Reinforcement and Rewards

What is positive reinforcement?

As reward based trainers all of our exercises are focused on teaching the dog specific behaviours and rewarding the dog when they get it right.

In a nutshell positive reinforcement means adding something pleasant in order to increase the likelihood of that behaviour being repeated. An example, you cue your dog to sit, their bum goes to the floor, we mark (“yes”) to tell them they got it right, and we reward with praise and a tasty treat or a throw of a toy.

Do rewards work?

In learning theory we can evaluate whether the rewards we are using are reinforcing by looking at whether behaviour has increased/improved or maintained. If the behaviour has not increased e.g. your recall is not improving, then this suggests that the rewards we are using to reward recall are not effective.

It is our job as handlers to evaluate whether the rewards we are using are actually reinforcing.

Different dogs value rewards in different ways. Some love to work for food, while others prefer to work for a game with a specific toy; some may enjoy both equally. We encourage people to write out a list of their dog’s top 10 rewards so that these can be used effectively when training them. Remember, it is not what we think the dog will find rewarding, it is what the dog actually finds rewarding, so take time to evaluate what really gets your dog going!

All dogs are individual and what works for one dog, will not work for another.

Types of Rewards

Food – including your dog’s daily diet and other tasty treats.

Toys and Retrieve Articles – test your dog’s preferences to see which they love.

Play – Explore games your dog might like to play.

Environmental – the most important to understand how to use to your advantage e.g. sniffing, hunting, retrieving, chasing, catching, play with other dogs etc.

Praise – both verbal and tactile touch.

Training – If we get it right, the dogs will find performing the behaviours rewarding.

How to use Rewards

We can make them “functional”.

When using food we don’t always have to post it to the dog’s mouth, when using a ball, we don’t always have to throw it for them to chase. We can use all the above rewards in different ways to fulfil your dog’s intrinsic needs and create functional, need-orientated reinforcers (we will do another blog on this later this year!). Think of it this way if we call our dog away from chase and they recall we need to reward them with a chase game to fulfil the desired outcome of wanting to chase e.g. chasing food thrown from side to side or chasing a toy.

Things to think about

Think about the delivery of food rewards, food can be animated in many ways, we can throw it for the dog to catch or chase, we can hide it for the dog to sniff it out, we can stalk up to it before allowing the dog to catch and consume it?

Can we hide the tennis balls for them to find rather that throwing it?

How can we control the dog’s access to environmental rewards to our advantage?

Is “good dog” really enough and does your dog really like to be touched?

Think about functional rewards, if we call our dog away from chase and they recall we need to reward them with a chase game to fulfil the desired outcome e.g chasing food thrown from side to side or chasing a toy.

What do I do if the dog doesn’t get it right?

If your dog fails to complete a behaviour do not let your frustration lead to punishment for the dog. Here are some things to think about when things are going a little wayward with your training. Review the situation and see if you can see why the exercise did not go well.

What changes need to be made to the environment to set the dog up to succeed?

Think about the 3 D’s – could we reduce the distraction, duration or distance slightly and then rebuild?

Do you need to go back to basics because the cues you thought were strong have not been tested in more stimulating environments?

Train a positive interrupter cue so you can use it when the dog gets it wrong.

Did you expect too much of the dog or choose a “leave it” on a chase, instead of trying a “recall”, would that have worked better?

Video your sessions so that you can watch your own handling to see if there are any subtle body language cues being given off that you weren’t aware of.