Training Methods
“Any animal can be trained to do any behaviour using any method given enough time”
Bob Bailey – World renown animal trainer
There are a number of ways to train a dog to do something or stop them doing something. For every solution offered there will be an opinion offered by someone else saying it is wrong and only their way is right. This will quickly be followed by someone else saying both of the first two were idiots. That is the joy of dealing with humans.
The joy of dealing with dogs is that they are just about the most flexible and trainable animal there is. As a consequence, they cause most of the debate by trying to make whatever method you use work whether it is effective or not. Most of the difference in response comes down to how badly we communicate clearly what it is we want and why it would be a good plan for him to work with us.
Of the below methods our club only supports the methods included in the positive, force-free and LIMA philosophies described below. This is because we believe that given the correct application of these techniques good behaviours can be trained and problem ones addressed. Sometimes more aversive methods may have quicker results with less owner effort for some issues but there may be a cost to your relationship with your dog or the creation of new problems. More positive options just require more patience and owner education in some cases. Some members may choose to train their own dogs outside those philosophies and this is not a problem for membership unless the committee considers specific cases abusive to the dog.
There is wide disagreement on what each definition actually is but here is a cheats guide to the different categories of training out there in broad and general terms:
Dominance-based – popularised by the popularity of Cesar Milan in the United States this tends to focus on establishing the owner as the “pack leader” so the dog will naturally defer to them and behave. Often this is characterised by more correction based methods and things like the “alpha roll”. Along with this is a number of habits to establish human pack leadership like eating first, exiting doorways first and rules like no dogs on furniture etc. The main issue is that the dominance hierarchy model we see as fighting to be the alpha now been totally debunked in wolves so this lacks any real scientific grounding. Wild wolf packs do have a hierarchy, but it is based more on family structure than who is the strongest. Dogs also are not at all like wolves in many behaviour areas – domestication is a real thing and it can change behaviour drastically with only small changes in DNA. There is a danger of treating the problem by trying to fix the imagined “dominance” of the dog rather than the specific behaviour cause and effect. This is us taking a simple model and linking every problem to a lack of clear boundaries and leadership.
Corrective based – use of corrections as aversive consequences for not performing a task correctly or bad behaviour. Most of the dog training in the 20 century was based around this model with a wave of trainers leaving the military after the second world war. This could include check/choke chain collars, prong collars or e-collars (shock, vibration, sound or scent electronic collars). This was the basis of modern dog training for the bulk of the 20th century and like most methods it can get results. It tends to be used because it is the easiest to apply and understand rather than the best for the dog and your relationship with it. It may work for many things but if we can do it smarter and with less force should we not try? Also it is hard to know in advance if the dog will understand the “lesson” you intended and instead the aversive may cause other issues including handler directed aggression.
Balanced trainers – use a mix of corrections and rewards in training. Reality is pretty much everyone uses a mix of both anyway no matter how hard they try to be purely positive or the alpha pack leader. Mostly these differ from reward-based trainers in that they do not accept that aversive corrections are unnecessary in some cases and will apply them without apology. These trainers fall on a spectrum of more punishment based to more reward based. Balanced here refers to drawing from all 4 quadrants of the operant conditioning model. The word “balanced” however infers a certain superiority over the others who by definition must be imbalanced. Just because it combines both main approaches (reward based and corrective/punishment based) does not make it better or “the right way”. The individual trainer is deciding what the “correct” balance of reward and punishment training is applied and the necessity for the aversive is their opinion based on their experience rather than a mythical perfect balance.
Positive trainers – do not believe corrections are necessary at all and rewards (often food based but any reinforcement for the dog such as toys or praise is included) should be all that is required. At the most positive end are trainers who are strictly positive and strenuously avoid any action that could fall under the heading of punishing for the dog. This could include a negative tone or removal of your attention.
Force-free training – this is largely positive training but accepts that sometimes words, tone or removal of rewards are punishing for the dog even if not intended to be. So rather get into the semantics of what is purely positive the term force free is used. Any physical aversives like choke/prong/e collars are not part of this training but depending on the trainer may be OK with head halters or front clip harnesses.
LIMA – Least Intrusive, Minimally Aversive is an approach which basically says the least aversive approach should be applied to create the behaviour change desired. it is generally adopted by more positive trainers who consider purely positive/force free styles are a great goal but not always possible in their strict definition. It may be disliked by the balanced camp because those trainers will often want to skip less aversive approaches as they have found a prong / e-collar / choke chain to be the quickest way to affect the change they want for a behaviour problem and have no philosophical issues with them so why not go directly there. Positive trainers may not like it because it leaves open the door to go to positive punishment (like the tools mentioned) if the more positive methods failed. They would believe the failure would have been due to a lack of patience or skill by the trainer and there should never be a need to go to aversive methods.