The following is a manifesto written by one of my good friends who is one of the most talented trainers in the world (literally, she travels the world and teaches and is a very popular sensation on YouTube).
This document is well thought out and worthy of a post on this site. It might be a bit advanced for those who are new to training, so I’ll try to provide a brief summary.
Summary: Emily Larlham is proposing a new vocabulary term for training that is very clearly positive-reinforcement only. She provides the details of what this is and is not. She proposes the new term because “positive reinforcement” has all but lost meaning due to overuse and misuse.
The type of training she practices (and that I always hope I am practicing) is what my mother described as this “Your dog receives either neutral (at worst), a reward of some type, or a jackpot for excellent behavior. Never a punishment.” It’s close enough, though ignoring unwanted behavior is “punishment” in the technical sense, but that’s getting into the nitty gritty.
I don’t care what it is called, but it, in my opinion (and unless science or some other very persuasive theory comes along to blast this one out of the water) is what I, personally, consider to be the highest level of training there is.
In our circle of training friends, we really don’t consider it ‘training’ unless you are challenging yourself at this level. Intimidation isn’t impressive: Anyone can do that. This tight set of constraints in training forces a trainer to be creativeknowledgeable, and skilled and is the ultimate in showing respect to an animal.
Because, in order to employ this type of training, you have to know what a specific animal likes/dislikes, fears, is annoyed by, what motivates the animal–experience and studying both play a part for a good trainer to either begin a new behavior or change an existing one.
“Do no harm” is the minimum requirement; “Help this animal to feel secure, even tempered, and confident” while learning is the ultimate goal. It’s the healthiest relationship we can have with animals.
Proud to call myself a Progressive Reinforcement Trainer.
Here is the manifesto in full from her website:
Progressive Reinforcement Training Manifesto
By Emily Larlham
The Need for a New Term:
A type of animal training exists that involves no forms of intimidation, confrontation, violence, reprimands, or domination.
This non-violent type of training has gone under many names: “Clicker Training,” “Positive Training,” “Positive Reinforcement Training,” and “Reward Training,” among others. There is a need for a more specific, more accurate, more inspirational term. The above terms have been used so loosely in recent years that they have lost their original meanings. How has this happened? Trainers who use compulsion methods may incorporate a clicker (a noise maker to mark desirable behavior) and refer to themselves as a “Clicker Trainers.” Trainers who use painful or intimidating methods may include food or toy rewards in their training and refer to themselves as “Reward Trainers” or “Positive Reinforcement Trainers.” It is already possible that a member of the public may seek the guidance of a trainer who claims to be “Positive,” only to find out that this trainer routinely uses physical violence towards animals.
I propose a new term that trainers and members of the general public can use to refer to this type of modern training – a training system that is not only humane, compassionate, and reliable, but is also based on the latest scientific studies. Because this form of training constantly incorporates the latest and most reliable scientific findings, and because it furthers an evolutionary progress toward a more harmonious relationship between humans and the animals who live with them, it shall be referred to as Progressive Reinforcement Training.
Progressive Reinforcement Training essentially means teaching animals by rewarding desired behaviors and excluding the intentional use of physical or psychological intimidation.
Progressive Reinforcement Training means:
1) Training by rewarding desirable behaviors so they will be more likely to occur in the future, while preventing reinforcement of behaviors that are undesirable.
An example: Letting a dog walk forwards while the leash remains loose to sniff a bush as a reward for not pulling, while not letting the dog reach the bush if the leash becomes tight (so that pulling on leash is never rewarded).
Another example: If you are training a dog to greet guests politely, you first reinforce the dog for calmly keeping all four feet on the floor (not jumping) in exciting situations, and then when the dog does jump up, you remove your attention briefly (by turning away from the dog- as attention is rewarding). However, if you simply tried to train a dog not to jump up by turning away from the dog repeatedly without rewarding him for the correct choices – the dog could become frustrated. It is true that if the dog figures out that the jumping is not getting attention, the dog will try an alternate behavior – however, a dog will more likely try jumping higher, barking, whining, and nipping over standing still or sitting for attention. By rewarding the dog for what you want him to do first, you give the dog a default behavior to try when what he is doing is not working.
Examples of Rewards:
Food, toys, attention, people, other animals, running, sniffing, swimming, going outside, coming inside etc.
Keep in mind the animal chooses what is rewarding, not the trainer. This means that if you give your dog a treat for sitting, and then ask him to sit again and he doesn’t sit, it’s very likely that the dog does not find the treat rewarding. Other things to keep in mind are that rewards will not be effective if the animal is full, or the animal is stressed.
2) Interrupting and preventing undesirable behaviors without physical or psychological intimidation, as well as rewarding an alternate response (training a behavior you find desirable in it’s place).
An example: If you want to train a dog not to lie on your couch, you train the dog to do what you want him to do first. That is, you train him to go and lie on his dog-bed. Then when he does try to go on the couch, you interrupt him and redirect him to the appropriate location (his dog bed) so that climbing onto the couch remains unreinforced. During the training process you, also use management and prevention: while you are away from the house, you block the dog’s access to the couch, as he would likely choose to lie on the couch – and be reinforced for it – in your absence.
You can interrupt an animal’s undesirable behavior so that he is not self-rewarded without using physical or mental intimidation. To do this, you can train the animal to respond to an attention cue or a recall: something that means, “stop what you are doing and look at me”, or “stop what you are doing and come here immediately”.
A very basic training plan for training an attention noise to interrupt behavior:
First you can make the noise that you want the animal to respond to (a whistle, or a kissy noise) and then feed a treat. Repeat this until the animal is expectant of a treat after the noise. Next make the noise while the animal is looking away from you and AS the animal turns to look at you (for the treat) mark that behavior with either a click (using a clicker) or by saying “yes”. Once you have repeated this step you can then add distractions. Have the animal on a leash so he cannot reach the distraction (perhaps a low value piece of food on the ground)- make the attention noise, and click or say “yes” and then feed a treat if the animal turns towards you after hearing the noise. If the animal does not turn towards you, do not click or say “yes”. The animal should not be allowed to reach the distraction that it is interested in. You can take a step backwards from the distraction to make it easier so the animal can succeed. You can condition this attention noise or a recall to muscle memory in the same way a driver responds to a green light traffic signal (green light means go!). Once you have created many different scenarios where your animal can disengage in what he is interested in to come towards you and look at you, you can start using the sound to interrupt behaviors that you find undesirable.
Keep in mind that if you ignore the animal and only pay attention to him when he is doing undesirable behavior, you will be training the animal to do exactly that which you do not want by providing your attention whenever the behavior occurs. So the GOAL is to reward the animals alternate responses to the same situations in conjunction with interrupting and preventing the undesirable behaviors.
Example: If your dog steals your underwear and runs around the house with them to get your attention, you have got to reinforce your dog with your attention when he is calm and doing NOTHING. When your dog is lying at your feet quietly, that is when you will reinforce him with MORE attention than when he runs off with your underwear.
3) Taking an animal’s emotional state and stress levels into account.
Trainers practicing Progressive Reinforcement read an animal’s body language to the best of their ability for signs of stress or arousal and adjust their training approach accordingly.
Example: Removing a dog that is offering stress signals from a situation where a child is chasing or pestering the dog.
4) Socializing and teaching an animal to cope with his environment using reinforcement.
You can use Progressive Reinforcement Training to socialize and teach an animal to cope with his environment by letting him experience low or non-stressful situations in which the animal is likely to succeed and earn rewards for desirable behavior. You can then increase difficulty and distractions as the animal succeeds, with the goal of creating a confident well-adjusted animal.
An example: Teaching an animal to be relaxed and calm while being handled or restrained by using reinforcement. Pavlov’s dog was trained to have a new emotional response to a bell because the sound of a bell was followed by food. You can train your dog to enjoy handling, very simply put, by touching the dog and then feeding the dog a treat, and increase the invasiveness as the dog remains unstressed by the situation. If the dog were to shy away, the trainer would have to go back a step to where the dog was comfortable (Classical Conditioning).
Another example: Feeding a dog a reward for remaining relaxed and calm around an exciting situation (perhaps a road with loud traffic), first from a distance and then as the dog succeeds from closer and closer. If the dog were to become too excited or stressed, the trainer could go back a step in the training process until the dog was successful.
5) Using a marker to train, whether it be a clicker, some other noise-maker, your voice or touch, or a visual marker. Or, on the other hand, not using a marker, and instead for example reinforcing an animal by feeding a treat directly to his mouth.A marker can be used to pinpoint behavior. It tells an animal that what he is doing at that exact moment in time will win him reinforcement.
For example: If a dog sits, the trainer can click as the dog is sitting, and then feed the dog a treat. Or the trainer can say, “Yes!” in a positive tone as the dog is sitting and then feed the dog a treat or release the dog to get a toy or go out the door.
Reinforcing behavior is also possible without using a marker. For example, you can feed a dog a treat for looking at another dog to change his emotional response to the other dog (Classical Conditioning). You can also reinforce your dog for calmly lying around the house or outside by tossing him a treat between his paws while he is not expecting the treat and he will be more likely to repeat the behavior in the future.
6) Employing humane, effective, respectful training based on the latest scientific evidence.
A commitment to Progressive Reinforcement Training means strictly following all of the above principles – not just in training sessions, but during 100% of the time spent with an animal.
Progressive Reinforcement Training does not mean:
1) The intentional use of physical or psychological intimidation.Using your voice, touch, body language, a device, or the environment to intimidate an animal for the purpose of continuing, initiating or ending the animal’s behavior.
Examples: staring at an animal, intentionally leaning over him, poking, jerking, shocking, squirting with water, startling with a noise, or using your voice in an intimidating way to suppress behavior (saying “no” or “eh!”).
2) Intentionally disregarding an animal’s stress levels or signals.
Intentionally putting an animal in overly stressful situations in which he cannot cope, rather than exposing the animal in a way that he is under his threshold (the animal can make choices and cope).
Example: Forcing an animal to meet a stranger while the animal is offering a wide range of stress and avoidance signals.
Example: Dragging an animal across a surface he is frightened of and refuses to cross, instead of teaching the animal to feel confident and calm crossing the surface using Counter Conditioning (rewarding the animal for choosing to take steps across the floor until the animal is confident to cross calmly on his own)
3) Holding selfish or uncompassionate goals for your training.
Intentionally putting an animal at risk for physical or emotional damage to satisfy ones own interests.
A commitment to Progressive Reinforcement means never intentionally using the intimidatory tactics above – never in training sessions, and never during any other time spent with an animal.
Why refrain from using Physical or Psychological Intimidation?
For scientific, moral, and ethical reasons. Using these forms of conditioning can produce unwanted side effects in addition to the basic trauma they do to an animal.
The many problems with using physical or psychological intimidation:
1) Without perfect timing, intensity, and consistency, the “training” amounts to nothing more than abuse.
2) The animal learns to avoid the punisher in order to indulge in undesirable behavior.
3) These techniques can cause irreversible emotional damage to the animal.
4) The punishment can increase stress hormones, arousal, and aggression.
5) Animals can habituate to the punishment – meaning that the intensity of the punishment must keep increasing to have any effect as the animal learns to endure it.
6) You cannot change an animal’s basic emotional response to find children, adults, or other animals (or anything for that matter) reinforcing by using intimidation; you can only suppress the dog’s punished behaviors.
7) Intimidation can cause dogs to hide their warning signs before attempting to bite.
8) Dogs trained with punishment can feel trapped by their handlers, since the decision to leave a ‘stay’ or to leave the handler’s side (to escape from a bothersome child, for example) can cause punishment. Animals who feel they have no escape tend to bite rather than move away.
9) Intended intimidation can actually increase the behavior you wish to extinguish, as intimidation involves giving a form of attention to an animal.
10) The presence of the punisher becomes less reinforcing for the animal. If you punish your dog using intimidation, it is harder to compete with the reinforcement value of other things in the environment. Your dog will find other stimuli in the environment more reinforcing than you as the dog increasingly associates you with punishment rather than reward.
11) Dogs who have been trained with physical or psychological intimidation do not offer behaviors on their own as readily when asked, making complex behaviors difficult to train
12) Handlers who use intimidation as punishment will punish their animals more readily in the future as punishment is rewarding to the handlers themselves (they get the result they wanted- hitting a dog made it stop barking, so they will be more likely to hit the dog in the future). In other words, using physical or psychological intimidation causes one’s own behavior patterns to change.
In conclusion, Progressive Reinforcement Training is not a permissive form of training. It requires providing consequences to all behaviors. The trainer takes on the role of a benevolent leader and guide using these ethical and scientifically based methods.