The temperament of your dog and what we are setting out to achieve in regard to your dog’s training and behaviour will dictate the methods and processes we utilise during training and behaviour modification. For dog behavioural cases I assist with, I need to understand the relationship between the dog and its owner(s) as this is usually where most unwanted behaviours stem from. The incorrect relationship between the owner and their dog can contribute to the relationship becoming unbalanced, causing conflict within the relationship and confusion in the dog.
Dogs, just like humans and every other species on the planet, learn what is and is not appropriate behaviour from experiencing consequences paired with behaviours. When we are teaching a dog or modifying behaviours, we use the principles of operant conditioning. There are 4 ‘quadrants of operant conditioning’ and these learning principles are utilised in nature for all life on the planet. Without all 4 of these quadrants no animal can survive. If man could magically remove any one of these 4 quadrants from nature life would instantly fall into chaos and very quickly cease to exist.
The 4 quadrants of operant conditioning are:
- Positive Reinforcement (R+) – Giving the dog something to encourage a behaviour to be repeated, such as praise, treat, toy, play. Anything that the dog considers a positive or pleasurable experience.
- Negative Reinforcement (R-) – Taking something away from the dog to encourage a desired change in behaviour. For example, we can remove something uncomfortable when the dog displays a required behaviour (such as sit) in the form of leash pressure. The reinforcer/reward for a correct behaviour is the discomfort being removed.
- Positive Punishment (P+) – Giving something the dog considers unpleasant paired with a behaviour to discourage the behaviour being repeated. For example, we can give a leash correction paired with the unwanted behaviour.
- Negative Punishment (P-) – Taking something away that the dog desires to discourage a behaviour from being repeated, for example removing our attention when the dog is demanding our attention.
My preference (and the preference of all dog trainers that have a more of a balanced approach to training) is to utilise all 4 quadrants of operant conditioning, as and when required. The majority of training incorporates positive reinforcement, as after all, we are usually teaching a dog behaviours that are desirable. You cannot encourage a desired behaviour unless your dog understands it produces a pleasant or desirable consequence.
Many mistakenly consider punishment as getting angry at a dog and therefore being abusive. This line of thought is far from the truth as anger should never be incorporated into your relationship with your dog or your dog’s training, ever. The term punishment in operant conditioning simply means it’s a consequence that the dog finds unpleasant paired with the current behaviour. An example of positive punishment; say your dog is playing with a rose bush and gets pricked by a thorn, which is unpleasant for the dog, your dog then after being pricked a couple of times very quickly learns to avoid playing with the rose bush. The rose bush did not get angry at the dog, it just produced an unpleasant consequence that the dog chose to avoid.
A dog should always associate any applied discomfort with the current behaviour and not as anger being projected by the owner. When a person gets angry at a dog all they are doing is conditioning the dog to avoid that person causing a fear response and therefore the dog not necessarily understanding it was due to the unwanted behaviour.
Negative Reinforcement is often mistaken as punishment…
Negative reinforcement as outlined above is not punishment, it is rewarding a desired change in behaviour. An example of negative reinforcement; your dog is laying in the sun and starts to get uncomfortable due to getting too hot, so the dog moves to the shade where it is cooler and therefore more comfortable. The dog learns that it’s more comfortable to go lay in the shade. The dogs way of escaping the discomfort was to go to the shade, which is rewarding for the dog.
Emotive Removal of 2 of the operant conditioning quadrants…
Some trainers prefer to use what is generally termed positive-only, purely-positive or force-free training methods. What this means is that they have chosen to remove 2 of the quadrants of operant conditioning from their training methodology. The 2 quadrants they remove are positive punishment (P+) and negative reinforcement (R-). By doing this, they have removed the quadrants that help a dog to very quickly learn what behaviours to avoid. This choice, in my opinion (and by all other balanced trainers), is more emotive than based on any logic or sound scientific reasoning. Refusing to utilise these 2 quadrants in our training if required does nothing more than slow down a dogs ability to learn, as the dog has no idea what behaviours to avoid. In cases where this methodology may work the learning process can be extremely slow, as it requires many more repetitions over a lengthy period to teach the dog. Removing those 2 quadrants also makes it virtually impossible to proof a dogs training around distractions that the dog finds more rewarding than the behaviour we want from our dog. This is explained in the next paragraph.
Trainers that remove those 2 quadrants (P+ & R-) from their training methodology have little understanding of competitive motivators and therefore find it extremely difficult to what we term ‘proof the training’ around distractions. An example of a competitive motivator is when a dog finds rushing over to another dog or person a stronger motivation than say receiving a treat for coming back to the owner when called. Whereas with proofing we can lower the motivation to carry out the undesirable behaviour by applying P+ or R- so that the motivation to please its owner or receiving a treat then becomes the stronger motivator. So, unless we include all 4 quadrants in our training, we have no way to override competitive motivators. As mentioned above, if we could remove those 2 quadrants from nature, all life on the planet would fall instantly into chaos and very quickly cease to exist. How would animals know what dangerous behaviours or situations to avoid?
Understanding Threshold of Discomfort
When we apply what is termed an aversive (P+ or R-), the discomfort level is dictated by the dog and not the trainer. We never apply more pressure or discomfort than required. In other words, we only take the discomfort to a level so that the applied aversive is just at or slightly above the dogs threshold of discomfort. This level can change dependant on the situation and the dogs current arousal levels and emotional state. The aversive is always paired with the dogs current undesirable behaviour. As an example, we have all had the experience of cutting ourself whilst busy focused on an activity – and haven’t realised we have cut ourselves until we notice bleeding. The cut in that instance when it happened was below our current threshold of discomfort and therefore we didn’t feel it. Whereas the same cut, when applied when we are in a more relaxed less stimulated state, causes instant discomfort the moment we cut ourselves because in the moment it was at or above our current threshold of discomfort.
Applying an aversive that is paired with an undesirable behaviour is never carried out with negative emotions such as anger or frustration. It is simply used to condition a dog to avoid an undesirable or dangerous behaviour. In other words, it is the behaviour that triggers the unpleasant consequence from the dogs perspective. Its more a mechanical response by the owner/trainer than an emotive response.
Dog Training Tools & Equipment
I am often asked by dog owners what training equipment I use to train dogs. Firstly, the desired or appropriate equipment is dictated by the dog and the behaviour we are seeking to modify. My go-to tool is generally a slip lead and for most dogs the only equipment required, other than food pouches, toys and praise to positively reinforce desirable behaviours.
For me the deciding factor on what equipment to use is based on what is the most comfortable for the dog. The tool we choose to use should apply zero discomfort unless pressure is applied. This is why I do not use halter-type collars or no-pull harnesses, as these in my opinion, apply continuous levels of discomfort even when the dog’s owner is not applying any pressure to the leash. In other words if there is no pressure applied there should be zero discomfort for the dog from the equipment being used.
The number one priority when helping a dog and its owner work through behavioural issues or obedience is to help get the quickest results with the least amount of stress for dog and owner. Positive reinforcement is of course the most used quadrant, as without positive reinforcement we cannot have a dog that ‘wants’ to carry out a desired behaviour for us.
I do utilise food in much of my training, however, we also should understand that we need to condition our dog to work for ‘us’ rather than triggering a feeding response each time the dog hears a command/cue, sees a particular item such as a leash, or when guided/lured to carry out a particular behavioural response. This of course is not to suggest that we should not use food with our training. The dogs current emotional state and the current behavioural situation we are dealing with will dictate whether it is appropriate or not to use food. My biggest issue with using food as the primary reinforcer when training companion dogs, is that through the process of classical conditioning (conditioning an involuntary response), each time we for example say a command such as “sit” and then immediately reward with food, we condition an involuntary feeding response in the dog. In other words, the dog’s digestive system is involuntarily triggered into action whenever it hears that particular sound or cue. Once we have created this involuntary feeding response in the dog it can be difficult in some cases to wean the dog off the food when we decide not to use food as the primary reinforcer anymore. I also feel in many cases, when we are using food as the primary reinforcer, training can become a lot more mechanical and less focus is placed on building an emotional connection with our dog during training. As the dog’s entire focus is on receiving the treat the trainer’s focus tends to be more about handing out food than building the important emotional bond with the dog during training.
I prefer to use praise and emotional energy when reinforcing or rewarding a desired behaviour. Dogs are very attuned to the energy we project so we can use this to our advantage. By using praise/affection to reinforce a behaviour we are in my opinion, in most cases, helping to develop a stronger emotional bond with our dog during training.
I do realise a lot of trainers (even balanced trainers) prefer to use food as a primary reinforcer, especially since clicker training has become so popular. This of course is a personal choice, and as I have stated previously, I do at times use food if I deem it is the best way to motivate a particular dog. However, my preference is not too. My aim when training companion dogs and maintaining discipline is to condition a calm state of mind in a dog and not a dog that is overly aroused because a feeding response has been triggered by a command or a particular situation.
We also need to understand that dogs are very attuned to our emotional state (and therefore our body language) and we need to be aware of how emotions and body language can have a major influence on our relationship with our dog. Modifying a dog’s behaviour is never just about training the dog. We should have a holistic approach, which means understanding the dog/owner relationship and the dogs environment. In many cases, modifying behaviours can be as simple as being aware of our own emotional state and the body language we are displaying that the dog is picking up on. By changing our own current state we can in many cases also affect our dogs current emotional state and hence behavioural responses.
Mark’s number one priority is first and foremost the welfare of his client’s dog. Secondly, maintaining or improving the relationship the client has with their dog. And thirdly, helping dogs to the best of his ability, to live in a more confident, stable and balanced state by guiding dog owners on how to offer their dog the best relationship possible – whilst also providing their dog with all of its instinctual and behavioural needs.