The Dog Training Field Is Currently Divided Into Two Groups - Only One is Humane

1. Aversives and Dominance:

This model is based on falsehoods and fallacious information.

Dog behavior is explained in terms of being problematic for people. In this model the explanation for why dogs do most behaviors is that dominance is the driving force. The dog is an alpha who is trying to take over a leadership role in the family, and therefore does all or most of his behaviors with this end goal in mind. If he can exert enough dominance and aggression over people, he can rule the household. When dogs’ behavior is labeled in this way, people commonly also phrase topics as if they need to be a pack leader, an alpha, a dominant personality who provides structure, boundaries, discipline, and demands respect from dogs. This is done by forcing a dog to comply. Stating “the dog should do it or else.” - “he should do a behavior because I said so” - “the dog needs to learn to respect people” - “he needs to learn right vs. wrong” - “sit means sit” - “he needs to learn his place” - etc.

Dominance theory comes from poorly designed studies on captive wolves from different families that fought and had problems that don’t exist in wild wolves. Their behaviors were described as them living in a dominance hierarchy, and then this was applied to dogs and their behavior (with no actual scientific evidence for this).

Sometimes this dominance-model gets called “dog psychology” - but in reality it ignores the actual dog psychology science that exists.

These types of ideas are really just excuses to use forceful, aversive, harsh punishments on dogs for all the behaviors they do that are considered problems. This includes jerking dogs on collars like choke chains, prong collars, regular flat collars; shocking dogs with remote-electronic collars; hitting and kicking dogs; grabbing dogs by the scruff and pinning them down (“alpha roll”); shaking cans containing pennies in dogs’ faces to scare them; throwing objects at dogs; startling dogs by stomping near them; blasting dogs with citronella scent (which dogs find unpleasant/aversive) or with ultrasonic devices to emit sounds that bother dogs; and all manner of similar things that scare and hurt dogs. “Balanced training” just includes using these types of tools and aversives on dogs for some of the training trials, which is just as bad as using them more often.

This is not training. This is abuse. It’s animal cruelty.

It’s bizarre that this has become normalized in human society. Aren’t we better than this as a species? Don’t we care about dogs enough to treat them better and kinder?

2. Reward-Training:

The other realistic model is one describing dogs’ behavior accurately - they are going to do normal species-specific canine behaviors such as barking, vocalizing/howling, running, moving quickly, digging, chewing objects, desiring to approach and interact with other dogs, mounting/humping, chasing moving things, etc. - and some of these behaviors conflict with what people want dogs to do - or at certain times or directed at certain individuals. Dogs are not wolves. They are not living in dominance hierarchies. They are not trying to assert dominance or be an alpha leader. They don’t view people as other dogs. Dogs are also not moral creatures. They don’t learn right vs. wrong in the sense people do. Dogs learn what is safe or dangerous to them. Dogs do behaviors that work for them.

In this model people know it’s their job as the trainer to provide appropriate motivation and rewards to dogs to modify their behavior. Trainers are educated and learn how to apply animal science concepts to modify behavior including operant conditioning, classical conditioning, social learning, etc. This is true dog psychology (studies done on dogs and other animal species to extrapolate results to humans). It’s a model that strengthens/reinforces dog behaviors with rewards (positive reinforcement), as well as decreasing unwanted behaviors by removing rewards/or the opportunities for them (both of which do no harm to dogs). It involves appropriately starting training with what a dog can currently achieve, and then building upon that in a gradual step-by-step plan, shaping dogs to perform more complex behaviors, and then generalizing those to new situations/environments. It involves using equipment such as soft harnesses that don’t hurt dogs’ necks.

This type of training additionally provides for dogs’ needs - giving them enriching experiences to satisfy species-specific biological necessities, taking care of veterinary/medical issues, providing healthy nutrition, and resolving underlying reasons for dogs’ behavior such as alleviating fears, anxieties, phobias, and stressors like frustration. This type of training provides for dogs’ mental and physical well-being. It involves teaching dogs at their individual pace rather than force compliance regardless of any other factors.

Rather than worry about irrelevant descriptors like providing structure, boundaries, respect, and leadership status, this model asks “what behavior do we want the dog to do?” and then trainers simply train the behavior by providing rewards for dogs performing it and putting it on cue - attaching it through association so the dog learns a signal for when to perform the behaviors, and when there is an opportunity to earn rewards.

Further Comparison:

In the first model, the dog does a behavior a person doesn’t like, and the person uses fear and pain to motivate a dog to not do that behavior. There is no concern for an accurate reason for why a dog is doing the behavior - for example, a dog might be in pain and react towards other dogs on-leash by barking, lunging, and pulling (which can lead to more pain from the leash/collar putting pressure on the dog’s sensitize neck). These people would just add more pain by applying aversives/punishers, which does nothing to resolve the underlying issue of why the dog is performing the behavior in the first place. And this obviously is going to harm the dog. Likewise if a dog is terrified of being left alone (with separation anxiety) and paces and barks in fear - these people recommend shocking the dog to try to eliminate the barking - adding pain to an already scary situation for the dog. Again this does nothing to resolve the underlying fear the dog has, and is actually confirming to the dog that being left alone is terrifying and now scary/painful shocks blast through his body.

This first model intentionally scares and hurts dogs to attempt to change their behavior.

The second model intentionally avoids scaring and hurting dogs, and uses things dogs enjoy to modify their behavior while providing for dogs’ welfare.

Despite claims that people have to train dogs using the first (aversive) model, we can absolutely train dogs using the humane, second (force-free, positive reinforcement) training model. But in order to achieve this, trainers need to be well-versed in animal behavior and training science - in other words, they need to actually know what they are doing - which should be expected of all professionals in their given field. Being uneducated/unskilled is not an excuse for mistreating and abusing dogs in the name of training.