How do you get your dog to "work for you not a scrap of food etc." - Page 7

Go Back   Dog Forum > Keeping and Caring for Dogs > Dog Training and Behavior

How do you get your dog to "work for you not a scrap of food etc."

This is a discussion on How do you get your dog to "work for you not a scrap of food etc." within the Dog Training and Behavior forums, part of the Keeping and Caring for Dogs category; Yay, happy to have contributed in a helpful way. I guess I sort of think of it as muscle memory; enough repetition and you don't ...

User Tag List

Like Tree65Likes

 
LinkBack Thread Tools Display Modes
Old 09-19-2017, 03:14 PM
  #61
Senior Member
 
AlwaysTomboy's Avatar
 
Join Date: Jun 2017
Location: near Reading, PA
Posts: 340
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Yay, happy to have contributed in a helpful way.

I guess I sort of think of it as muscle memory; enough repetition and you don't have to think about it any more. The brain recognizes a specific set of criteria and tells the body what to do at the subconscious level and the body responds automatically. The best example I can think of for myself is playing a musical instrument. If I see a 'C' note on the sheet music, I don't think, "Ok, on the clarinet, a 'C' is thumb, index finger, middle finger, ring finger here, here, here and here." I just see 'C' and play the note. My fingers know where to go without me consciously thinking about it.

My boyfriend does martial arts and fencing and would probably have a better combat type example.

I'm just spit balling here, but is it possible automatic behaviors like that may not be rewarding as such, but something the dog learns to accept and do as proper etiquette when in the company of a particular human? I guess I'm thinking of it maybe the same way that dogs learn proper social interaction with dogs from other dogs, maybe they can also learn proper social interaction with people from people.

Maybe 'bike stops, dog sits and waits for command' from the dog's perspective is just one of those weird things a dog has to do when hanging out with humans (or a particular human), like tolerating hugs or smiles or staring, all of which have opposite meanings to dogs than they do to people, so (some) dogs learn to accept those things as normal human interactions and not as hostile or threatening.

Or am I giving doggy brains too much credit? Maybe those automatic behaviors just amount to another way a dog does something just because it's what the owner wants and expects. The reward is the approval of the human.
leashedForLife likes this.
AlwaysTomboy is offline   Reply With Quote
Old 09-20-2017, 10:48 AM
  #62
Banned
 
Join Date: Aug 2017
Location: Boston metro-area, USA
Posts: 1,885
Mentioned: 46 Post(s)
Tagged: 0 Thread(s)
Arrow well-rehearsed, fluent behaviors: Association, maintenance, & rewards internal & ex-

Quote:
Originally Posted by AlwaysTomboy View Post

... maybe the same way that dogs learn proper social interaction with dogs from other dogs, maybe they can also learn proper social interaction with people from people.

Maybe 'bike stops, dog sits and waits for command' from the dog's perspective is just one of those weird things a dog has to do when hanging out with humans (or a particular human), like tolerating hugs or smiles or staring?...
all of which have opposite meanings to dogs than they do to people, so (some) dogs learn to accept those things as normal human interactions, and not as hostile or threatening.

Or am I giving doggy brains too much credit?
Maybe those automatic behaviors just amount to another way a dog does something just because it's what the owner wants and expects. The reward is the approval of the human.
_____________________________________________

I think in general, humans give dog-brains much too little credit.

Since we learned that dogs care capable of inferring the name of a novel object by eliminating the *known* named-objects & hypothesizing that the novel name must belong to the novel object, a process that starts only when humans reach approx 4 to 5-YO, researchers at least have considerably more regard for dog cognition & learning than they had before.

To get back to rewards / dog behavior, I'd say these fluently-learned, familiar behaviors have become, at least to a degree, self-rewarding - they've been positively reinforced many, many times, over months or years, & are now part of the default repertoire of that dog - at least, in the company of that human, & in that context [rider & bike + dog].
They are so well-known that they are nearly automatic - & they are highly associated with external rewards, so that they generate 'internal rewards' - a pleasant happiness while performing them.

That doesn't mean they don't merit praise, & i'd still maintain an intermittent reinforcement with external rewards - winning the prize periodically on a random nonschedule is the best way to keep a fluent behavior; the gambler's hook is the best maintenance tool we've got in the toolbox, according to research, & IME / IMO.

There's plenty of data on the random-rewards / maintained AKA persistent behaviors connection - 'persistent' is used for those behaviors we don't want, 'maintained', for those we do.
One relative who visits no more than once every few months, who persists in signaling the dog to jump-up, is sufficient to maintain the unwanted behavior. That's persistence.
One highly-desired reward in a dozen times while bike-jogging may be sufficient to maintain auto-stops / auto-sits at curbs, but that doesn't mean i wouldn't praise the action each time, occasionally cue it B4 it occurs, & reward the dog MORE OFTEN than mere maintenance requires.
If a single high-value reward for a single auto-stop / auto-sit of the many during just 1 in 12 bike-jogging trips will "maintain" the behavior, i would personally better than double that rate, & cue plus reward highly, at least once in every 5 trips.
It keeps the cue fresh in the dog's mind, not hazy in memory, & it provides an impetus to perform the behavior fluently, to a high degree of accuracy - since such a savvy dog would already have realized that "doing it REALLY well" gets the goody.

Dogs learn to work us, too - & there's nothing bad in that.
Doing a behavior to excellent standards in order to get something excellent back, is a simple quid pro quo, & the dog has certainly still earned their reward.

- terry

leashedForLife is offline   Reply With Quote
Old 09-20-2017, 08:57 PM
  #63
Banned
 
Join Date: Apr 2015
Posts: 1,404
Mentioned: 104 Post(s)
Tagged: 0 Thread(s)
The idea is that what the dog should learn/execute to do per some of my expectations is not what I would consider "excellent". Yes, maybe up front when the behavior is being developed and nailed for the first times but after that expectations become more "expected" and the "quid pro quo" is tempered but still commensurate with expectations based on the dog's abilities and acknowledgement of understanding/ displaying the particular behavior. Today, I spent some time revisiting some of our basics both in training and play and I don't recall making anything more of it than some general reinforcing reciprocation because the dog knows it better than I probably do. I'd like to think the dog simply worked with me in a wonderfully compatible fashion which gave the dog as much enjoyment as it did me. I'd also like to think that this enjoyment was as much a reward as the mint chip ice cream I shared with her this evening for no "reward" reason at all. I made sure to give her bites with minimal chocolate pieces in it.
DriveDog is offline   Reply With Quote
 
Old 09-20-2017, 10:06 PM
  #64
Senior Member
 
Markie's Avatar
 
Join Date: Sep 2017
Posts: 1,485
Mentioned: 74 Post(s)
Tagged: 1 Thread(s)
Dogs are a lot like children. LFL would you give your child a piece of chocolate Everytime they went to the bathroom?? Or would you expect them to learn and understand without something being there all the time to promote it. Kids like surprises and things and so do dogs. The cow joint bone my dogs get who knows when is a complete surprise to them and they enjoy the heck out of it. I have my seven year old sister who lives with me also. I expect her to clean her room and go to bed she doesn't get a reward for that behavior. She just does it.
DriveDog and AlwaysTomboy like this.
Markie is offline   Reply With Quote
Old 09-20-2017, 11:47 PM
  #65
Banned
 
Join Date: Aug 2017
Location: Boston metro-area, USA
Posts: 1,885
Mentioned: 46 Post(s)
Tagged: 0 Thread(s)
Unhappy Gee, what a fun elder-sibling, & how very un-rewarding.

Quote:
Originally Posted by Markie View Post

Dogs are a lot like children.
LFL, would you give your child a piece of chocolate Every time they went to the bathroom??
Or would you expect them to learn and understand without something being there all the time to promote it.
...
_____________________________________
"going to the bathroom" is a physical inevitability - they're gonna VOID, in any case; the Q is, do they void when & as we would prefer they do?..., rather than anywhere / anytime.

Whether "they" are pups / dogs voiding outside vs indoors, or toddlers / children using a toilet vs their underwear, we adult humans definitely have preferred places & procedures for voiding.


"To promote" what? - what, precisely, are the children in this scenario spozed to "learn & understand"?

I've housetrained children as well as dogs, BTW - & yes, rewards are given to the toddler who pees in the potty vs their diapers or pull-ups, who washes her or his hands after, who uses toilet tissue as needed, & so on.
Yup - kids get goodies, praise, fun activities, etc, JUST * LIKE * PUPS.

I also model apropos behavior - i use the toilet, then wash my hands; i use my paper-towel to dry my hands, & then to grab the bathroom-door handle; I carry it out with me, & i discard it in a trash-can, not on the floor.


Quote:
Originally Posted by Markie View Post

... my 7-YO sister ... lives with me. I expect her to clean her room & [to] go to bed;
she [gets no] reward for [those behaviors]. She just does [them].
Well, aren't U just a barrel of fun? // Does she get an allowance, this 7-YO paragon of virtue?

Does she get anything for an A in a school-subject? // Does she get anything for improved reports in deportment?
Is she verbally praised for getting the bedspread neatly squared to the bed?
Or for tucking the linens under the pillow, & indenting the spread before covering the pillow with it?

Do U test how taut the bedding is, by bouncing a quarter off it?
All the drill-sergeants do that, in boot-camp. I think U should start doing that, if U don't, already; she also needs to master hospital-corners ASAP.


For heaven's sake, don't start praising her, now! - it will only fluster & confuse her.
Just criticize her severely when she does it the least-bit wrong, & ignore her utterly when she does it right -
no matter how excellently she does it, correctly, it's not worthy of praise or rewards. // ................ /sarcasm.

- terry

leashedForLife is offline   Reply With Quote
Old 09-21-2017, 01:41 AM
  #66
Senior Member
 
AlwaysTomboy's Avatar
 
Join Date: Jun 2017
Location: near Reading, PA
Posts: 340
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
We've kind of circled back around to the original topic of the thread, I think, and I like the kid analogy, so I'm going to stick with it.

Kids need to learn life skills, things like making the bed in the morning, putting dirty clothes in the hamper, washing hands before dinner, asking to be excused from the dinner table. At first, to get the child to learn the appropriate behavior and to provide incentive for the child to do it regularly, the parents might provide a reward every time the child performs the task when asked. The reward might be something material, like a piece of candy or a sticker, it might be an emphatic "good job, son/daughter. I'm proud of you." It might be something related to the task that the child wants to do anyway, as in washing hands before dinner or asking to be excused from the table. In those cases, getting to eat dinner sooner rather than later and getting to leave the table to go play are the rewards.

As the child gets older, though, the expectations are higher. The child might need to do certain tasks automatically without being asked. The reward might come once a week as an allowance rather than immediately after the task. For some tasks, like washing hands or brushing teeth, there may not be praise or reward at all. It's something that needs to be done to stay healthy whether you like it or not. If the child is old enough to understand that, the behavior doesn't necessarily still need a reward to be performed reliably.

At some point, maybe when the kid is in high school, or gets a summer job, or maybe not until the kid goes off to college or moves out of his /her parents' house, there are no more rewards for basic life skills.

So the question, I think, isn't reward vs no reward. It's whether it's possible for dogs to get to the point of reliably responding to trained cues even without regular or even sporadic rewards. My personal feeling is that it is possible to do away with material rewards, at least in some cases. I have no first hand experience, but in researching how service dogs are trained, I know it's possible for at least some dogs to be trained with praise as the only reward. Dogs don't have the same level of intellect or cognition as humans, though, so my personal opinion, again, not first hand but based on research I've done, is that a dog will always need some sort of reward (even just an intermittent one) to maintain training. If there's absolutely no reward, then you're setting the dog up to extinguish that particular bit of training the same way you can ignore an undesired behavior until the dog realizes it's a wasted effort and extinguishes that behavior.

The other part of the original topic of this thread is, of course how to go from reinforcing the learned behaviors with material rewards to having the same amount of reliability without a material reward. I don't know enough about dog training to have an answer, unfortunately.

Quote:
Originally Posted by leashedForLife View Post
...& yes, rewards are given to the toddler who pees in the potty vs their diapers or pull-ups, who washes her or his hands after, who uses toilet tissue as needed, & so on.
Yup - kids get goodies, praise, fun activities, etc...
But at some point, the rewards stop, but the appropriate behavior is expected to continue. A toddler gets rewards for proper bathroom etiquette, but does a 7 year old, a 13 year old, a 20 year old?

Quote:
I also model apropos behavior - i use the toilet, then wash my hands; i use my paper-towel to dry my hands, & then to grab the bathroom-door handle; I carry it out with me, & i discard it in a trash-can, not on the floor.
I might be misunderstanding. Does that correlate to dog training as a way to get the desired behavior without rewards? I've heard of modeling in the context of dog training, so that's why I ask. Apologies if I'm reading more into your reply than is there.
AlwaysTomboy is offline   Reply With Quote
Old 09-21-2017, 02:28 AM
  #67
Banned
 
Join Date: Aug 2017
Location: Boston metro-area, USA
Posts: 1,885
Mentioned: 46 Post(s)
Tagged: 0 Thread(s)
Talking Um - No.

Quote:
Originally Posted by AlwaysTomboy View Post

I might be misunderstanding. Does [modeling apropos behaviors in the bathroom for a toddler] correlate to dog training as a way to get the desired behavior without rewards?

I've heard of modeling in the context of dog training, so that's why I ask. Apologies if I'm reading more into your reply than is there.
Nope. But i like it!
seriously, modeling "this is how we do it..." for a toddler is just that. // It saves constantly verbalizing.
Showing - b/c we naked apes are so prone to mimic behaviors that we see - is highly-effective in transfer.

Humans learn best via a sequence of learning processes:
- See the action or read the description.
- Do the action, or Perform as described.
- finally, explain what U did in the order it was done, & demonstrate it for the class - or for the instructor.


dogs can learn by watching other dogs perform; they can also learn from watching humans.
Research found, for Ex, that dogs who watched a person open a door with their foot when their hands were full, inferred that the person used their foot b/c their hands were occupied, & did not imitate the gesture - they used their noses, not their paws, to push the bar.

OTOH, dogs - unlike humans! - are not quick to mimic novel actions. // Teaching a dog only by modeling the desired behavior would markedly slow the process of learning, in most cases; shaping, capture, & luring are all more-efficient options than modeling the new behavior.

Learning such actions as pushing down a lever-style doorknob are exceptions to the rule - dogs can see that, & fairly quickly realize "this is how to open that door".
But U can 'model' SIT for a dog for a long time, if said dog is happily engaged in sniffing, & s/he won't feel impelled to imitate. They'll sit when they're tired of sniffing, but not b/c U sat.


"Mirror Method dog-training" is a specific process of ASKING the dog to 'do as i do, now'.

It's quite different from conventional training, altho it shares a pos-R basis with reward-based training that uses 'lure, capture, or shape' as the modi operandi.


My UK-buddy MultiAnimalCrackers uses a form of shaping to get her dog to 'mirror' her respective leg's action:
... but i hasten to clarify that's not 'Mirror method' training; she simply wants him to move his left leg the way she moves her RIGHT, & his right leg the way she moves her LEFT, thus imitate the action 'as if' in a mirror.
[Cliff's Notes:
If U want to cheat & see the desired cross-leg behavior, FFwd to 4-min:30-secs, & watch Barney, the black Lab, in action.]

cheers,
- terry

leashedForLife is offline   Reply With Quote
Old 09-21-2017, 06:19 AM
  #68
Senior Member
 
Markie's Avatar
 
Join Date: Sep 2017
Posts: 1,485
Mentioned: 74 Post(s)
Tagged: 1 Thread(s)
LFL my little sister gets things but probably not away that's okay with you. She gets a huge Christmas every year, Easter presents, Valentine's sent to her school, Birthday presents, her Adoption day gifts. She doesn't get a reward for everything she does. I've raised my sister for five years and see no point in that. She doesn't have an allowance because she doesn't want it for what it's for. She is more interested in the president on the bill instead of it's use(which I love). Dogs can be trained off of materials things I've seen military dogs in training and they don't get a reward for everything they do. The thing with the potty breaks in toddlers I meant all ages. Would you give a 17 year old a piece of candy for going to the bathroom? No you wouldnt.
Markie is offline   Reply With Quote
Old 09-21-2017, 06:44 AM
  #69
Dog Forum ModeraTHOR
 
kmes's Avatar
 
Join Date: Dec 2009
Posts: 11,910
Mentioned: 486 Post(s)
Tagged: 0 Thread(s)
A little food for thought...
The examples of service and military dogs working without food or toy rewards. Gotta remember that they are often still trained via balanced methods. In dogs trained with aversives either primarily or even just in the mix, negative reinforcement (removal of aversive) is a large driving force in their behavior. They are never working for just that praise.
Posted via Mobile Device
cookieface and leashedForLife like this.
kmes is offline   Reply With Quote
Old 09-21-2017, 06:54 AM
  #70
Junior Member
 
Join Date: Sep 2017
Location: France
Posts: 21
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quote:
Would you give a 17 year old a piece of candy for going to the bathroom? No you wouldnt.
Nop, but you'll reward him until this behavior is automatic and fully understood. At some point, you won't need to reward simple tasks anymore. (and I think we agree on this )

Kaya ALWAYS had a treat when I asked her to come back (this is a hunting breed and it's been difficult for a while), as time passed she was only rewarded randomly, just for her to remember that MAYBE she will get a piece of cheese. Today she is never rewarded for coming back, but I might reward a much more complicated sequence (like come back-sit-down-up-walk behind-stop). That's my way to keep her motivated at all time while still teaching her that a reward has to be earned, it's not a gift.

You could say she is only motivated by the eventuality of a reward, but aren't we all ? You work because you get paid, you get involved in charity because it makes you feel good about yourself, I can't think of anything that's not rewarding in any way. Dog's are no exception, they won't do anything if there is no potential reward doing it, be it food, a toy, a hug, your happiness...
Zelis is offline   Reply With Quote
Reply



Thread Tools
Display Modes

Similar Threads
Thread Thread Starter Forum Replies Last Post
Puppy Suddenly Guarding Food From US! (kinda long) TheRubyCavalier Dog Training and Behavior 9 07-21-2017 08:18 PM
Dog is (suddenly) afraid of its food dish yukina Dog Training and Behavior 2 05-09-2017 01:35 PM


Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On

All times are GMT -5. The time now is 05:45 PM.


Powered by vBulletin® Version 3.8.8
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
SEO by vBSEO 3.6.0 PL2
vBulletin Security provided by vBSecurity v2.2.2 (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
User Alert System provided by Advanced User Tagging v3.1.0 (Lite) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd. Runs best on HiVelocity Hosting.