Aargh! I started a response to this last night, did a bit more before work this morn, then came back to it tonight & just about finished it... then my computer shut down on me & lost the lot!!! Bloomin time saving devices!!
Maybe similar to a classroom of kids. You have to go at it differently than if your favorite niece or nephew is visiting for the summer.
Yes, I reckon that definitely has a bearing on different methods. One reason I never wanted to be a 'horse breaker', churning out horses, altho I'm happy taking on individual horses for training.
But on the note of clicker training(speaking of principles that govern, not necessarily specifics like noisemakers & lollies), did you know that there are
schools and institutions that use it with people & kids? Funny that in that scene, I've heard the opinion 'sure it works with kids but how could it work for animals??' I've heard dog people say they don't believe it would work with people, horse people say 'horses aren't dogs so it won't work on them', people that have seen it used in zoos say it wouldn't work with domestic animals, and vice versa...
Both Monty and Parelli said to never ever punish a horse for biting. They claim it just makes it worse. They claim a horse does not understand punishment in human terms.
You may not have been around here long enough to hear my... diatribes
on punishment. I'm not big on it at all. I think it has limited value in training, and tends to come with many undesirable 'side effects' if not used very judiciously. However I'm not at all against it's use(have also warned people about *inappropriate*
treats/+R too, but...) I don't agree with the above statements. I think it depends
on the reasons
for the behaviour and how
you punish, as to whether it's appropriate or not, with biting or whatever. As for understanding punishment differently, well, yes, of course we think
differently, and different things motivate us. And people tend to make a lot of assumptions of animals, based on their own way of thinking, but basically, behaviourally speaking, we all, whatever animal, learn
essentially the same way. The 'laws of learning' apply to all. We repeat & get better at things we're reinforced for, and are less likely to repeat things that don't 'work' for us & have unpleasant consequences.
*But it is also important to consider there's always a lot more going on than that one particular behaviour you want to effect, for eg. if the animal is 'distracted' by fear. I feel both punishment and positive reinforcement when an animal is frightened are usually inappropriate, because of the association between the treat/jerk/whip/spur... whatever, and the emotion
. Which tends to make your consequence ineffective and the 'bad' behaviour worse.
association between a food reward and the desired behavior the food must be given within one second after the desired behavior occurs.
Don't recall Temple commenting on c/t but not surprised. Must dig out that book again...
Yes, if there was one thing to pick as most important, most 'misused' I'd say timing. The animal should be reinforced/punished *at the time of* the behaviour, or within a second(or few, depending on species) at worst. You may have heard of The 3 Second Rule, aka 3 seconds of Death
, aka Come To Jesus
. The idea is that you have 3 seconds after the behaviour to punish a horse. This somehow became, to some, that you should 'make him think he's gonna die' for 3 seconds
The above idea came about from behavioural studies that showed it was far more effective/understood to reinforce/punish at the time of a behaviour, but dogs could frequently associate cause & effect for up to 3 seconds abstracted. The same tests done with horses showed I think a '1 second rule'. Obviously the whole story didn't get across to everyone.
'Clicker training' - that is, the clicker or noise/signal part - is known behaviourally as a 'bridging signal', which can also become a 'conditioned or secondary reinforcer' like praise or a pat. You associate the signal with an actual positive reinforcement - something desirable, food treat for eg - and when that signal is strongly linked, you can use it effectively as a precise 'marker', without having to instantly give the primary reinforcement. It 'buys you time'.
Oh & pre empting a question of 'then you always need to give treats/click every time' with this sort of training. Actually, it's more the opposite. While as with any other form of reinforcement, you do indeed reinforce promptly & frequently, for even small stuff when you start teaching something, once the lesson is learned, it's more effective in continuing to 'strengthen' the behaviour/response, to reinforce less frequently, and randomly.
...& now there are so many more replies, can't see where I was up to, but just want to say about 'is a scratch in an itchy spot reinforcement'.... Absolutely! Another pet peeve is people who don't understand the principles get hung up on the food treats. It's not about food!! That is just a convenient 'tool'.
In the same way, a food treat isn't always a +R. If it is desirable, wanted, then it's a potential reward. Just that food treats are a convenient and generally strong one(but not always, if you're full, sick, don't like the flavour...). If your horse really enjoys a good scratch, and if you find it's practical, then great! If he doesn't like being touched or couldn't care less, or is more concerned with something else, then it's not going to cut it.