ok I'll admit it I've been occasionally lurking on here for a while - a refugee from the IHDG looking for a bit of informed debate. I have really enjoyed some of the thought provoking stuff on here so I hope you don't mind me joining in.
I have been using clicker training with my two gees for about a year now and have had some good results with the simple stuff (ground tying, backing, turn on the fore-hand, etc)but I'm not getting very far with more complex behaviours. To try to get over my failings in this area I have been investigating other training techniques that would work well in conjuncion with +ve reinforcement. I have found Klaus Hempfling's (sp??) work really fascinating and (before I launch into a long discussion on the subject) wondered if anyone else has tried his techniques and what you thought?
Hi Clare - glad to meet you! Have watched a video (think it was dancing with horses) and at the end can honestly say that I was none the wiser about his techniques! Would be interested in knowing more about them. I would say though that you have to be very careful to be aware of mixing +ve and - ve reinforcement in training.
As for CT - what exactly are you having problems ie what are you trying to train?
Jan
Thanks for the welcome Jan. I'm interested to hear that you've come across problems combining -ve and +ve reinforcement. I would like to hear your experiences.
I sometimes use mild -ve reinforcement to initially prompt a behaviour. For eg when first asking for turn on the forehand I use strong body language towards the hindquaters or gentle pressure on the horse's body to get the initial shift of weight that I can then ct. After the horse has offered the behaviour a couple of times I would reduce the pressure until it becomes the cue. I never escalate the pressure, just repeat or try alternatives until I get the required response. I have found this method a lot quicker and less frustrating for the horse than just waiting for the bhr to occur.
Klaus Hempfling uses a combination of body language and reward to train horses. He differs from Monty and co in that he never uses physical pressure when on the ground. He uses minimal classical weight and leg aids when riding. He is in the strictest sense using -ve reinforcement in his body language but this is refined into very subtle cues. I don't know where the dividing line between -ve reinforcement and cues can be drawn.
I have tried to use +ve reinforcement to encourage Jazz (my partner's gelding) to work in a more correct outline. I find his tries and therefore opportunity for reward are so far apart that he gets frustrated and doesn't suss what is being rewarded. This is obviously my fault as a trainer not his - but I have really failed to acheive anything other than at halt either ridden or on the lunge.
Enter Klaus' book. He uses a small areana to free school. The horse is not restrained by physical pressure and is therefore able to balance better. Jazz worked in a perfect outline in the second session - I was absolutely gobsmacked!
This got me thinking that a combination of Klaus Hempfling's method plus properly delivered +ve reinforcement could work very well. Anyone any experience of using his methods. What are the draw-backs - there have to be some!
Have you tried to incorporate voice cues? If you introduce this on the ground initially with the click and treat. ie. click + voice cue + treat. You can use any word you like and then once the voice cue is learnt you can take away the treat occasionally. Say every other click to begin with. So that you can do your ridden work without having to stop every few strides to give him his treat!
The other option is to chain some behaviours together. For example say you wanted to open a gate from his back then you would break down the sequence of behaviours and teach them one by one, so that you gradually build on from the inital behaviour until he is performing several different behaviours for a single treat.
Each time it gets too difficult for him and he gets frustrated go back a step and check that he has learnt the previous behaviour properly and also break down the next step even further.
Great to see you on here (apologies for the belated welcome - don't seem to have enough time to keep up with posts these days!). And just out of noseyness, are you the same Clare with Esme who lives on Anglesey and has JaneB trimming for you - or have I mixed you up with someone else?
I really liked Dancing with Horses too and agree that it would mix really nicely with CT. I've read the book and watched the video - the video isn't very "how-to" but the book goes into much better detail.
I think for me this book was what gave me the stepping stone between IH and CT. In IH we were always encouraged to lead a horse on a loose rope but as I'd always done this anyway it didn't have much impact on me. But Hempfling's approach of encouraging you NEVER to let the horse come to the end of that loose rope really helped me become aware of how much we (I!) nag our horses. I realised that if Jak was never going to take up the slack in the rope then I had to train him one hell of a lot better! Even without using CT for leading, just by stopping dragging him along and letting him choose his pace he re-became a horse who was a pleasure to lead rather than one I needed to carry! aaaaarrgghhh, it's just all so obvious now!
My only problem with Dancing with Horses is the way he gets the movement in the first place, ie non-contact P/-R. But since you can achieve movement with +R anyway it you can just replace the bits you don't like with things you do like. It's a lovely book but the bulk of it condenses down to using unambiguous cues to encourage behaviour and I get the feeling that he is a sufficiently empathetic horseman that the horses' motivation for complying is just his presence. At least - that is what we are shown in the book/video and the horses seemed to have a real enthusiasm about them. I'd love to hear from someone who has done a course with him and find out whether he practices what he preaches.
I guess the other potential problem with his method is that it comes across as a one-size fits all method without taking into account all horses being different. But then it depends on how open minded you are and how you adapt it to fit your horse.
As for have I done this?! Well, we keep getting sidetracked with other games but once I have walk, trot, canter and lateral work sufficiently on-cue then we will definitely see if we can put some more of his exercises into practice.
The difference between subtle cues and -R? Well this is a thread in itself! (Likewise the implications of mixing +R and -R - will try to find an older thread for you and bring it up) But for me it comes down to how the subtle cue was taught in the first place and what is motivating the horse to perform the behaviour. If it started off as a "less-subtle" cue which was then diminished then for me that is still -R/P. For example, at the moment I ask my horse to walk on by squeezing my legs. Years ago he learnt that if he ignored my gentle squeeze then he got a harder squeeze or a tap with the whip. So for me his motivation is avoiding the higher level of pressure. But I am now free shaping his paces using +R and am about to start doing ridden CT work. Once he is walking nicely, motivated by the treats, then I will add the leg-squeeze again (and obviously never escalate it). Same cue, but different motivation. Maybe this is just semantics but deep down inside me it is something I know I need to do, maybe more for myself as I feel guilty about the lessons with the "proper" instructor I subjected him to. Whatever!