Working to get my dog to get more serious - Page 18

Valk, that is the whole game, reward,reward,reward, and that one day at trail the dog performs because he expect the reward, than next training reward, reward and reward again

but if you stop rewarding, or reward at wrong times etc, your dog will stop performing as desired

i was asking in regard of that remark:

if the decoy doesnot show up for 5 or ten times the dog will get more confused and eventually stop reacting to your "Pozor"

I guess most of those hoŕible accident are caused by stupid LE handlers, owners, people as whole ( like uneducated children running to every dog on leash they see, be it a Chihuahua or a Rottweiler.

susie, dogs used to go with us everywhere. on the beach they were been left (off leash) to watch after our stuff. there were plenty of people walking, running, jumping, playing with balls, etc. never ever even a single accident happend that dog left his spot and chased running kids or attack someone who didn't posed threat or didn't attempt to take a stuff, guarded by dog.

valk, that is not about who will do that, but about what will happen, proving that the command doesnot trigger the drive, the command only creates expectation, by pavlov way of learning, so if the command comes, but expected follow up will not come multiple times,it will confuse the dog, as by Pragers example, if you command a dog in an empty room, it will not start barking at nothing, but will wait for what he expects to happen after that command

I guarantee you that dogs that are competing at the top levels of IPO and PSA are trained with food and toys, along with compulsion, and they don't go down in drive during a trial or else they would not be winning.  Food and toys are used to teach behaviors.  At some point there will be competing motivators for a dog's behavior which results in disobedience after the dog has learned the behavior.  That is where compulsion comes in.  It is not either/or.  Either all positive reinforcement with no positive punishment or strictly compulsion.  Praise and petting will not get the level of drive needed out of a dog with very high drive.  Toys and food teach a dog how to learn through self discovery which is why when used to lay the foumndation, they lead to more reliable obedience.  However, the dog also needs to learn he has to obey, but he has to understand what is expected of him first.

Re: Juno's comment about teaching the out, a simple less conflicted approach would be to teach the dog to out for another bite object.  Same with getting a dog to out a ball.

Interesting discussion.

Of course toys and treats should be used as reinforcers for a long long time to learn a new behavior until it is very firmly entrenched in muscle memory.
Without the use of reinforcement, the behavior will extinguesh over time as we all know and Duke mentioned. Since every dog is different, and must be seen as such, it will extinguish faster in some dogs than others. (What do we learn: Always watch the dog...his reaction, don’t train by only one method or by rote.)

To LEARN a new behavior or sequence, always reward with a high value. Always.
Once the new behavior or sequence is learned and totally PROOFED, then...and only then....we should not forget about the importance of intermittent reinforcement.

Intermittent reinforcement just means the reward is typically random and unpredictable. Spaced apart, not to far apart in the beginning. But will keep the dog guessing.

Intermittent reinforcement is proven by many lab studies to work, and is likened to gambling at slot machines. How long will we stand there and throw in coins, just in the hopes of a win or a jackpot. We can be conditioned on expectation of a reward alone. Studies have shown that intermittent reinforcement is often more effective than constant dependable reinforcement.

Then down the road, later, one can space them apart after a particular sequence of commands or cues ...always adding a few cues or activities to the sequence....until we can get a whole program going without the need for reinforcement or reward during that sequence. You can do a whole routine lasting for a long time if this is done slowely, fairly and correctly, so that the dog knows, the reward/toy will appear after the complex sequence is over...whether that be 2 minutes, 5 minutes  or 30 minutes of constant work.

Not only intermittent reinforcement works, but one can also mix up rewards and reward delivery. Hid the toy in the vest or your shirt and bring it out only after a sequence. Lay the toy on the sidelines or in the car so the dog knows its there and at the end of the sequence you will walk the dog, calmly over to it and give the OK to grab it and tug. Switch it up with a game of tug or a throw. Change your voice to more excitement or less...”free” the dog to run as he likes as a reward, for a few minutes before calling him back to do more...all these are options to mix things up a bit.

If it is your dog, I think the difference sport or real life can be easily trained, and as a matter of fact, I think that most intelligent dogs can tell the difference by themselves, as Juno wrote in his last post how his dog reacts differently in practice with decoy or helper vs walking and seeing a suspious person. (In LE it might be different if the dog is used by a lot of different handlers.) But a normal dog living in a family, give the dogs the credit they are due as thinking animals. Let them have a part of your everyday life, off the training field (but still train them...in every walk of life whether in the Home Depot, at the beach or walking downtown with lots of people, on a close relaxed heel. Everything is training, and you dog will differentiate the circumstances.

You seem to have it down JJ. It amazes me how many people don't understand and utilize operant learning. Continuous reinforcement then shift to intermittent reinforcement. The four pillars are positive and negative reinforcement and positive and negative punishment. A stimulus is only a reinforcer if it increases the frequency of a behavior. The stronger the reinforcer that faster and better the dog learns. That approach is to get the best performance out of your dog. Your relationship with your dog is a whole other matter. Are you a fair and consistent pack leader? How well can you read your dog and adjust your training approaches? Do you properly praise and pet your dog? Are you honest with yourself about your dog's strengths and weaknesses? I have never been one to create an impoverished environment for a dog to build drive, such as crating him for long hours before training. I think the GSD breed ( and all dogs) deserve a quality lifestyle. I will manipulate food drive, but not in an extreme way. Teach a focused heel on the trial field and then a casual heel for daily heeling with the dog on the opposite side of you when you do the focused heel. A well bred dog will know when a threat is real and a dog trained in sport can protect for real. Also, train beyond sport to test your dog.

Jesse, there are thing like bond and hierarchical loyalty between handler and dog.
one of quality, which places GSD breed somewhat above other it's an innate urge always be close to masters and willingness to interact, cooperate with them.
when dog owner wasn't able to establish such bond, into play comes contrivances to tricks the dog into cooperation and to substitute absent personal connection.

Thanks Apple. Yes, your are right...there is so much more involved than training only on the field.
Valk-
You are right that the GSD innately want to always be with his person (or should want to) and interact and cooperate, which creates a bond and loyalty. Which is what IMO make them the perfect companion for me and all of us here.

But not every dog will give you his loyalty, or at least right away...there are also aloof dogs.
And in training a puppy, well, there is just no better way than with food.
But, I see reinforcement (play, treats, toys) as the icing on the cake...its fun for me and the dog. I love seeing them think “what should I do now to get this treat”...But, most importantly, it is an easy way to teach the young pup how the world he is going to live in (with me) will work for his whole life. We are just classically conditioning the pup to always do THIS when given THAT stimulus. Soon, you will not need treats or toys for daily life.
But, having said that, I always have some old kibble in every single pocket of every piece of clothes I own (a real drag when I have to do laundry, BTW) or a small tug in the car...so that I will always now and then, surprise him with a reward when I particularly like what he has done for whatever reason. I do this until the end of their life. This creates an unbreakable bond. Will my dog do things for me without rewards? Sure he will...but I like to reward sometimes. It rewards me too, seeing him happy.

We humans also attach ourselves to people that enhance our lives, and not to those that are neutral, aloof or don’t return our affections (rewards)...right? Or at least we should. But we humans are more complex emotionally and no doubt, do weird things, or attach to people, that don’t behoove us at times. But technically most of the time, we are no different. Only the rewards are more hidden or higher staked.

PS:
In sport, I don’t even know of another way to train EFFECTIVELY than with a reinforcement system. If you do Valk, let me know. I would really be interested in your opinion.because I think I have read this from you in the past, that you are very against rewards.
... even each bite is a reinforcement - reward for some action.