Originally Posted by
wilbar
Well said Wys
.
In the context of dog training, frustration is may a negative emotion, but people that try to use pos R for the majority of their training, especially in the context of training NEW behaviours, will do their best to ensure that there is minimal frustration. The whole point of using something like clicker training or other forms of pos R is to set things up for the dog to succeed by using shaping techniques, incremental steps &, most importantly IMO, being aware of
the context in which you are doing the training.
In the context of dog training, the
temporary witholding of a reinforcer can be used to up the ante & ask the dog to perform “better”. But also in the context of dog training, anyone trying to use pos R would ensure that the frustration is minimised by not asking too much too soon ~ allowing the dog to succeed. Provided that the ante is upped slowly & gradually, any frustration would soon be overcome by the opportunity to succeed. Dogs that have been taught using pos R are not only being taught to do a certain behaviour, or not do another behaviour, but have also learned that training is fun, that the whole “learning” session is fun, they are eager to learn & clearly enjoy the whole process.
Clealy using OC principles there will be some frustration if the reinforcer is witheld when you are trying to get more accurate & instant behaviour etc. Similarly if you want to stop a behaviour that was previously reinforced there will be an extinction burst while the dog tries harder to get the reinforcer. But pos R trainers will do their best to minimise this by teaching an alternative & incompatible behaviour.
I am currently witnessing an extinction burst at home with one of my cats. The cat makes a piteous sounding miaow (just the way she sounds, the “piteous” bit is husband’s anthropomorphic interpretation of the noise!) so husband gives her attention which shuts her up. When he stops the attention, she miaows again ~ cat has learned that this gets her attention. Husband moans that he can’t understand why the cat keeps miaowing, what it is that she wants & he hates the noise, but still continually rewards the cat by giving her attention. I told him to ignore the noise & give her attention when she’s quietened down & is sleeping in her bed. Husband tried this & says she just miaows even more when he ignores her. I explained about extinction bursts
.
So in the context of my cat, she may be experiencing negative punishment inasmuch as she’s frustrated by not getting the attention she wants & used to get. But in the context of her whole life experience, this is hardly a major event. She gets plenty of attention at other times, just not when she’s sitting in her bed in the sitting room, while we’re eating or watching TV. IMO this is not anywhere near the same as using pos punishment by shouting at her, or hitting her, or picking her up & removing her from the room, or shutting her away. Inevitably there’s a bit of neg pun at the start ~ I don’t think any of us that profess to be use pos R in animal training would deny that there can be an element of neg pun now & again. But what we try to do is minimise that side of things & try to find alternatives.
I love this quote from Kay Laurence in “Learning About Dogs”. It sets out quite clearly the ethos of pos R training. But more than that, it makes it quite clear that relying purely on the science of OC principles is not what dog training is all about. OC principles are just a small part of the process, just a tool in the trainer’s tool box.
"If I am teaching a dog, I avoid every atom of punishment or removal of something good to get the behaviour. It is not a question of how aversive, it is the thought that aversive is a method to get a behaviour. The actions are an indication of the thought process that aversives are part of the teaching process. I will say, 'Let's just find another way.
We all think we're "positive" trainers. But training with reinforcement involves more than just being nice, and more than using reinforcers. It involves creating a climate of security in which it is safe to learn new things, and safe to rely on what you've already learned. In this climate an animal can learn to control itself, rather than being controlled by you. In this climate, rather than just reacting to the environment like an untutored shelter dog, barking at every noise, plunging towards every attraction, jumping on everyone and everything, mouthing and smelling and grabbing—an animal becomes confident and calm. In this climate, having confidence that your cues are meaningful and will lead toward pleasant goals, the dog is trusting and—this is very unscientific—the dog is happy."
IMO the use of harsh punishment such as electric shocks, prong collars, lead jerks etc says far more about the person using them, & not in a good way! If they try to convince others that using these methods is kinder because it’s quicker, or it’s kinder because the alternative is euthanasia, then they clearly don’t have the ability to empathise with animals, to consider the emotional element of training, or to consider the welfare impact on the animal as a whole.