Reinforcement Primer

Training a dog is about controlling the reinforcement.  So if you want to train your dog, you need to have a basic understanding of reinforcement and how it works.

Dogs don’t process information and learn the same way humans do.  It serves as a disservice to dogs to expect them to think in complex human terms like “right” and “wrong.”  Dogs, and all animals, repeat behaviours that are reinforcing for them.  Think about what dogs like to do – sniffing a tree, for example.  They do this because they learn so much from smelling smelly stuff.  

Do you have a dog who nudges your hand when he wants a pat?  Or a dog who barks or paws the door when he wants to be let outside?  Maybe your dog pulls like a sled dog when he wants to go sniff a tree or greet another dog?  These are all examples of a dog who has learned to do something that gets him what he wants.  If you froze in your tracks every time your dog tried to drag you up to meet another dog, over time your dog would stop pulling on the leash to get to where he wants to go.

When we reinforce with food, it’s a way of manufacturing a situation where a behaviour we would like to see more frequently – such as not jumping on people – becomes a more reinforcing behaviour.  However, if your dog jumps on people and every once in a while someone laughs and welcomes this type of greeting, it becomes difficult to distinguish the behaviour of jumping on people because it becomes like a slot machine for your dog – he will jump on people because every once in a while, it pays off with fussing and attention.

It’s nearly impossible to completely control every, single outcome of your dog’s behaviour.  Some mornings when you take your dog for a walk, you may be running behind schedule.  So rather than putting on the brakes every time he pulls on leash, you let him pull on the leash so you get through your walk quickly.  The secret to training success lies in two key concepts:

  • Introduce the behaviour – in this case, not pulling on leash – in situations like a dog class where you are able to control the outcomes until the behaviour becomes strong.
  • Think “consistent.”  If you are able to maintain your desired criteria – I stop moving as soon as the dog pulls on the leash – 70% of the time, you’re gonna be ok.
So, why use food as a reinforcement?

Because food is typically very high on your dog’s list of reinforcers, food can be easily transported with you (if your dog’s idea of a life reward is being allowed to sniff a fire hydrant, you can’t take the fire hydrant with you) and food allows you to get a lot of reinforcement for a behaviour in quick succession.

Initial training should ideally be done with food.  It allows you to build a deep reinforcement history, quickly.  For example, value for being on your left side when on leash.  Feed ten cookies for position in 30 seconds and, BOOM!  You’re already well on your way to building value for your left side.

Also, initial training should be started in a “low distraction” environment – like your basement, backyard or at a dog class.  Low distractions means things you might use a life rewards such as fire hydrants aren’t handy.

What are some good food rewards?

Kibble is often a great food reward if you’re training in your home, but you may need a higher value food reinforcement if you are in the backyard or in a dog class.

Every dog is different but here are some common food rewards:

  • Rollover food rolls, cut up into pea-sized pieces.
  • Freeze dried liver or chicken.
  • Cheese.
  • Zuke’s brand Minis.

You will have a good idea whether your dog finds your food rewards reinforcing if he stays engaged with you.  If he chooses to wander away, you may need to try another food reinforcement.