The Dangers of Aversives

You’ve probably heard the words ‘punishment’, ‘correction’, and maybe even ‘aversive’, but what do they actually mean?

Science defines ‘positive punishment’ as something you do to your dog that they find unpleasant (painful or scary, for example), which results in the reduction in frequency of that behaviour.

An aversive is something your dog doesn’t like; something he would like to avoid having happen again. If your dog pairs an action, behaviour, or event, with an aversive, there is a chance he will avoid that situation or that action in future.

The terms ‘correction’ and ‘punishment’ can be misleading. People use them incorrectly or don’t understand what they mean. For example, some people think a correction is fine and a punishment is horrid. It all boils down to this – if you do something to your dog and he does not like it and would like to avoid it happening again, then that is aversive. If it serves to reduce the frequency of a behaviour, then it is a punishment. It doesn’t matter what you call it, or what your intentions were; if it’s aversive, and it makes your dog think twice about performing that behaviour, then that is what it is. I have yet to find anyone using the word ‘correction’ who is actually telling the dog what the correct thing is. They’re delivering an aversive every time.

However, and here we come back to that scientific definition of a punishment, it needs to reduce the frequency of a behaviour. If it doesn’t
or if it did and then it stops working but you continue doing it, it is no longer a punishment. It may, however, still be unpleasant for your dog!

What would you call something unpleasant, painful, or scary that someone keeps doing to you even though it isn’t improving your behaviour? I’d call it abusive.

And even if the aversive is altering behaviour, that’s still not OK. That is still a problem. To understand the reason for this, we first need to look at human behaviour. The principles of learning theory apply to us as well as dogs. If we do something that works or that is rewarding to us, we do it again. That’s positive reinforcement. So, Dave jerks Fido’s lead to stop him from pulling when walking, Fido stops pulling for a bit, Dave thinks, ‘well that worked’, and he has been positively reinforced for delivering an aversive to Fido.

Let’s say that this works as far as Dave is concerned for a week. At the end of the week, however, Fido sees a squirrel disappear over a wall and so he pulls, and this time, when Dave jerks, Fido doesn’t stop pulling. What does Dave do now? Well, he’s always been rewarded for jerking Fido’s lead, so the logical thing for Dave to do is jerk the lead harder!

Now, if you’re saying ‘no, no, Dave would think of something else, surely’, let me ask you this:

• What do you do if your pen stops writing? Get out a new one or scribble a bit harder?
• What do you do if the key won’t turn in the lock? Call a locksmith immediately or shove the key in and out, wiggle it a bit harder, and rattle the handle a bit?
• What do you do if the kids upstairs don’t hear you yell them down for tea? Go upstairs or yell a little louder?

Be honest, we all experience this. It’s called an extinction burst. If a behaviour we’ve previously been rewarded for stops being rewarding, we will try harder and harder before we eventually admit defeat.

So, Dave jerks Fido’s lead harder, and it hurts Fido, and he squeaks and walks nicely for a bit. Now that Dave has been rewarded for jerking harder, he’s more likely to do it again.

But Fido’s neck is becoming thick with extra muscle, and the nerve endings are deadening as a result of all this tugging and jerking. So, when Fido pulls even harder, and Dave jerks harder, and Fido doesn’t respond… Now what?

Well, there’s a strong chance Dave will use two hands and yank Fido off his feet, but he might also look at something like a prong collar or even a shock collar. Why? Because he thinks he needs something stronger to get Fido to listen to him.

Once you’ve started working your way up that scale of aversives, it is very difficult to stop. There are risks involved beyond ‘just’ increasing the aversiveness:

• Pain. It hurts, that is why it works
• Fear. It’s scary, that might also be why it works
• Physical damage. Many aversives are designed to cause pain, and with pain comes damage; collapsed tracheas, bruised necks, burnt skin, damaged joints, the list goes on
• Mental damage. Subjecting anyone to aversive experiences they cannot escape from will cause psychological damage. Sometimes, just a one-off experience will have a negative effect. Other times, the damage occurs after repeatedly using the aversive. In humans, we’d call this post-traumatic stress disorder (PTSD). There is no guaranteed cure for PTSD in humans. There is scientific evidence on the effects of prolonged stress and repeated trauma on dog behaviour. If we can’t fix PTSD in humans who can talk to us, what chance do we have in dogs?
• Incorrect associations. This is where the dog links the aversive with something other than his own behaviour
• Worsening or new behavioural problems due to the suppression of behaviour

Using aversives does not treat the underlying cause of the problem. The behaviour you are seeing is a symptom of a physical or emotional state or condition, or it is just something dogs do that is inconvenient for people. Simply stopping that behaviour is not addressing the cause of that behaviour.

Let’s go back to Fido and Dave. Fido naturally moves faster than Dave. His natural and preferred pace for a walk would actually be a trot, around 5-7 mph. Dave’s preferred pace is a leisurely walk of around 2.5 mph. If you’ve ever tried to halve your walking pace to match a child or elderly relative, you will understand why Fido struggles.

So, to begin with, we need to teach Fido to walk slowly without pulling. He doesn’t know how to do this. It isn’t natural for him. We need to give Fido a reason to walk with a person. Instead of teaching him ‘it is painful if you go ahead of the person and pull’, we could teach him ‘it’s really rewarding to look at the person and stick next to them’. We can simply drop treats down on the floor as we walk.

But Dave has gone for the aversive, so what happens next?

Since Fido is wearing a collar, all those lead jerks are happening around his throat. Dave asks someone in the pub about this, and they show him a thin slip lead that is placed high up on the dog’s neck behind the ears. This, Dave is told, is the ‘proper’ way to use a slip lead. It’s more effective this way. The person telling Dave this is an ex-police dog trainer, and they are not wrong, but they haven’t explained why. It is more effective because a thin slip lead, rather than Fido’s broad collar, coupled with the placement high up behind the ears and under the throat, will cause Fido more pain when he pulls or when Dave jerks the lead.

So, Dave tries this out. He gives it a go on a Saturday morning walking through town.

First, they meet Mrs Bunn the Baker. Fido pulls forwards to say ‘Hi’. He likes Mrs Bunn… JERK! OUCH!

Then they meet Mr Bones the Butcher. Fido pulls forward to say ‘Hi’. He really likes Mr Bones… JERK, OUCH!!!

Then they meet Mr Smith the Fishmonger (what?!). Fido pulls forward to say ‘Hi’. He really likes stinky Mr Smith… JERK! OUCHHHH!!!!!

Then they meet Little Jimmy the Loveable Child. Fido pulls forward and then… Growls at Little Jimmy, bares his teeth, barks loudly and backs off… And gets a jerk and slap on the nose for growling and making Little Jimmy cry!

What happened here? Well, Fido wasn’t really aware that he was pulling on the lead to greet people. He was focused solely on the people – his favourite people! He was only thinking of people each time he got a painful jerk to the neck. So, when he met Little Jimmy, he associated seeing a favourite person with feeling pain. He did what a dog does when a situation occurs that he feels worried or threatened by. He growled, bared his teeth, and barked to get Little Jimmy to back off… And he still got a jerk in the neck and a slap on the nose.

Now he has learned that people are bad news, whether you approach or not! He may also have learned that warning people about how you feel (by growling) is a bad idea, because that earned him more pain.

The following week, Dave takes Fido into town to do the shopping. It’s Fido’s first walk since he growled at Little Jimmy as Dave has been too busy. He is quite pleased as Fido seems to be sticking closely to him today, not leaving his side. The training must be working!

Here comes Little Jimmy! He’s sneaking up behind Dave and Fido as they wait at the bus stop. He’s going to surprise them by jumping out in front of them to say ‘Hi’. He’s totally forgotten about Fido growling and barking at him last week because he’s 10 and this is a story.

Little Jimmy jumps out, ‘HI! SURPRISE!’

Fido is shocked! Here is the person that meant a jerk and a slap last week! He knows he can’t growl; it will get him a slap. He can’t back up either as they’re standing against a wall, and he can’t run as he is on the lead. So, as Little Jimmy bends down to pat Fido’s head, Fido snaps and catches Little Jimmy’s hand with his teeth… Oh dear, Fido. Oh dear.

While this story is fiction, it is based entirely on real events. Fido is an amalgamation of several dogs I have been called on to help, and who had bitten people as a result of using aversives and positive punishment.

There are lots of aversive devices and training techniques out there, and I can’t possibly list them all. There are collars and harnesses designed to tighten and choke or squeeze a dog’s sensitive areas. There are collars designed to really hurt, with prongs on the inside or electrodes to shock. Some are intentionally mislabelled as delivering a ‘tap’, a ‘stimulation’, or a ‘buzz’, but if it’s got electrodes, it’s giving a shock. There are slip leads, and even braces to hold them up over the sensitive structures behind the ears.

With some of them, it’s obvious that they are designed to be aversive. Some are more subtle, because when it comes down to it, it is the dog who decides what he or she finds aversive – not you, and not me. So, your dog may hate water being sprayed in her face and my dog might love it. It is an aversive for your dog, and a reward for mine.

Some aversives may look innocuous, like a can of pebbles or pennies, a ribbon loop with metal discs, or a bunch of keys or a chain. When thrown down on the ground, these all make a startling sound. They might not ever touch the dog, but startling someone to stop their behaviour is unkind, ineffective, and can have negative effects, such as sound phobias, which are very hard to fix.

Some aversives are awful on a psychological level, which can be every bit as damaging as the physical punishments.

The most horrific thing about aversives is that they don’t need to be used at all! Dogs, like Fido, experienced them unnecessarily. There are ways of teaching that could have been fun and enjoyable for Fido!

If Fido had been put on a harness, so that pulling never hurt him, he would have had a very different experience. If he had been taught using rewards, and even more rewards when Little Jimmy or Mr Bunn appeared, so that he understood staying next to Dave was a good thing, then he would almost certainly have bounced back from his startle at Little Jimmy jumping out in surprise. He would have been happy to see Little Jimmy the Loveable Child. He wouldn’t have been worried that he couldn’t run and couldn’t growl as he would not have needed to!

Next time someone suggests a training method or device to you, ask them (and yourself ) these questions:

• How does this work?
• What does my dog get if he gets it right?
• What are the risks?
• Why is the behaviour occurring in the first place?
• What could I teach instead of the unwanted behaviour?

Thinking about these things will help you decide whether you should follow the method. In case you need a bit more help, my final rules of thumb are these:

• If it starts with ‘this sounds awful but’, then it is awful. Don’t do it
• If it starts with ‘you have to be cruel to be kind’, then it is cruel, and no you don’t. So, don’t do it
• If it would get you locked up (rather than, say, raised eyebrows) if you applied the method to a toddler, then don’t do it!