Puppy Forum and Dog Forums banner

1 - 20 of 34 Posts

·
Registered
Joined
·
28 Posts
Discussion Starter #1
I've started working with my 9 week old lab (Charlie). He isn't the first puppy I've trained, he's my second (LOL) so I am not too experienced at all. I didn't clicker train the first puppy because I was working through a seeing eye dog group and had to follow their training, which was mostly luring behavior with treats then fading treats.

I really want to use capturing his behavior as much as possible with Charlie and have been charging the clicker the past 2 days. Short sessions with him. But I don't know if he really gets it. He looks at the clicker hand when I click then looks at the treat hand and I give him a treat. Does it sound like he's got it?

On the other hand, he is doing amazing with Zen. I'm kind of in shock that he gets it already. :clap2:

Thanks so much in advance for any advice :)

Lisa
 

·
Registered
Joined
·
4,350 Posts
It sounds like he's getting a connection right now: first comes the click, then the treat. That's a good start! The thing Harper has been doing lately, is focusing more on the sound, and looking at me. He doesn't look to the other hand to get the treat. The treat doesn't necessarily have to follow immediately, because he's confident he'll get it in a second. So, it's kind of like the click is really what he's after, because he knows what will eventually come. It doesn't have to be immediate with him looking at both hands! Does that make sense? Probably not, lol!
 

·
Registered
Joined
·
28 Posts
Discussion Starter #4
I've been making sure he's already focused (more or less) on me when I'm doing this so I will try it next time when he's not looking at me and see what happens. I'll also try giving it a second before giving him the treat and see if he looks at my face. Pretty much I've been following the advice from Pat Miller's The Power of Positive Dog Training: "Click! the clicker and pop a treat into your dog's mouth."

:sigh: I just want to get this right, you know? BTW, my avatar is of my two dogs. Charlie is the puppy, and the older lab is Kiwi.
 

·
Registered
Joined
·
869 Posts
Actually, there's no point at which he "gets it", meaning there will never be a point when he's at 100% forever. He can have a strong charge, weak charge, or non-existent charge. There's no absolute charge. Your goal is to keep the charge strong, and you'll know when you click at he makes a beeline for your hand.
 

·
Registered
Joined
·
4,977 Posts
you'll know when you click at he makes a beeline for your hand.

That would sound like the dog "gets it" to me.

Yet you say there's no point where he'll get it?

What part of the above behavior would be the dog NOT getting it?
 

·
Registered
Joined
·
869 Posts
That would sound like the dog "gets it" to me.

Yet you say there's no point where he'll get it?

What part of the above behavior would be the dog NOT getting it?

In training, we're not concerned about knowledge and understanding. I'm not saying dogs don't understand stuff, it's that until we know how the brain works, we'll treat them as a black box where we simply observe frequency of behavior. If you started clicking everyday and nothing happened, eventually the click's charge will become weaker and weaker. We do know what he "gets", so we don't care about that.
 

·
Registered
Joined
·
3,722 Posts
I usually hold the clicker behind my back, because I don't want the clicker to be a visual thing, I want them to respond to the sound, not my finger moving. This is mostly because when I do shaping with them sometimes my finger twitches on the clicker, because I almost clicked them, but then decided it wasn't good enough and didn't click after all, but because the dogs saw me 'almost click' they think they are getting a treat.

Wait for your puppy to be playing or do something away from you, then click, and see if he looks up at you or comes running.

Shaping puppies is sooo much fun:)
 

·
Registered
Joined
·
28 Posts
Discussion Starter #9
Thank you!!! Will definitely put this one to use as well!

I usually hold the clicker behind my back, because I don't want the clicker to be a visual thing, I want them to respond to the sound, not my finger moving. This is mostly because when I do shaping with them sometimes my finger twitches on the clicker, because I almost clicked them, but then decided it wasn't good enough and didn't click after all, but because the dogs saw me 'almost click' they think they are getting a treat.

Wait for your puppy to be playing or do something away from you, then click, and see if he looks up at you or comes running.

Shaping puppies is sooo much fun:)
 

·
Registered
Joined
·
3,109 Posts
I don't worry too much about charging the clicker. Barring special circumstances (noise sensitive dog for example), this is not a particularly challenging concept for the dog and over the course of the dog's life he is going to experience so many click-treat pairings that he's going to "get it" deep in the doggy fibers of his soul. I would say you have enough of a correlation to move on.

I understand wanting to get it right, I'm very picking in my training mechanics as well, but one of the great things about clicker training is that it's very resilient to fixing mistakes you might have made. You just change what gets reinforced!

If you haven't come across it already Susan Ailsby's Training Levels is pretty much the best thing on the internet, and an updated book form is due very soon.
 

·
Registered
Joined
·
28 Posts
Discussion Starter #11
Thank you RaeganW for the link, it's one I haven't seen. GAHHH!!!! I have so much to read LOL!!

I know I'm probably over-doing this - I am currently loading the Protocol for Relaxation days and each step into my iPhone's todo app - but I'm having fun. I also created a database to track the puppy training and my older dog's continuing education on my iPhone as well. You can call me a nerd, it's ok :) I was going to use a notebook to journal the progress but I figured since my iPhone is my third hand and most of my brain I might as well put it in there.
 

·
Registered
Joined
·
5,406 Posts
In training, we're not concerned about knowledge and understanding. I'm not saying dogs don't understand stuff, it's that until we know how the brain works, we'll treat them as a black box where we simply observe frequency of behavior. If you started clicking everyday and nothing happened, eventually the click's charge will become weaker and weaker. We do know what he "gets", so we don't care about that.
LOL. Probably the reason I'll never get to be a total clicker purist (though it is a tool I use and like an awful lot). I simply can't visualize a dog as a black box, and training as simple i/o would bore me to tears. I am very concerned with knowledge and understanding, and I look at training as a conversation between me and the dog. Frequently the most interesting part of the training session is not the dog increasing his frequency of behavior, but the questions he asks and the suggestions he makes. That's what keeps me involved - not charting frequency and calculating reinforcement schedules. I train because I LIKE dogs and I like being able to set up a common language with them. I guess I'll always be an artist instead of a scientist.

I don't worry too much about charging the clicker. Barring special circumstances (noise sensitive dog for example), this is not a particularly challenging concept for the dog and over the course of the dog's life he is going to experience so many click-treat pairings that he's going to "get it" deep in the doggy fibers of his soul. I would say you have enough of a correlation to move on.

I understand wanting to get it right, I'm very picking in my training mechanics as well, but one of the great things about clicker training is that it's very resilient to fixing mistakes you might have made. You just change what gets reinforced!

If you haven't come across it already Susan Ailsby's Training Levels is pretty much the best thing on the internet, and an updated book form is due very soon.
I second Sue Ailsby's Training Levels! It's a great, easy to follow step-by-step program to a well trained dog, and also to a good foundation for any kind of competition. And the list is very supportive (also she has some great articles on her website www.dragonflyllama.com) She's also blogging about her new puppy right now! Sue thinks that charging a clicker is un-necessary, and that while it sends the message "you'll get something good for that!" it misses the other very important part of the message "I like when you give me that behavior". I don't charge, I just start with a simple behavior like the name game or targeting fingers, and usually the dog is starting to put it together in a few repetitions.
 

·
Registered
Joined
·
869 Posts
LOL. Probably the reason I'll never get to be a total clicker purist (though it is a tool I use and like an awful lot). I simply can't visualize a dog as a black box, and training as simple i/o would bore me to tears. I am very concerned with knowledge and understanding, and I look at training as a conversation between me and the dog. Frequently the most interesting part of the training session is not the dog increasing his frequency of behavior, but the questions he asks and the suggestions he makes. That's what keeps me involved - not charting frequency and calculating reinforcement schedules. I train because I LIKE dogs and I like being able to set up a common language with them. I guess I'll always be an artist instead of a scientist.
I'm concerned with knowledge and understanding as well. Only, the knowledge and understanding is happening on my end, not the dog's end.

I'm not a clicker purist by any means, I'm not too experienced at shaping, but I set myself to a pretty rigid set of scientific principles. There's a common misconception that clicker training is cold, boring, lifeless, uninvolved. That's exactly the opposite of how I see it. In shaping behaviors, usually done with a clicker, the dog is 100% involved in his learning. The dog is offering new behaviors, readily and willingly. The human is equally involved - the human is observing behavior and perfecting his/her timing. There is totally a conversation happening in clicker training, and it's every bit as rich as you would like. I find that approaching training in a scientific way first allows me access to the artistic side.
 

·
Registered
Joined
·
4,350 Posts
I understand what qingcong is saying about a strong, weak, and no charge, but I still believe you can tell if a dog is basically "getting" it. We started Doggy Zen, and have done some other trainings, too, that say, when there is ___% accuracy you can move on, or when they get 5/8 right, you can move on.
I agree there is no total "complete" charge, necessarily, but maybe because I'm not a purist in anything, I feel like I can make the generalization that my dog "gets" it.
I also agree with Raegan not worrying too much about charging the clicker....if there is a lot of C/T going on, the dog will see the connection there, easily enough.
 

·
Registered
Joined
·
5,406 Posts
I'm concerned with knowledge and understanding as well. Only, the knowledge and understanding is happening on my end, not the dog's end.

I'm not a clicker purist by any means, I'm not too experienced at shaping, but I set myself to a pretty rigid set of scientific principles. There's a common misconception that clicker training is cold, boring, lifeless, uninvolved. That's exactly the opposite of how I see it. In shaping behaviors, usually done with a clicker, the dog is 100% involved in his learning. The dog is offering new behaviors, readily and willingly. The human is equally involved - the human is observing behavior and perfecting his/her timing. There is totally a conversation happening in clicker training, and it's every bit as rich as you would like. I find that approaching training in a scientific way first allows me access to the artistic side.
Then I'm guessing you really don't look at dogs as black boxes? I think training black boxes would be boring. And, if I were trying to sell someone on the idea of this kind of training, I suspect telling them to look at their dog as a simple input/output object would be extremely off-putting. Hopefully learning and acquisition of knowledge goes on both with you and the dog, and of course, like with any learning, if you don't use it you lose it. So nothing is ever completely "learned." It is always a work in progress.
 

·
Registered
Joined
·
4,977 Posts
I'm concerned with knowledge and understanding as well. Only, the knowledge and understanding is happening on my end, not the dog's end.
I'm the exact opposite.

I already know what I know. I want to know what Wally knows or doesn't know and how he's expressing the "gaps" in his knowledge, the parts of the tasks he doesn't understand and why he might be making those incorrect-in-my-mind choices.

If that makes me unscientific, so be it.

I use enough of the science to help understand what's in his mind. That's why I understand operant and classical conditioning and am trying to go beyond that as well.

I believe in using behavior to see inside the "black box". To me, that's the real thrust and point of behaviorism.


In shaping behaviors, usually done with a clicker, the dog is 100% involved in his learning. The dog is offering new behaviors, readily and willingly. The human is equally involved - the human is observing behavior and perfecting his/her timing.
You say trainers aren't considered about if the dog is getting it - yet the whole point of shaping is seeing the dog getting parts of the whole in progressive sequence towards the goal you have in mind, and then pushing the dog to extend what he already understands to develop the next step.

So I would say a trainer using shaping absolutely is concerned about the dog getting it for each and every step and then the dog's getting it in putting the pieces together.

If you started clicking everyday and nothing happened, eventually the click's charge will become weaker and weaker.
In other words, the dog is becoming more uncertain about the association he made originally.

It's the same principle as why desensitization works. The original X = scary becomes more uncertain when X = good stuff starts happening.

If click starts equalling nothing, the dog becomes less certain that it means food is coming, so you get less "sharp" behavior (the whiplash, etc) because he's not as certain.

Uncertainty comes out in behavior, so seeing less frequent/less intense/conflicting behaviors is seeing uncertainty in the mind.
 

·
Registered
Joined
·
5,406 Posts
I'm the exact opposite.

I already know what I know. I want to know what Wally knows or doesn't know and how he's expressing the "gaps" in his knowledge, the parts of the tasks he doesn't understand and why he might be making those incorrect-in-my-mind choices.

If that makes me unscientific, so be it.

I use enough of the science to help understand what's in his mind. That's why I understand operant and classical conditioning and am trying to go beyond that as well.

I believe in using behavior to see inside the "black box". To me, that's the real thrust and point of behaviorism.

.
That's sort of where I am at. I see the science as necessary, but a "means to an end." And, as Bob Bailey is fond of saying, clicker training is a MECHANICAL skill.
 

·
Registered
Joined
·
1,598 Posts
You will know when your significant other is a doofus and clicks the clicker absentmindedly one night and the puppy comes BOLTING to you for a treat.

>.>
 

·
Registered
Joined
·
4,350 Posts
You will know when your significant other is a doofus and clicks the clicker absentmindedly one night and the puppy comes BOLTING to you for a treat.

>.>
LOL! OR, when the little kid brought to training class to watch randomly starts clicking right next to you!
 

·
Registered
Joined
·
1,598 Posts
LOL! OR, when the little kid brought to training class to watch randomly starts clicking right next to you!
Or when you're watching a dog training show that involves clicker training and you find yourself rapidly muting the television b/c puppy will come ZOOMING at you if he hears even the faintest t.v. clicker click!
 
1 - 20 of 34 Posts
Top