Ouch! You grit your teeth, steel your resolve and turn your back…again, as your 35 kg, over exuberant young dog bites your elbows, grabs your jacket, bounces off your shoulders and generally uses you as a punch bag/trampoline…
Your ears are ringing and there is a sledge hammer pounding the inside of your skull in time to the high pitched, incessant screech barking of your dog demanding attention. So far, her record is 20 minutes without taking a breath – can she breathe through her ears? You pray she can’t go on for too much longer…
You sit there staring steadfastly at the TV screen. Poke. Snuffle. Whinge. Poke poke. Bop. A toy lands in your lap. Then another one… and another. You’re gonna be buried alive under spit covered dog toys. You do your utmost not to look; not to move a muscle and definitely not to respond. “It’s nearly bed time! Please, just go away and settle down” you think silently to yourself…
Been there, done that?
If you’re nodding your head in sympathy and understanding, you’ve probably been given the well-meaning advice to ‘ignore’ your dog when he’s misbehaving or being a nuisance and only interact once he’s stopped. And maybe you tried. Really tried. The examples above might strike a chord simply because you gave it your best shot repeatedly and these were the results you got for your efforts. If you’re like most normal people, you will have got to the point where you just couldn’t ignore your dog anymore; either yelling at them, looking at them or touching them. And the next time you tried ignoring them? They once again worked themselves into a frenzy of effort until you, again, couldn’t help but acknowledge their existence.
Introducing the ‘fifth quadrant’
All the examples above have one thing in common. They are behaviours that reward-based trainers will frequently try to reduce by ignoring them. There are four other consequences for actively changing behaviour, sometimes referred to as “the four quadrants of operant learning”. If you want to know more about these and how to use them (or why to avoid using them!) you can find out more here, but for now I want to look at the other consequence – that of no consequence at all. The dog behaves and NOTHING changes. There’s no result. Nothing added, nothing taken away. Just…nothing.
This is known in technical jargon as extinction: letting a behaviour die out because it has no effect on anything at all. I call it the ‘fifth quadrant’ because it changes behaviour by providing a sort of ‘non-consequence’ consequence, if that makes any sense! Not so many years ago it was the ‘go to’ of the reward-based trainer for getting rid of unwanted, nuisance behaviours. In some places it’s still recommended and used frequently. The unfortunate thing is, it’s a tricky tool to use. Often it has no effect on reducing the problem or it actually makes the situation worse. Bugger.
Pitfalls of extinction
So why is ignoring a problem behaviour so frequently ineffective? Well, there are a number of reasons: Extinction only works if the behaviour has no ‘intrinsic’ value. So, if a behaviour isn’t fun to do and it doesn’t result in any benefit for the dog, it’ll get dropped from the dog’s repertoire. That’s extinction.
The issue is that lots of problem behaviours ARE fun for the dog! Barking can be inherently pleasurable for many dogs. They just enjoy doing it. They do it for emotional reasons but they also do just because it feels good to express themselves. Jumping up is fun! Carrying toys can be fun! You can’t remove the reward of doing these behaviours if the pleasure is internally generated – which means that the behaviour still DOES have a consequence.
1) The hallmark of extinction is frustration.
Imagine you usually get your morning coffee and snack from a vending machine. Today you put your money in the slot, hear the satisfying clunk-rattle as it hits the machine’s innards, push the usual buttons and eagerly await the delivery of your sustenance. Only it doesn’t happen. Nothing happens. Frowning, you give the buttons and levers a bit of a rattle and shake. Still nothing. You push more buttons; harder this time. You feel your annoyance growing and building like a storm cloud on the horizon. You start to mutter and swear under your breath, then louder. You hit the machine, maybe bang the sides and contemplate giving it a quick kick. Nothing happens. You can almost taste your favourite snack; there it is, sitting there on the other side of the glass, thumbing it’s nose at you. Frustrated? You bet!
2) Frustration leads to escalation
So now you’re good and frustrated, did you notice what happened to your behaviour? Yep, you tried harder… and harder. Your efforts got both more intense and less inhibited. You’ve just experienced an ‘extinction burst’ first hand. Pleasant experience? Feeling full of the milk of human kindness? No? I’m not surprised. Frustration is a horrible emotion and it’s very close to anger/rage. And beings who’re experiencing anger/rage don’t usually make good decisions. Are you SURE that’s what you want your dog to go through?
3) It’s tough on the person
Now you know what your dog is experiencing when you try to ignore a behaviour that has previously worked for him. You know he’s going to try harder. Things are going to escalate and he’s going to be less thoughtful about what he’s doing. His efforts will get stronger, more intense and less inhibited. And if those efforts are focused on getting your interaction, he’s going to be VERY difficult to ignore.
4) The problem gets bigger
Now, depending on your dog, their size and how long they’ve been practicing the behaviour will dictate how much strength and determination they’re likely to put into their efforts. If he’s a small puppy, just experimenting, ignoring something might work just fine. However, if you have a full-grown dog with a history of the nuisance behaviour working well for them, I’d be very impressed if you can ride out the resulting frustrated outburst. There’s a very good chance you’ll give in and respond somehow to their behaviour – even if it’s only in self-defence! And guess what? You have now taught your dog that when in doubt, try harder and the extra effort will pay off. Damn. Bet that wasn’t what you had in mind at all.
What to do instead?
So, what can we do about those attention seeking, nuisance behaviours? If ignoring them doesn’t work or is potentially counter-productive, and we’ve taken punishment out of our training toolbox for ethical reasons, what’s left? How on earth do you stop a dog doing something with reward-based training methods? I mean, we use rewards to strengthen behaviours, don’t we? Yep. And that’s exactly what we’ll do.
The well-kept secret
The secret to getting rid of unwanted behaviour is to get in early, BEFORE it happens, and reward something else instead. This takes a wee bit of planning but it works VERY well in most cases. Keep a note of the situations when unwanted behaviour happens. Note the events that invariably trigger it. After a few days you should have a nice long list 😉 Prepare your tool kit. That means either having toys or food hidden in the location or about your person for quick access. Just as the trigger event happens, pull out your food or toy and reward your dog for NOT doing the nuisance behaviour. (Technically this is known as ‘non-contingent reinforcement’ but you don’t need to worry about labels.)
Let’s look at some examples.
Jumping on you when you get home:
AS your dog charges towards you in a state of ecstatic glee, lower your hand and stuff a handful of tasty food in his face. Don’t feed the whole lot in one go! Let him snuffle and nibble at the food in one hand while you stroke, pat and scratch him using the other hand. And don’t forget the cooing and kisses while you’re feeding.
Why does this work?
Eating usually lowers arousal. That in itself helps the situation. The low presentation of the food redirects his attention to your hands rather than your face. He still gets the lovin’ interaction he wants; you get to avoid being mugged. After a number of repetitions, he’ll start aiming for your hand instead of your face and be generally calmer in his greeting. Over time you can reduce the amount of food you need in your hand until you no longer need it at all; good lovin’ itself will reward the calmer, more polite greeting.
Jumping on Guests
If your guests are willing and able to follow instruction, use the same protocol by arming them with very tasty food before letting your dog loose on them. If your guests aren’t so amenable, have your dog on a leash and feed him yourself. Control his access to people by either using barriers or a leash until he’s calmed down. Alternatively, provide something to occupy him until he’s calmer.
Barking at you
First up, are you sure your dog is JUST barking for attention? If he has biological or emotional needs, make sure those are met. If you’re sure that you’ve built the barking, and it’s directed at YOU, note the situations in which it predictably happens. Now in those situations give your dog something to do instead of barking – BEFORE the barking starts.
As an example, if I have dog barking at their owner in a class I will provide a food stuffed rubber toy or appropriate chew item as soon as the dog enters the training area. In some cases, that’s meant asking the owner to provide the toy and lead the dog into the area with the dog licking the toy! The aim is to get the dog in without barking happening. Once we have that we can work on getting rid of the toy as we progress with training. (Note that barking is a complex problem and will usually need a multifaceted approach to reduce it consistently.)
Why does this work?
If the dog is barking because he has no idea what else to do in a given situation, filling the void with something fun takes away the need to bark. It also avoids the dog associating the location/situation with mindless barking as they don’t get do it. 🙂
Pestering for play
Make sure your dog’s needs for exercise and mental stimulation have been met. It can help if you consistently use some signals to indicate the start and end of a play session. I use “ready?” and the presentation of a toy to start a game and “all done” and a few treats scattered on the floor to end it again.
At the time when your dog usually starts to be a pest (usually through your favourite TV program!), pre-empt him by putting him on a leash, and have him settle next to you with a tasty chew, food stuffed toy or similar. Keep the leash low by either tying it to the chair/couch leg or something heavy. While he’s calmly occupied, praise, stroke or quietly drop the odd extra treat between his feet. If you do this consistently you’ll find that the situation (or theme tune!) that used to signal pestering now comes to cue ‘settle down with a chew and go to sleep’. Once that has happened you’ll no longer need the leash or the attention from you.
Why does this work?
By providing clear signals of WHEN play and interaction with you is available, it’s easier for your dog to learn when it’s not. By setting up the trigger situation so that pestering is difficult, but the wanted behaviour of settling is almost unavoidable, you change your dog’s expectations of what they do in that setting.
In a nutshell, don’t use extinction unless you have a young puppy that’s experimenting to see what happens if they do ‘X’. As long as there is no intrinsic reward for the behaviour, and no extrinsic reward from the environment (i.e. the trash lids falls off and a party is suddenly on offer!) the behaviour will extinguish on its own. If your dog has been doing the unwanted behaviour successfully for a long time, extinction is likely to be painful and unpleasant for all concerned, if it works at all. Use a different approach such as those given in this blog.
Sarah and the Gang
PS Want to read more like this? Find out how here. No spam, no gimmicks, just blog posts and the occasional update about what’s happening at Hotdogs.
PPS Like what you have just read? Don’t forget to share so others can also enjoy it too 🙂