Should You Use a Secondary Reinforcer (Clicker) in Counter Conditioning?

Counter conditioning is a powerful technique to help reactive and fearful dogs--but it doesn't need a clicker. Here's why, along with some tips on teaching fearful dogs to no longer be afraid.

Photo: Angyalosi Beata/Shutterstock

By Zazie Todd, PhD

A question I am often asked is, should you use a clicker (or secondary reinforcer) when using counter conditioning with a fearful dog?

The quick and easy answer is “No”, but since this is an interesting question I thought I’d take a closer look. 

Even though so many of us are familiar with the story of Pavlov’s dogs there is still, surprisingly, a lot we don’t know about exactly how classical conditioning works. Nonetheless, most of us know this is a powerful method we can use to help dogs (and cats or other animals) get over their fears by learning to like the thing they previously found scary.

The main things to know are:

• When teaching dogs behaviours using positive reinforcement, some people like to use a clicker. The click marks the behaviour and is always followed by a treat.
• When doing counter conditioning with a fearful dog, the aim is to change the dog's emotions, so we are not looking for a behaviour. That means there's nothing to click.
• In classical conditioning, it's important to keep a really clear relationship so that the dog knows that the thing they consider scary is followed by wonderful snacks. A click would be an extra complication and distract from that relationship. 

Let's look in more detail at some of the reasons why we don't use a clicker when using counter conditioning with a fearful or reactive dog. 

What is counter conditioning?

First of all, let’s remind ourselves what happens when we’re using counter conditioning to help a dog with their fears. We use something the dog already likes, which is not conditioned, such as delicious food. This is called the unconditioned stimulus (US).

Technically speaking, during the training we make sure the conditioned stimulus (the CS) is always immediately followed by the unconditioned stimulus (US, tasty food). 

Over time, the dog learns that the CS (scary thing) predicts the US (ultimate snack). When working with fearful dogs, we can remember the terminology by saying that the thing that's Considered Scary predicts the Ultimate Snack. 

The dog’s response will change from fear to a conditioned response of liking the thing they previously considered scary (the CS).

Most of the time this training will happen with the scary thing at an intensity where the dog is perfectly happy, because the most important thing with a fearful dog is to help them feel safe.

What is a secondary reinforcer?

Now let’s look at a secondary reinforcer. The most common secondary reinforcer that we use in dog training is, of course, a clicker. It is typically used to mark the moment the dog does a behaviour we want and is always followed by a reward (tasty food).

For example, you cue the dog to sit and the dog sits, so you click and then reinforce the behaviour with a treat.

This is a classically conditioned relationship: the click predicts a treat.

A clicker is especially useful when you need to bridge a gap between the dog doing the behaviour and you giving them reinforcement. Many people absolutely love clicker training, but some people prefer to use a specific word (like "Yes") as a secondary reinforcer instead.

The click is a secondary reinforcer because it tells the dog that a primary reinforcer (tasty food) is coming. Without that relationship, it wouldn’t mean much. 

Another example of a secondary reinforcer is money; we can use it to buy things (primary reinforcers), but on its own it doesn’t have that intrinsic value as it isn’t necessary for survival.

Why secondary reinforcers don’t belong in counter conditioning

So the question is, does it make sense to click before giving the food when counter-conditioning? 

Well, no. 

In counter conditioning we want to make it very clear to the dog or cat that the CS (the thing they consider scary) is now predicting the US (the tasty treats). 

A click would be an extra thing happening between the CS and the US, and it’s something that’s already trained. Technically, this would be called a compound stimulus, and having compound stimuli can weaken the response. So it’s best to keep it simple and not use the clicker. 

As well, because the dog is used to hearing the click when they do the right thing, they may associate the click with their own behaviour at the time rather than with the CS.

What does the science say?

Using a secondary reinforcer in counter conditioning (instead of a primary reinforcer) has been tested with people, and although it worked, it didn’t work as well as it might have done (Keller et al 2020). 

Counter conditioning studies with people typically start by conditioning a fear, and then testing ways to get rid of it. They probably aren’t the most fun experiments to take part in given that conditioning a fear often involves an electric shock.

In one study, people were first taught that moving a joystick in particular directions (CS) would be associated with a painful electric shock (US) (Meulder et al 2015). This is the fear conditioning part in which a fear is induced.

Then the scientists tried to get rid of that fear response either by extinction (stopping the shocks from happening) or by counter conditioning (instead of the shock, people were promised money).

In this case, there was no difference between the two groups in terms of the emotional response. Both counter conditioning and extinction led to the same level of reduction of fear of a painful shock. 

This is surprising because you would expect counter conditioning to work better.  Maybe this is because money, the US in this case, is a secondary reinforcer. It could be that the experience of being promised money is not a pleasant experience in the same way that eating nice food is. Perhaps a primary reinforcer would have worked better.

The scientists did not test whether giving people tasty food like chocolate when they made the joystick movement would have worked better. They suggest that future research investigate “the subjective appreciation of the reward”. 

Of course, there is another possible explanation that people are suspicious of psychologists doing experiments and maybe didn’t realize they really would get the money at the end. That’s one reason why it’s difficult to compare studies on people to those on animals. 

But despite these results, the scientists still recommend counter conditioning as an approach likely to work better than extinction in the long run.

Getting counter conditioning right

If you love clicker training, keep the click for teaching behaviours. In dog training, we would not expect a click on its own (without food) to work. It's important to use really great treats in counter conditioning, such as pieces of steak, chicken, or sausage. 

Of course, we use a clicker (if we use one) when wanting to change behaviour. In counter conditioning we are trying to change emotions, not behaviour, so it doesn’t make sense to click. In counter conditioning, the dog gets the US (tasty food) regardless of what behaviour they happen to be doing at the time; it’s simply for seeing (or hearing) the scary thing. (For more on that, see if your dog is afraid, avoid these two mistakes). 

Get Bark!

Although counter conditioning aims to change emotions, it's still important to pay attention to the dog's behaviour. If the dog is barking, growling, or lunging when you are trying to do classical conditioning, that's a sign that they are struggling. Put some distance between them and the thing they are scared of (and still offer the wonderful snacks). 

It can feel very difficult to offer your dog wonderful food if they are lunging and barking. While it’s best to avoid those situations in which the dog goes off, in real life it is sometimes hard to do. If it happens, it’s still important to do the counter conditioning and offer the food. Offer the food even if you don't think your dog will eat it; they will still notice it. You can read a story about me doing this with my late dog, Bodger, in Bark!. 

Summary

So the answer to the question is that even if you love to use a clicker, save it for when you’re training a behaviour. It’s best not to use the clicker in classical conditioning because in this situation, it may not work so well.

If you'd like to know more about how to help your reactive or fearful dog, check out my book Bark! The Science of Helping Your Anxious, Fearful, or Reactive Dog, which is full of tips to help with reactivity, resource guarding, and other behaviour issues. 

References

Keller, N. E., Hennings, A. C., & Dunsmoor, J. E. (2020). Behavioral and neural processes in counterconditioning: Past and future directions. Behaviour research and therapy, 125, 103532. 

Meulders, A., Karsdorp, P. A., Claes, N., & Vlaeyen, J. W. (2015). Comparing counterconditioning and extinction as methods to reduce fear of movement-related pain. The Journal of Pain, 16(12), 1353-1365.  

As an Amazon Associate, I earn from qualifying purchases.