[QUOTE=Equino;6585942]
I agree completely. I look at the clicker as a training aid to help teach/shape behaviors and then move on.[/QUOTE]
that is how most use it. Marker/reward training is for .teaching. the initial behavior so you have something to put on cue.
that is how most use it. Marker/reward training is for .teaching. the initial behavior so you have something to put on cue.
yeah, that’s what MOST people do- you use the clicker during stage I of training and then rapidly move on, so you don’t need the clicker later on or throughout most of the dog’s life, invalidating people’s concerns that they won’t have their clicker around when they need it. If you’re setting up the kind of controlled training session during which a clicker is most useful, you’ll have time to go get it, or even run out and buy a new one if necessary.
However, not everyone uses clickers the same way- I once met an idiot at a dog park whose dog was, like the OP’s, afraid of the clicker sound, so he proceeded to use the clicker as punishment- any time the dog did something he didn’t like, he’d chase the dog around and repeatedly click the clicker at the cringing dog. This use of the clicker didn’t seem to be very effective in producing a well-mannered dog. Yes, even clickers can be used abusively if you try hard enough.
Have to say, in response to this thread I went to PetCo on the way home and bought a clicker for $2 and tried that with my horse instead of my former practice of saying “There!”
He “got it” a lot more quickly with the clicker, and almost instantly trained up to the new reinforcement sound. It made enough of a difference that I am now officially willing to Carry One More Thing.
JME.
[QUOTE=meupatdoes;6587704]
Have to say, in response to this thread I went to PetCo on the way home and bought a clicker for $2 and tried that with my horse instead of my former practice of saying “There!”
He “got it” a lot more quickly with the clicker, and almost instantly trained up to the new reinforcement sound. It made enough of a difference that I am now officially willing to Carry One More Thing.
JME.[/QUOTE]
Cool. 
Not to hijack the thread, but what kinds of things are you working on with your horse?
I was thinking about this thread today after my phone pinged, alerting me to a text message.
I don’t get a lot of texts, but my brain instantly recognized and wanted me to respond to that ping. I didn’t, because I was working with my horse, but it took a conscious decision to ignore it!
On the other hand, I know for a fact that people can say hello or even talk at some length to me while I’m doing the same thing with my horse, and I don’t even notice they spoke–much less want to reply. (Not sure what this says about my manners, but anyway…:lol:)
I think that sort of unique, short sound really is recognized in the brain in a different way than verbal sounds, which are so common in the environment of any human or domestic animal.
[QUOTE=threedogpack;6586056]
that is how most use it. Marker/reward training is for .teaching. the initial behavior so you have something to put on cue.[/QUOTE]
One of the most difficult things for me in using positive reinforcement is fading the marker and reinforcer. I think I find it so reinforcing myself to mark and deliver a treat for a correct attempt!
It’s so counter-intuitive to have faith in the fact that random reinforcement (after the task is learned) actually increases the motivation. It’s one of the most valuable tools in PR, and completely unknown to most people unfamiliar with PR methods. Yet you can see it happen in the first session in something as simple as teaching a horse to turn their head away from you.
[QUOTE=MelantheLLC;6587960]
One of the most difficult things for me in using positive reinforcement is fading the marker and reinforcer. [/QUOTE]
I have to make a deliberate plan to click every other response when I begin to fade. I also limit the number of reinforcements available so I know how many clicks vs how many not clicks I do, even when I reinforce every correct response.
I cannot do random, it has to be deliberately thought out or I click every correct response and never fade the clicker.
For example, right now I’ve got a foster Puggle that is learning a retrieve. I count out 6 treats, click 1, 3, 5 and don’t click 2,4,6. Any more than 8 or 10 and I am hopelessly lost as to where I am, click or no click.
This does have another benefit as well, as it keeps each trial very short and increases the desire of the dog to work (he mostly thinks people are speedbumps on his road to whatever his wee wicked heart desires).
[QUOTE=MelantheLLC;6587940]
I was thinking about this thread today after my phone pinged, alerting me to a text message.
I don’t get a lot of texts, but my brain instantly recognized and wanted me to respond to that ping. I didn’t, because I was working with my horse, but it took a conscious decision to ignore it!
On the other hand, I know for a fact that people can say hello or even talk at some length to me while I’m doing the same thing with my horse, and I don’t even notice they spoke–much less want to reply. (Not sure what this says about my manners, but anyway…:lol:)
I think that sort of unique, short sound really is recognized in the brain in a different way than verbal sounds, which are so common in the environment of any human or domestic animal.[/QUOTE]
This is so true! I feel like I’m being trained by technology. Every time I hear my phone ‘ping’ my immediate instinct is to check my phone. I can pick out the sound of my phone getting a text from across the house or in the midst of a noisy crowd, when much louder noises don’t get noticed at all.
[QUOTE=Sithly;6587782]
Cool.
Not to hijack the thread, but what kinds of things are you working on with your horse?[/QUOTE]
Just silly things.
I recently retired him from “competitive work” because he wasn’t staying sound enough so rather than bute/injections/etc he is juat going to be a trail/hack around/clicker horse.
Currently we’ve been working on picking a towel up off the floor, and also “statue”. I’d like to be able to do some at liberty stuff with him too, bit towels first. 
I gave up on “being random” and went for “reward the better responses only” instead. I find it’s much easier and more intuitive to do it that way- the only thing you have to remember is what your subject’s mean performance was last time you worked on something. I think this approach produces better results than actual randomness, but have no actual proof.
Intermittent reinforcement
[QUOTE=threedogpack;6588104]
I cannot do random, it has to be deliberately thought out or I click every correct response and never fade the clicker.
For example, right now I’ve got a foster Puggle that is learning a retrieve. I count out 6 treats, click 1, 3, 5 and don’t click 2,4,6. Any more than 8 or 10 and I am hopelessly lost as to where I am, click or no click. [/QUOTE]
So what exactly do you do on the 2, 4, and 6 no clicks?
[QUOTE=Bicoastal;6592029]
So what exactly do you do on the 2, 4, and 6 no clicks?[/QUOTE]
I either mark it with a quiet good and treat, or just treat. I used to just treat, but with the Impossible Puggle I’ve got right now I have to mark it or he’s off to do something else.
he’s making me a better trainer, but he is also really stretching my patience…
anyone want a bad dog who doesn’t seem to retain what you teach him?
Intermittent reinforcement rewards
[QUOTE=threedogpack;6592463]
I either mark it with a quiet good and treat, or just treat. I used to just treat, but with the Impossible Puggle I’ve got right now I have to mark it or he’s off to do something else. [/QUOTE]
So the dog still gets a reward? If this wasn’t an Impossible Puggle…maybe it was an experienced clicker dog with an established behavior you are perfecting, would those ‘off’ ones still get a mark & reward?
Let’s say you’re working with a clicker to get a closer front…
And thank you threedogpack and wendy for your thoughtful answers. I cannot get this anywhere but COTH. I’ve tried!
ok, If I were trying to get a closer, straighter FRONT, using my clicker, I’d first note the “baseline” level of performance, and then when I called the dog FRONT, if the dog was just “average” in his front I’d probably just continue on and ask for a finish and/or other known behaviors- proceeding on with a new behavior tells the dog he’s doing just fine- and if his finish or next behavior is excellent, that gives you another opportunity to reward excellence (and keep motivation up) without making too much of a fuss about his sub-par FRONT. If his FRONT is better than usual, I’d click and treat. If his FRONT is fabulous, I’d click and jackpot and end the training session.
If you’re just teaching a front, I usually move from click n treat to verbal marker n treat to verbal marker with gradually fading out the treats. You want the click to be a strong cue so it always gets rewards, so you have to move away from the clicker to a verbal marker before you can start fading the treats.
So some people try to be “random”, but I can’t do that, so I try to give treats on the better efforts- so I’ve moved from clicker to verbal cue, and now I don’t feed treats on fronts unless I like the front a lot- the dog that comes in faster, straighter, closer, or with better attitude, gets the treats. You should fade out the verbal markers as well as the treats once the dog clearly grasps what to do- remember in formal competition you can only praise the dog at the end, so work towards that in practice. You’ll get a dog who will keep on eagerly working as long as you keep the commands coming because the dog knows a) that if you issue a new command, he did fine with the previous command, and b) at the end will come the payoff, and c) if you reward brilliance when it occurs, the dog will try to offer brilliance.
If you’ve made it a regular practice to reward only the best efforts in training, in competition where you can’t reward the dog you’ll get a dog who works harder and harder trying to achieve brilliance that will convince you to cough up a reward.
This is very counter-intuitive to many- I’m sure you’ve heard people say if you train with treats the dog will quit on you in the ring because you aren’t handing out treats. But not so.
-oh, and sometimes you have to do things differently with certain dogs. So watch your dog carefully. If the dog doesn’t seem to be responding “well” to what the “conventional wisdom” says you should do, try something else.
yes, I am quick to fade the clicker, less quick to fade the reward. And I only fade one criteria at a time. If I am going to fade the clicker, then I do NOT fade the food until I have already conditioned another reward to take it’s place. That would be a quick game of tug, a touch (of my hand) or something else the dog likes.
If this wasn’t an Impossible Puggle…maybe it was an experienced clicker dog with an established behavior you are perfecting, would those ‘off’ ones still get a mark & reward?
no, I would set the dog up to get what I wanted. I might use shaping if the dog is high and needs to work in order to calm down, or I might use a target if the dog is in the game, then fade the target. In this case, you are still training, so I’d use the clicker. I only fade the clicker when I am no longer training the behavior.
Let’s say you’re working with a clicker to get a closer front…
And thank you threedogpack and wendy for your thoughtful answers. I cannot get this anywhere but COTH. I’ve tried!
if you are training a closer front, I’d use the clicker and treats, as you need the precision of the clicker and you are still training. Does the dog shape well? If not I’d use my hand or a finger as a target to get the dog where I want him/her. Once you have an 80% success rate, fade the clicker, mark with a verbal and reward in position.
You are welcome! And BiCoastal, I have not forgotten the door behavior but need a new camcorder and don’t have the money right now. (and I’ve been working with Mr. Impossible on other things in order to place him as quickly as I can)
I am loving all the discussion on this thread! Very informative.
Heinz the GSD and I went to our first “Good Dog Manners” class last night. He was very good. Excited, but I could get him to pay attention to me as he is VERY treat driven. In fact, now I have to work on making eye contact instead of staring at the treat.
One of the things the instructor is working on is “loose leash walking”. I understand the progression involved in the training, but now my question is how do I integrate this into our routine? In other words - lets say I do 15 minutes of "click(cluck) treat reward for walking at my hip. Then we continue on our “usual” long walk. Does this not tell the dog that “once the weird having to stay close business is over I can go back to being a butthead on the leash” and pretty much negate the training?
I can’t just run this guy in the backyard while I work on the leash skills - I don’t have a backyard! So long walks every day are necessary.
I’m assuming at some point the behavior becomes ingrained, so polite walking takes over from pulling, but I guess I’m wondering how this happens in stages when we have to walk and train loose leash walking at the same time?
[QUOTE=Sunsets;6593095]
One of the things the instructor is working on is “loose leash walking”. I understand the progression involved in the training, but now my question is how do I integrate this into our routine? In other words - lets say I do 15 minutes of "click(cluck) treat reward for walking at my hip. Then we continue on our “usual” long walk. Does this not tell the dog that “once the weird having to stay close business is over I can go back to being a butthead on the leash” and pretty much negate the training? [/QUOTE]
I don’t think so.
Imagine a horse:
You spend 5 minutes at the walk.
Then you spend 5 minutes at the trot.
This does not negate your walk training, it just means that sometimes the human wants the horse to walk and sometimes the human wants the horse to trot.
Sometimes you want the horse to do the horse show step and sometimes you want him to do the add in the same line. He has to listen to what YOU want, not just regurgitate the same behavior at all times.
Sometimes the horse is supposed to follow you closely on a 6’ rope, other times cantering around you on the longe is desired.
So sometimes you want the dog to walk HERE, and other times it is ok if he walks THERE. Sometimes he has to be in a very contained leash jurisdiction, other times he gets a larger leash jurisdiction. Indicate as such with an “OK!” or a change in body language or whatever and the dog will be fine. The same way if your horse tried to canter off past you leading down the aisle you would be like, OH HAYULL NO, and if he cantered off on the longe based on the “Canter!” cue and sending-him-out body language you would be like, good boy!
Ooof. LLW is a tough, tough concept for many dogs. Dogs just do not seem to generalize this well.
To make it as easy as possible on all parties, I’d probably use different equipment for “pull all you want” and “don’t pull on me”. I’d use one collar for one set of criteria and another (perhaps a head collar?) for the other.
That said, if you want him (?) to begin to generalize this, I’d start with the “don’t pull” collar for X number of steps or till X point, then switch to the “we aren’t working anymore” collar.
Or.
You could work on LLW in a particular place (say a parking lot) and when you get back out to the sidewalk, switch equipment and that can be an environmental cue.
I think this is not the way to see it. I really, really doubt a dog is going to generalize this well.
Instead, I think of it more like a stud who gets ridden.
When you use this halter, we go breed.
When you use that halter, we’re riding.
Also, Sunsets? I do nearly all LLW work in my house, in my living room, off lead walking a small circle with the dog on the outside.
By submitting your email, you agree to our Terms of Service and Privacy Policy. You may also receive promotional emails from The Chronicle of the Horse. You can opt out at any time.
For Customers
Company
Terms of Use
