Table of Contents
The term operant conditioning was developed by behaviourist B.F Skinner who is considered to be the most influential psychologist of the 20th century. Operant conditioning in dog training provides the learning framework to increase or decrease behaviour and works by pairing the behaviour with a consequence.
This consequence may be perceived as either positive or negative to the dog and it’s important to note that the dog decides whether it is positive or negative, not the trainer. This is important because what some trainers classify as positive or neutral, may in fact be perceived as ‘bad’ by the dog. Using electric shock collars as an example, many dog trainers will claim that they do not hurt the dog and will place them on their own arms to prove their point.
However, just because a dog doesn’t communicate or perceive pain in the same way humans do, we shouldn’t assume that we’re doing no harm. Dogs can be in a significant amount of pain and simultaneously convey very little to us about their condition. Furthermore, due to commonly held beliefs, humans tend to empathise more with certain dog breeds in terms of their apparent pain sensitivity or lack thereof.
Operant Conditioning in Dog Training – Consequences
The consequence otherwise known as the reinforcer, utilises what dogs perceive as either good (treat, play) or bad (tight check chain) and pairs that reinforcer with a behaviour. The aim of the exercise is to increase the duration, strength or likelihood of the desired behaviour occurring in the future.
Operant conditioning is at play in the lives of all animals including humans. For example if you go to work and get paid for doing your job well you are likely to go into work the next day and repeat this behaviour, because you’re being rewarded or reinforced. If however you go into work and talk with the person sitting next and as consequence your boss yells at you for not completing the work to the right standard, you are less likely to repeat that behaviour and more likely to complete your daily tasks in the future, in order to avoid the negative emotions you felt as a result of being yelled at.
If you’re anything like me however, I wouldn’t feel great about having been yelled at and I’m likely to have less respect for my boss who perhaps should have found out why I don’t enjoy completing my work. In contrast, the employee who completes their work and gets showered with praise and rewards (bonuses, pay rise) will be much more motivated to complete their work. They might even go out of their way to offer extra desirable behaviours because they know that their good choices will be rewarded.
It’s the same with our dogs, once you have built a history of positive reinforcement (the good stuff) for desirable behaviours, it becomes almost like a game and you will see your dog offering behaviours in the hope of receiving praise or treats.
The Four Quadrants of Operant Conditioning in Dog Training
There are four quadrants of operant conditioning that influence how dogs learn. Many dog training methods use positive reinforcement, negative reinforcement, positive punishment and negative punishment. Force-free dog trainers prefer to use positive reinforcement and occasionally negative punishment, they have been shown to be the most effective and least harmful dog training methods.
Negative reinforcement in dog training – An outdated dog training method whereby a stimulus that the dog considers unpleasant is used to increase the likelihood of a more desirable behaviour occurring. The driving motivation for the dog in carrying out the trainer’s desired behaviour lies in avoiding or preventing the bad thing (aversive) from being inflicted in the future. The dog is motivated by fear and anxiety which can negatively impact the relationship between the trainer and the dog.
Examples of Negative Reinforcement –
*An everyday example that you may be able to relate to involves getting up early to AVOID being awoken by your alarm clock (aversive) or removing the aversive i.e turning off the alarm clock using the snooze button to STOP the alarm and remove the annoying stimulus.
*A dog training example which is sadly all too common in obedience clubs – using a check chain or prong collar to stop the dog from pulling. When the dog pulls, the collar tightens around the dog’s neck and when the dog slows down and comes back into heel position at your side, the chain is loosened by the handler. Therefore the dog learns that in order to loosen the tight chain he must return to heel position, thereby increasing the frequency of walking in heel position, to prevent the ‘bad thing’ that is the unpleasant and tight chain/prong collar feeling around his neck.
Why Negative Reinforcement can fail
This training method does not develop a dog who feels empowered to be involved in the learning process (an intrinsic learner) and will not nurture a creative puppy that offers behaviours. Ultimately the dog has to be bad to be good in this situation, that is when the dog pulls, he is checked back into position and some dog trainers will even reward him at this point.
For example, they will give the dog a reward when they return back into the heel position after giving into the pressure of the lead tightening thus they are combining negative reinforcement with positive reinforcement. This combination will develop a behaviour chain whereby the dog learns that when he pulls he ultimately gets rewarded i.e he has to be bad to be good.
Operant Conditioning in Dog Training: Positive reinforcement – The dog is given a reward (reinforcer – treat, play, toy etc) to increase the likelihood of a desirable behaviour occurring in the future. For example when training a dog for loose leash walking, while in the heel position, we reward heavily. The dog begins to associate being in the heel position with ‘good’ things, this increases the likelihood of the behaviour of walking next to us occurring in the future.
It is important to note that when using positive reinforcement, the dog does not have any negative consequence applied when they do not perform the desired behaviour. Wrong choices are simply not rewarded and we control the environment to increase the likelihood of good choices occurring, thus increasing the rate of reinforcement for desirable behaviours. If the dog strays and starts to pull on the leash, you have raised your criteria too quickly and you should practise the exercise in a low-distraction environment such as inside the house.
When your dog is reliably offering the behaviour in your home, slowly increase the distraction level by working on the exercise in your backyard, front yard and so on. This is called proofing the behaviour. In the early stages of training any dog behaviour we should always endeavour to set our dogs up for success. When my dog has a good grasp on the behaviour and has generalised the behaviour to new environments, I will test for failure.
Failure is not a bad thing in any learning process, providing a solid history of reinforcement exists for the behaviour you are training. Failure is an integral part of learning, success is not built on success, it’s built on failure but we need to maintain our enthusiasm for success and the process of building a history of positive reinforcement in controlled environments helps with this.
Punishment in Dog Training
Operant Conditioning in Dog Training: Punishment – The use of punishment in dog training undermines the trust in the relationship and is not conducive to learning new behaviours. In fact, the phenomenon of classical conditioning can have the undesired effect of you being perceived as ‘bad’ in the eyes of your dog.
- Positive punishment – The worst form of punishment involves adding bad (aversive) stimuli (consequences) to decrease a behaviour for example hitting your dog when they get something wrong. I don’t think this needs further explanation, this method has no place whatsoever in a dog trainer’s toolbox.
- Resource guarding training example – Your dog growls whenever you walk past its food bowl. In this instance, you would be better off counter-conditioning the dog rather than punishing them. Basically showering the dog with treats when they have their food bowl and they are quiet and eventually, the dog will associate our presence and the food bowl as a predictor of good things and not something to be feared or reason to be on guard. If the dog growls at you, you are too close and need to increase the distance between you and your dog for this exercise. Slowly you can get closer but it may take days, possibly weeks depending on the dog. You must read the dog’s body language for signs of a stress response otherwise known as going over their threshold.
This example highlights how damaging the application of positive punishment can be in the context of training dogs. Dog trainers that still use positive punishment, will often be applying fear and intimidation to a dog that is already stressed. This is detrimental to the well-being of the dog and isn’t a learning environment in which we can train the behaviours we want. Similarly, punishing a dog for communicating when it is uncomfortable can increase the risk that the dog will bite without warning in the future. Such an environment is clearly destructive and has been shown to increase the likelihood of a dog developing avoidance behaviours and aggression.
- Negative punishment – In dog training negative punishment involves removing something the dog or puppy wants. This could be denying a dog your attention when they bark or taking away their long-lasting chew or toy. The dog barks expecting your attention (reward) and you take that away by leaving the room, eventually, the dog will stop this behaviour as it becomes inherently punishing.
The most ethical dog training methods, utilise positive reinforcement and might occasionally use negative punishment. However, when resorting to the use of negative punishment it’s important to first consider what the dog’s needs are and if the dog’s needs have been met. For example, it’s a hot day and your dog is barking at you. The dog might be thirsty, in which case water should be provided and negative punishment would be inappropriate.
Summaries of the four quadrants of Operant Conditioning in Dog Training with examples
- Positive reinforcement – adding rewards for behaviour that we’d like to see more of in the future. For example – your dog sits and you give them a treat. If the dog enjoys the tasty treat and associated it with sitting, they would be likely to repeat this behaviour in the future.
- Negative Reinforcement – removing a stimulus that the dog finds unpleasant, with the aim of increasing the likelihood of a behaviour occurring in the future. For example – the check chain or prong collar tightens when the dog pulls, so the dog returns to the heel position by your side. The dog would theoretically be more likely to walk next to you in the future.
- Positive Punishment – adding something the dog perceives as punishing, in order to decrease the likelihood of a behaviour occurring in the future. For example – applying an electric shock collar and shocking the dog when they bark. If the dog found this sufficiently unpleasant or punishing, it would be less likely to do this behaviour in the future for fear of the same thing happening again.
- Negative Punishment – removing something the dog wants in order to decrease the likelihood of a behaviour occurring. For example – the dog barks and you turn your back to them, so the dog learns that barking caused something it wanted (your attention) to go away and would possibly be less likely to do that behaviour in the future.
Operant Conditioning in Dog Training – Conclusion
Of the four operant conditioning quadrants, the most useful and beneficial to dog trainers are positive reinforcement and negative punishment. At the heart of my dog training philosophy is building a relationship that is based on trust. We can successfully train our dogs using solely positive reinforcement (rewards only, no bad stuff). My dogs are a testament to this and for those that argue you can only use positive methods with ‘lap’ dogs, my Dobermann thrives on force-free training. This video shows Kaiser and me working as a team. Note his willingness to engage with me and that he chooses to participate in the obedience exercise despite being free to roam.
Sadly, there are many dogs that struggle with behavioural issues that are made worse by dog trainers and owners who apply negative reinforcement and positive punishment. No dog, let alone any living thing, should ever have their freedom of choice withheld or feel pressured into doing behaviours that for whatever reason they don’t feel empowered to do. Humans are meant to have the larger brain of the two species and yet so many trainers fail to be creative, resorting to quick fixes and blaming the poor dog for being dominant, instead of asking, does the dog actually know what behaviour we would like them to do?
Let’s give our dogs a chance to learn appropriate behaviours and if they get it wrong, let’s use this valuable feedback to revise our training plan. Operant conditioning in dog training is not complicated when we focus on using reward-based methods. It gets complicated when trainers introduce pressure and force because adding in stress, fear and anxiety is not conducive to learning or the well-being of your animal.
In effect through the use of aversive training methods, learning decays and this will ultimately lead to problematic behaviours which will impact the welfare of the dog. Let’s empower our dogs to make their own choices firstly in controlled environments. By reinforcing (rewarding) those behaviours that we wish to see more of, those behaviours can then be generalised to new environments. An approach that utilises the positive reinforcement quadrant of operant conditioning in dog training creates a dog who blossoms in the ‘game’ of learning.