Post by oksaradt on May 3, 2011 15:00:42 GMT -5
Probably the most important lesson I ever learned and internalized in dog training is timing and consistency. Both sound simple. Both can take years or longer for some handlers to achieve and some just never do.
When it comes down to it, a dog handler/trainer is really a linquist with a willing student that wants/desires to please him/her. Granted this desire is self-serving on the dog's part, but so is our desire to convince the dog to do our bidding.
Now, how the handler/trainer chooses to communicate with the dog is a menu thicker than the NYC phonebook. The variations the handler creates by his/her interpretation makes this is even more so. It really doesn't matter though as long as the handler/trainer is CONSISTENT and has very good TIMING. If you use an ear pinch/e-collar to tell the dog when it's doing wrong, to convey this to the dog you need to stim it before the ill-thought is formalized and sent to the muscles, i.e. if a dog has started a chase the proper stim will be too late. Knocking the dog to the ground with a strong stim simply implies the handler/trainer was slow and has to use power as a crutch.
On the reverse, if the dog does a desired act, the reward must come as the act is happening to convey to the dog "yes, this is what I want when I say WHATEVER." It helps if "WHATEVER" is used every time (consistency) as the request to teach the dog what "WHATEVER" means. "WHATEVER" may not even be a word, but a hand signal....even if the handler/trainer is oblivious they use the same hand signal unconsciously every time.
(Example, my first working dog Dax and I were waiting in a line to do a building search. The line was crowded due to circumstances and I was trying to take up less room and would pull my right arm in causing my right hand to go up to my chest. Dax would bark for (what I thought) was no reason. After three of these barks at different times, I realized she was barking each time I brought my hand up to my chest in a swinging motion. She did this because I often teased her with the reward (when I was teaching the bark) by clutching it close to my chest. - I'd created a "cue" by being consistent. The dog (an alien creature that considers Human as a second language) had concluded that my act was a request for a bark.)
Realize that a working dog really does want to work and wants to do well at it. Many humans just expect their dogs to know what they want. ....well duh....the supposedly less intelligent creature has evolved to do just this by reading our consistent behavior with our intent. Unfortuantely, the guessing game doesn't always work with more complicated tasks.
So, we humans have worked out hundreds/thousands of "THE Way" to train dogs from the clicker to the stim collar to everything in between. If our response time sucks, no method is going to work. Contrary to common opinion, a dog does not have associative memory to the past. The comment of "He KNOWS he's done wrong, look at that expression." Tranlastion: (dog knows human goes psycho whenever dog and a poop pile on the white shag rug appears. Dog shows submission in hopes the crazy human isn't going to shove dog's nose in the pile, beat dog profusely, and throw it out on it's butt in the backyard while yelling gibberish.) That the dog created that pile on the white shag three hours ago because the owner didn't know how long a puppy can go without pooping....well, that's irrelevent to both parties. This would definitely be a case of "poor timing", not to mention clueless human.
So, let's say we use a "marker system" where something good happens when the dog does a desired behavior after a verbal stimulus......
wait wait....how can we expect the dog to know the correct behavior for the verbal stimulus.... such a quandry.....well, a clicker trainer would start out with the dog sitting and click while saying "good sit"..... me, I just go "Sit" in a nice tone....ya know sort of like taking someone from "east side of really nowhere" and walking around your house pointing to items and saying THE RIGHT TERM (or the english term...could be the spanish term...british term....french term....all depends on the redneck flavor of the month) for the item. You know......table ....TAYYYBELLLL..........window....WINNNNNNDOH...we do tend to yell and stretch the words in hopes this assists in understanding. Many handlers do the same thing...sit.....SIT.....SSSSSSIT...."I SAID SIT DAMNIT! YOU DEAF DOG?! Don't run away on me......
As with humans, children learn foreign languages much faster and so do puppies.....again as long as the humans have a clue.
Dogs trained by foreighn handlers learn foreign words in hopes that the thief won't know to use "plotz" for "down" on the malinois racing at his behind. Makes one wonder if the Deutsche law enforcement officer doesn't use southern texas words for their commands.....making sit a two-syllable command.
Have I gotten the point across that it really doesn't matter what identifier you use to elicit a behavior from a dog? So, what does matter? That you use the same ones for the same desired behaviors again and again....consistency.
Consistency is not only in language but in expectations. If you don't think so, just lose a scent source like a bone (less than an inch long ) in a large pile of light brown twigs and you'll suddenly realize how dependent you are on the precision of your HRD dog. Realize that for months you've been training your dog that within six inches is good enough. Now, you want that same dog to find the gnat on the elephant's behind. Not only do you want the dog to find it, but you want the dog to show you definitively where it is such that you can pick it up and repossess it or worse, have the dog be able to point to it such that the semi-blind geriatric anthropologist can see it and collect it.
If you are inconsistent, ye shall reap what ye sow.
So, Timing....The quicker the handler/trainer can provide a response stimulus to a desired and/or undesired behavior, the more likely the dog will comprehend their wish. With dogs, this is reported to be within 3 seconds. My observations are that three seconds is too long. The sooner you can reward the dog for the right behavior, the sooner it's going to learn what you expect it to do.
I seriously advise some handlers to get a video game where they have to respond faster and faster to win to improve their timing. I've advised handlers to practice throwing the ball with someone giving a trigger to improve their timing. There's nothing like a spouse saying, "throw" and then nicely responding with "no offense, but you suck dear" to give you incentive to improve. It does help if the ball lands right where the dog is doing the desired stimulus like touching a nose or paw to the source. Hitting the dog is not advisable, but my first dog did get versed at catching the ball when my aim was off.
If your timing is off and the dog "finds" the source while you respond two seconds later with "goooooood", the dog now thinks you like it when it watches the fruit fly zip by. If you have a really fast dog, then you have to really be quick because it not only watched the fruit fly go by but counted the eight others in formation around it. Oddly as it sounds, there are dogs that react much like humans in that if the handler doesn't "get it" then the dog yells louder and louder the same bark over and over again in hopes that you'll finally get it. Unfortunately, they have the same affect on us as we humans have on the dogs. Both wish the other would chill. Such dogs are aching for a definitive communication system. Such dogs bark in scent but can have a hard time targeting for a multitude of reasons, but primarily it is due to frustration with their handler. I knew one handler (back in the day) with FEMA that was just happy that his dog barked incessantly when it was in scent. No one could tell where the victim or source was, but he knew they were there somewhere. Usually such dogs make finds a multitude of times but have not been given a way to tell their handler. This can mean the dog doesn't know to stay at the source until the handler recognizes it because this was never built up.
Staying at the source implies commitment. There are several ways to teach commitment, but they all start with the dog being rewarded with the find until the dog feels comfortable with that part of the process. That means the handler has to reward fast at the initial finds until they feel the dog gets it. For me that's like 100 finds all in a row in all sorts of different situations. Then and only then I might intentionally slow down my reward time just a bit then a bit more then a bit more until the dog decides to leave. At that point my timing again must be fast to correct the dog for leaving the source. Dog goes back, tells me about the source, I reward fast. I'm conveying to the dog that I knew it had it right, but I chose to wait on the reward. I'm conveying to the dog, "yes, you got it right. No, don't leave it until I reward."
If your dog has a down alert and you didn't see it make the find, you better hope it has a lot of committment to the source so that you can find it before it gives up on you. If it leaves the source to find you, then you may not ever have a clue that it made a find as conditions may have changed such that the dog doesn't repeat its find for you.
There are lots of analogies I can tell as to why timing is so critical, but it comes down to this. The faster a handler/trainer responds to a desired behavior, the quicker the dog learns what it is supposed to do or not do. Be consistently slow and the dog will likely invent its own game and its own rules.
Probably my biggest weakness when training someone else is I'll reward their dog if they don't do it quick enough.
And people, I'm old, fat, and slow..........
Jim
When it comes down to it, a dog handler/trainer is really a linquist with a willing student that wants/desires to please him/her. Granted this desire is self-serving on the dog's part, but so is our desire to convince the dog to do our bidding.
Now, how the handler/trainer chooses to communicate with the dog is a menu thicker than the NYC phonebook. The variations the handler creates by his/her interpretation makes this is even more so. It really doesn't matter though as long as the handler/trainer is CONSISTENT and has very good TIMING. If you use an ear pinch/e-collar to tell the dog when it's doing wrong, to convey this to the dog you need to stim it before the ill-thought is formalized and sent to the muscles, i.e. if a dog has started a chase the proper stim will be too late. Knocking the dog to the ground with a strong stim simply implies the handler/trainer was slow and has to use power as a crutch.
On the reverse, if the dog does a desired act, the reward must come as the act is happening to convey to the dog "yes, this is what I want when I say WHATEVER." It helps if "WHATEVER" is used every time (consistency) as the request to teach the dog what "WHATEVER" means. "WHATEVER" may not even be a word, but a hand signal....even if the handler/trainer is oblivious they use the same hand signal unconsciously every time.
(Example, my first working dog Dax and I were waiting in a line to do a building search. The line was crowded due to circumstances and I was trying to take up less room and would pull my right arm in causing my right hand to go up to my chest. Dax would bark for (what I thought) was no reason. After three of these barks at different times, I realized she was barking each time I brought my hand up to my chest in a swinging motion. She did this because I often teased her with the reward (when I was teaching the bark) by clutching it close to my chest. - I'd created a "cue" by being consistent. The dog (an alien creature that considers Human as a second language) had concluded that my act was a request for a bark.)
Realize that a working dog really does want to work and wants to do well at it. Many humans just expect their dogs to know what they want. ....well duh....the supposedly less intelligent creature has evolved to do just this by reading our consistent behavior with our intent. Unfortuantely, the guessing game doesn't always work with more complicated tasks.
So, we humans have worked out hundreds/thousands of "THE Way" to train dogs from the clicker to the stim collar to everything in between. If our response time sucks, no method is going to work. Contrary to common opinion, a dog does not have associative memory to the past. The comment of "He KNOWS he's done wrong, look at that expression." Tranlastion: (dog knows human goes psycho whenever dog and a poop pile on the white shag rug appears. Dog shows submission in hopes the crazy human isn't going to shove dog's nose in the pile, beat dog profusely, and throw it out on it's butt in the backyard while yelling gibberish.) That the dog created that pile on the white shag three hours ago because the owner didn't know how long a puppy can go without pooping....well, that's irrelevent to both parties. This would definitely be a case of "poor timing", not to mention clueless human.
So, let's say we use a "marker system" where something good happens when the dog does a desired behavior after a verbal stimulus......
wait wait....how can we expect the dog to know the correct behavior for the verbal stimulus.... such a quandry.....well, a clicker trainer would start out with the dog sitting and click while saying "good sit"..... me, I just go "Sit" in a nice tone....ya know sort of like taking someone from "east side of really nowhere" and walking around your house pointing to items and saying THE RIGHT TERM (or the english term...could be the spanish term...british term....french term....all depends on the redneck flavor of the month) for the item. You know......table ....TAYYYBELLLL..........window....WINNNNNNDOH...we do tend to yell and stretch the words in hopes this assists in understanding. Many handlers do the same thing...sit.....SIT.....SSSSSSIT...."I SAID SIT DAMNIT! YOU DEAF DOG?! Don't run away on me......
As with humans, children learn foreign languages much faster and so do puppies.....again as long as the humans have a clue.
Dogs trained by foreighn handlers learn foreign words in hopes that the thief won't know to use "plotz" for "down" on the malinois racing at his behind. Makes one wonder if the Deutsche law enforcement officer doesn't use southern texas words for their commands.....making sit a two-syllable command.
Have I gotten the point across that it really doesn't matter what identifier you use to elicit a behavior from a dog? So, what does matter? That you use the same ones for the same desired behaviors again and again....consistency.
Consistency is not only in language but in expectations. If you don't think so, just lose a scent source like a bone (less than an inch long ) in a large pile of light brown twigs and you'll suddenly realize how dependent you are on the precision of your HRD dog. Realize that for months you've been training your dog that within six inches is good enough. Now, you want that same dog to find the gnat on the elephant's behind. Not only do you want the dog to find it, but you want the dog to show you definitively where it is such that you can pick it up and repossess it or worse, have the dog be able to point to it such that the semi-blind geriatric anthropologist can see it and collect it.
If you are inconsistent, ye shall reap what ye sow.
So, Timing....The quicker the handler/trainer can provide a response stimulus to a desired and/or undesired behavior, the more likely the dog will comprehend their wish. With dogs, this is reported to be within 3 seconds. My observations are that three seconds is too long. The sooner you can reward the dog for the right behavior, the sooner it's going to learn what you expect it to do.
I seriously advise some handlers to get a video game where they have to respond faster and faster to win to improve their timing. I've advised handlers to practice throwing the ball with someone giving a trigger to improve their timing. There's nothing like a spouse saying, "throw" and then nicely responding with "no offense, but you suck dear" to give you incentive to improve. It does help if the ball lands right where the dog is doing the desired stimulus like touching a nose or paw to the source. Hitting the dog is not advisable, but my first dog did get versed at catching the ball when my aim was off.
If your timing is off and the dog "finds" the source while you respond two seconds later with "goooooood", the dog now thinks you like it when it watches the fruit fly zip by. If you have a really fast dog, then you have to really be quick because it not only watched the fruit fly go by but counted the eight others in formation around it. Oddly as it sounds, there are dogs that react much like humans in that if the handler doesn't "get it" then the dog yells louder and louder the same bark over and over again in hopes that you'll finally get it. Unfortunately, they have the same affect on us as we humans have on the dogs. Both wish the other would chill. Such dogs are aching for a definitive communication system. Such dogs bark in scent but can have a hard time targeting for a multitude of reasons, but primarily it is due to frustration with their handler. I knew one handler (back in the day) with FEMA that was just happy that his dog barked incessantly when it was in scent. No one could tell where the victim or source was, but he knew they were there somewhere. Usually such dogs make finds a multitude of times but have not been given a way to tell their handler. This can mean the dog doesn't know to stay at the source until the handler recognizes it because this was never built up.
Staying at the source implies commitment. There are several ways to teach commitment, but they all start with the dog being rewarded with the find until the dog feels comfortable with that part of the process. That means the handler has to reward fast at the initial finds until they feel the dog gets it. For me that's like 100 finds all in a row in all sorts of different situations. Then and only then I might intentionally slow down my reward time just a bit then a bit more then a bit more until the dog decides to leave. At that point my timing again must be fast to correct the dog for leaving the source. Dog goes back, tells me about the source, I reward fast. I'm conveying to the dog that I knew it had it right, but I chose to wait on the reward. I'm conveying to the dog, "yes, you got it right. No, don't leave it until I reward."
If your dog has a down alert and you didn't see it make the find, you better hope it has a lot of committment to the source so that you can find it before it gives up on you. If it leaves the source to find you, then you may not ever have a clue that it made a find as conditions may have changed such that the dog doesn't repeat its find for you.
There are lots of analogies I can tell as to why timing is so critical, but it comes down to this. The faster a handler/trainer responds to a desired behavior, the quicker the dog learns what it is supposed to do or not do. Be consistently slow and the dog will likely invent its own game and its own rules.
Probably my biggest weakness when training someone else is I'll reward their dog if they don't do it quick enough.
And people, I'm old, fat, and slow..........
Jim