So, I am getting my puppy in 7 days and I am overly excited to say the least. I want to make sure he will have a fantastic live so the plan is to give him lots of love, experiences, exercise and train him well so he will be loved also outside the house I have been studying a lot and training principles sure have changed for the better since I had a dog last time (when I was a kid 25 years ago) and I truly believe in and want to apply reward-based/positive training. That said, I do believe there will be times where the dog will need a correction beyond the ones applied in purely reward based training (light vocal correction, distraction, removal of rewards, timeouts). Imagine that the dog sees another dog, maybe a female (he is a male). I am sure the temptation to go there will be very big and I imagine that he will think "I am fine about missing out on my treat if I can just say hello to that sweet female lab overthere" and off he goes. I know that is probably not exactly the thought process but I do fear and imagine that the corrections above will not be sufficient to get a fully reliable dog 1. Am I right? 2. If yes, what sort of corrections would you apply and when? My own thoughts on 2: I imagine that for basic obedience (sit, stay, walking at heel, etc) and basic manners (like not begging at the table for food, not cheewing), the light vocal correction, distraction, timeouts, removal of rewards will be perfect along with patience To get a fully reliable dog (i.e one that does not chase another dog or birds without permission, that dont suddenly chew on something dangerous that he has not seen before I would think that the following could work: -- using different words and sharper tone of voice than for the typical vocal correction -- light leash corrections (would only use flat collar) -- a very light nip like his mother would do I would never ever apply physical correction beyond a very light nip or a light leash correction. Anyway, would lover to hear more thoughts and experiences on when corrections are applied?
Re: Reward-based training -- which sort of corrections do you apply? [quote author=andreasjuuls link=topic=9344.msg134108#msg134108 date=1420226119] 1. Am I right? [/quote] No, I don't think so. It is not necessary to apply any corrections (more straightforwardly called punishments, even if you consider them "mild"), and the evidence is that dogs behave better without these everyday punishments (not abuse, just the stuff you are talking about) often dished out to dogs. There is lots more information here: http://totallydogtraining.com/the-downside-of-punishing-dogs/ http://totallydogtraining.com/the-evidence-for-positive-reinforcement-training-for-dogs/ http://totallydogtraining.com/dog-training-methods-should-i-punish-my-dog/ And an explanation of how and why positive training works here: http://totallydogtraining.com/modern-dog-training-methods/ You are not the dog's mother when it comes to nipping your dog - you do not have a dog to dog relationship with your dog.
Re: Reward-based training -- which sort of corrections do you apply? [quote author=JulieT link=topic=9344.msg134115#msg134115 date=1420227073] [quote author=andreasjuuls link=topic=9344.msg134108#msg134108 date=1420226119] 1. Am I right? [/quote] No, I don't think so. It is not necessary to apply any corrections (more straightforwardly called punishments, even if you consider them "mild"), and the evidence is that dogs behave better without these everyday punishments (not abuse, just the stuff you are talking about) often dished out to dogs. There is lots more information here: http://totallydogtraining.com/the-downside-of-punishing-dogs/ http://totallydogtraining.com/the-evidence-for-positive-reinforcement-training-for-dogs/ http://totallydogtraining.com/dog-training-methods-should-i-punish-my-dog/ And an explanation of how and why positive training works here: http://totallydogtraining.com/modern-dog-training-methods/ You are not the dog's mother when it comes to nipping your dog - you do not have a dog to dog relationship with your dog. [/quote] Thanks a lot. I had thought I would have t be 98% positive and 2% correctional but if it can be 100/0 that is great -- for me and the dog ;D Thanks again
Re: Reward-based training -- which sort of corrections do you apply? It is completely possible - and if you are starting with a new pup, then you have every chance of succeeding. If you are interested in training your puppy for gundog work using positive only methods, you could check out "positive gundogs" on facebook too. There are people on there training hunting dogs using positive only training.
Re: Reward-based training -- which sort of corrections do you apply? A significant part of positive reinforcement training is managing the environment and the ability of your dog to self reward. It can be a tough thing to do but to gradually increase the distractions (your female dog for example) is key, try not to ask too much of your dogs ability to respond and set him up to succeed. It can be easier said than done but it's worth the effort I think My current example is my young spaniel puppy is doing his loose lead training in the middle of a quiet road because 1 it gives me more space to work with and 2 if I'm on the footpath the smells of all the people and dogs that have walked the path is immensely distracting, the nose goes down and the lead goes tight. I hope to make the safety of the pavement soon I think there is usually an element of removal of rewards or preventing self rewarding that goes with positive reinforcement training.
Re: Reward-based training -- which sort of corrections do you apply? [quote author=JulieT link=topic=9344.msg134128#msg134128 date=1420227717] It is completely possible - and if you are starting with a new pup, then you have every chance of succeeding. If you are interested in training your puppy for gundog work using positive only methods, you could check out "positive gundogs" on facebook too. There are people on there training hunting dogs using positive only training. [/quote] No hunting plans but will check it out - nice with some inspiration. If a dog can be trained to do gundog work with positive methods only, then I am sure there is something there to learn from.
Re: Reward-based training -- which sort of corrections do you apply? [quote author=bbrown link=topic=9344.msg134130#msg134130 date=1420227830] A significant part of positive reinforcement training is managing the environment and the ability of your dog to self reward. It can be a tough thing to do but to gradually increase the distractions (your female dog for example) is key, try not to ask too much of your dogs ability to respond and set him up to succeed. It can be easier said than done but it's worth the effort I think My current example is my young spaniel puppy is doing his loose lead training in the middle of a quiet road because 1 it gives me more space to work with and 2 if I'm on the footpath the smells of all the people and dogs that have walked the path is immensely distracting, the nose goes down and the lead goes tight. I hope to make the safety of the pavement soon I think there is usually an element of removal of rewards or preventing self rewarding that goes with positive reinforcement training. [/quote] That makes a lot of sense. It sounds like slow progression and setting the dog up for success is one of the keys to success
Re: Reward-based training -- which sort of corrections do you apply? It's true that you don't need to use corrections aka punishment. Removal of rewards (aka 'negative punishment') to reduce behaviours is fine though - examples of this would be: - Turning away in response to jumping up to remove the reward of getting to your face - Stopping dead in response to pulling on the lead to remove the reward of getting to where the pup wants to go Actually, there is one exception to the 'no correction' thing and that is the use of time-outs. A lot of people find young pups very bitey (as you may have gathered from reading threads!) and some times the only this that works to stop this is to put the pup in an unoccupied but safe room for a short time (like, half a minute) as a consequence for the biting. That is fine. But put away all thoughts of shouting, jerking the lead, tapping the nose and the like - there is no need for that As Barbara said a lot of it is managing the environment. This can apply in very straightforward ways. You don't want your dog to steal the loaf of bread off the kitchen bench? Don't put it there. You don't want your dog to run over to every human being on the planet? Use a lead. Prevention is better than cure. The other thing in your armoury is training behaviours that are incompatible with the things you don't want. Don't want your dog to hassle you during meals? Teach him to lie on a mat instead. Don't want your dog jumping on people? Teach him to sit for a treat instead. Don't want him to chase birds? Expose him to birds gradually, on lead, and reward for being calm and/or looking at you instead. Practise walking past birds calmly for lots of great treats. So, be proactive and instead of reacting with a negative when undesirable behaviours happen, think ahead and train the positive alternative instead so the negative doesn't happen in the first place You should be commended for wanting to embrace positive, kind methods and for learning as much as you can about them before you get your pup Reward based training will create a great relationship with your dog, and will create a happy, confident dog who wants to learn
Re: Reward-based training -- which sort of corrections do you apply? [quote author=Oberon link=topic=9344.msg134203#msg134203 date=1420242102] Actually, there is one exception to the 'no correction' thing and that is the use of time-outs. [/quote] Yes, interesting. I recently came across a trainer that explained time outs as punishments, which they most certainly are. I don't use them - not that I ever really did (easy for me to say with a dog long done with the puppy biting stage). They need to be really, really short if used I think.
Re: Reward-based training -- which sort of corrections do you apply? Time outs can be thought of as either negative punishment (removal of a desirable thing, attention, to reduce the performance of a behaviour) or positive punishment (introduction of an aversive, isolation, to reduce the performance of a behaviour). A lot of things kind of fall into two baskets that way, don't they, depending on which part you focus on
Re: Reward-based training -- which sort of corrections do you apply? I just wanted to say hi and welcome to the Forum and good on you for getting so well prepared for your pup. We have Dexter ,he's our first dog and is now 2.5 and I wish I'd found the Labradorsite before he arrived with us. There is a lot of great advice and experience on here ,some of which has been shared already ,and Pippa has just launched a new site wwwthehappypuppysite.com which will be a great resource for new puppy owners wanting to embrace positive training methods Best wishes Angela
Re: Reward-based training -- which sort of corrections do you apply? [quote author=Oberon link=topic=9344.msg134203#msg134203 date=1420242102] Actually, there is one exception to the 'no correction' thing and that is the use of time-outs. [/quote] Interesting, because the few times I used this with Willow, I didn't even consider it to be an aversive. I was simply putting her in her crate to help her settle down - it was used invariably because she was over-tired and needed to realise that fact. Putting her in her bedtime environment calmed her down almost immediately and she would drop to sleep within a minute. I suppose that may be different to what you're saying, when it's used as more of a punishment for bad behaviour? I'm certainly not a 100% positive trainer, although I try my best to be. One example would be using "enough" when my two start getting too rambunctious with one another. Occasionally (I suppose a couple of times a week), it gets too loud and boisterous to carry on inside, so I put myself between them, say "enough" and don't let them reach one another for playing. Is that negative? Now, the word "enough" is enough to settle them down and send them to separate beds. Another would be when Willow is chewing holes in her bed, I tell her "no" and remove it from her mouth, shoving in a chew toy instead. I know this isn't ideal because it certainly won't stop her chewing her bed when I'm not there to tell her! The use of extinction is a slow but useful method - as ours grew tall enough to reach the kitchen counters, they would jump up. Again, I used a "no" and would tell them to leave the kitchen (they have been taught to sit at the threshold to the kitchen, and I have put this on cue as well, but it's not perfect yet). But, as a couple of the others have mentioned, if they learn that there is never anything of interest on the counters, they will eventually give up and stop jumping up. I suppose initially they are self rewarding just by having a look - it's a new and exciting view for them - but as they learn that that view never changes and there's nothing for them to get from there, it just becomes boring and pointless for them. It does mean having to train the people in the household not to leave things on the surfaces, though. Double bonus - a dog that doesn't counter-surf and a tidy kitchen!
Re: Reward-based training -- which sort of corrections do you apply? I also use 'enough' but I train it using treats - so 'enough' means stop, sit, and get a treat. I expect them to stay stopped, of course! I use 'no' with mine when I don't want them to touch something which they don't already know is not for them eg the Christmas tree. I have just done it now as the tree lights are off the tree and on a low coffee table. Tatze went and had an interested sniff so I said 'no' firmly then when she came to me a pat and 'good girl'. It works, she didn't go back to the lights. Gypsy is different, she's not really interested in 'things' (not even toys - she will play if Tatze or me present her with a toy, but she's not bothered otherwise and isn't nosey about new stuff - proving she's only 50% Lab!)
Re: Reward-based training -- which sort of corrections do you apply? [quote author=snowbunny link=topic=9344.msg134241#msg134241 date=1420269073] But, as a couple of the others have mentioned, if they learn that there is never anything of interest on the counters, they will eventually give up and stop jumping up. I suppose initially they are self rewarding just by having a look - it's a new and exciting view for them - but as they learn that that view never changes and there's nothing for them to get from there, it just becomes boring and pointless for them. It does mean having to train the people in the household not to leave things on the surfaces, though. Double bonus - a dog that doesn't counter-surf and a tidy kitchen! [/quote] Yes, this is what we did, our two never bother at all with the counters now. But my friend's dog does - she made the mistake of feeding him bits of tuna from there - bad bad idea!
Re: Reward-based training -- which sort of corrections do you apply? [quote author=Oberon link=topic=9344.msg134211#msg134211 date=1420249524] Time outs can be thought of as either negative punishment (removal of a desirable thing, attention, to reduce the performance of a behaviour) or positive punishment (introduction of an aversive, isolation, to reduce the performance of a behaviour). A lot of things kind of fall into two baskets that way, don't they, depending on which part you focus on [/quote] I suppose it's determined by how the individual dog views it. If (in Fiona's example) the puppy is put in his crate and thinks "great I wanted a snooze anyway" it's not punishing at all (and presumably won't do anything to stop the biting either). If the pup feels isolated and a bit scared after being bundled into a bathroom, then the dog finds it punishing.
Re: Reward-based training -- which sort of corrections do you apply? Exactly. If the dog in fact wanted to be alone or in the crate then it would be a positive reinforcement - the addition of a desirable thing (being in the beloved crate) that increases the performance of a preceding behaviour (biting). It's all about the motivation, desires and perceptions of the dog. But mostly biting is about wanting interaction, so removal of the opportunity for interaction would tend to decrease the biting (ie it'd be a punishment).
Re: Reward-based training -- which sort of corrections do you apply? I don't think biting is really a particularly good example as they grow out of it anyway, so it's hard to tell which method works - we used all of them. Substituting for a chew or a toy, popping dog in crate if tired, stopping all play, squealing and walking away. Wearing wellies But I think the thing that really worked was them simply growing past the bitey puppy stage. Looking at that list only one is reward based 'tho, isn't it?
