Dog Forum banner
Status
Not open for further replies.
1 - 10 of 10 Posts

·
Premium Member
Joined
·
7,969 Posts
Discussion Starter · #1 ·
I will preface this by saying, I will ask my trainer about this on Tuesday but I'm looking for suggestions to try in the meantime.

As I'm working with Aspen on some new stuff, I've found that I've made some mistakes in her basic stuff.

When I first got her, she jumped a lot (like all puppies). "Sit" was one of the first things that I taught her. From there, I taught her to stop jumping by having her sit before she got pet/played with. When she sat and didn't jump, I'd give her a treat.

Now, when we're learning something new, she wants to sit. I'm trying to teach her to "stand" but she parks her fanny right on the ground. I'm also trying not to give her treats every single time she does something, especially a command that she knows, like "sit". She's catching onto that and won't always do it if there aren't treats to be had.

So, after all of that rambling, I guess my questions are:

1) How do I get her to stop assuming I always want a "sit"?
2) How do I get her responding to me when I do not have treats?
 

·
Registered
Joined
·
719 Posts
Aspen, personally with my pups I found it easier to teach " down" from sit as a second command. But the trick for stand when she's sitting take the treat hold it in front of her nose, but in such a way she can't take it from you. Then slowly pull it forwards, hopefully at some point she should lift her butt and that's the moment you click/ treat.
As for not responding if she doesn't see treats, have you tried variable reinforcement? She must think there's at least a chance that she'll get a treat, even if she doesn't see it.
 

·
Premium Member
Joined
·
7,969 Posts
Discussion Starter · #3 ·
@Bordercollie - thanks, I'll look up variable reinforcement!

She does know down.

Her list of commands that she knows are: sit, stay, down, heel, come, watch me, around and touch.

This week, we're working on "stand" and "back".
 

·
Premium Member
Joined
·
11,875 Posts
1) How do I get her to stop assuming I always want a "sit"?
My guess is that she doesn't really quite understand what it is you want. Often if a dog is confused, they'll guess and default to a behavior that has been highly rewarded in the past.

My puppy sits and sometimes downs as well. I normally get him up with a hand target or just taking a step or two away encouraging him up often with me patting my leg and an upbeat ''hey pup pup'' as he's a really sensitive guy. With him that sit or down tells me that I need to adjust what I'm doing to make him successful. Much of the time this issue pops up when I've made to big a jump in criteria. Also keeping a high rate of reinforcement during the initial learning phase is super helpful with my boy.

I actually want a default stand for training purposes so I'm more or less just capturing and rewarding standing near me as well as staying in a stand. I watch my reward placement to encourage stands. Sits aren't rewarded unless asked for. And like I mentioned above, if he does sit or down I just get him back up with a hand target or moving away.
2) How do I get her responding to me when I do not have treats?
Take a look at how you're using food. Make sure you've made the switch from lure (food in hand causing the behavior) to reward (behavior first then food coming out of pockets or other put of sight location). If the switch isn't made then the dog normally doesn't respond without seeing the food first.
Also once the switch has been made, be sure that you do not reach, get out the food, and ask again if she fails to do as asked the first time! Super common mistake that also often results in a dog that will only respond when they see the food.

And start switching up your rewards for known behaviors. So for sits, sometimes a cookie, sometimes praise and petting, sometimes play, etc. :)
 

·
Registered
Joined
·
719 Posts
Aspen, wow sounds like you're doing really good! Someone on this forum once sent me a link to her blog where she has an amazing post on phasing out treats. I'll see if I can find it and send you the link.
 

·
Registered
Joined
·
2,711 Posts
I like to start with known commands and treats to reinforce that I have treats, you can earn them and also "yay! you know things!" (but my dog has a very low tolerance for frustration, so he has to have 99% success rates).

So I start every session with sit, treat, look at me, treat, touch, treat, then I move onto new things.

Now, she may not be trying to get treats out of you. It's possible she just doesn't know what you want, so she tries a sit. She wants to be right, you've previously established you like sit, so she tries a sit. So I'd try teaching stand a different way and see if that makes a difference.
 

·
Premium Member
Joined
·
7,969 Posts
Discussion Starter · #7 ·
Thank you, everyone!

Aspen REALLY wants to make me happy so I know this is something I'm doing wrong and it's not her being stubborn! Haha. She also honestly enjoys training. Her ears are up and her tail is wagging the whole time :)
@kmes - your pup sounds a lot like Aspen. I'll try your suggestions tonight. Thanks!

I'm SO HAPPY that I decided to use PR with her when I got her. She's such a sensitive, biddable dog that I really think she would just break down if I tried any other methods with her. I feel bad if I've made mistakes and confused her while I'm learning.
 

·
Registered
Joined
·
316 Posts
1. Reward a whole bunch of other behaviors! Asking dogs to try something new is asking them to take a gamble...and gambling is more fun when you "win" more frequently. Not only does a high rate of reinforcement boost confidence, it reminds your dog that working with you is really, really fun.

My dog also defaults to a "sit" or "down," for the same reason -- they are behaviors I have reinforced to an astonishing degree in a huge variety of contexts. So they are pretty safe bets to make, and usually a sign that she has no actual idea what I'm asking of her. It reminds me to take a step back and break the behavior up into smaller pieces, and to break our training up with more movement in general...I like to train for just a minute or two and then run around playing, or throw in some tug, or scatter a few treats on the ground and tell her to "find it," or otherwise take off the pressure for a bit.

If it helps to hear, learning "stand" after being reinforced heavily for sits and downs is hard for a lot of dogs :)

2. Variable rate of reinforcement, for sure. And try to prevent her from knowing whether you have treats in advance anyway. For me, I try to have treats nearby all the time (and if I don't, I usually have some other form of reinforcement I can plug in), so that my dog learns not to pay attention to whether I have treats. Dogs are geniuses when it comes to figuring out predictive patterns, and if they learn that the best predictor of getting a cookie is you having a cookie, that's the thing they'll pay attention to. I want my dog to learn that the best predictor of getting a cookie is her behavior, so she learns to think about that instead! That means paying attention to my training mechanics, so that I have the cookies out of sight (and hopefully out of mind) until after the behavior has finished...training myself not to reach for a cookie halfway through a successful behavior is harder than it sounds!

To vary your rate of reinforcement, start by making just a small change. If you're currently reinforcing every time she sits, for instance, start by skipping only every sixth time (but make it less predictable than that. Skip the sixth, then the fifth, then the seventh, then the sixth, then the fourth, then the seventh...we humans are creatures of habit, and tend to fall into predictable patterns no matter how hard we try!). Reduce that rate very slowly, and only as you see her enthusiasm remain high. I like kmes suggestion to mix up rewards too...there are so many ways to keep an eager dog happy about playing training games! Another way to vary reinforcement is to only reinforce the "better" behaviors -- the top 50% of "sits," for instance. Your trainer can probably help you figure out a good schedule for your specific needs.

Don't feel bad about making mistakes. Mistakes are how we learn! Training is really just a conversation between us and our dogs, and figuring out how to clear up a miscommunication can be even more rewarding than magically getting it right the first time. Sounds like you and Aspen are a great match.
 

·
Premium Member
Joined
·
11,875 Posts
Lol
Yeah, Leggs drops his butt into a sit or into a down rear and tail wagging too. Very much a guess. He'll pause and clearly think. Then guesses. Will this work? What about this? He also sometimes reverts to pawing if there is an object involved.

In one way it's a good thing. Them offering different behaviors to see what works means they aren't too terribly frustrated and are still working. It's a good thing imo for these soft guys to learn that it's no biggie if they get it wrong and that they just need to try again. Now if she stops trying, loses her interest in training, starts excessive frustrated vocalizations... then it's time to worry.

And def don't feel bad at all. If she is doing this for reasons similar to my guy, just consider it feedback for you that what you're doing isn't quite right or what Aspen needs to understand what you're trying to teach. As long as you listen and adjust it's not a problem.:)
 

·
Premium Member
Joined
·
7,969 Posts
Discussion Starter · #10 ·
Thanks so much!

It's normal that I want to leave work early and go practice some of this stuff, right? haha
 
1 - 10 of 10 Posts
Status
Not open for further replies.
Top