Are you being too Operant?

Operant conditioning is one of the cornerstones of animal training.  It is the first thing that I teach at Beckman's Academy for Dog Trainers and often the first thing I go over with group class clients and most clients that want to do any type of basic obedience training.  However, I constantly see trainers and some clients overusing operant conditioning.  While OC is important for basic obedience it rarely helps with fear, reactivity or aggression.  In fact I have seen dogs become more fearful and aggressive by a trainer who is so committed to OC that they missed the window of opportunity to desensitize or counter condition a young dog.  

The textbook example is when a client shows up ay my facility with a bag full of treats and clicker. They come in with their dog barking and lunging, they wait for a split second after the dog stops, and then they click and treat.  Seems like the right thing to do right?  Wrong, they are being too operant.  They are only thinking in terms of operant conditioning, mainly positive reinforcement.  But even positive punishment (corrections)  isn't the right thing to do either.  So I say to them, "you are being too operant.  I want you to forget about the clickers and treats, it hasn't worked for you this far".  These clients and dogs need a total shift in direction.  They need to focus on the other main type of conditioning, Classical Conditioning.  I take the leash and start to walk the dog in a big circle, then I let Bosco out within view, the dog may bark and lunge, and I just keep walking, once the dog has calmed down, I may let another dog out or let Bosco get closer.  Eventually I usually let the dog meet Bosco.  This is done without treats or corrections (both operant in nature).  

Classical Conditioning is the second principle I teach at the Academy.  And it is by far more "powerful" than operant conditioning. It's elicited responses vs emitted responses, and I'll take emitted responses over elicited all day.  

I know a trainer at a zoo.  His area is trying to get a hyper dog to be calmer around the big cat that it's suppose to live with.  But instead of constant desensitization and learned helplessness around the big cat, they are clicking and treating for every little calm behavior.  If they are very skilled trainers they may achieve what looks like calm behavior, but since it was done operantly I'm not sure that the dog will really be calm or that it will happen at all.  Operant conditioning isn't good for conditioning calmness.  (And for all you prospective trainers reading this, about half of your clients just want their dog to be calm)  Constant desensitization and learned helplessness (see previous blog on what learned helplessness is) is the only way to truly achieve an emitted calm behavior.  

Also I was at a event with arguably the most well know "Force Free" trainer,  we were talking and watching an agility event.  For the whole time that we were hanging out, her little dog was offering behaviors; rollover, wave, sit, shake, the dog simply couldn't calm down, it was very difficult to watch.  The dog had been trained with so much operant conditioning, (I'm sure she even tried to train calmness with OC), that the dog's little brain just couldn't relax.  This trainer was so good at OC that she believes that it's the answer for everything. 

But why doesn't operant conditioning work for calmness?  If you were to reinforce and animal for being calm wouldn't that behavior occur again?  Yes, but you are just going off of the physical "look" of being calm, often the dog's "mind" is not calm.  Also just the presence of reinforcement (treats) makes it impossible to achieve true calmness. And lastly, a principle called superstitious behaviors, this is "an accidentally reinforced behavior", so you are trying to reinforce a look away a half of second before the lunge at another dog but instead you reinforced the dog "thinking" or even doing the lunge.  

Why do so many trainers focus so heavily on OC?  It's complicated but here are a few reasons.  1) Trainers believe that the better they are at OC, the better trainers they are.  When in fact simplifying things for clients is the true skill. 2) All positive reinforcement trainers look up to the leaders in the industry, and all the leaders in the industry look up to the marine mammal industry.  But the truth is the marine mammal industry doesn't really need calm animals for a long period of time, they are with the animals very little, and when they step away from the animals the animals are no longer near the trainer, essentially they are "out of the trainers hair".  Whereas dog owners are often with their dogs all the time.  3) The marine mammal's environment rarely changes whereas we need our dog to be calm in ever-changing environments.  Basically, while marine mammal trainers are some of the best trainers in the world, but it's a very different thing entirely, so following the marine mammal doctrine like many leaders in the dog training world, is a losing proposition.  

My point of this post is that there is no substitute for desensitization, there is no substitute for learned helplessness and there is no substitute for counter conditioning.  And all dog owners and trainers should do these 3 things way more than they think they should.  Understanding and being skilled at operant conditioning is very important, but you have to know when to use it and when NOT to. 

Using a dominate animal to help with aggression

Our method of using Bosco, my helper dog, to help me with aggressive, fearful and reactive dogs is nothing new, in fact I've seen "dominate" animals used for this purpose throughout my career with exotics.  

When I was at Seaworld Orlando, I will never forget the conversations that were had about what to do if there is an aggressive incident towards a trainer in the water.  Let me give you a teaser, it involves bring another whale, not under stimulus control, into the situation to help. We will get back to that in a minute.

Dog aggression has become a epidemic in our society.  It's rarely talked about in this "epidemic" way, but I get calls everyday about dogs, people and children being bitten or attacked.  And what further complicates the issue is there is no agreed upon method to "fix" or even minimize aggression. The shock collar folks have their way, the positive reinforcement folks have theirs and the dominance folks have theirs.  And all these groups discredit the other group's methods of training and dealing with aggression.  And in fairness, there's no easy way to deal with aggression because there are so many types of aggression, motivations and by nature aggression is hard to extinguish.  So no method works for all.  But at Beckman's Dog Training we feel like we have the best way out there.  This is due to my unique experience and perspective having worked with so many "aggressive" animals in the past.  I have taken from all my experience to develop the "Beckman's way".

The Beckman's way to deal with aggression, fear and reactivity is "dogs helping dogs, incremental desensitization, mental and physical exercise and an owner confidence approach".  It's no accident that dog's helping dog is first.  Our goal is to always get the fearful, reactive or aggressive dog with my dog Bosco as quickly and safely as possible.  Bosco is a confident, non aggressive and likable dog.  We use muzzles, leashes and slow meetings to get the dog with Bosco.  After the dog "likes" Bosco we have achieved what most trainers cannot or will not even attempt.  Many of my clients are already amazed that their dog can be friends with another dog.  But we usually don't stop there, we then take what we have learned about the best way to meet Bosco and apply it to meeting another dog, then another, and so on.  And oddly even "dominate" dogs seem to respect Bosco's "dominance", confidence and playfulness.  A note about dominance, I am using "dominance" in anthropology terms because people can relate and understand what it looks like.  There have been articles on how dominance does't apply to dogs, and while I don't know if I agree with that totally, I do believe that the term is overused, so again, I'm using it as a umbrella term because lay people can understand it.  

How we use Bosco in a group of dogs is also very important.  At the Beckman's facility we let dogs play in groups multiple times a day, and often we have a reactive dog or dogs involved in the playtime.  We only allow this after we have done the previously mentioned process with the dog. So all the dogs are running around, if Bosco see's a dog do something he deems as antisocial, reactive or aggressive he will go over to the dog and bark or growl, this alone will often snap the dog out of what it's doing.  We have also, but rarely, had fights during a playtime or private session.  The aggression is rarely towards Bosco but instead towards another dog. I tell my staff what to do during a fight like this, then I say "don't call Bosco out of the fight let him do what he feels is right" and what he feels is right, is to go at the aggressor and bark, growl and often put his mouth around the top of their neck, this is all happening as the aggressor is pinning the other dog, this alone will rarely break up the fight, but as I'm breaking it up, Bosco there as the dominance dog helps tremendously.  In fact, when Bosco is not involved in the group playtime, we see more fights.  When Bosco is gone there is a noticeable difference in the group dynamics and some dogs feel they need to fill the vacuum. 

Now back to the seaworld story.  Let's say a trainer is in the front pool when a "non dominant whale" takes the trainer under wanter or something else unwanted.  If the whale doesn't respond to a recall and we feel the trainers life is in jeopardy, it was talked about that we would let in the dominate female.  This is of course a risk, but most likely she would come in with the attitude of "knock this nonsense off" and most likely the subordinate whale would do just that. Most likely the dominate female would even take the trainer back to safety. In the same way that Bosco tells the other dog "knock this nonsense off, that's not what we do here".  

As trainer we are paid to get results, and while I never use shock collars and prong collars (because they can actually cause aggression), I take my job very serious.  People are paying me their hard earned money so that they can live a normal life with their dog.  This attitude is what has made Beckman's Dog Training so successful.  I say this because I feel too many trainers have adhered to only methods that they read about or saw at a conference, but have never tried to think out of the box in order to help their clients.  At Beckman's Dog Training and Beckman's Academy for Dog Trainers, we don't adhere to any one method, but instead incorporates all the good parts of a specific method and draw on my unique experiences.    

Using a "dominance" animal to correct or intimidate a subordinate aggressive animal is not widely used in the animal industry.  However, if you have the right animal within the appropriate social structure, it can be the first or last line of defense.  Beckman's Dog Training is one of the most successful dog training companies in the country.  This is in part due to my understanding about using other animals to help reduce the very difficult behavior of aggression.  

Please let me know your thoughts.

Joel Beckman, Beckman's Academy for Dog Trainers and Beckman's Dog Training

 

 

 

 

 

Who is smarter? Dogs, cats or whales?

Last week during our first Academy for Dog Trainers course, one of the students asked me "which breed is the most intelligent?".  Instead of going to the standard response of border collies, Australian shepherds, dobermans, etc., I saw it as an opportunity to dive into the big picture of animal intelligence.

Let's first establish that animal intelligence is almost impossible to measure, and people further complicate the discussion by attempting to compare the intelligence of different species. Think about it...we can't even come up with a reliable measure for human intelligence--let alone a test for different species that have completely different requirements for hunting, breeding, communicating, etc.  So the first step is to understand that measuring animal intelligence simply cannot be done. We need to ask if what we're measuring is truly an animal's "intelligence", but once we acknowledge the shaky ground we stand on--it's still fun to dive into the discussion. 

So let's look at some of the "smartest" animals in the world, most of which I have worked with, and see why they are considered to be so "smart".  

 
wolves-1388558_1920.jpg
 

There is an extraordinary amount of data out there on animal "intelligence", most of it very interesting.  Some common ways to measure animal intelligence are: tool use, communication, response to operant conditioning, etc.  While it's fun to figure it out and write about the intelligence of animals, in the end animals are as smart as they need to be to survive and pass on their genes.

In my opinion there are two reasons animals "become" intelligent (or exhibit behaviors people classify as "intelligent"). There are others, but I believe these are the two main intelligence producing traits.  However it's still the chicken or the egg argument.  I'll let people smarter than me figure that out.

1) Natural Hunting Behavior.  Basically if an animal has to hunt over a long distance or period of time, they have the ability to learn long chains of behaviors.  Hence, why dogs are often considered "smarter" than cats.  Dogs come from wolves. Wolves will chase a herd of prey for hours sometimes, in order to wear them out.  So this is essentially a long "chain" of behaviors with reinforcement at the end. An animal's ability to problem solve or complete a long series of events in order is often a measure of intelligence. Similar to a dog doing a long agility course with a ball being thrown (reinforcement) at the end.  Since all cats are pounce hunters, they don't need the capacity to learn long chains of behaviors--it would be a useless function and would take brain power away from something more important.  

Another animal that is considered to be very intelligent, and I agree, are killer whales.  Some groups of killer whales hunt large whales over hours and hours.  Like wolves they will slowly wear down their prey over a long period of time.  This is why at SeaWorld, with our experienced whales, we could do a whole show with no reinforcement and then jackpot the whale at the end of the show.  

2) Sociability, is the other main factor in intelligence.  Look at some of the most "intelligent" animals: humans, monkeys, apes, dolphins, killer whales, elephants, crows, parrots.  They all have one thing in common: they are all highly social animals.  Chimps are widely regarded as the most intelligent animals, and they are one of the most social species. All these animals have very different social structures, but are intelligent none the less.  

We need to be careful about what we assign intelligence to--whether it's human characteristics or even a physiological trait. When we anthropomorphize an animal it's our tendency to associate these "human-like" behaviors with intelligence, when they are often being exhibited for a specific reason that is not necessarily an "intelligent" reason. Being able to see a variety of colors is not a sign of intelligence. I was at an elephant facility while I was in school and one of my classmates asked if the elephants see color and the elephant keeper said "I would think they do because they are so intelligent." The problem is seeing color has nothing to do with intelligence, but rather helps them to survive in the wild.  Baboons probably see color because they need to see the female's rear end to know when she is in heat. Monkeys and parrots need to see color to know when fruits are ripe.  

The wild is a violent and brutal place and each animal that is able to survive is as smart as they have to be.  And dog breeds are just as smart as they need to be to do their job effectively.

So, when we compare the intelligence of dog breeds, we should first take into consideration what intelligence is, how we are measuring it, and perhaps a breed is not more "intelligent", but rather more adapted to what they were bred for. 

These are my thoughts, and I would be interested to hear yours and how you think dog "intelligence" relates to their breed.

THE POWER OF LEARNED HELPLESSNESS

Hello Everyone,

Joel Beckman here, I'm going to continue our series on Behavioral principles and how they relate to training your dog (or monkey or dolphin) or being a dog trainer. Today I'm going to talk about Learned Helplessness and how it can be the missing piece in your dog training repertoire. Sorry for the long post, please stick with it, I guarantee you won't hear this explanation anywhere else!

Today I'm going to talk about a VERY interesting and overlooked concept called Learned Helplessness. I don't know when I first learned about LH, I think it was at Moorpark College Exotic Animal Management and Training Program or at SeaWorld, but it wasn't until I got into the dog training industry that I really saw how it applies to animal training every day.

One definition of LH is "when an animal is repeatedly subjected to an aversive stimulus that it cannot escape. Eventually, the animal will stop trying to avoid the stimulus and behave as if it is utterly helpless to change the situation even when presented with the opportunity to do so." I'm going to refer back to this definition so try to remember it. Or as I say they "give up the fight" Sounds bad right? It certainly can be when applied to abuse or imprisonment, but it's a very helpful tool when working on two of the most challenging behaviors for dog owners and trainers, Recall (come) and Loose Leash Walking.

The best example I can give of LH in action is something we all see constantly. Ever seen a service dog attached to a wheel chair? You have probably thought, wow, that dog is really well trained. So how is this LH? Here goes, The dog has constantly and consistently walked next to the chair, the leash (the adversive stimuli) has never changed lengths (cannot escape) due to the wheelchair being so heavy and not letting the dog change it's course and rarely do people pet the dog (no reinforcement which adds to the adverse nature of the whole situation). If the person took the leash off would the dog say "I'M FREE" and run around, NO, they would mostly likely just keep following the person and chair as if the leash was still there (utterly helpless). BTW- This is not an indictment on Service dog training in any way. I believe in using learned helplessness for service dog training and in other forms of training. This is simply an explanation of a behavioral principle. If you disagree with me thats fine, but don't let emotions about the wording get in the way.

So how can LH apply to you and your dog? There are many methods to train a Recall (come) it would take me a day to write out all of the different methods, most of which are good and important. But often one piece is missing, the Learned Helplessness aspect. When you ask the dog to come and they don't, GO GET THE DOG EVERY TIME THEY DON'T COME. We do this constantly at my facility. Once the animal understands the cue and what the criteria is, if they ignore the cue, we head out and get the dog, EVERYTIME. We do it very calmly but it always ends in me getting the dog, there is never punishment applied, but to some degree I am the "aversive stimuli" in this situation. Now, of course we will jackpot and throw a party when they do come, this is just the method we use when they don't come. Once this has been done a few times, the dog will simply start to come. The dog will basically think "I can go to that guy and get a treat or he's going to come get me and not give me a treat, but either way I'm going to end up where he originally asked me to come". You might be asking yourself, but Joel, how can I do this at a large dog park? 3 things 1) Start this at home or small dog park so you can go get them, then the LH should kick in and they hopefully do it even when they can avoid you, but they choose not to. 2) Still try to get them, go with a significant other so you both can calmly track them down and get them, if you do this even once you may be surprised at your dog's response the next time you ask them to come at the dog park. 3) If you don't think you can follow up with the recall, don't ask it. If I'm at Disneyland with my kids and I'm foolish enough to say "if you don't stop crying we will leave" and then they don't stop crying, I will leave. You may hate to lose hundreds and/or thousands of dollars but it's what you have to do, you have to follow up or people/kids/dogs don't know what's real and what's not. 
 BTW- a reliable recall can be trained with out LH, but using these principles can speed up the process.

Learned Helplessness is also very important for Loose Leash Walking. Think about why the service dog walked so well next to the person. Then think about how you walk with your dog. Does your dog ever pull you to a spot they want to pee, or even pull enough so your arms extend out, before you pull them back? The dog is learning that walking next to you is really only a suggestion and pulling will on occasion get them where they want to go.

So in conclusion, Learned helplessness is a tool for complex basic obedience behaviors like LLW and recall. It is a small but important part of any training program. I welcome your feedback.

THE IMPORTANCE OF FADING

Hello Everyone,

Joel Beckman here and I'm going to start highlighting some of the unique content prospective students will learn at Beckman's Academy for Dog Trainers. I hope this will be helpful to professional dog trainers and dog owners alike.

On Day 3 of the Academy, one of the things we will go over is Fading. Fading is the process of gradually removing reinforcement, cues or the bridge. Most dog trainers know what this is, but I see dog trainers constantly struggle with it. Some don't even seem to have the ability to fade treats, let alone fade props, bridges, and hand signals. First let me tell the Positive Reinforcement trainers, none of your clients want to use treats for very long. They have to know that you are committed to helping them Fade them out. Using treats too long is the number one complaint against positive reinforcement training.

The main reason that I am good at fading is due to my experience as a killer whale trainer. Imagine having to use a 30ft long target pole, then completely fade it out so the behavior can be performed during a show. The best advice I ever got was from Dawn Brancheau. I was training a break spin with two target poles and she told me "remember the more props you use the more you have to fade out." This is one reason I don't use target poles with clients, because they don't have the skills to fade them out and they don't want to use it forever. Also, I see trainers use high raised beds for a Stay. I get it, it makes training a Stay super easy, but remember, that tall bed needs to be faded out, so why not use a smaller bed that is easier to fade and easier to carry around.

Personally I like to fade bridges, treats and hand signals almost simultaneously or alternate fading them quickly. This is good for fading everything very quickly, but difficult for clients to learn. Clients need to be shown the fading process much slower or you will blow their mind and they will be confused.

Let's start with fading the bridge. I use a verbal bridge "YES". I believe a bridge is very important for a behavior like "Touch" while less important for a behavior like "Sit". This is due to touch only happening for a split second while sit happens for much longer so essentially you can just give a treat without bridging the behavior. I use the bridge much longer for "Touch" and fade the bridge for Sit within a few minutes. Basically once the animal understands the behavior I fade out the Bridge and just reinforce.

Fading the treat can be a bit more difficult. The start is to go to a variable schedule very quickly. I believe that trainers stay on a one-to-one schedule much too long. Right after the animal "understands" what the criteria is, put everything on a variable schedule. Sometimes jackpot, sometimes do nothing, sometimes just look at them in a nice way, sometimes release them, sometimes ask a behavior with high reinforcement history (Premark Principle). Basically get out of the rut of using a on-to-one schedule of reinforcement, the dog will love you for it. Doing this strengthens the behavior and allows you to fade everything quickly.