Archive for the ‘positive reinforcement’ Tag
Trust is a central theme of soap operas, TV dramas and political relationships. It’s lauded as being the keystone of good marriages and partnerships. Teenagers are reminded that they will not be allowed to stay home on their own, or out late, or have the keys to the car until they can be trusted. For many people the realization that trust has been “broken” can lead to a lengthy or impossible reconciliation.
If one was inclined to look at the importance we place on trust in a marriage from a biological point of view, the risk of raising someone else’s offspring, or losing the support of a good partner to someone else, could impact the long-term success of one’s own off-spring. But mostly, when someone discovers that their partner was not “faithful,” babies aside, it feels really bad- poem-writing, sad song singing bad.
Dogs are among the few species on the planet who allow us to break trust with them, and not make us pay for it, consistently. Yell at or physically reprimand a cat and you might not see them again, or are at least likely to have to clean out the scratches you received in return. Few believe that the lions in the cage being kept under control with a whip are to be trusted to safely snuggle on the couch with their “tamer.”
We can and do break the trust with our dogs routinely and there is a price. It’s bad enough to wonder if your partner is trustworthy when they call claiming another late night at the office. It’s another to wonder if the person approaching you is going to physically restrain, hurt or scare you. Being at risk physically, even if it’s done leaving no marks, is not something one forgets or puts aside easily.
The risk of losing trust with a dog is greater the shorter the relationship or the smaller the existing trust account. If we, from the moment we meet and handle a dog demonstrate that we are safe and worthy of their trust, and should we have to withdraw from the trust account we’ve built, we are less likely to lose it all. We are less likely to get bitten, or growled at by a dog and more likely to have them come when we call them. A dog’s behavior can tell us as much about our relationship with them as it tells us about them. Trust counts.
I received this message on Facebook. I thought I’d share it and my response.
“And while I was writing this in response to someone’s question … I was booted from that group ..lol nice Let me tell you what happens every time I state something I believe… I’m abused. I’ll be part of a discussion and before you know it I’m getting bombarded with nasty emails and become everybody’s example of ” what not to do” you’d think instead of ridiculing me for some of my beliefs there’d be someone curious to know how I came to those beliefs… You think I enjoy being hated by my peers? Trust me I don’t but there’s no way around it. When i pose the question ” is it possible to get 100% reliability recall using all positive ” I’m actually hopping to learn something… I’m not looking for a fight. My dogs are reliable with recall , and as a result they have lots of freedom . However the only way I can ensure reliability is with remotes.. Even though the remote NEVER goes past 1. (The lowest setting ) it’s still technically still a shock. I can’t imagine getting 100% reliability without the remote as backup. However if there was a way I’d absolutely change my opinion and switch over.. It’s not about ego for me it’s about making my dogs as happy as I can nothing more… Sorry for the babbling.”
I’m not sure why I have received this message from you. I don’t know you and so wonder if it was meant for someone else, is spam or trolling. But I will take it at face value, on good faith, and will attempt a response.
A question that comes to mind is why you persist in behaving in a way that culminates in a consequence you profess not to like or want? Are you honestly seeking answers to your question about training reliable recalls without aversives? Are you an e-collar advocate out to espouse its benefits anywhere, to anyone, at any cost? If it’s the latter I return to the first question. Do you also go to dinner parties and try to convert people to your religion or diet regime, continuing even as other guests pick up their drinks and scurry away from you? Either you want to be at the party or not.
There will be different cultural “rules” at different parties. Frat house parties will allow for certain behaviors that a party with the table set with 3 forks and 2 different spoons would not. I for one would not go to a frat party and try to get the guests to put napkins on their laps before they eat their pizza. Nor would I put the table center piece on my head at a formal gathering regardless of how funny I thought it was. Not if I wanted to be invited back anyway. If I felt that not being able to put the table center piece on my head cramped by artistic freedom or inner child, I would have to seriously consider my choice to attend such parties and the judgement I was using to come to that decision. If I felt very strongly about dining etiquette practices for Delta Phis (or whatever the heck those frats are called) I would consider how well my approach to disseminating information was working or likely to work. Obviously not very well if I was having the door slammed shut and locked behind me after being pushed out of it. If you are indeed looking for answers and not advocating e-collars you probably need to look at how you are saying what you are saying if so many people are missing your point.
The question you are asking, regarding achieving 100% compliance is a disingenuous one. There is no 100% in behavior. We can only predict the likelihood that we will see more or less of a particular behavior based on the consequences for the animal. The makers of e-collars are aware of this and many include a “bump” button that will raise the level of the shock (or stim if you prefer) a set number of levels. If the dog does not respond to level 10, hit the bump button and they get level 30. If we could get 100% at level 10 why the need for a bump? What happens when we don’t get 100% at level 30? Onward and upward?
I would also need clarification as to whether we are talking about using an e-collar for training or management. Many trainers and most pet owners are using them as a management tool, not a training tool. If what you mean to say is that you use an e-collar to replace a longline or fence, I think we should make that distinction. There is no 100% even if an e-collar has been used in training (or for management, dogs are learning when being managed). Any behavior, in the absence of a historic reinforcer or punisher is subject to extinction. The trainers who I have seen who use e-collars more effectively and humanely than most (and there are unfortunately few of them around if what is routinely posted online is any example), rely on positive reinforcement to build and maintain behaviors, i.e., they don’t need to use shock to teach or maintain the behaviors they’ve taught the dog. Many trainers and owners can’t get a reliable recall from their dog because they have failed to provide or stopped providing positive reinforcement for the behavior. They may be “rewarding” the behavior, but they are not reinforcing it and they don’t understand the difference.
I recently attended a training seminar and had the opportunity to learn from people who train birds. The birds are fully flighted and many are taken outside for exercise and enrichment. Flocks of parrots without a bit of hardware on them (nor were they trained using any), and not starved, are allowed to fly around as they like. When they are cued to return to their handlers they do, close enough to 100% of the time for the handlers to continue the practice. Pretty impressive for a creature not touted as “man’s best friend,” the animal who is saddled with an increasingly available array of equipment in order to teach or maintain behavior.
The other fly in the ointment as far as 100% goes is handler error. Batteries go dead, contacts aren’t made, remotes are turned off, a behavior occurs and by the time the handler reaches into their pocket the opportunity for training is long gone and they are instead merely providing a consequence to whatever behavior happens to be occurring at the time they press the button. It may suit their needs or not, but it’s not good training. I also have to question an e-collar which is designed so that level one works for a dog. One would assume that using the lowest level of shock possible to get what they need would be preferable. If there is no lower setting available how can they be sure they are doing that? I can’t help but wonder why the commitment, in the face of social media shunning (if that’s as aversive to you as you seem to trying to say), to promoting an aversive with a high potential for misuse as a training tool, or advertising the use of one?
If you are looking to learn how one achieves high levels of compliance without the use of an e-collar I would look to people who are training dogs to perform at competitive or professional skill levels, trainers like Denise Fenzi and Steve White, to name only two of the many who are out there working with dogs. The way it is done is by arranging antecedents, managing the economy and providing positive reinforcement as consequences in a skilled manner.
Please do not feel that you need to respond to this post publicly. My goal is not to single you out, insult or shame you. I just thought that it was a worthwhile question to answer, even if you were only trying to “get my goat” or sell something. Ultimately we all have to make choices about what we are willing to do to an animal to get the behaviors we decide we need or become skilled enough to get using the least amount of invasiveness.
I returned home yesterday from a multi-day workshop on training birds at Natural Encounters in Florida. Watching and learning from the best bird trainers on the planet (and that is not hyperbole) was inspirational along with educational. One of the take-aways for me was new language to use when talking about training, any animal.
Many of the participants at the workshop were zoo keepers. People working with animals who have the potential to injure or kill them, i.e., large, wild animals, use the term “protected contact” to describe training in a setting in which the animal can’t touch you. At first glance it looks like a set-up designed with the human’s safety in mind, but it also provides the animal with the information that the human can’t get them either.
The first step we need to take when working with a fearful dog is to provide the dog with an environment in which they feel safe. How we do this depends on what is scaring the dog. Many of the dogs people contact me about are afraid of people. Unless we are able to manage the dog so they consistently feel safe in the company of people, we are not likely going to see progress in their ability to interact with us, or that progress will be painfully slow. It may be so slow that the conclusion is reached that the dog is unsalvageable. We may need to find ways to work with our dogs using “protected contact.” In the following video you will see how I created an environment in which I was able to work with a new foster dog (and yes he is now my dog) to help him learn skills while maintaining his ability to choose how much contact we had. You don’t need to watch the entire video to see how I set it up to make sure that he did not have to worry about me trying to touch him.
It will be easy to find excuses as to why providing this kind of protected contact is not possible with your dog or the dogs you work with. Those excuses will not change the reality that an animal who has to worry about their physical safety is not going to learn new behaviors as easily as one who knows they are safe and can begin to build a new repertoire of skills and behaviors.
At my first appointment with a new dentist after I moved to Vermont I asked if he’d like me to have my charts sent from my previous dentist. His reply was, “I don’t need them, I have your mouth.” Everything he needed to know about my teeth was in front of him.
When we begin to work with fearful dogs it’s not uncommon for us to think that we need to know the dog’s past in order to help them. It’s not that the information would be superfluous, but it likely will not change how we are going to work with the dog. We have their mouth, so to speak. Their behavior will guide us. Whether it’s an 8-week old pup or 8 year old dog who won’t come out from under the bed, our approach will be the same–help them feel safe. The same would be true of a dog growling, we don’t need to determine whether the dog is fear aggressive or aggressive and not fearful, our response to the situation will be the same–do what we need to do to end or prevent the growling without punishing the dog. We take away any perceived threat, desensitize and countercondition, and teach the dog to do something else using positive reinforcement-based training.
I have had clients spend the majority of a consult describing in great detail everything that happened to their dog. They think that something is going to inform me about the exact “fix” their dog needs in order to stop being fearful. If there was a sudden onset of the dog’s behavior it would indicate the need for a vet visit, and even with that, we’d prepare ourselves to work on any newly added fears that occurred due to pain or illness. We’d do this the same way if the dog had been displaying fearful behavior for years.
“Why” can get in the way of developing humane and effective plans for working with a dog. Decide that a dog is being aggressive because they are trying to dominate you and respond in a way to thwart this attempt and you’re likely to start brewing trouble. Knowing whether our dog was timid from birth, spent years in a cage at a puppy-mill, was tied up in a yard for most of their life, was beaten by a man with a hat and a beard may satisfy our curiosity, but it won’t change our training plan. Make sure they feel safe, DS/CC and teach them something.
My dentist did take x-rays. It would be nice to have a machine to look into a dog’s past, but don’t worry that we don’t.
The subject of using medications to treat dogs with fear and anxiety issues is a controversial one among pet owners and trainers, and one I frequently feel inclined to address in regard to working with fearful dogs. Drugs have been a blessing and curse for humans. They can both save and destroy lives. Deciding to give a scared dog medications is often a struggle for owners. An incomplete understanding of why they are being used is often at fault.
There is an immediate emotional response to the idea of giving a dog a medication for a behavior issue, and for some people it’s a bad response and for others it’s more neutral. There are few pet owners who thrill to the idea. That some people mis-use medications with their dogs, and by this I mean that they assume that training challenges, or the failure to provide a dog with enough stimulation and enrichment on a daily basis will be remedied with a pill, does not take away the benefit these pills can have for many dogs.
A common misunderstanding about the use of behavioral medications is that they are being used to sedate a dog, this is especially the case when a dog is fear aggressive. Owners assume that the dog will be “doped-up” and spend the rest of its life in la-la land, unable to function. People often worry about potential side-effects of medications, but have given no consideration to the impact chronic stress (which a medication might alleviate) has on their dog. And if a medication does not prove to be effective or there are negative side-effects the option always remains to stop using them. There are a different medications available, and one might work better for one dog compared to another.
If you step on a rusty nail and suffer a deep puncture wound, even if you develop an infection there is a chance you will survive. Antibiotic medications will likely play a role in this. If you wait too long to take the drug the infection may progress to a point where the drugs are not effective or your life can be saved, but not your leg. While we are hoping that our dog’s problems can be addressed with soap and water, a kiss and a bandage, the infection may be setting in. We know what normal, healthy dog behavior looks like. If you are unsure as to whether or not it’s time to stop hoping the problem will resolve on its own find a trainer** who understands the challenges of working with fear-based behaviors and talk to a vet or vet behaviorist to explore ways you can ensure you save the leg.
**Any trainer who recommends the use of force, coercion or punishment to help a scared dog “get over” their fear should be avoided. At no time during training should a dog be handled in ways that are designed to elicit fear in your dog.
*This is also the name of my upcoming book.
When we ask a dog to do something in exchange for something they want it’s not about “no free lunches” or that dogs need to learn to work for what they want. Every organism on the planet, if they are going to be around for long already has this information. If they didn’t there would be no beaver dams, no mice caught by cats, no webs woven by spiders, no carcasses cached in trees by fisher cats, no acorns hidden away by squirrels. When we teach a dog to respond to a cue, rather then making them work for treats, or their food bowl, or an open door, we are making the world a more predictable place. The behavior of humans makes more sense to a dog when they are trained to respond to our cues.
When we use positive reinforcement to teach a dog behaviors we are infusing the whole process, all the pieces of it, from the appearance of the person to the performance of the dog, with “feel good.” We are the ones who label a particular behavior as “work” whereas for the dog it might just be sitting, fetching a ball, waiting or rounding up sheep. It’s something to do, and doing something is often better than doing nothing, especially when good outcomes are the result.
John asking Sunny to target his hand is not being used to lure Sunny closer, he had already shown he was quite happy to be that close. It was not to make him “work” for the treat. Asking Sunny to target his hand enriches the relationship between the two because the human’s behavior now has meaning instead of being a random, possibly threatening, gesture. It was a cue Sunny already had learned from me and performing behaviors to make treats happen is an enjoyable experience.
Build a vocabulary with a dog and you’ll be amazed at the conversations you can have with them.
For dogs without fear-based challenges it may only take one introduction for the dog to feel safe with you. For others it might require a dozen, and for another hundreds. This is likely the reason many fearful dogs are able to be ok with their primary caregiver(s). There are enough repetitions of positive interactions to tip the scale in their favor. We can’t know how many reps it will take for an individual dog so just keep on adding to the total.
Today on our woods walk I watched as the scale began to tip in John’s favor. For years John rarely joined me and the dogs on our daily walks. But lately that’s changed. Initially I handed out the treats when Sunny stopped and eyed John warily. One day I started doling out handfuls of treats for John to dispense to the dogs, tossing them on the ground for Sunny. I took my chances the next week and handed him a bag of treats and watched as he increased the number of times he stopped and fed the dogs. Sunny began to spend more time close to John looking at him expectantly for treats.
The years of coaching (ok, call it nagging if you will) John on how to avoid making eye contact, leaning toward, or reaching for Sunny actually paid off as John noticed how something as simple as turning his head away from Sunny made it possible for him to take a treat from his hand. Both players were being reinforced for their behavior. Sunny got treats and John discovered that fabulous feeling of watching an animal begin to trust you.
Today I noticed that Sunny decided to sit, granted a safe distance away, and wait for John while he worked on building a log bridge over a small stream. I had continued on with the other dogs and was surprised to find the only one missing was Sunny. Even after Sunny caught up with me he returned to John three separate times.
The journey isn’t over for them but gone are the days when Sunny had to shadow us in the woods, too afraid to stay on the trail with John. Call it a milestone or a miracle.
There was a blog post going around recently that could have easily been parody as serious. It was written by a volunteer trainer at a shelter who was declaring he was never going to volunteer at an animal shelter again because of his recent experience at one. The reason for his defection? Were dogs being mistreated? Were the conditions gross and unsanitary? Were too many dogs being euthanized? Was the staff being disrespected? Nope. The reason he was never going to volunteer at an animal shelter again was because the dogs were going to be clicker trained. Apparently this was so distasteful to this trainer that he was out the door never to return.
His litany of reasons for this decision included many of the old faithful, and incorrect I will add, missives regarding why force-free training and the use of food in training kills dogs. They included:
Force-free trainers sit around flipping through magazines waiting for good behaviors to reinforce while a dog gobbles down the food off the counter, chews the leg off the dining table and pees on the Oriental.
Dogs in shelters are there because their owners, patient saints each and every one of them is, never told a dog “NO!” or yanked on their leash. Never having been reprimanded the dog has become uncontrollable and dumped at a shelter.
Dogs in shelters are there because force-free trainers advised the aforementioned saints not to prevent or interrupt their dog’s inappropriate behavior, but instead to grab a magazine and wait for an appropriate behavior to reinforce with the bits of filet mignon they are schlepping about in a treat pouch. That studies of dogs left at shelters have shown that upwards of 94% of pet owners never even consulted with a trainer about their dog’s inappropriate behavior is insignificant. Of the 6% (or less) of the pet owners who did contact a trainer all were handed a clicker and treat pouch (and a magazine) and instructed on how to create a dog they will want to dump at a shelter.
As I read I kept waiting for “SURPRISE! only kidding.” I am not going to walk away from shelter dogs for similar reasons that boys walk away from their sports team, because someone (who never knows as much as they do) decided to let girls play. Or quitting the typing pool because your typewriter is being replaced with a computer. This guy was leaving because the shelter management decided to introduce a training method that is less stressful to dogs. Full stop. Period. It’s like the fellow at a hearing about our local nuclear power plant declaring that he’s, “Lived next to the plant for 30 years and he’s ok.” But what about the spent fuel rods sitting in a pool on the property sir? Any suggestions as to what to do with them? Or someone boasting they’ve been smoking a pack a day since they were 15 and didn’t get cancer and see this as enough reason to keep on smoking and recommend Marlboros to friends.
I admit I only read the post once, glanced through the comments, shook my head and left the page. The requisite “all dogs learn differently” myth was tossed out as further evidence that one dare not attempt to use force-free methods with them. I’m assuming that this is because dogs, unlike any other organism on the planet, NEED to be trained using aversives. There were the head spinning, tail chasing comments about how trainers who advocate positive reinforcement also use punishment by depriving dogs of food treats when they don’t do the right thing, justifying the use of any number of collars designed to hurt or impede oxygen intake, or whatever form of positive punishment they prefer. Terms like “balanced” were used to describe trainers who I’m assuming do not have the mechanical skills in place to put behaviors on dogs using primarily positive reinforcement, tipping the scales to unbalanced in favor of force-free.
For those of you unfamiliar with the world of force-free dog trainers, allow me to share this with you. Many of us spend hundreds to thousands of dollars a year on continuing education. We study with some of the world’s best animal trainers. We practice the mechanical skills of training with some of the great names in animal behavior. Do you honestly think that trainers who can teach a multi-behavior chain to a chicken, hamster, fish, horse, lion or dolphin can’t teach a dog to walk nicely on leash, sit on a mat when people come into the house, watch quietly as other dogs go by, without using positive punishment? When “balanced” trainers choose to call us “treat dispensers” I want to channel Jack Nicholson in A Few Good Men and shout, “Show me some goddamn respect!”
If you want to pack up your toys and go home because the 21st century of animal training has arrived on your doorstep, go. You have as much to lose as the dogs you claim to care so deeply about.