How Many Repetitions To Teach A Required Behaviour?

[whatevah]

I think Moses took about 10 reps to learn to do this. But also depends how polished etc you want it to be. With this movement I wanted Moses to do on just voice and no hand signals. Because very hard to do hand signals for this trick.

Edited March 11, 2007 by CrashTestDummy

[Lablover]

CTD,

Fun youtube.

Pushups, WOW. Impressive on your part also. I do a pushup every day too, getting out of bed.

[Erny]

I do a pushup every day too, getting out of bed.

:rolleyes:

[whatevah]

Actually the push ups are very pathetic as I my arse in way too high in the air like the harbour bridge, but I am doing this to help the dog out. At the start he kept, jump over my back because this is what he has learnt previously. Hope nobody is looking at my cleavage, tried to tuck my singlet in. LOL

[Lablover]

Looking at your cleavage........good grief, no way. As per always, I was focused on Moses. People ask me did I see that guy (good looking with big shoulders apparently) with his dog, and I reply what guy.

[Erny]

Hope nobody is looking at my cleavage, tried to tuck my singlet in. LOL

I wouldn't have any problems there!!

Will trot off for a look at the video clip now ...... oh ..... and I am NOT interested in your cleavage, just in case you were wondering!!! :rolleyes:

[Working_Setters]

There are many variables including, the dog, the handler, the nature of what you're trying to teach and where you're trying to teach it. However I suspect the biggest variable is what people consider "trained". Does it mean you get the behaviour first time, every time (which is my definition of a “required behaviour” - ie the dog perform the behaviour whenever and wherever I require it to do so), or does it mean you get the behaviour in a distraction free environment when you're luring with treats?

If it only takes 3-5 reps to train a behaviour, why when you attend retrieving trials do you see Restricted/AA dogs ignoring sit whistles and directional commands? These are, for the most part, very good dogs with very good trainers, that have had far more then 3-5 reps, so what gives?

IMO it takes 100s of reps and many, many hours of training to reliably train a dog to perform even simple (eg sit to whistle) required behaviours. The key word here is reliably - ie first time, every time.

[Tangwyn]

Hmmm... I'm going to throw a spanner in the works here. Personally, I don't think a command is taught until it is generalised and proofed. Just because the dog seems to "get it" with a few reps doesn't mean that it will still "get it" at a different time, place and with distractions. So I would have to say many repetitions in many different environments before I believe a command is taught IMHO

Just noticed your reply W_S - I concur!!

[Erny]

I don't think a command is taught until it is generalised and proofed.

This is where interpretation can greatly influence and manipulate the questions and answers given here.

I think of "teach" as creating a (eg) word association.

I think of "training" as showing/revealing to the dog that there are consequences on the choices it makes.

I think of "proofing" as building on what has been trained, generalising that training and increasing distraction intensity.

Edited March 12, 2007 by Erny

[Tangwyn]

Too true, Erny! Interpretation is the key (and sometimes problem) with most types of communication. When would you say a behaviour is "learnt" then? This is the aspect of Lablover's OP that sent me down the path of proofing and generalisation.

Might I add, also, that dogs who have "learnt to learn" take far less reps to establish an association with a command than dogs who have not had previous training. My experience has always been that the hardest thing you ever teach a dog is the first thing you ever teach it. After that its easy-peasy

ETA Erny, I must be channelling you... I just re-read one of your earlier posts and noticed you also used the phrase "learnt to learn"! Been one of my favourites for years and I thought I was the only one Ha Ha!

Edited March 12, 2007 by Tangwyn

[Tonymc]

I can in this case see some similaritys between Dog learning and Human Learning.We have all met People who have never studied or had to learn much before.When they are faced with the situation of learning,initially it is harder for them.

I have found in some cases the same with Dogs that had little exposure to learning. Tony

[dogdude]

I sometimes wonder if saying that dogs learn in patterns/sequences is that simple. I have noticed that in the proofing stages, when you break the pattern, dogs often display understanding, sometimes not. I think their brains are more complex than that.

[Lablover]

I had to check in before driving to training.

Thanks for all the replies.

How dogs learn is completely fascinating to me. And whywhy why certain drills/exercises and methods work.

[Tangwyn]

Tell me about it. If someone can explain why indirect pressure works it would be a load off my mind!

[whatevah]

I agree with working setters. I thought gee it took Moses 10 reps to learn that exercise, he mustn't be as smart as those dogs that only took 5 reps. If Nova only takes 5 reps to learn something then how come it took him 3 years to learn flyball?

Flyball is a good test, they may know it on their training ground, but take them to a comp. Completely different lots of new dogs heaps of distractions. One minute they are lying in their crate the next minute they are in the comp and have to switch on.

Josh and Tia take a lot more reps then Moses.

[Sayreovi]

]

I thought gee it took Moses 10 reps to learn that exercise, he mustn't be as smart as those dogs that only took 5 reps.

If Nova only takes 5 reps to learn something then how come it took him 3 years to learn flyball?

Cant let that go can you?

Nova is smart enough too have figured out many many ways too challenge me and "forget" things. Nor did i train him any more then once a week. He is fast becoming one of the most reliable dogs in the team who wont run over and interfere with the other team because of all the work i did with him.

Why do you care how "smart" Nova is or not? That has nothing too do with you at all.

Dont forget intelligence doesnt always equal trainability.

Why are you trying too compare two completely different dogs? That is pretty stupid especially when they are nothing alike.

Edited March 13, 2007 by tollersowned

[Erny]

I sometimes wonder if saying that dogs learn in patterns/sequences is that simple. I have noticed that in the proofing stages, when you break the pattern, dogs often display understanding, sometimes not. I think their brains are more complex than that.

I may have misinterpreted what you mean, here. If I have, please correct me.

However, by the proofing stage, dogs have already learnt what action is required of them by the patterns/sequences taught in the teaching phase. The pattern of (eg) "you say sit; I sit; I get good thing" OR "you say sit; I don't sit; I get things I don't want" is already well and truley practiced and learnt. Well, by proofing stage it should have, anyway ..... else proofing is occurring too soon.

[dogdude]

Hi Erny

If you break a learned pattern, ie, proofing, it is no longer a pattern.

Dogs are certainly capable of independant troubleshooting (thinking outside a pattern)

But I suppose that because of the lack of speech, everything has to be learned from previous experience

Not sure if i'm making sense lol.

Edited March 13, 2007 by dogdude

[m-j]

Hi

In addition to all the above suggested, how close together the reps are and if using treats what the dog is actually doing when the treat is being put in the dogs mouth, even with using a marker and the dog is marker savvy. How many successive approximations go into teaching the behaviour.

cheers

M-J

[Erny]

If you break a learned pattern, ie, proofing, it is no longer a pattern.

Hey DD. Puzzling this.

Even in proofing, the same pattern applies. (Referring back to my VERY basic example in earlier post.)

Hi M-J

Depending on the level of distraction it would otherwise cause, I like to treat at the time the dog is giving me the behaviour I seek. (ETA: Especially if the dog has not been able to pick up the "pattern" easily.) At least whilst the dog is still in the teaching phase. Sometimes this is too hard for novice dog handlers to do though, because they often don't have the observational skills or co-ordination to do this smoothly and swiftly before the dog changes to an unwanted behaviour.

Personally I don't have a set "rule" on how close the reps are. Much depends on the dog in question and its emotions at the time as to whether it needs tiny, small. medium or large breaks inbetween. Don't know what anyone else thinks on this.

Edited March 13, 2007 by Erny