Training Talk: Primary vs Secondary Reinforcers

People really like to get their panties in a twist when it comes to using treats when training their pets. It seems as though treats make you a lesser trainer, or that it is somehow insulting that your pet will work for food. This instalment of “Training Talk” will explain why food makes such a great reinforcer during training, as well as looking at other reinforcers we can use and when to use them.

In the first “Training Talk” blog we looked at reinforcement, and how when we reinforce a behaviour that our pet is offering we will see it more often. But how do we reinforce a behaviour? When training our pets we can use what we call either primary reinforcers or secondary reinforcers.

primary reinforcer is something that is essential to our pets survival, and therefore is a very motivating thing to work for. Common primary reinforcers are food, water, air, shelter, and sex. Now personally I’m not into depriving my pets of water or air, nor am I likely to put them out in the hot sun and make them train for the opportunity to get some shelter. I’m definitely not going to offer them any “special favours”, so to speak, for a job well done either. That leaves us with food. All animals eat. In my blog post “Is Your Pet Food Motivated” I look at reasons why your pet might not want to eat during a training session, as well as how to remedy the problem. I recommend you read that post if you feel your pet won’t work for food.

dog treats
An assortment of dog treats

Now, many people who oppose positive reinforcement suggest that for food to be an effective reinforcer you must first deprive your pet of food so they are hungry. It is true that most animals won’t work for food if they’re full, but we don’t need to starve them to train either. I like to use food rewards that are either part of your pets daily ration, or something extra tasty that only comes in a training session.

Secondary reinforcers are things that aren’t essential to survival, but over time and by being paired with a primary reinforcer (like food) have come to be motivating and enjoyable to your pet. Common examples are petting and praise, or a great game with a favourite toy. For many pets these things don’t mean much initially, but over time and when paired with things your dog really enjoys, they come to be reinforcing to your pet.

Secondary reinforcers are not as motivating as primary reinforcers, which is why trainers recommend teaching new behaviours using food rewards. Once that behaviour is learned it is actually to your advantage to start mixing it up with how you reinforce your pet – you would work harder too if a bonus could be just around the corner!

toy play reinforcer
Wilbur loves training for a game of tug!

Take care when training with secondary reinforcers. Remember that reinforcement is an individual thing. If you are offering a pat or praise, make sure you pet is offering the behaviour more often. If not, it is not being reinforced – try something more fun or tastier!

Advertisements

Training Talk: Punishment

This instalment of “Training Talk” is going to look at punishment. This might seem an odd topic to find on the blog of a force-free trainer who utilises positive reinforcement, but it is my belief that it is important to understand not just what you recommend, but also what others recommend.

Punishment is a word that holds a lot of unpleasant associations. When we hear the word we tend to think of a smack or harsh words. In the world of training it isn’t quite that simple.

Dog Training

Punishment is simply something that decreases the chance of our pet repeating a certain behaviour. When our dog barks or our bird squawks and we yell “NO!”, we are hoping that they will stop the noise and hopefully not do it again. This example, and many other commonly used “punishments” are rarely as effective as we would like.

Just like reinforcement, there are two types of punishment we can utilise in training. The first is the most well known, positive punishment. This sounds contradictory, but all the positive means is that we add something to decrease a behaviour. We might add a loud noise to stop a barking dog, a squirty bottle to stop a scratching cat, or a yank on the leash to stop a pulling dog.

We can also use negative punishment. This sounds silly too, but like the positive means to add something, the negative means we take something away to decrease a behaviour. When we use negative punishment we take away something our pet likes in response to their misbehaviour, it’s like a time-out. This is what people are hoping to achieve when they “ignore” their pet for doing something naughty, they are taking away their attention, but often there are environmental factors outside our control that may be undermining our efforts.

Timing is everything in training, and this is true of punishment too. If we want our punishment to affect our pets behaviour, it needs to happen immediately following the behaviour we want to see less off. This means if you discover your pets misbehaviour hours after the event, you just need to move on. Punishing you pet will be damaging to your relationship, and completely irrelevant to your pet. Also consider how hard it is to properly “time-out” your pet immediately following something naughty – by the time you get them to the naughty corner or back in the cage, 10 things have happened since the behaviour you want to punish, and the lesson is lost!

If we get our timing right we can stop unpleasant behaviours, but we have to realise why punishment works to decrease a behaviour. It works because our pet doesn’t enjoy the outcome, most often because what we do is aversive (or unpleasant). Training with aversive’s create stress and anxiety for our pets, which can lead to bigger problems long term. We would do better to look at why the problem behaviour is occurring, and looking for ways to prevent it happening again while teaching our pets more acceptable behaviours that we can reward. This is much more enjoyable for both us, and our pets.

Train smart, not tough!

Training Talk: Reinforcement

Welcome to the Training Talk series! We have decided to put together a series of blog posts to introduce readers to the basics of learning theory. Understanding the simple rules that cause animals to behave in different ways will help you to be able to work with your pet more effectively. Any trainer or owner, whether they are increasing good behaviour, decreasing bad behaviour, or working on an animals emotional response to something, is working with the principles of learning theory (whether they know it or not). We want you to understand how the different methods of training work, so that you can make informed decisions about how you train your pet.

In the first instalment of the Training Talk series we are going to look at reinforcement. What is it, how can we use it, and why does it work?

In the world of animal training and behaviour modification, reinforcement is simply something that increases the likelihood of the target behaviour occurring again. Notice how there is no mention of treats or praise? Reinforcement doesn’t exclusively apply to something good, it just means we are doing something that will make our pets repeat a behaviour.

Modern pet trainers generally focus on using what we call positive reinforcement. This means that the “something” that increases the chance of our pet repeating the behaviour is good. Our pet is repeating the behaviour to earn something, like a treat, and because they enjoy it they will do it more often.

positive reinforcement cat

Positive reinforcement is easy for us to use when training our pet. All that matters is that our pet wants what we have on offer (whether it is a treat, a toy, or praise) and we will keep seeing that behaviour. Even better for us, because our pet wants what we’ve got, they will look at training time as a fun game. Their behaviour can lead to fun consequences.

Traditional pet training (especially for dogs) focuses on a different kind of reinforcement. Negative reinforcement is still increasing the chance of our pet repeating the behaviour, but rather than earning something good they are avoiding something unpleasant. A mild example would be teaching a dog to sit by pushing its bottom down. By sitting, the dog will stop the pressure. They will learn that in future, if they sit promptly, they can avoid the pressure altogether.

You can see that while the end result is still that the behaviour increases, the experience is very different for our pets. Rather than seeing us as goody dispensers, and learning that they can earn the good stuff by behaving in certain ways, our pets learn that we can sometimes create stress for them, which they can avoid by behaving the way we want. This is less fun for both our pets and ourselves, and long term can create strong avoidance behaviours in our pets.

A feature that is crucial to understand about reinforcement in general is that it is unique to each animal or person. If you are trying to teach your pet to do something using praise or pats as your reinforcer but your pet is not repeatiang the behaviour, then praise or pats are not reinforcing to your pet in this context. Try using treats or a game with their favourite toy instead. Remember, by definition, reinforcement increases behaviour.

At Treat. Play. Love. we focus on teaching owners how to use positive reinforcement when training their pets. People bring pets into their lives for companionship and fun, we want to make training an enjoyable experience for everyone!