Jump to content
BC Boards

Annoyingly slow sits and downs, any tips?


Chanse
 Share

Recommended Posts

Archer is doing relatively well with all of his training, right now we are mostly just working on being able to relax and calm down, but that is a different problem. The issue that I want to bring up here is Archer's slow movements in response to a command.

 

Archer has no problem following the command and each command is only given once. So he sits when I say sit and goes down when I say down. He doesn't have a delayed response (i.e. the problem isn't that I give the command then he has to think about it before doing it). The issue is that he is just so slow with executing the command. When he sits he slowly lowers his butt to the floor. This happens all the time and is even worse when I try to train with toys. Mostly because he seems to not want to go into a full sit or down in anticipation of me throwing the toy. With a sit he seems to think that I might throw the toy early so sometimes he will pause half way down. This is even worse for downs. He will often just try to get away with going most of the way down but holding his chest off the floor.

 

I have never rewarded him for these partial sits or partial downs and I never release him until his butt or chest hits the floor. Also, if he does pause when doing a sit or down he won't be rewarded for that, he will have to be released and do another one.

 

I've tried increasing the pace of our play a lot to see if he would start going faster, but this hasn't worked at all. I will run/move quickly then give a command for sit...then everything slows down and I have to wait a few seconds for his butt to hit the floor. I've tried rapidly firing commands at him. So it would go sit - wait for butt to hit floor - release - sit -wait -release -sit - and so on. One right after the other and I was hoping that I could reward the faster sits. This failed miserably because he just got slower with each command and started to pause more frequently.

 

The only time he gives me really fast responses is once I get home and let him out of the crate. He is of course super excited and when I say sit his but slams to the ground. If I say down he slams himself to the floor as flat as he can be. I reward these a lot because that is what I want.

 

So I am hoping someone on here might have some advice for making my pup move a little faster. Right now my only idea is to go back to rewarding every sit and down, and maybe using a clicker to make him realize that butt on the floor is what gets the reward.

Edited by Chanse
Link to comment
Share on other sites

If you know when you are going to get a quick response I would not ask for a sit or down at other times when you know his speed will be under par. He sounds confused and he's getting opportunities to practise what you don't want.

 

How old is he? I'm finding it much easier with my youngster to work to his schedule of what he is ready to learn when rather than my own preconceptions of what he ought to be able to do.

 

Having trouble getting through to him with something? Shelve it for a while and come back to it when he's ready.

 

Lighten up. If you try too hard over something you make it more important than it is. Release the pressure on yourself and on him.

 

Do you clicker train?

Link to comment
Share on other sites

He's 9 months. Sits and downs are something that he has known for a while and there are only a few scenarios that I wouldn't trust him to not perform a sit. For example, passing another dog on a walk. But I don't try to ask for this behavior if I know I won't get it.

 

Also, I want to mention that sits and downs are pretty strong default behaviors already. If he wants something then he will usually go to a sit or a down, or keep trying both to see which one gets him a reward. And by default, I mean without a command.

 

Right now my training with Archer is mostly CU stuff to help his focus, impulse control, and to take care of his reactivity towards some dogs and skateboards. I haven't really been teaching any new tricks lately. I just try to reinforce behaviors that he already knows during play sessions (mainly sits, downs, stays, and a few tricks). I should mention that even for the slower sits or downs and for the ones that he pauses for, he still gets some verbal praise once his butt hits the floor. I try not to get upset with him for being slow, and since he did follow my command I still want to inform him that he did good. But for these I will withhold the larger reward (the toy) until he gives me a better response.

 

I'm trying not to stress it too much with him, but I of course get a little frustrated when I'm not getting the result that I want. I do my best to not transfer that to Archer, but he still probably realizes that something is wrong. So mum24dog, I should probably take your advice to lighten up.

 

I do clicker train, but haven't used it on sits and downs for months.

Link to comment
Share on other sites

I had the same thought as mum24dog - right now, only ask for the fast sits and downs when you know that he is going to give them to you. Also, if he is only going to give you one or two fast sits, then only ask for one or two, jackpot each one (as you are doing) and then either release him or go on to something else. When you think he might give you a third fast sit, then ask for it and see what he does. Work on increasing the # of sits and downs gradually.

 

Kiefer also will occasionally not want to put his butt on the ground if he thinks I am going to reward with a toy. He thinks a squat is good enough since he knows that if he sits, it will take him that microsecond longer to get to the toy vs. from a squat. Usually food treats work better for me if I want a faster sit.

 

Have you ever done puppy push-ups? It is a rapid paced sit-down-sit-down, etc. I make it into a game, and in the beginning, I may lure with a food treat and will also move my body to help him understand to go down and come up. Happy voice, great treats, luring, body motion. Approach it from the aspect of a game, and not obedience.

 

Also, he sounds like he might be bored? If he is working on the same few commands all the time, maybe you need to mix it up a little?

Link to comment
Share on other sites

Have you worked a splat down on mat? That might be helpful for the downs.

 

For the sits, a sit platform might be helpful.

 

Can you clarify splat down? I am assuming it just means getting to the mat and dropping on it, but I am wondering how you train the 'splat' part. I do plan on doing more mat work. His downs for his 'place' command are good. I generally do this with his dog bed and he will plop right down.

 

A platform seems like a cool idea. That might help turn sits into more of a game. I'll look into getting/building one.

 

He thinks a squat is good enough since he knows that if he sits, it will take him that microsecond longer to get to the toy vs. from a squat.

 

That definitely sounds like Archer.

 

 

ave you ever done puppy push-ups? It is a rapid paced sit-down-sit-down, etc. I make it into a game, and in the beginning, I may lure with a food treat and will also move my body to help him understand to go down and come up. Happy voice, great treats, luring, body motion. Approach it from the aspect of a game, and not obedience.

 

Also, he sounds like he might be bored? If he is working on the same few commands all the time, maybe you need to mix it up a little?

 

I've heard of puppy push-ups, I am almost certain that I did them when he was younger, not sure why I stopped. Maybe I should reintroduce them.

 

I was also worried that he might be getting bored. I try not to work him on these commands too hard during play, but sometimes I probably forget that we are playing and not training. I try not to focus on any one command when we are playing. We will do touches (he seems to love that one, he will poke my hand with his nose then whip his head to me looking for a reward), shake, and through (weave between the legs). Touch and shake are definitely his quickest commands. Through is still pretty new and he is getting better. Oh and I will occasionally throw in some heeling, no verbal command, just pat the side of my leg and I will have him heel for a few steps before restarting play.

 

It sounds like I should probably back off on these commands and train something new and exciting for him. Then probably work on sits and downs in a different context like the mat/platform training.

Link to comment
Share on other sites

If it were me, I'd try getting the clicker back out and clicking only when he responds promptly to the cue. If he's slow to respond, I'd use a no-reward marker like "too bad" or "oops" in a slightly upbeat tone (i.e. it's not scolding, just noting for him that he didn't do what you asked him to do, so he gets no reward) and walk away. A few minutes later you can try again. When he does respond promptly, click, treat and make a big fuss of praising him.

 

I expect you'll have good sits and downs quickly as a result.

Link to comment
Share on other sites

One thought - because this is something I have been working through with Rook. His response to arousal may be to go slower, glazing over a bit. I don't know if this is in any way related to being "sticky", which is a term I've heard used for dogs working sheep. But, anyway, it is a response to arousal, not being difficult. The key of course, is to get enough arousal that they are engaged with you, but not too much. It's a balance - one I'm still fine tuning.

 

Some ideas

 

1. Don't reward slow sits or downs, just say, "that was slow" or "you can do that faster", you can lightly push them out of postion or release them, move a bit quickly and ask for another one. Use a light, upbeat happy voice and attitude.

 

2. Play tug-sit-tug. Tug then ask for the sit/down and the second the rear hits the ground release and tug again. Again your motion and quick happy voice can be catching and the quicker he sits the sooner the fun begins again. If he is really slow - see # 1 - you can use it in this context.

 

3. Along the same lines but with food, "sit", give cookie as soon as rear hits the ground, release, run off a distance with puppy chasing and "sit" again, quick release, run, rinse and repeat. Quick and upbeat voice and movement so it is a fun game vs a drill.

 

Here is a little clip of Rook practicing tug, drop, tug. We also have some "stickiness" releasing the tug at times which you can see in this clip also - it happens more often when I ask for the release vs just popping it out of his mouth. BC are very interesting dogs to train as I'm finding out :)

Link to comment
Share on other sites

We do a foundation agility exercise of 'fast downs', where you have a toy out, are playing with it and getting the dog hyped up. You cue a down, then selectively reinforce with short play (tug is perfect for this game) if it's in the top 30-50% of speed for the behavior. If it's slow, you just say 'too slow', release and start again immediately. It's kind of like GentleLake's suggestion, but it keeps the game fast-paced (which is important when you want a fast behavior). The top 30-50% of reps is a rolling target; he should start getting faster and you can increase your criteria, but you still want to reward fairly often so he doesn't disengage.

Link to comment
Share on other sites

Here's an article on a method of improving latency: http://www.clickertraining.com/node/2129

 

Their use of averages (adding all the trials and dividing by the #) may not be entirely appropriate if the distribution of times isn't symmetric, unless you selectively throw out outliers. I would use the median or a fixed percentile instead.

Link to comment
Share on other sites

Chan - thanks for the video, those always help. Also your description of "stickiness" seems to be what I am encountering. This does seem to be a response to arousal because he is 'stickier' with toys than with treats (he isn't exactly fast with treats, but less 'sticky' would be a good description. He tends to be more relaxed when I am using treats, probably because he just doesn't care for food treats that much).

 

I like the idea of trying it as a fast paced game, but so far this hasn't yielded any results. I have also tried giving him a neutral 'too bad' (same use as the 'too slow') a few times and resetting. But this seems to result in Archer giving up on the sit entirely and he will just start pausing in a squat. If I wait long enough he will sit, but the waiting period gets longer between sits with no significant reward. I think I will have to start with GentleLakes idea of just aiming for one sit and if it is too slow then try again later. Trying the fast paced games may be frustrating Archer and that might be why he is slowing down. We may have to work up to the fast pace game.

 

As for latency, that's not really the issue I am having. Archer's latency is basically zero. Once I say sit his butt will start moving to the floor. The problem is the speed that he moves his butt :). He is good with verbal and hand signals. I am actually pretty impressed with his attention on me during play. His focus is on getting the toy, but when I have it he is alert to any signal that I may give. I may be able to incorporate the same idea of improving latency to improve the speed of the sit though.

 

Thanks for all the ideas so far!

Link to comment
Share on other sites

I like the idea of trying it as a fast paced game, but so far this hasn't yielded any results. I have also tried giving him a neutral 'too bad' (same use as the 'too slow') a few times and resetting. But this seems to result in Archer giving up on the sit entirely and he will just start pausing in a squat. If I wait long enough he will sit, but the waiting period gets longer between sits with no significant reward. I think I will have to start with GentleLakes idea of just aiming for one sit and if it is too slow then try again later. Trying the fast paced games may be frustrating Archer and that might be why he is slowing down. We may have to work up to the fast pace game.

 

I get the feeling that your criteria may be too high with the fast-paced game if you're resetting multiple times without rewarding.

 

You could also try making the game involve a sit/stay component, with the time he waits to be released directly proportional to how fast he sat.

Link to comment
Share on other sites

Splat Down

 

 

I train it by sitting on the floor with the mat in front of me. When the dog runs to the mat, I click/treat, and toss the treat away. After a few reps with the dog just getting to the mat, I start to wait to see if the dog will offer a sit or down. A sit gets a click and one treat. A down gets a click and a jackpot. No cues at this point.

 

Once the dog is running to the mat and going into the down immediately (history of jackpots), I repeat the process with myself in a chair in front of the mat, and then standing. Once the dog is running to the mat and going right down, I add the cue.

 

Video above is the finished product.

Link to comment
Share on other sites

Rereading your original post and the others these thoughts came to me.

 

For a week or two, only ask for a sit/down when you first get home and he's excited and you know you'll get a prompt response. Be prepared to give a really good jackpot (think chicken, steak, liver, something really good and 5 or 6 tiny pieces one after the other.

 

After a week or two of really good, prompt responses with jackpot rewards (and no asking for a sit/down at other times) then, ask for a sit/down at a time other than when you first get home. If you get a really slow response, "bummer", walk away, opportunity lost. Try again much later and see what you get. Be prepared to lower your criteria for the initial response in this new situation and gradually up your criteria (quicker butt to the ground). Remember to just ask once and either jackpot or "bummer". You can repeat a few times throughout the day but spaced well apart. The impact of one big jackpot can create a high desire to do the behavior, more so than repeating it several times, even if you're rewarding each time.

 

I saw this one jackpot reward method really work with Rook and the teeter bang game. After about a week of the one time big jackpots he would go over and want to do it when we were working on something else.

 

Chandra

Link to comment
Share on other sites

I get the feeling that your criteria may be too high with the fast-paced game if you're resetting multiple times without rewarding.

 

So I have definitely been trying to think about where my criteria is too high and I may have found a mistake I am making when doing the fast paced game. As I mentioned, I try to reward the fastest sit, but I think my reference sit is the problem. For example, when I ask for multiple sits in a row, I think I am basing each successive sit on the first sit in the series, instead of sits as an overall behavior. So maybe Archers first sit is relatively fast compared to all sits he has ever done, but I don't notice because I then ask for another sit to compare to that first sit. This is probably confusing him and might be why he slows down more during fast play. I didn't realize that I was doing this until I really thought about it. I definitely knew this was a result of me doing something wrong, but I couldn't figure out where I was going wrong.

 

I think I will start with Chandra's and GentleLake's idea of jackpot rewards for the very occasional sit and hopefully build up the value of this behavior. I will also bring the clicker back out to help communicate with Archer that the reward is for butt hitting the floor. Once I build up the value of this behavior more then I will reintroduce them to our fast paced games. Also, I will definitely work on downs in a different context (i.e. mat work), where I don't have to use the command 'down'.

Link to comment
Share on other sites

Chanse, do you ever reward with tugging? Would that give you a faster response?

 

Yeah, tugging is generally what we use for fast-paced play. I was doing quick 3sec tugs and then a command, with the occasionally longer tug session and/or toy toss. You are definitely right though, tugging does give a faster response then something like a ball. The ball makes Archer even more "sticky" like Chandra was saying.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...