Shaping vs. luring in training

[TrulyReady]

I will start this off with I know the difference between the actual training methods of shaping vs. luring. My question is for those who do both methods in simple exercises what are the benefits to take the time to shape a behavior? For example, something simple like 'Spin' right and left I can effectively teach my dog by luring in just a few sessions and have a pretty solid behavior performance in the end. Where if I took the time to shape it, to me it take considerably longer. I understand also that as the dog learns to offer behaviors that shaping new things gets faster/easier but my border collie in particular has never been one to offer behaviors and have VERY low drive for treats and looses interest very quickly even with very frequent clicks/rewards when shaping mostly because she's not interested in the treats (it can be steak, hot dogs etc she just doesn't care) She is VERY toy motivated and I try to mix it up with a toy reward every few clicks which is very time consuming. Luring seems to work much better for her but I'm wondering if the end result is really that different?



My newest edition is a Staffy bull terrier puppy who is over the top food motivated and has no toy interest (totally opposite!) but will offer behaviors for a cookie til the cows come home!



What started my thoughts on this topic was trying to teach the Silvia Trkman Heeling method (spinning front paws on a bowl to start) My 5 month old Staffy puppy picked up on feet on the bowl = treat within 3 minutes. My border collie no matter how much I treat for a head shift in the right direction etc etc etc will connect it with putting her feet on the bowl, and because of her lack of food interest getting the treat isn't that important. If I resort to luring her for the initial behavior am I losing anything in my end behavior result? I will add that my border collie is brilliant and excels in agility and have over the top drive for other things that she can go "do" something. I just can't seem to get her to stop and think about non-agility activities no matter how exciting I try and make them!

[mum24dog]

I will start this off with I know the difference between the actual training methods of shaping vs. luring. My question is for those who do both methods in simple exercises what are the benefits to take the time to shape a behavior? For example, something simple like 'Spin' right and left I can effectively teach my dog by luring in just a few sessions and have a pretty solid behavior performance in the end. Where if I took the time to shape it, to me it take considerably longer. I understand also that as the dog learns to offer behaviors that shaping new things gets faster/easier but my border collie in particular has never been one to offer behaviors and have VERY low drive for treats and looses interest very quickly even with very frequent clicks/rewards when shaping mostly because she's not interested in the treats (it can be steak, hot dogs etc she just doesn't care) She is VERY toy motivated and I try to mix it up with a toy reward every few clicks which is very time consuming. Luring seems to work much better for her but I'm wondering if the end result is really that different?

 

 

 

 

If you have a dog that isn't interested in food as a reward but does value toys why are you using food to clicker train at all?

 

I don't like luring especially for a dog that is very focussed on food because I prefer the dog not to lose concentration on what it is doing, but I will use it if I have to. But if I need to prompt a dog I prefer to do it by my own body position. It's hard to explain but positioning can influence what a dog does without seeming to. More subtle than luring.

 

I had a collie mix that was a crossover dog and very worried about doing the wrong thing. He never readily offered behaviours. I had to lure him for some things but not others. What he learned was good and reliable but it could be hard work sometimes.

 

My hound mix I shaped to turn on cue in 2-3 minutes while I was waiting for my turn in an obedience class. We were both bored and I noticed he looked towards the door whenever there was a noise from that direction so we worked on that. I never lure him because all he would think would be Food Food Food Gimme Gimme Gimme.

 

Our BC was taught to turn using a toy but not clicker trained to do that. It took longer to train than with the hound and several sessions as opposed to one but the end result was similar as a single behaviour but better on an agility course because his reactions are quicker.

 

I do sometimes lure the BC because he is so easily distracted if he isn't concentrating on a task that really engages him, mostly for heelwork.

 

You just have to take the dog in front of you and work with it in the way that suits its personality. If you need to lure use it as a prompt just to get the behaviour started rather than a crutch that you may have difficulty fading. I learned that the hard way with my first dog.

[Root Beer]

TrulyReady, on 06 Apr 2013 - 03:33, said:

I will start this off with I know the difference between the actual training methods of shaping vs. luring. My question is for those who do both methods in simple exercises what are the benefits to take the time to shape a behavior? For example, something simple like 'Spin' right and left I can effectively teach my dog by luring in just a few sessions and have a pretty solid behavior performance in the end. Where if I took the time to shape it, to me it take considerably longer.

 

My experience has been that for simple behaviors, such as a spin, there is no real end benefit to shaping over luring.

 

Just in the way of background - when I first learned how to train, I learned lure/reward. What I did not learn immediately was how to effectively fade a lure, so at first I ended up with a dog who would only work with food present. Of course, at that time I blamed the method, not my ability to effectively carry out the method, so I started to look for other training tools, which, ultimately was a good thing.

 

From there I learned that luring is "ineffective" because the dog "doesn't learn how to think" but "just to follow food". That sounded very sensible, so I learned shaping, which is absolutely a skill that all of my dogs learn, and that I consider to be invaluable.

 

I ran into some problems with shaping, as well, though. I didn't know how to effectively transfer an offered behavior to cue at first. I found that my dog wanted, after a time, to offer the same reperitiore of behaviors over and over and over, etc. So, I found that shaping wasn't the be all end all either.

 

Then, I started working with my mutt girl, Maddie, and at first there was no shaping happening with her. She was very much a "show me the food or I'll just lay on the floor and do nothing" kind of gal. With Maddie I learned how to lure effectively. I learned how to fade the food and get a lured behavior on cue. I learned how to incorporate a clicker into the luring process as a marker to help the dog gain comprehension and fluency faster. Eventually she did get to a point where I could shape and capture with her, and she adored targets, but quality experience with luring got her to that point

 

I also learned that luring, when done properly, does not inhibit the dog's ability to think. Yes, the dog is following food at first, but after being allowed a chance to rehearse the behavior with help (kind of like training wheels on a bike), the time comes when the dog must think, get it right sometimes, and make mistakes, and work out exactly what is desired on a particular cue.

 

At that point I readily accepted both methods, along with targeting, which I use very heavily, as highly effective ways to introduce new behaviors. There are some things I usually start off with a lure (spins, sits, stays, sit pretty, etc), some things I shape (paw work, nose targeting, backing up, etc.), some things I capture (LAT, mat, beginning recall, etc.), and some things I target (leg weaves, moving down, distance work, etc.). At the same time, if one tool doesn't work particularly for one dog, I change it up. Dean learned paw work by me picking up a paw and popping a treat in his mouth. After I did that twice, literally, he started offering paw. It was on cue after several reps in one session - both paws. Speedy, on the other hand, learned paw work through quite an elaborate process of shaping that took weeks and many steps. Tessa learned, basically, with a lure. You can't tell the difference in the end behavior. They all understand the behavior, carry it out readily on cue, and enjoy it.

 

Quote

If I resort to luring her for the initial behavior am I losing anything in my end behavior result? I will add that my border collie is brilliant and excels in agility and have over the top drive for other things that she can go "do" something. I just can't seem to get her to stop and think about non-agility activities no matter how exciting I try and make them!

 

It has not been my experience that anything is lost in the end behavior, as long as the lure is effectively faded, once the dog is showing evidence of comprehension of the behavior. This is something that I gauge as my dog is in the process of learning something.

 

In fact, I have found that my dogs have become more fluent in certain behaviors through luring than shaping. I did quite a lot of experimentation with this with Dean - that was back in my "luring doesn't teach a dog to think" days and I tried to shape everything I could with him. Some of those behaviors turned out just fine, but I did have to go back and retrain a few of them with a lure to get the results that I was looking for.

 

The bottom line is - I have ceased to be worried about whether I introduce a behavior with luring, shaping, capturing, or targeting. As long as the dog moves from no clue of what is desired to an ultimate fluency and comprehension, I'm happy with any of them. Also, I mix and match them very freely. I might introduce something with capturing, and once the dog is starting to get the idea, throw in a visual target to reinforce the concept, then, once I see basic comprehension, shift to shaping for a bit.

 

Experiment - have fun.