Self rewarding

Discussion in 'Clicker Training' started by Jen, Dec 15, 2013.

  1. Jen

    Jen Registered Users

    Joined:
    Aug 30, 2013
    Messages:
    2,897
    Hi

    I know self rewarding is a problem with clicker training outdoors especially with dogs like mine ::) . Although I have to say they don't self reward anywhere near as much as they did since I've been using a clicker. The C&T certainly helps keep their focus on me and what I'm asking them to do.

    One of my self reward problems are smells.

    For example I will send the dog for a retrieve. The majority of the time it comes straight back C&T. Occasionally though the ball might be by an interesting smell or not far from one or just near an interesting twig! ::)
    I will give him a few seconds chance to remember what he's there for. If that doesn't work I blow my recall whistle. 8 times out of 10 that breaks the spell of whatever it is he quickly grabs the ball and races back. C&T he responded to my recall. If that doesn't work I retrieve the ball myself. Ignore the dog but make sure he knows I've got the ball and return to where I was. He will come running after me. I've got the ball so game back on. No C&T obviously. I will then do an easier retrieve, away from the distraction, so he can succeed and earn his C&T.

    I don't know if I'm dealing with this correctly particularly using my recall like that. ???

    I would love to hear your opinions, suggestions and tips.

    Jen :D
     
  2. JulieT

    JulieT Registered Users

    Joined:
    Jun 15, 2013
    Messages:
    20,186
    Re: Self rewarding

    I don't know, Jen, sorry - once Charlie got distracted by objects around the dummy and retrieved them instead of the dummy, but that's the limit of my experience with this. Hopefully someone will be along soon. I saw Barabara correct Riley when he was about to wander off with a no, no, and Riley sort of shook himself and went for his retrieve. She might be able to help.

    Your post touches a point that I've been wondering about:

    [quote author=Jen link=topic=3669.msg41338#msg41338 date=1387092436]
    The majority of the time it comes straight back C&T.
    [/quote]

    This use of the single C&T rather than C&T for repetition of actions working towards a certain desired behaviour. So I can see that a single marker for good behaviour at a point in time has its uses - eg silence in between noise, or the fastest point of a recall, or the instant a dog's bum hits the ground at a distance. But I've been wondering about the general use of the marker before a treat at the end of a string of actions forming a more complex task I wonder if it does much good or even is a bit confusing for the dog? This is related to a point you raised a while ago, that also got me wondering, about the use of the clicker in the recall. (You always prompt such interesting thoughts and discussions!).

    Must read more...
     
  3. heidrun

    heidrun Supporting Member Forum Supporter

    Joined:
    Feb 10, 2012
    Messages:
    2,626
    Re: Self rewarding

    Jen, it sounds as if your dogs are blinking the dummy, it means they are deliberately ignoring the dummy in favour of hunting. If you are clicker training the retrieve then you need to get the behaviour rock solid in areas with absolutely no or very few distractions and then slowly proof against distractions. Pretty much like you would proof any behaviour in the face of distractions.
    Here is an interesting article written by Pippa with some helpful tips. http://totallygundogs.com/when-gundogs-refuse-to-retrieve-dummies/
     
  4. Jen

    Jen Registered Users

    Joined:
    Aug 30, 2013
    Messages:
    2,897
    Re: Self rewarding

    Thanks Heidrun. Very interesting article.

    I think your right what they do is very similar to blinking the dummy. Also it only happens towards the end of a training session when they are tired. I start and end training with a few retrieves as they enjoy it so use it like a reward. I rarely get any blinking at the beginning usually only at the end. I guess the solution is to not do retrieves at the end of the session. ::)

    I know your dogs are beautifully trained but how did you deal with self rewarding in general ?

    Jen
     
  5. Jen

    Jen Registered Users

    Joined:
    Aug 30, 2013
    Messages:
    2,897
    Re: Self rewarding

    Julie what sort of complex task were you thinking of?

    It is confusing sometimes and I still don't know if I'm doing this clicker thing right (hence the questioning about self rewarding ).

    When you break a behaviour down into small pieces to clicker train (I'm still not sure of the correct terminology) eventually you put them together and only C&T at the end. That obviously works and doesn't confuse the dog.

    We need a clicker expert Julie ???
     
  6. heidrun

    heidrun Supporting Member Forum Supporter

    Joined:
    Feb 10, 2012
    Messages:
    2,626
    Re: Self rewarding

    [quote author=Jen link=topic=3669.msg41369#msg41369 date=1387109015]
    Thanks Heidrun. Very interesting article.

    I think your right what they do is very similar to blinking the dummy. Also it only happens towards the end of a training session when they are tired. I start and end training with a few retrieves as they enjoy it so use it like a reward. I rarely get any blinking at the beginning usually only at the end. I guess the solution is to not do retrieves at the end of the session. ::)

    I know your dogs are beautifully trained but how did you deal with self rewarding in general ?

    Jen
    [/quote]

    The only way you can stop your dog self rewarding as a clicker trainer is to completely control the rewards your dog is getting. That means micro managing the dogs environment. I am in danger here of opening up the debate about taking your dog for a walk ;), but a walk will offer a dog hundreds of opportunities to self reward and to learn behaviours you might not really want. So that might be something to bear in mind when clicker training.
    I am not a pure clicker trainer, I do use traditional methods in certain situations. One of those situations would be one of my spaniels self rewarding in a field full of pheasants ( self rewarding = chasing in this situation :eek:) It would be catastrophic to allow that sort of self rewarding and I wouldn't hesitate for a second to use an old fashioned correction to stop the dog. ;) :)
     
  7. JulieT

    JulieT Registered Users

    Joined:
    Jun 15, 2013
    Messages:
    20,186
    Re: Self rewarding

    [quote author=Jen link=topic=3669.msg41372#msg41372 date=1387109972]
    Julie what sort of complex task were you thinking of?
    [/quote]

    I was thinking of the retrieve in this example - so I can see why you would click at the end of very fast repetitive retrieves when you have put a string of behaviours together to form the action you are repeating. So doing 10 or 15 short clicker retrieves for example. But I'm wondering about the value of one click, for one retrieve. In this case it's not part of a fast repetition and it's not marking one precise action. If you see what I mean. The same would apply for a recall - your previous post got me thinking about this. What in the recall, exactly, would I reward? And I came back to the conclusion that now the behaviour is trained, I'm marking speed - the point when the dog is running the fastest. So the clicker is used as a precise marker.

    Probably getting muddled again... ::)
     
  8. Jen

    Jen Registered Users

    Joined:
    Aug 30, 2013
    Messages:
    2,897
    Re: Self rewarding

    Thanks Heidrun .

    I did think after I'd asked the question the answer is training in an environment with low distractions then gradually increasing it, as youd already said in your first post. Unfortunately for most things i train with the clicker I go from low distraction indoors to full on outdoors I haven't really got an area inbetween.

    I understand about the walking thing ;) and other ways a dog will pick up behaviours that aren't wanted. A middle ground would be nice but unless you've got a dog with that sort of temperament I don't think there is one. :(

    Although I much prefer clicker training and the positive effect it's had on my dogs is obvious. I have to admit with some things I do revert to the old fashioned method because with time it could be clicker trained but right now it's not. For example this morning scout and my cousins dog had a row through the fence. I had to get hold of him to move him and I told him NO. It worked he didn't go back but I also did a quick sit stay C&T unti pip had gone away so I combined the two sort of ::) . Now if I'd got hold of pip !!!! ;) only joking . My NO even on the wrong side of a fence works on all the families dogs ! ???
     
  9. Jen

    Jen Registered Users

    Joined:
    Aug 30, 2013
    Messages:
    2,897
    Re: Self rewarding

    I see what you mean Julie

    Like you said I guess you C&T the behaviour that's most important to you.

    At the moment with the retrieve I'm clicking the delivery whether it comes fast or slow. (haven't quite mastered that bit yet :()

    The recall I click when they get to me. I'm still not sure if that's right but I want them to know 'well done, that's what I wanted you to do' . From what I've read about clicker training you C&T the thing you want. When I recall I want them to come back to me so to me it makes sense to C&T when they get to me. My concern clicking when they set off in my direction would be that they might deviate on the way. Then what do I do ? ??? If you see what I mean. Once I'm confident in their recall then I will probably change the criteria. ( I've just used a technical term in the right way I think. ;D)

    I hope that makes sense.
     
  10. bbrown

    bbrown Moderator Forum Supporter

    Joined:
    Jun 13, 2011
    Messages:
    9,435
    Re: Self rewarding

    When we're out and about I look for a variety of environments...
    open, short grass
    short grass with longer grass at the sides
    natural corridors and straight lines
    a variety of long grass and cover

    then the stuff we do well we do in the more difficult or new environments, new stuff we do in open space on short grass until I'm really, really sure we've got it. Riley can be easily distracted so sometimes needs a reminder that he's working for me but I only correct him if I'm sure he knows what he should be doing :)
     
  11. charlie

    charlie Registered Users

    Joined:
    Sep 29, 2012
    Messages:
    12,217
    Location:
    Hampshire, UK
    Re: Self rewarding

    [quote author=Jen link=topic=3669.msg41381#msg41381 date=1387115218]
    Thanks Heidrun .

    I did think after I'd asked the question the answer is training in an environment with low distractions then gradually increasing it, as youd already said in your first post. Unfortunately for most things i train with the clicker I go from low distraction indoors to full on outdoors I haven't really got an area inbetween.

    [/quote]

    Jen, do you have a driveway or maybe a park close by to move onto? I move from indoors to our driveway which is less distracting than our garden. Just a thought :)
     
  12. Jen

    Jen Registered Users

    Joined:
    Aug 30, 2013
    Messages:
    2,897
    Re: Self rewarding

    We have two long drives Helen unfortunately one has no gate and the other has tractors up and down it. :mad:

    I shouldn't complain we are very lucky.

    I try and keep to the top of the garden which is less distracting than the bottom and gradually move down as we proof until I can do everything at the bottom that we can at the top. A bit like barbara and different lengths of grass. It doesn't always go that easy of course. ::)

    I think my two would manage to get distracted if I trained them on concrete Barbara. ::)

    I also only correct something I know they know and like I said earlier the need for this has become a lot less and is still improving since I've been using the clicker. I'm still amazed at how much they've changed and improved since I've been clicking. :D
     

Share This Page