Do you train using pos reinforcement, neg reinforcement, or both? And how do you correct your dog for disobeying a command, let's say "STAY" for example?
Some Less Obvious Benefits of Dog Training - Help For Owners of Older Dogs
Recommended Answer:
I train my dog by positive reinforcement and corrections.
There are two sides of dog training and neither side gets along well with the other. Here is how I categorize dog training.
There are the alpha/dominance old school/traditional dog training people and the positive reinforcement dog training people. I like to think I train by strictly positive reinforcement. Let me try to explain.
I've found that the alpha/dominance old school/traditional dog training people are extremely close minded and set in their ways. They scoff at ANY and ALL positive reinforcement training. They believe that corrections all the time with a quick pat on the head as praise is the way to train a dog. They tend to be OBSESSED to the point of just borderline crazy (just my opinion) with corrections and being the pack leader. Now, I'm not saying ALL traditional trainers do this, but some of them use physical abuse as “training.” Some trainers, hit, slap, punch, spank, alpha roll, pinch, strangle, and the most popular hang their dogs as discipline. This is considered animal abuse to me and I WILL NEVER do that to my dog. To be blunt, these people are very hard to be around (I personally despise them) and I definitely would not want to associate myself with them. The end.
I much prefer the positive reinforcement people. These people really care about building a relationship with their dogs and training them in a humane way. Positive reinforcement includes training with clickers (clicker training) treats, praise and toys. The only thing that most positive reinforcement trainers DO NOT advocate are the use of tools like choke collars and prong collars. This bothers me a bit because I've seen with my own eyes that these tools do not cause pain or harm to the dog when used properly. Notice how I didn't mention shock collars because in my opinion, this tool does NOT go under positive reinforcement. How can it? You can't change the fact that you're still issuing an electric current to your dog.
My dog has a prong collar and I do issue corrections when necessary but those corrections do not hurt her in any way. This is the only place where I differ from the positive reinforcement dog trainers.
I really don't want people to dump me in the alpha/dominance old school/traditional dog training
category just because I use a prong collar and issue corrections.
I really think I'm a positive reinforcement person because I LOVE all aspects of positive reinforcement training and that's how I mostly train by. I guess some people would say I'm in the middle though.
I hope this all makes sense lol!
Do NOT get me started on Cesar Millan! Once I start, I can't stop ranting about him! I despise him! He is the WORST thing that has happened to the dog training world! I can post a very detailed answer on why I hate him, but for now, I'll just post these websites for Cesar Millan fans to look at. http://beyondcesarmillan.weebly.com/ http://www.4pawsu.com/cesarfans.htm
Dog Training Collars and Harnesses - Making the Proper Selection is Essential to Training Success
- depends on the dog.
a softer dog may not need a physical correction. sometimes merely giving them a look is enough.
a hard dog needs a higher level of correction... you need to find yourself a trainer who understands that the method of training and correction hinges more on the dog than on their personal preference.
my own dogs need a very high level of correction, such as a firm shake of the collar or a collar pop. otherwise, they could care less. they're high drive game dogs, and very little can break their focus.
if one of my dogs broke a stay, i'd pop the prong collar. BUT ONLY IF HE KNOWINGLY DISOBEYED.
if the dog doesn't understand, there is no correction other than a verbal cue to let him know that he's not doing what you want, like uh-uh or no, then he should be made to do it until he fully understands it.
corrections aren't just about HOW- not at all. WHEN and WHY are more important.
TRAINING should be done with positive reinforcement. CORRECTION should be done with something the dog doesn't like, and something appropriate for the dog.
it also depends on what level of obedience you hope to obtain, and what you use the dog for.
my own dogs are super hard. shaking a finger at them, or merely putting them back where they were is NOT enough to CORRECT them. when i correct, i do so in such a way that i will not have to repeat the correction again.
i also have need for my dogs to have a higher level of correction. for example- this morning i ran into an emergency while working Security, and had to leave Tank (avatar) outside an elevator for nearly two hours. how would i come back every few minutes and check to make sure he stayed? how would i make sure he didn't wander off, or bother dogs that were passing by with the Residents? training. training with effective rewards, and equally effective corrections.
he doesn't consider disobeying. - For disobeying a command, he gets a collar pop and a verbal correction.
If it is a stay command and i don't have handy or don't want to be fiddling around with a collar and lead to put it on him to then correct him i scruff him upwards with my hand and give a verbal correction also.
He has been trained like this all his life and shows no injuries, no fear and is not hand shy because of it, although that being said i rarely have to resort to correcting him at this stage for not obeying commands.
The majority of corrections i give now are 're-directions' to get him away from an object or an area of the room/home which he is not allowed and that only requires a firm 'No' and a command for him to 'come here'.
As for peoples views on physical compulsion on a dog in here... you may be interested in a previous question i asked entitled ''How do you proof a dog without physical compulsion?''
http://answers.yahoo.com/question/index;…
These type of corrections were around LONG before science gave them names and Cesar Milan, and they will be around long after Mr Milan has faded into obscurity. - I'm not a professional, but I use positive reinforcement. If my dog disobeyed "stay", I would say "Ah Uh!" in a stern voice, and make them go back and sit. Then I would walk away at a lesser distance that what I did before. If they got up again, I would make them go back and sit until they understood that they need to stay there, and then give them a treat and praise.
- As soon as she breaks her stay, I exit the building and with a slightly disapproving correctional tone say, "I told you to stay!" Then I replace her into the spot you had previously told her to stay.
I use positive reinforcement. If the dogs disobey, I correct with "No" or depends on the command but yeah. - For something liek disobeying a command don't grab the dog. Just simply look in his face face and say "ANT!" and then turn around and dont look at him. Wait five seconds, and then turn around and try again. For something like barking, tap his nose lightly and say "ANT!" nothing more. Don't use his name as a command to stop something, and never use "NO!" always "ANT!"
- I use a verbal interruption to stop the behaviour while it's occurring and a leash correction at the same time. We wait 5 seconds and then try again. If the correct behaviour or action occurs, I use praise and a bait reward. As the behaviour or action becomes more predictable, the bait becomes smaller until there is none.
- First off, *I* am not a dog trainer.
For corrections on my dog, I use a prong collar and my voice. A pop correction with the prong and a firm "No" with my voice.
For rewards I use treats, the disc, my voice. A "Yes!" (to mark the behaviour) followed by a "Good boy" or a treat/toy.
My dog is a soft dog, so I honestly only need to say "No" if he gets up during a stay, and he will go back to his sit/down stay.ADD @Bells, it looks like your TD fan club is giving you a hard time today xD LOL. I gave you a TU. - while my dog was a pup i used positive providing him with treats when he did something good. if he wandered off while he was meant to be in a stay position i walked back over to him didnt speak to him put him in the position he was meant to be in and started again. when teaching him not to bite i would say 'no' and completely ignore him. if he got too boisterous i would pin him to the ground no harshly just so he knew i was boss. this is what they would do in the wild. you ignore the dog but hold them down. this is what i learnt from caesar milan too.
- Correct...with a leash pop...preferrably attached to a prong collar.
I've done it with "Stay" before. I'd say the command, he'd look at me like I had 3 heads. Needless to say, prong went on....he yelped and hit the floor...staying there. I get 100% out of the Stay command now. - You should use positive reinforcement. Never hit or do physicaly hurt your dog, you can correct your dog by saying NO, and trying again.
- It really depends on the dog. I have 1 dog who is pretty soft and a harsh sounding "Cut It Out" will do. I train using a buckle collar, I never needed more.
My little female, on the other hand, has a bit of a mind of her own. With her, I do give collar correction and have used a prong collar on her.
When introducing a new command, I always use food or toy. I continue using food and toy, until I see they clearly have gotten it. Then I will food or toy ramdomly or after following a serious of commands.
As for the Stay command. If I told the dog to stay and they moved, layed down or whatever. I quietly and non-emotionally take them back to where we started and we start the exercise again, this time, I stay closer. If a dog is "new" to the Stay command, they are on a 6 foot leash or on a Flexi. This way, I can correct immediately if necessary. I also proof the stay, by dropping food just out of reach, bouncing a ball, playing with my other dog, moving around or whatever. The dog learns that Stay means Stay. When I return to the dog to release from the Stay, if it's off leash, I pick up the leash, pat the dog twice on the rump and quietly say "Free" and which point I give the dog a treat or throw the ball.
My dogs have also learned, that if I don't give the 2 pats on the rump, but still give the " free" command, they do not move. If I do give the 2 pats and don't say "Free", they still should not move. It's only the combination of 2 pats and Free that releases them from the Stay. - Sorry, now you are recommending to people what to watch regarding dog training when you are the one asking the question? Don't ask questions if you have already made up in your mind what is right or wrong.
Using a pop on the leash as a form of correction is fine. That's how I trained my dog.
Some dogs may not need a physical correction if they are submissive enough.
Grabbing a dog by the collar and yelling at it is ineffective.
There always has to be some form of positive reinforcement in any kind of training because that is how you get your dog to WANT to work.
For instance - I ask my dog to "sit" and if he doesn't he gets a pop. When he does then sit I give him verbal praise and calm physical praise eg. calmly petting him while calmly saying good boy. I use positive reinforcement in the form of treats when I am teaching my dog a new behaviour but I slowly phase it out so that he will want to work for the tone of my voice and therefore proofing my training as I may not always have a treat on hand. And really - only working for treats is not helping the dog use its brain as much as its using its stomach.
As far as something like "stay".
I put him in the "sit" or "down" position ask him to "stay" and then walk away. If he releases his position without my release command I bring him back to his original spot, give him a pop and tell him to stay. How else is he going to really learn what I'm asking him to do? When he does successfully stay and I release him on "okay" then I give him positive reinforcement.
You should never have to repeat your command more than twice (really only once). If you have to keep saying "stay" "stay" "stay" - your dog has learned nothing.
I also use a prong collar as part of my corrections - if people actually took the time to learn how to use the correctly and learn why people use them as a training aid, then there wouldn't be so many ignorant comments about them.
ADD: No "s" at the end of my name, thanks.
I don't need to watch the videos - I don't know those people, never heard of them. Plus I would never make a judgement on training based on a video and I certainly would never train my dog via video. I train with trainers who work with people one on one that have proven track records in their fields.
ADD2: No offense to Z - but why do clicker training advocates always make the comparison to training whales/dolphins? lol
Dogs are not whales/dolphins. The comparison shouldn't be made.
And some dogs have higher drives than others. That doesn't mean people don't want to use things like clicker training - it just means it might not work for their dog.
Not every dog can be trained by a clicker just like not every dog responds to a training aid like a prong collar. I just rather my dog respond to my voice than a clicker.
ADD3: I'm here to help people - this isn't a debate site, it's not supposed to be at least.
I offer advice and opinions.
Also, please don't assume that the trainers I have talked to are only situated in my immediate area. I have done TONS of research to make sure I applied the best possible method for MY dog - which is why I always say that everyone needs to figure out the right method for THEIR dog. - Ok, firstly, i'm not a dog trainer. I use both pos. and neg. reinforcement. When i first start out with a command i use treats and praise as the reward. After a few days (or hours depending on how easy that said command is) that's when the corrections come in. A simple POP on the leash is usually the best way i've done it. However, for things like when i trained recall if my dog ignored me she got grabbed by the scruff of the neck and dragged to where i asked her to come to. Also, if they have been extremely bad and are not responding to the basic leash corrections i will give a stronger, physical correction. If they do not sit when i ask (regardless of what they're doing) they get a nudge on the hind with my foot, not hard, in fact i doubt it brings any uncomfortable feeling to them at all. I'm just reminding them that if they don't sit/lie down immediately pressure is applied(in this case it's my foot nudging against the side of their rear).
I've used corrections with stay before. I asked my dog to down and stay and she started following me when i walked away, so, seeing as i didn't have a collar or leash on her at the time, she got grabbed and shoved to the ground (that sounds much harsher than it is) with a sharp NO, then given the command stay again. I made her stay for longer. Did she stay after that? She sure did and always does stay, now.
My dogs do have a release command. In fact they have a few (is that possible? haha). When they have done something right they get a "Good", "Yes" and/or a "thankyou". When they release from their position after i have given the release command they get pos. reinforcment. They get the "good girls" and belly rubs like nothing on earth, as if they have just saved someone from a burning building. If they release without my permission/command then they get a correction and no praise. They have not pleased me and they are told that.
These corrections are used with my Border Collie because she needs slightly more harsher/stronger corrections to my Spaniel. My Spaniel does need corrections but is a much softer dog that my Collie, so needs softer corrections. I think a training method should be formed for the dog and not what you thinks best. A dominant dog with only pos. reinforcement? Yeah, your going to have a hard time training that dog. A very dominant dog does NEED corrections.
Please note that i only use the physical corrections when my dog knowling disobeys. If they don't know the command yet and disobey me then i make the sound "ach!" that just tells them that that isn't what they are to do and to try again.
If they go to do something bad, say steal food of a plate when i get up to do something (i fully expect to leave a plate of food on the chair when go into a different room without my dogs having a feast on it- my BC doesn't touch it, but my Spaniel tends to have her greedy moments) they get a stern "Don't you dare!" They then know that i'm correcting them for what they are thinking about doing, when they do leave the food when i leave the room, again, they get a hero's praise! - Well, it depends on the offense...
Heeling - the dog knows how to do it, but insists on lagging. forging, or sniffing, they're going to get a quick pop back into position, and then rewarded once they are doing it right for a while.
Basically - I don't correct other than a little "uh uh" and start over until the dog KNOWS the command. Once the dog knows it and has it down reliably, if they don't perform it when told to then I will generally tap them on top the nose with two fingers, or pop the lead.
In general though, I try and keep training sessions fun - like a game. This encourages the dog to want to learn.
If a dog bites or snaps at someone, they are going to get a harsh "NO" and restrained by the scruff. If they try it again, they are going to get scruffed and popped under the chin. Rarely EVER does a dog try something like that after getting popped under the chin. Funny thing is, I will bop them around HARDER when I am playing with them, lol, but if the same, or lighter bop is combined with the correcting intention, they know, from the expression on your face, or the sound of your voice. - A correction stops the undesirable behavior. It can be a slight 'no' or a leash correction. It should not be done unless your are SURE the dog understands the cue, and has had a chance to learn the behavior and cue in increasingly distracting environments. A 'stay' in the kitchen before feeding is way easier than a 'stay' out on a walk with squirrels, kids and other dogs.
Correcting a dog when they don't understand what you want is unfair. Ever been hit and you didn't know why? Then you understand.
Personally, I use positive reinforcement. Clicker training is one method, is intensely powerful, and the behaviors last much longer than those trained with traditional training. The click marks the behavior. The reward is food and praise. When the dog knows the behavior, phase out click and food, and just go with praise.
You can't give a killer whale a leash correction (even with a prong collar ;P) but you CAN teach very complex strings of behaviors - with clicker training. Actually they use whistles, but same idea.
Yelling at a dog shows more about the owner than the dog - unstable and stupid. Correcting a dog until he yelps - unless it was a GSD and you mistakenly over-corrected - is abusive. To brag about it is twisted. What a sad thing to be proud of.
Finally, to correct a stay, I calmly bring the dog back to the spot and make the stay from a closer distance and shorter time. Then go a little longer, from farther away.
Always end on a good note!
Add: My fellow trainers at work have all made the change to positive reinforcement, specifically clicker training. It can be a difficult transition for people with decades of traditional 'kick and treat' to move over to 'click and treat'. Some were unable to fully do it, and some use a little of both. I do both and use flat, martingales, slip , prong (very rarely, now that I do clicker training) and haltis collars.
Add2: Hi Aphrodite - I make the comparison to marker-based training in whales to demonstrate that forceful correction is not always necessary. Yes, dogs and whales are different, as are cats, camels, chimps and chickens. And clicker training is effective for all of them.
The reward need not be food based - I've used a tennis ball for GSD's, and a squeak from a toy for Goldens. Whatever works for every team. Every person and dog is different. And I only use it to train new behaviors or to brush up on shoddy work. I haven't used a clicker with my personal dog for months - she knows the cues she was taught. But to teach new behaviors - she just loves the clicker game! And my friend's dog is afraid of the clicker, so there's that. :)
Clicker training is certainly not a be all and end all for training - it is just a very positive way to train, and whenever anyone asks, I will suggest that someone learn it and try it. It teaches timing, observation and respect for the animal. It teaches trainers how to train. It certainly made me a better trainer.
ClickerExpo is a great resource for learning how it works. Ken Ramirez often presents there - he's got some wild stories. - I'm very firm with my dog. I do *NOT* want a dog that sees me as a weak owner.
My dog disobeys, I tug her ear and put her back into position. If she's on the lead, then I'll pull it back and say NO.
Too be honest, she doesn't need a lot of correcting, because I am firm with her and have been since she was 8 weeks. She is not a toy, or a baby or a child. She KNOWS I am in charge, and so that is how it will remain.
There is no messing around.
.....................................
Bells I am with you.
I don't treat- she gets to play with a tennis ball if training has gone well- which it ALWAYS does because I ALWAYS end on a good note- so my dog is eager to train again.
I treat my rescue- he has pure positive training because he's a nervous wreck and I discovered that the negative training makes him worse.
But if you have a dominating and confident dog there is no reason NOT to use a *pop* of the collar and firm handling- puts them right in their place. - I'm no trainer, but here are some of the corrections I use:
If I'm walking out the door for instance, and I tell my dog to "Stay" but he still tries to go infront of me, I immedietely walk infront of him blocking the way of the door and I say "NO. Stay..." and I walk out the door still telling him "Stay.... stay..... good stay.... COME!" and when I tell him "Come!" he runs to me and I praise with a "GOOD BOY, JR!".
If we're in the yard, and he see's somebody on their bike and tries to run up to them while I tell him "Stay" I just tell him "NO" and maybe a quick pop on the leash.
I use positive reinforcement. If he does something bad such as pull the leash, I say "NO PULL." and stop walking. I begin to walk as soon as he releases the tension on the leash. I taught him everything by positive reinforcement, the tone of my voice and simple leash corrections with a plain ol buckle collar. - I will occasionally use a No Reward Marker (NRM) to mark a mistake, but mostly I let the dog figure out that something went wrong because they didn't get a reinforcer. The vast majority of my communication with my dogs is positive reinforcement, with a bit of negative punishment, and positive punishment in the form of an NRM making up the remainder.
To be clear, because it seems from your question that there may be some misunderstanding, here are the definitions of the 4 quadrants:
Positive Reinforcement: Positive refers to adding something, and reinforcement means that whatever you did is likely to increase the future likelihood of the behavior occuring again. So positive reinforcement is generally the addition of something the dog likes- if you give a dog a cookie for sitting, sitting is likely to increase in the future.
Positive Punishment: Again, positive refers to adding something, and punishment means that whatever you did is likely to decrease the future likelihood of the behavior occuring again, in the future. So you do something like a leash pop to decrease the likelihood of the dog jumping up on you.
Negative Reinforcement: Negative refers to you taking something away, and reinforcement means that the likelihood of the behavior occuring in the future increases. This one is sometimes hard to wrap your brain around. Trainers who use e-collars by shocking the dog until the dog comes are using -R: the cessation of the aversive stimulus (the shock stopping ) is the negative part, and the dog working to avoid or turn off the shock increases the likelihood of the behavior increasing in the future. One thing to keep in mind, however, is that in order for the shock to be removed (the negative reinforcement part) it must first be applied (which is positive punishment) so the timing must be very good, or you can end up punishing the dog for starting to come.
Negative Punishment: Negative again refers to the removal of something, and punishment to decreasing a behavior. The prototypical example of this is removing your attention when the dog jumps on you..
For a list of good trainers who choose to focus on dog friendly training methods, see:
http://www.trulydogfriendly.com/blog/?pa…DeeDawg: I don't carry a clicker around forever, just in training sessions until the behavior is learned. There is increasing evidence that non-verbal markers are more efficient when teaching things that require precision and multiple steps. - pos reinforcement for training, corrections for a dog that disobeys when i know it knows what the command means. I harsh pop on the prong collar for disobeying a heel command or being distracted by another dog. Usually good behavior that follows is rewarded. if its a down stay and i have a helper training with me, the helper will pop the dog back into proper position if i am 2 far away to correct. again good behavior is than rewarded
I would love for you to spend a day with my dog when he was around 12 month of age training him and survive without giving him a correction. that would have been something
added: the guy in the video really doesn't make much of a case. he says punishment suppresses behavior than he says he uses punishment to change behavior? make up your mind dude. If you can't articulate it you probably shouldn't be teaching it. Haven't had a dog that needed so much correction that it would build up a tolerance. you correct once but you mean it and the problem should not surface again. If you are wishy washy about your corrections thats when the problems arrise because the dog ignores the correction and hence builds up a tolerance.
No comments:
Post a Comment