Prep for a Gary Wilkes Seminar…cont’d.

“The most sane voice in the dog training world is Gary Wilkes. His life experiences have qualified him from a unique perspective to point what is right and wrong in the dog training world. The man that brought clicker training (positive) to the dog world is also an expert and proponent of well-time, safe and effective corrections.”
Mark Fulmer: Owner, breeder, master trainer: Sarah Setter Kennels, Aiken, SC

To get you ready for one of my seminars, there are some simple things that can dramatically improve your experience. They are simple and require very little time. I have included links to blog posts and videos that you can examine in your spare time. I offer two types of information – basic and advanced. The basic stuff covers foundational knowledge that will get you up to speed, quickly. The advanced stuff won’t likely make sense unless you have experience that examines the topic in a manner similar to mine. Don’t be bothered if some of the information doesn’t make sense. It will.

Take a look at this blog post. It lays out the basics of pairing the clicker to treats and introduces targeting as a fundamental skill. If you and your dog are already beyond this, so much the better. https://clickandtreat.com/wordpress/?p=1972 This video shows a short session using targeting with a dog that had barely paired the clicker with treats – and the handler is a restaurant owner and former client, not a trainer. https://www.youtube.com/watch?v=y2C7kP4Fo2g

Lesson 1: Building Duration and an ‘End of Behavior’ Signal
Effective training requires that you sometimes limit the amount of information the dog considers as important. This is the opposite of generalization. Instead of getting the dog to ignore a great part of the environment, we want to get the dog to consider only a select amount or type of information. This requires a dedicated signal that limits the dog’s focus by teaching a unique format, including a clear ‘end of behavior’ signal. That is the clicker’s purpose. While this simple statement still draws argument, it is inarguable that unless you have a clear end to the information the dog is left in learning mode and may pick up signals that are not productive. Try this experiment. Get your dog to lie down. Use the clicker to end the behavior and deliver a treat. Do about ten repetitions. The amount of time the dog holds the down isn’t important. Try to get about five seconds.

(If your dog is beyond this point in its training, just click at three seconds and then toss a treat on the ground to get the dog up. That resets the behavior. As soon as you can, stop saying “down”. The dog will offer the behavior as the one most recently reinforced.)

Now try this – once the dog is down, withhold the click for a few extra seconds. If you are working at three seconds, try to stretch it to five. If the dog breaks before eight seconds, say ‘wrong’ in a normal tone of voice (or oops or uh-uh or whatever you choose) and don’t give a treat. If the dog does remain down, click and toss a treat to get it back up again.

Once you have your eight seconds, start mixing up the length of time. Arbitrarily end the behavior at a randomized time. Do not consistently increase the time, such as five seconds, ten second, 20 seconds. That will teach the dog that the task is only going to get harder and longer. Ping-pong the time this way – 5 sec. – 2 sec. -12 sec. –4 sec. You are trying to teach a subtlety that means “the play is still going”. This is the equivalent of sports players, ‘playing to the whistle’. You are teaching the dog that they should continue to pay attention until the clicker sounds. That is the signal that ends the behavior. To use a different metaphor, you are trying to put parentheses around the learning event. That limits the amount of information the dog has to filter to make correct associations. Tangentially, it succeeds in limiting the environment in the same way that behavior analysts used tiny boxes to keep the world out. That is totally unnecessary because teaching the animal to focus in a robust environment isn’t difficult.

Lesson 2: Things may not be what they seem –
expect the unexpected.
To the dog, learning appears to be a very random process. Each successive repetition is different from the last. You are asking the dog to step off into the unknown. This causes low-level stress for a naïve dog. To combat this, modern styles of training focus on never making the dog confused or frustrated. This is done by teaching in tiny increments that may actually perpetuate the dog’s stress. There are two reasons why this is a short-sighted approach. First, not all behaviors can be taught in a tiny increment style. Some of them require that the dog give extra effort to achieve the new standard. A dog that is never challenged in training may never achieve its potential. Second, some behaviors are complex – like precise scent discrimination. There are going to be more mistakes than successes in the path to skill. Teaching a dog to persist in the face of difficulty will insure that more complex learning does not cause the dog to shut down.

Expect the Unexpected:
1) Ask the dog to do a simple task. Click the clicker. Give a one-click, one-treat outcome for correct responses. Repeat about ten times.
2) On repetition 11, click the clicker and don’t give a treat. Watch closely to repetition 12. Is it slightly different than 11? The dog can’t possibly know why it didn’t get a treat. This will trigger a subtle change in the dog’s attitude.
3) Now, on repetition 13, click and give the dog four or five times the normal amount of treats, be very vocal and add some vigorous physical affection and animated praise. Watch closely what happens on the next repetition.
4) On the next repetition, no matter what the dog does, say “wrong” and give no treat.
5) Now go back to your straight expectations. Every correct response gets one-click, one-treat.

Yes, this will cause your dog’s behavior to wiggle. I would recommend you don’t do it with behaviors that are part of a precise obedience routine. This is an introduction to a process that eventually teaches perseverance in the face of difficulty. If it is used while teaching a new behavior, the wiggle actually fosters faster learning that formulas that require baby-steps. If you find this daunting, borrow a dog that is not a competition dog. This exercise is as much for you as it is for the dog. J

Reactive Dog Work: Halter to the Rescue
At the seminar I will be teaching how to control reactivity using a head-halter. If you have never used one, or never used one for that purpose, it will help to at least familiarize yourself with the tool. My halter of preference is a Gentle Leader. It is the only halter that is convertible – you can use it as a halter or slip the nose loop down and it’s a regular buckle collar. Take a look at these two videos. The first is with a young Husky whose owner is over 70. As you see at the start, she is typical – highly interested in other dogs and pulling her owner toward them.
The second video shows how to fit a Gentle Leader and includes its use with a large Cocker Mix that was the typical reactive dog. He would stand still until the other dog got very close and then he would explode. I work with this a great deal.

Why a halter?

For several thousand years, humans have controlled strong animals by isolating the muscles that connect the neck to the head. This blocks the animal’s ability to thrust forward with all its strength. Once the animal is conditioned to the head-control, a child can walk at 1,000 pound horse with ease. This same process works on dogs, only better. With dogs we get the same mechanical advantage we get with horses. However, dogs are highly visual animals whose instinctive repertoire requires them to approach threats. We see this whenever we observe a reactive dog. They see the target and automatically move to engage it. Once aroused, the senses of smell and touch are submerged and vision becomes the master. A fully aroused dog will pull forward on a buckle collar, choke chain or prong. They do not ‘feel’ the pressure on their throat as a limitation on behavior.

One of my first clients was a vet whose German Shepherd had impaled herself by lunging through the spikes of a prong collar. A halter prevents the lunge in a very elegant way – by controlling the dog’s vision. As vision seizes the dog’s brain, the halter easily pulls the dog’s eyes away from the target. They hate that. It is more important for them to see the target than it is to actually contact and engage the target. You can prove this with any reactive dog. Get a piece of cardboard that can block the dog’s vision. Put it between the dog and the target. Watch the dog instantly attempt to move so that it can see the other dog. The point is this – the dog will change its behavior almost frantically to keep the other animal in sight. That’s where the halter comes in. (And the proof that this is true) A halter, used correctly, teaches a dog to back up. That is because when the dog lunges forward it can no longer see the target. It must keep the leash slack in order to keep seeing the other dog. That is exactly what they do. Added to the lesser strength necessary to turn the dog’s head, this is an invaluable tool for actually teaching a reactive dog to be passive. The behavioral effects involved are depending on the dog’s instinct and correctly applied negative reinforcement. https://www.youtube.com/watch?v=d7oVZm_2jNg

The “P” Word: Punishment and behavioral control
The last 70 years, social scientists, behavior analysts and now, modern dog trainers, have attempted to foster a false dichotomy of ‘reinforcement vs. punishment.’ We are now at a stage where the word punishment is conflated with abuse. That is not an accident. One of the best ways to shut down logical discussion is to associate a word or topic with bad things – even if there is no real connection. In the case of punishment, if it means ‘abuse’, no one will speak in favor of punishment – because that would advocate abuse. You see this use of the word virtually everywhere. Not here. Not at my seminars. I do not allow the distortion of language to demean others and squelch logical discussion. That is devious and destructive ideology at work.

What punishment means:
There are two basic meanings for the word punishment; retribution for an antisocial or dangerous act and the use of something aversive to cause a behavior to decline or stop. The latter definition was crafted by the people who appropriated the word ‘reinforcement’ to describe a behavioral effect. They created a simple, logical polarity between reinforcement and punishment – and then proceeded to distort the words to mean ‘good vs. evil’. I use the logical definition. I do not assume that punishment causes trauma or damage. I assume it describes a natural behavioral effect that causes the suppression, weakening or absence of a behavior. Period. Abuse is abuse. Punishment used correctly benefits the human or animal. Without our sensitivity to aversive things, we could not survive. Labeling a necessary survival ability as innately damaging is stupid, deluded or a devious and dangerous misuse of language.

A basic example of punishment is the effect of a thorn or cactus needle that controls your behavior immediately and leaves a lasting effect. It is perhaps the best example because it tells you the function of the effect and some of its great advantages. Here they are…
1) You only touch a cactus needle once. You will adapt your behavior for your life-time from that one experience.
2) A cactus needle does not cause damage beyond the immediate. At worst, you shed a drop of blood. It quickly heels. This is far less invasive than a diabetic who must jab themselves with a needle, daily, to find out their blood sugar levels. To complain about shedding blood as a general rule is stupid.
3) You will develop a mild caution that allows you to live around cactus and not be punctured. Like all learned caution, it is the result of adapting to aversive events…touching a hot stove, driving your bike into a tree and using a knife to chop vegetables.
4) You may create a new device or behavior that makes the event neutral or beneficial. For instance, wearing thick gloves allows you to harvest the prickly pear cactus to make delectable jellies from the fruit.

What will not happen –
1) You will not develop a deep-seated fear of cactus.
2) You will not move to an area that does not have cactus.
3) You will not lose sleep at night because you touched a cactus.
4) You will not hate cactus or start destroying cactus because one of them pricked your finger.

Obviously, our behavioral reaction to punishment is a natural advantage in almost all settings. It is designed to keep you safe. It is much better to be spanked seriously as an infant than to fail to learn to stay out of the street when cars are coming. The spanking will be forgotten but the inhibition and threat of danger will fortunately stay with you for life. If you think a Wal Mart parking lot isn’t a place to be vigilant you are mistaken.

The most common goal – creating an inhibition:

The primary practical goal of punishment is to stop a behavior immediately and leave a lasting prohibition. This is not always accomplished elegantly, however there are rules that can aid you in this task. For instance, it’s a good idea to limit the possible causes of a punishing event. Was it the sound of water, the time of day, the color of the sky or the force or direction of the wind? Virtually all punishment starts in ignorance, includes confusion and rapidly causes an inhibition. Depending on the nature of the punishment, the inhibition may be small and precise or big and wildly mistaken. Sometimes the goal isn’t complete suppression but a lessening in intensity that allows for control to allow or disallow the behavior.

Origins:

If you have been in the US over the last 50 years or so, you’ve seen the unrelenting growth of anti-punishment sentiment and activism. The trail is long and devious. It started in the late 1900’s with the growth of ‘humane societies’ that heralded the widespread awareness and opposition to animal abuse. Then it made entrees into academia and the creation of ‘social sciences’. From there it filtered into the mainstream of society. People were employed as social workers whose job was to improve life via the discoveries of social science. In the tradition of humane societies, humane educators and enforcers started influencing society. That’s where it got problematic – academic masterminds teaching their student-minions to meddle with the population. Ironically, those who decried any form or coercion used it exclusively – to create a better world. Their weapons were considerable. Within academia, those who attempted to present a reasonable discussion of aversive control were marginalized and shoved into bizarre pigeon holes. In society, criminal prosecution for child or animal abuse became law in all 50 states. Eventually, the most powerful sledgehammer is the accusation of being an abuser of children or animals. Since few people in society actually abuse children or animals, the saints needed a foolproof way to destroy anyone who disagreed with them. Their hammer was ‘imagined harm’. It works to allow them to discredit or undercut anyone who disagrees with their agenda by offering objective observations.

If we wanted to truly attack any objective discussion or use of aversive control we must first neutralize rational thought. This has always best been done by using terrible insults. Since abuse requires some reference to harm, trauma or actual damage, the anti-punishment saints had to disconnect that requirement from their litany. That started in 1944. A behavior analyst, W.K. Estes, tried to create a comprehensive theory of punishment. He offered findings based on what has become the giant smoking gun of anti-punishment ideology. He used ‘non-contingent shock’ to monkey with captive rats and pigeons. Non-contingent punishment means it’s not connected to any specific behavior. In the minds of most people, shocking an animal independent of any consistent rules is, ironically, abuse. Within a few years, this intentional mislabeling of ‘punishment is abuse’ became the standard throughout behavioral science. It is as old as the 1940’s and as current as 2014.

What It Means:
The reason this seemingly trivial piece of information is important is that it defies logic and common observation of reality. This hogwash has been sold as ‘scientific’ when it is actually ideological nonsense – and once you examine it, you will know that for a fact.

The Lowly Cactus Needle Tells the Tale:
In the early days of behavioral science, punishment described the behavioral effect responsible for causing behavior to diminish, stop and/or never return. We know this effect well. We can look at it logically and describe personal experience that confirms that such an effect exists in day-to-day life. Consider the cactus needle.

A single touch of a cactus needle teaches you not to touch them. Having touched one, only once, you will never voluntarily shove your finger into one. If someone grabbed your hand and tried to force it onto a needle, you would resist. This proves that you can learn a life-long inhibition on a single occurrence of beneficial punishment. If ‘beneficial’ and ‘punishment’ seem to clash with your sensitivities, it’s only the result of the fictions spread by behavioral scientists, social workers and modern animal trainers. In reality, punishment functions to keep us safe. The normal influence of punishment on behavior brings obvious benefit. Only the abuse of the effect causes harm. Now you know the dirty little secret. When a behavioral scientist or modern trainers says “punishment”, they are describing abuse. They predict imagined harm from its use – the same behavioral effect that taught you to ride a bike. It is the same behavioral effect responsible for all fine motor skills. If you have no punishment for failure, you have no brain surgeons, fighter pilots, emergency medical technicians or concert pianists. Without some ‘downside’ to failure, we don’t worry about making mistakes. If the outcome of our behavior is the same whether we succeed or failure, failure becomes, not a option, but a practice.

 

Dumping the Dichotomy:
I have no bias in favor of any behavioral effect without first acknowledging a specific context. To do otherwise is mindless. For instance, I do not prefer ‘water’. Water to someone in the desert is precious but may be deadly to someone drowning from a flash flood – in the desert. I do not prefer left over right or up vs. down. These preferences must be plugged into a real-life setting before they can have any meaning. For instance, is it better to alphabetize your client list or number them? To answer that you must decide what you are trying to accomplish. Perhaps you can do both or alternately rather than one or the other. The same is true for reinforcement vs punishment. Unless I know the goal, preferring one over the other or picking a specific sequence is, again, mindless. I am not mindless in my work. This video shows ten minutes of a dog’s life – a dog that had bitten five people. You will note that I use positive reinforcement as prelude to punishment for threatening me and then back to positive reinforcement when the inhibition is learned. p.s. The ‘punishing stimulus’ is a loose piece of terry cloth towel. https://www.youtube.com/watch?v=tbn1DjDEB-k

Basics: Rewards build behaviors pleasantly. Punishment stops things.
We tend to
move toward things we want and move away from things we don’t. It doesn’t get any more simple than that. However, there is a tendency on the part of modern trainers and behaviorists to forget a very important point – at any given time, the same experience may trigger attraction or avoidance. Consider the effect of a campfire. If the night is cold you will be attracted to the fire. If you stand to close, you will be repelled or find some other way to deal with the heat. This might include gradually playing a vertical human rotisserie to be generally warm. However, campfires usually afford haphazard warmth that requires dynamic adaptation. That’s life. In the summer time, a campfire is not something you want to hug – it makes your heat quotient worse than no fire at all.

Merely looking around you can teach you about the absurdity of thinking that it’s all about pain vs. pleasure.

“Your mother she was raised way down in Texas,
Where the jimson weed and sand-burrs grow;
Now we’ll fill you up on prickly pear and cholla
Till you are ready for the trail to Idaho.” –
Whoopee-Ti-Yi-Yo-Git Along Little Dogies


Yes, cows can eat cactus. They learn to consume them, needles and all. The simple truth of that is that pain doesn’t automatically inhibit behavior. It also means that getting what you want may include considerable discomfort. For the ‘all positive’ or ‘all negative’ trainer, this should give pause to remaining in an ‘either/or’, one effect fits all ideology. (It doesn’t)

The nature of working to ‘get’ something.

Why do cows eat cactus, despite the pain of ingesting cactus needles? There are two answers. First, because they can. That should be good enough for everyone, yet isn’t a good enough explanation for most people. The second reason is that life is often a matter of making deals – often with the Devil. Consider the cactus fact in context…

Behaviors maintained by positive reinforcement are based on the dog ‘working to get something’. This assumes that all behaviors are fundamentally attempts to ‘get something’. In the case of the cow, the animal is attempting to gain sustenance. Pain in the acquisition of nutrition is a common rule of nature for both predators and prey animals. It shouldn’t surprise people that to ‘get what they want’ all manner of behaviors can be triggered. That is because the nature of behavior is dependent on an animal’s physiology, including instinctive behaviors. These behaviors give the animal a ‘leg up’ on survival from the moment they are born. The will use their innate tools and talents to get what they want and to get away from things that may hurt them. This simple fact is often missed. That is why the explanation of ‘because they can’ is rarely examined. Instead, humans attempt to reduce behavior to a set of rules that are often contradicted by common observation.

Let’s take the whole ‘positive reinforcement’ mantra for a moment. To train with positive reinforcement we must discern what the animal will work for. However, the nature of the work is assumed to be designated by the trainer drawing on a clean slate. There’s your first error. Animals are never ‘clean slates’. There are born with the ability to do things without prior learning. Consider this possibility…

What if the behavior is instinctive and the dog wants to get something via violence? We know that resource guarding is innate. How about ‘resource acquisition? Gee, that’s a definition of the violence used by predators. That means that predatory violence is innate. What if you want the dog to sit and the dog bites your hand to get the treat? Oops. It’s a perfectly natural behavior. That requires that we find a way to stop the perfectly natural behavior before we can have a practical way to use ‘positive reinforcement’. That is rarely considered yet is perfectly logical – a behavior might have to be suppressed/stopped/inhibited in order to establish a civil repertoire. If a perspective ignores instinctive reactions and makes all behaviors a matter of reinforcement vs. punishment, it is woefully incomplete. Life is far more complex than simple one-dimensional explanations.

If you have gotten to this point, I would suggest that you look at three things – my YouTube channel https://www.youtube.com/user/wilkesgm1, my blog https://clickandtreat.com/wordpress/ and my facebook page https://www.facebook.com/gary.wilkes.39?ref=br_rs If you have any questions before the seminar, you can contact me at wilkesgm@aol.com. To register for my Scottland and England seminars, go to my blog and hit https://clickandtreat.com/wordpress/?post_type=product.

Leave a Reply

Your email address will not be published. Required fields are marked *