I was having probs with my little exmoor beasty stopping at junctions (he wouldn't!) and so I set about using CT to get him to 'Stand' on command. Worked absolutely perfectly, in one session too, so I was really pleased.
However, on the very next ride I found that he kept stopping for no apparent reason, until I realised that whenever I moved my hands (scratching a leg, signalling to turn off road, pretty much anything!) he stopped dead, bending his head round to take the treat which he thought I was reaching for. He's obviously a bit too smart. I have now relinquished all treat giving since last Friday, although the problem was worse on Sunday, when he stopped in the middle of a busy staggered junction (where someone smacked into my car last OCtober!) and went on strike. I had to use mu whip, reserved for rabid dogs and weirdos in the woods normally - not good!
Has anyone got any ideas? I spent the remainder of the ride in the woods giving signals like a loony, whilst sending him forwards at the same time. I also moved my hands to my pockets and legs, ignoring his head bending. He definitely recognises the cue, as when I came to the next road and said 'Stand' he stood like a good boy (and got a scratch as reward). It's almost as though the bridge between cue, bhr and click/treat has been lost, although I've never treated him out on rides for anything other than a stop after the command.
is to actually ask for a stop whilst out hacking as you go along, somewhere safe and not hindering anyone, and click and treat for that behaviour. Can you put the treats somewhere else? Is he thinking that hand in pocket is the bridging signal rather than the click? If so you must have been inadvertently doing something in the training that makes him think like this so you might have to start all over again.
Hi Sutton,
In fact, that's how I taught the stand, asking for it at random times, a bit like when you train a dog to come to call. As soon as he munched the treat, I asked for a walk on and no probs.
I had only done one session with treats full stop, all rides after that didn;t even involve a jacket, let alone pockets. He would stop if I literally moved my hand away from his neck e.g. to signal or open the rein for a turn.
I'm still convinced that while you were teaching him with cue/command, click, treat, something got identified in his brain that if you moved your hand that was the bridging signal and treat would follow. I have had the same problem with getting my horse to back, I just face him and walk a step towards him and he automatically goes back. So now I find that if I want to stroke him I have to appear from the side otherwise we go continually back as that is what he thinks I want. I've miscommunicated something and I think you have too so we have to find how we've done it and try to put it right. I've now found that if I say stand he will stand and face me and if I say back he will back up so I've just used different words. Try to think of every step that you did and see how you think he might have misconstrued it.
Afraid I can't help agreeing with Sutton - it sounds like a "Clever Hans" situation. You've taught the behaviour and put it on cue extremely quickly, especially if you haven't done much CT work before. One thing that we always have to remember is that the horse is doing what he's been trained to do and we always have to compare what we think we've trained with what the horse is showing us we have actually trained. It's difficult to suggest specifics without seeing what Henry is doing but possibilities are that "stand" is acting more as a bridge than a cue, you may be cueing/bridging him with some subtle body-language you are unaware of or his training is incomplete and he doesn't actually understand the behaviour. It certainly isn't properly on cue or he wouldn't offer the behaviour in the absence of the cue.
The other thing that seems relevant are the rules of shaping (have you read Karen Pryor?) - you have a lot of variables going on. You're training two behaviours really, stop and go. You're also expecting him to generalise to a hacking environment after just one training session and incorporating a variable schedule of reinforcement - ie sometimes you stop for a longer period, sometimes you stop after more strides of walking etc etc You're also combining +R and P - sometimes he stops and gets a treat but on one occasion he stopped and got a smack - I doubt he understood what the smack was for because in his mind he was doing what he thought you'd asked him to do. (I can appreciate why in a traffic situation but CT is about trying to see it from his perspective)
Sorry, I don't mean to sound critical - just trying to point out that you've started with something pretty complicated! Been there done that! I originally tried a bit of CT out hacking as well, trying to mark a nice walk with lots of impulsion. But I soon gave up when I realised that to do it properly we would need to do lots of hacks with only a couple of strides before stopping dead!
In order to progress I would go back a bit and start with an easier behaviour. That way you can really get a feel for achieving a behaviour using +R, then stretching it a little using a vsr and then putting it on cue. Something like targetting or head down would be ideal - something he can offer of his own volition. Once you've both got the hang of this then you can start to retrain other things using +R (sorry if you're already been doing some of this - I understood from your post that this was the first thing you've tried). The easiest way to re-introduce the "stand" cue might be on a loose lunge in the field. Teach "walk" and get it on vsr and he will obviously stop to get the treat. You can then start to lengthen and click/treat the stopping periods and eventually pair them with a cue. And all the "relax old criteria before introducing new criteria" type rules apply. When you're transferring this to ridden work you'd be much better off starting off in the same place you did the lunge work before generalising it to other locations. Try to be strict with yourself about not mixing +R and -R/P as it really makes a huge difference to motivation and reasons for achieving the behaviour.
I know this sounds horribly long winded and tedious but it is all about putting the groundwork in place. I've still not really done any ridden CT although we are just about to start. That's after about a year of CT playing - although to be fair we could have started the ridden work a while ago but we're just having lots of fun doing groundwork. Please don't let what I've said put you off - I promise it's worth persevering with! There's nothing like that "lightbulb" moment when you can really see your horse's brain whirring!
Hmmm, I could go with the 'Stand' being a bridge theory. Agree totally about mixing P and +R. The junction in question though really is unsafe. Believe it or not, I had moved out starting to turn right and yet people were STILL overtaking me, coming from the other side of the staggered junction! I had one idiot nearly take Henry's nose off the other week as I turned right onto the bridleway which is immediately after the same junction. Still was mentally kicking myself though and was not happy!
Yes, I have done some groundwork, but not much I have to confess.
As for seeing Henry's brain work? That pony's got more brain than I have, for sure.
I can't offer much help I'm afraid but I can reasure you that you are not the only one this has happened to. I have been training Crunch to 'freeze' so that I can get on him bareback without him moving and me ending up headfirst on the other side
I still ride him out conventionally which was working fine with the ct ridden training being left only to work in the school. However once I started training 'Freeze' he doesn't want to move....I know he is getting confused and probably testing the boundaries a bit too and I think I am going to have to take the plunge soon and transfer ct work to all aspects of our riding.
I'm trying my hardest with the 'freeze' work not to get the behaviour onto a long duration variable schedule, making sure he moves somewhere else before I ask for the behaviour again...I guess the best thing to do is not to get into the trap of rewarding a long duration 'Stand' so that after a couple of minutes he realises that the bhr isn't achieving anything and does another bhr that gets him lots of praise (be careful not to set up a chain though!)
what you've both said reminded me of the Alexandra Kurland video (3 or 4) in which she says to train behaviours in pairs. ie head down and head up, stop and start etc. I found this quite a useful concept, especially if you're trying to train halt. When you train walk it is easy to get the halt because the click ends the behaviour and the horse stops for the treat. But if you start by training halt then it's harder to get the movement in order to halt. So maybe it is easier to start with moving (assuming of course the horse understands that the click ends the behaviour and WILL stop!!)
Yep, that make sense...trouble is, I have trained both walk and stand together and he can transfer between the two - it's just that I think Crunch is trying to take the ct work to riding out and as freeze was the last thing that I started to train him, he thinks that standing still will always get him treats (which of course I make sure it doesn't). It's a pitty becasue if the bhr was trot I'm sure he would give up quicker than freeze in antici[ation of a treat...think I am going to take the plunge and transfer ct to all our ridden work now...ahhh!
I haven't had time to do any groundwork, but can tell you all that I had a VERY long and uninterrupted ride on Saturday, where the command 'Stand' was enough for stopping, without a click OR treat, so I shall continue to leave the CT for a while whilst I'm on board! We even used THAT junction without incident, phew!
