Poodle Forum banner

1 - 20 of 21 Posts

·
Registered
Joined
·
2,320 Posts
Discussion Starter #1
I prefer positive reinforcement (R+) training as opposed to negative reinforcement (R-) or positive punishment (P+) training, however, I have not been very successful using R+ methods with Mitchell. He's so focused on the reward that he is not focusing on learning the command. I have taken to using P+, and have had amazing results. This site explains the difference between training methods Examples of operant training quadrants
Loose leash walking is a breeze now, and he is starting to sit, shake paw, and lie down on command without a treat in sight.

Are positive reinforcement methods supposed to work with every dog personality?

I know it's impossible to tell me if I was not training him correctly, but was I too impatient? Should I have continued with R+ despite the slow/no progress?
 

·
Registered
Joined
·
1,905 Posts
Sounds like maybe he needs some impulse control training. Teaching him commands like leave it and turn away will help with that.

I can imagine that R+ might not work for every dog. I'm sure it's possible to use a combination of P+ and R+, or maybe even a combination of all three, depending on the dog's personality.
 

·
Registered
Joined
·
2,537 Posts
I say use what works! Since youve had Mitch for awhile now, and are just getting him to this point in training then i would say he needed a change. I use mostly positive training, but im not opposed to a "sharp mom voice" or a small collar pop. I think when people get way hung up on one method and one method only is when they have problems. Every dog is different.
 

·
Registered
Joined
·
2,314 Posts
I think when people get way hung up on one method and one method only is when they have problems. Every dog is different.
I couldn't agree more. Every dog is different and has different needs. I love positive reinforcement/ clicker training new behaviours. But sometimes you need small aversives to correct problem behaviour. So if what you are doing works for Mitch, then that's great!
 

·
Registered
Joined
·
1,905 Posts
I really can't get the hang of clicker training, but it's okay because I think we're doing okay without it.

I agree with you guys--use whatever works with your dog. You know him better than anyone else.
 

·
Registered
Joined
·
2,686 Posts
I can't tell you because I don't use only positive reinforcement. However, I think it is wrong to get so stuck on one method that it becomes like a religion. You have to use what works.

Try using R- or P+ and see if you get the results you want.
 

·
Registered
Joined
·
2,320 Posts
Discussion Starter #7
Okay, that's great to hear.
I've been using R-/P+ methods, and we're making big strides.

Thanks!
 

·
Registered
Joined
·
2,686 Posts
Glad to hear it. Positive training is great, but I have to tell you that I don't know a single Obedience competitor who uses only Positive training. Sometimes the dog has to know that there is a negative consequence for noncompliance.
 

·
Registered
Joined
·
114 Posts
Hi Locket, as you say we do not know exactly what your doing. I use Clicker training (R+ with event marker) it is good to teach incremental steps. Here is an example:

---------------
If your dog won’t go all the way down:

If your dog only lowers his head part way as he follows the treat, you can use this movement to shape the behavior. First, click and treat whenever he makes a partial movement down. Then only click and treat when he moves a little further down. Build on each movement in the right direction until he offers the complete behavior.
-----------------
I think some people expect too much when training and the little steps help meet your overall goal. Since he is interested in treats, you can get his attention & build on that. Good luck
 

·
Registered
Joined
·
17 Posts
cbrand, dogs trained by (good) gung-ho R+ trainers do know their actions can have undesirable consequences. R+ goes hand in hand with P- in positive dog training. Faith used to forget the ball in flyball. The consequence to that mistake was she had to leave the race and have really boring time outside. It took a little while, but once she figured out that no ball meant no fun, she stopped making that mistake.

There are also obedience folks who use mostly R+. Morgan Spector has written an excellent book for anyone interested in training for competition obedience using R+. Faith and I don't show yet in traditional obedience, but we have earned her RE, RL3, and ARCH with great performances and no collar pops. Why should she have to learn that failing to obey causes her pain when thus far she's always learned that it's fun and worthwhile to play my game?

I believe that it's possible to use P+ and R- wisely and fairly, but that some people are not that good at training. I'd rather mess up by giving my dog an undeserved treat or play session than an undeserved correction. I want my dog to not worry that trying new things or not knowing the correct response is going to get her into trouble. I think R+/P- can absolutely work with any animal, provided the trainer can figure out what motivates that animal. Susan Garrett has some good impulse control games--check out "It's Yer Choice" on YouTube.
 

·
Registered
Joined
·
2,686 Posts
Hey I'm all about using methods that work. However, understand that Competitive Obedience is a very different sport than Rally or Fly Ball or even Agility. In Rally you can give multiple commands and your dog always works close to you. In Agility you get to give multiple commands and you have a wide latitude in what is correct (you just have to have 1 toe nail in the contact area). In Obedience you get one single command and have to be exact the very 1st time or you flunk out.

When you start to train in Obedience and your dog fails to retrieve.... what will you do? Obedience retrieves aren't as hyped up fun as Fly Ball retrieves. I'm not sure that sitting out is going to be enough insentive. :) If your dog gets up during the long sit/down how will that be addressed with only positive training?

Anyway, if you are using only Positive training and you are getting the results you want. More power to you.
 

·
Registered
Joined
·
17 Posts
APDT rally (my favorite venue) penalizes multiple commands and also includes a formal retrieve--one that to my mind is more difficult than the Open retrieve (although the scoring IS more lax). But what counts as a failure or deduction in the rules and what counts in the handler's mind are two different things. If my dog earned a perfect score in a rally run while I played the cheerleader, that is not a success to me. If my dog touches the yellow with one toenail, we might Q but that is not a success to me. Just because the rules may allow sloppiness does not mean that the competitor trains for or accepts sloppiness.

P- in obedience works pretty much the same as in flyball. What goes away for Faith is the opportunity to earn a reward. However, you are right in that obedience is a less inherently enjoyable sport for the dog, and because of that I think R+ is far more useful in that context--not only because it effectively trains the dog but because it makes obedience fun, and what is the point if both team members aren't enjoying themselves? As for what do I do if my dog fails to retrieve or hold her stay? Simple: I do not reinforce that failure. I figure there's been a hole in our education somewhere or that my dog is not yet ready to perform in that environment. Something else to think about: if my clicker-trained dog fails to retrieve (or perform whatever exercise) in the ring, my response has to be the same as that of a so-called "balanced" or "motivational" trainer, which is to not correct it. In that scenario, my dog has not really learned anything. The dog trained with corrections, though, has learned that the ring is a safe place to goof off. I hear pro-correction competitors complain about their dogs "blowing them off," but I don't hear that so much from R+ people.

Sorry for rambling on, especially on this tangent! Like I said before, I am not completely against P+/R-. But I definitely wish that clicker training and other R+ training was much more prevalent in competitive obedience circles.
 

·
Registered
Joined
·
2,686 Posts
APDT rally (my favorite venue) penalizes multiple commands and also includes a formal retrieve--one that to my mind is more difficult than the Open retrieve (although the scoring IS more lax).
How is the retrieve done in APDT Rally?

Sorry for rambling on, especially on this tangent! Like I said before, I am not completely against P+/R-. But I definitely wish that clicker training and other R+ training was much more prevalent in competitive obedience circles.
Well you have to ask yourself why it isn't. Perhaps it is not the most appropriate training method for the sport. I also know that clicker training is not typically used in Hunt competitions and Field Trials. Why is that? I never used the forced fetch on my 1st bitch but my 2nd bitch had a habit of quitting or not wanting to pick certain items up. I taught her a forced fetch and now I'm sold on it. I feel like I get a much more reliable working retrieve.

I will be interested to hear how your Competitive Obedience work goes. I'm still curious though. If you are training sits and downs, what do you do if your dog gets up from its sit? What do you do if your dog sniffs or makes a move towards another dog? For me it is simple. I don't say anything (that just gets the dog waiting for a second command). I walk back and give a firm collar correction. I have to say that using this method, our sits/downs are really, really good!
 

·
Registered
Joined
·
1,648 Posts
what a lot of people miss IMO is that clicker training isn't OMG here's the whole new wy to do it.

it's a "here's a way to let the dog know when he's doing it RIGHT" now go do it which eve rway makes sense/works for that dog.

iE some clicker people will NEVER lure ever
well unless you wnt to sit there for a million years- that dog might never ever lay down. ever. for you to click it- lure way- click treat.

when i started clicker working (10? 12 years ago) i did it with horses nd dogs. And the line "The rules still exist" was the one tht mde me really get it- yes we were using the clicker- but the RULES that i hd before still exist- jumping up STILL gets 'NO" horse stepping on me still gets "OMG MOVE OVER OUCH MY FOOT!!!!!!"

impulse control nd teching new 'instnt behvoirs" is another thing. Your not teching the dog to do nything your just helping them learn to offer a behvoir tht is not what they did before. IE my little danish dog cme to me- his reaction to everything stressfull ws to snap and scream

we did controll unleashed exercises- and now his dfault behaivors are to look at me- nd sit... and to lay down.

cbrand- in your example- with the sit- i'm like you- if they blow it nd they KNOW how to sit- there's a correction. again clicker training doesn't and shouldn't mean tht corrections no longer exist. It's only to replce what used to be a 'good boy" nd to encourge more 'want to work' with the dog
 

·
Registered
Joined
·
17 Posts
In APDT, the retrieve article is placed in front of the last sign by the judge or steward while the team is elsewhere on the course, so that the dog must be directed to mark before being sent to retrieve. I think it is harder because the dog does not see the handler throw the article as in the Open retrieve.

I know nothing about field work, so I won't venture an opinion there! I know many herding trainers use plenty of aversives as well, and honestly I am sure it's worth it to the average working retriever or herding dog; they seem to live for their job. Most dogs do not love obedience the same way, though. I do believe that in general a dog trained with a forced retrieve cannot enjoy the exercise like a clicker-trained dog, and I also know how some dogs (like my mom's whippet) would respond to a force fetch--it would not be effective at all because she would simply shut down.

Like I said before, my response to my dog breaking her stay would be to not reinforce it and possibly to remove further reinforcement opportunities for a bit or simply ask her to take her position again. An example of effective P- in this scenario: if my dog eats from a bowl for her dinner, she does a sit-stay before she is released to eat. If she breaks, bummer, kid--bowl goes away, Mom ends the pre-dinner ritual, and Faith gets woe-filled eyes. She comes to understand that maintaining the stay is the only way to get her dinner. However, because I try to always set my dog up for success and increase the difficulty of an exercise in increments, she generally understands how to earn the reward. Since she wants the reward, she generally does not fail. The same process pretty much goes with any obedience issue.
 

·
Registered
Joined
·
114 Posts
<snip>
Like I said before, my response to my dog breaking her stay would be to not reinforce it and possibly to remove further reinforcement opportunities for a bit or simply ask her to take her position again. An example of effective P- in this scenario: if my dog eats from a bowl for her dinner, she does a sit-stay before she is released to eat. If she breaks, bummer, kid--bowl goes away

<snip>
How long do you wait to give the bowl back for feeding?
 

·
Registered
Joined
·
2,686 Posts
In APDT, the retrieve article is placed in front of the last sign by the judge or steward while the team is elsewhere on the course, so that the dog must be directed to mark before being sent to retrieve. I think it is harder because the dog does not see the handler throw the article as in the Open retrieve.
Interesting. That would be a fun retrieve. How is it scored? Do you flunk if you mark and send and your dog does not retrieve the 1st time? Do you get multiple opportunities to mark?

Like I said before, my response to my dog breaking her stay would be to not reinforce it and possibly to remove further reinforcement opportunities for a bit or simply ask her to take her position again.
The problem I see there is that repeating a command is the kiss of death in Obedience. This is the downfall of many handlers because their dogs start to wait for the 2nd command.
 

·
Registered
Joined
·
17 Posts
Taxi, the time before a retry varies. A few minutes, at least.

cbrand, I am actually not sure about the scoring for that exercise. It is online on the APDT website if you are really interested. I am sure you can ask your dog to mark multiple times, but you might get a deduction. I don't know if you can resend your dog if she fails on the first attempt; I suspect not just because I have never seen anyone do that. You can't NQ on that exercise, though, unless your dog does something really wrong like leave the ring--the worst that happens is you don't get credit for it.
 

·
Registered
Joined
·
2,686 Posts
Taxi, the time before a retry varies. A few minutes, at least.

cbrand, I am actually not sure about the scoring for that exercise. It is online on the APDT website if you are really interested. I am sure you can ask your dog to mark multiple times, but you might get a deduction. I don't know if you can resend your dog if she fails on the first attempt; I suspect not just because I have never seen anyone do that. You can't NQ on that exercise, though, unless your dog does something really wrong like leave the ring--the worst that happens is you don't get credit for it.
Can you pass then even if you don't complete all the exercises? I'll go look at the site.
 

·
Registered
Joined
·
17 Posts
An APDT rally course at any level consists of the main course plus a bonus exercise. The main course is out of 200 points, and the bonus exercise is worth an additional 10 points. An IP at the bonus just results in 0/10 for that exercise and does not impact the score you already earned in the main course. The retrieve is one of the bonuses of Level 3. Therefore you can mangle that exercise and still Q. An IP of any other sign on the course (in other words, any non-bonus exercise) results in an NQ.

The bonus thing is kind of weird compared to AKC rally...but it's a very fun venue, especially for any dog sport newbies like me!
 
1 - 20 of 21 Posts
Top