Why reward based training is best

[JoanneF]

There's nothing ground-breaking in this article but it's a good summary of why force free training is more effective.

Seven Reasons to Use Reward-Based Dog Training

 

[Mad Murphy]

Im glad people are finally letting go of the whole dominance alpha male rubbish.. Stabijs as a breed are known to shut down when faced with harsh treatment or training , the breed club actually says a firm but gentle hand is needed.
Just because you use rewards it doesnt mean you cant be firm in training and firm does not mean harsh, something many fail to understand.

 

[DixD]

Reward based training is the way to go, but timing and circumstance ARE very important, and, if somebody doesn’t understand the complexities tgey can end up rewarding unwanted behaviours. For example, I trained in a fabulous A frame contact with my little dog in the garden. When I put a jump in its place, the wee dog went over it, stopped and refused to move. Turned out, I had trained him to stop when his front feet touched the artificial grass in front of it. Transfer that to somebody rewarding their for stopping an unwanted behaviour for a second, say barking at another dog, rather than carrying out a positive action, and you can actually train your dog to bark and then stop.

 

[DixD]

Forgot a sentence at the end. I meant to add... Whereas if you trained a “watch me” for that circumstance, the dog is rewarded for carrying out a positive action, rather than carrying out a negative action and then stopping it. For another example of unintended consequences, a dog comes to visit next door. It barks all the time. The owner blows a whistle to distract it (not sure if that comes under positive dog training or aversive). The dog stops barking for a wee while, and then starts again. The owner blows the whistle...) Seems, from this side of the fence, the dog has trained its owner to blow a whistle.

 

[JudyN]

Transfer that to somebody rewarding their for stopping an unwanted behaviour for a second, say barking at another dog, rather than carrying out a positive action, and you can actually train your dog to bark and then stop.

Agreed, but it's important to add that often, what we are really trying to address - or what we should be trying to address - is the emotion behind the unwanted behaviour rather than the behaviour itself. So, say your dog doesn't like/is scared of other dogs so barks at them. You aim to give him a treat when he sees another dog, but sometimes he gets a bark in first. That's not really a problem... keep at it, and eventually your dog will see another dog, bark, and say 'Do I get my treat now?' Once the emotion behind the bark has changed, it's relatively easy to reduce the volume of the bark and eventually it will disappear altogether.

So yes, if you're specifically working on a behaviour, as in agility, you have to be careful what you reinforce, but if the underlying cause is an emotion that causes an unwanted behaviour it's more important to focus on the emotion without worrying too much about the behaviour.

 

[DixD]

That’s basically what I was trying to say, reward based training can be complex. It seems simple, if you understand the theories behind it, but a lot of owners haven’t studied dog behaviour to that extent, so it can sometimes nit achieve tge desired effect.. It was really brought home to me, when I started clicker training my horse, how complex the whole thing is, how humans and other mammals have to build up a common language and understanding in their relationships. But, once tge language is there, it’s a very strong and positive method.