I am often asked by dog owners and prospective clients what methods and tools I use when training dogs, or helping to modify unwanted behaviours. This article will provide you with most of the answers.
Your dog dictates the methods and process we utilise during training and behaviour modification. Not an ideology or methodology. In most cases, we need to work through the relationship between the dog and its owners, as this is usually where most unwanted behaviours stem from. This is due to the relationship becoming unbalanced, causing conflict in the relationship, and therefore confusion in the dog.
Dogs just like humans and every other species on the planet learn from consequences. When we are teaching a dog or modifying behaviours, we use the principles of operant conditioning. Mother nature utilises all 4 quadrants of operant conditioning for all life on the planet. Without all 4 quadrants, no animal could have survived. If man could magically remove any one of the 4 quadrants from nature, life would instantly fall into chaos, and all life would very quickly cease to exist.
The 4 quadrants of operant conditioning.
- Positive Reinforcement (R+) - Giving the dog something to encourage a behaviour to be repeated, such as praise, treat, toy, play
- Negative Reinforcement (R-) - Taking something away from the dog to encourage a behaviour to be repeated. For example, we can remove something uncomfortable when the dog displays the required behaviour, such as leash pressure
- Positive Punishment (P+) - Giving something to discourage a behaviour being repeated. For example, we can give a leash correction paired with the unwanted behaviour
- Negative Punishment (P-) - Taking something away to discourage a behaviour, for example removing attention when the dog displays unwanted behaviour
My preference (and the preference of all dog trainers that have a balanced approach to training) is to utilise all 4 quadrants, as required. The majority of training incorporates positive reinforcement, after all, we are usually teaching a dog behaviours that are desirable. You cannot encourage desired behaviours unless your dog understands it produces a pleasant or desirable consequence.
Many mistake punishment as getting angry at your dog, and therefore considered abusive. Anger is never incorporated into your dogs training. Anger has no place in training or even owning a dog. The term punishment in operant conditioning simply means its 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 creates some discomfort for the dog, your dog then after being pricked a couple of times very quickly learns to avoid playing with the rose bush. 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 lay in the shade. The dog should always associate the discomfort with the behaviour, and not as anger being projected by the owner. When we get angry at a dog, all we are doing is conditioning the dog to avoid us, and not necessarily the unwanted behaviour.
The 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. 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 based than based on any logic or sound reasoning. Removing these 2 quadrants does nothing more than slow down your dogs learning ability extremely, as the dog now has no idea what behaviours to avoid. In cases where this methodology may work, the process is extremely slow, as it requires many more repetitions to teach the dog. Removing those quadrants also makes it virtually impossible to proof training around distractions that the dog finds more rewarding.
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. Whereas with proofing we 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 is then the stronger motivator. So unless we include all 4 quadrants in our training, we have no way to override competitive motivators. And as suggested above, if we could remove the 2 quadrants these groups refuse to use in training from nature, all life on the planet would fall instantly into chaos, and therefore very quickly cease to exist. As how would animals know what behaviours or situations to avoid? They wouldn't.
Thresholds - discomfort and pain
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 that the aversive applied is just above the dogs threshold of discomfort. This level can change dependant on the situation and the dogs current arousal and emotional state, and 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 our self until we notice the cut bleeding. The cut in that instance was below our threshold of discomfort, and therefore we didn't feel it. Whereas the same cut when we are in a more relaxed state, causes instant pain the moment we cut our self, because it was above our threshold of discomfort in that moment.
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, its the behaviour that triggers the unpleasant consequence from the dogs perspective.
Dog Training Tools
I am asked by many people what tools I use to train dogs with. Firstly the tool is dictated by the dog and the abilities of the owner. My go-to tool is generally a slip lead, for the majority of dogs I train, and for the majority of dogs the only tool required.
For me, the deciding factor on what tool to use is based on what is the most comfortable for the dog. So 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 dogs owner is not applying any pressure. So if there is no pressure applied, there should be zero discomfort for the dog.
My 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 it we cannot have a dog that 'wants' to carry out a behaviour.
Positive Reinforcement methods
I tend not to use a lot of food in my training, as I prefer to condition a dog to work for its owner than have a feeding response triggered each time the dog hears a command, sees a particular item such as a collar and leash, or guided to carry out a particular behaviour. That is not to suggest I never use food. The dog and situation we are dealing with dictates whether I include food in my training or not. My biggest issue with using food as the primary reinforcer when training companion dogs, is that through the process of classical conditioning, each time we for example say a command such as "sit" then immediately reward with food, we actually condition an involuntary feeding response in the dog. In other words the dogs digestive system is involuntarily triggered into action whenever it hears that particular sound. Once we have created this involuntary feeding response in the dog, then it takes a lot of work to wean the dog off food, when we decide not to use food any more. I also feel in most cases, when we are using food as the primary reinforcer, training becomes a lot more mechanical and less emotional. The dogs entire focus is on receiving the treat, and the trainers focus is about handing over food, and therefore emotion tends to becomes less of an influence on our training.
I do prefer to use praise and emotional energy when reinforcing or rewarding a desired behaviour. All you need to do is look at the expression on Bosco's face in the photo at the top of this article, when I told him to stay in position to take the photo. He is not even thinking of food, nor has my command triggered a feeding response. 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 personal opinion helping to develop a stronger 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 own emotional state and the types of energy we are projecting, and therefore these have a major influence on our relationship with our dog. Therefore, modifying a dogs behaviour is never just about working with the dog. We need more of a holistic approach, which means understanding the dog/owner relationship, and the dogs environment, if we are to be successful. In many cases, modifying certain behaviours can be as simple as being aware of our own emotional state and the body language we are displaying.
This article is not completed. I will be adding a lot more information shortly.