaviour ‘sit’ on command
1 | You say sit - Fido sits - you reward |
2 | You say sit - Fido sits - no reward. Fido thinks ‘huh, why no reward, I know that was right’ |
3 | You say sit - Fido thinks ‘better do it quicker this time, she might reward me’ - Fido sits - you reward - Fido thinks ‘ahah, I must have to do it quicker’ |
This will work if you reward in a random fashion, picking out the best performances and importantly rewarding before your dog gets totally fed up. To start with you should still reward quite often. If however you have been rewarding your dog every time and you suddenly stop, he will quite quickly stop sitting for you. If you have taught him that there will always be a reward eventually, he will keep doing it just in case this is the time the reward is given.
By rewarding like this we can refine the behaviours that we have taught. For example, if we want to teach our dog to sit more quickly we would only reward a sit within 2 seconds of asking, and then only within 1 second of asking and so on. In this way we can steadily improve the dogs responses as he tries harder in order to get the rewards.
It is not by any means vital to reward your dog intermittently. If you choose, you can go on rewarding your dog for sitting every time for the rest of his life and he will still sit happily for you.
Something for you to reward:
To start you off with clicker training try the following:
To start your dog on something completely new and very simple begin by teaching him to touch your hand with his nose.
This is a task that is very easy for you to click & treat because it is clear whether you dog is, or isn’t touching your hand. With such a black-and-white behaviour, with no grey areas, it should be easy for your dog to understand exactly why the click is occurring.
Begin by holding out your empty hand near to your dog. If he looks towards it click & treat. Next wait for him to go towards it - click & treat. Then aim for him to nudge your hand - click & treat. If you want to you can teach him to nudge your hand quite hard, by not clicking the moment he nudges and waiting to see if he is more persistent. Don’t rush through this task, try to get several repetitions of each stage before moving on and expecting a little more from your dog. Once he starts to get the idea then increase the difficulty gradually.
Once your dog has learned to touch your hand in an easy position, begin to move you hand around and teach him to touch it regardless of where it is. Again, take it slowly so that you do not knock your dogs confidence. If your dog is finding things difficult and begins to give up, make things very easy for him again and then build back up to the level you had previously achieved.
As a general rule only accept the same or better behaviour than you have clicked before except if your dog is losing confidence in which case you should make things easy again for a while.
|