Page 1 of 2 12 LastLast
Results 1 to 10 of 14

Thread: Chris Connell - reward based training for IPO

  1. #1
    Join Date
    Aug 2009
    Location
    Adelaide
    Posts
    12,583

    Default Chris Connell - reward based training for IPO

    Hi all

    I watched an interview (c/o Susan Garrett) this morning with Chris Connell whose previous job was training Navy seal soldiers to train their dogs. He's now full time people and dog training and also he has a new puppy that he's training with all reward based training - no aversives (apart from you don't get a reward for that...) for IPO and tracking.

    He talked about how he deals with people who think it can't be done. Mostly by the doing... He's doing Susan Garrett's puppy peaks and recallers and learning heaps about the current animal training science - and his mechanics (timing and placement of rewards) are amazing.

    Here's two videos.

    Chris Connell trial performance prep - various breeds of dog
    using reward based training (no aversives) to train for IPO and tracking - busting traditional aversive based methods.
    https://www.youtube.com/watch?v=HAD4u9quYyI

    Chris Connell at Bailey chicken camp - mechanics supremo
    today he said that even a shoulder twist would be seen as a cue by the chicken
    https://www.youtube.com/watch?v=tglNXx8vzK8

    Chris's youtube channel The Silent Option
    https://www.youtube.com/user/TheSilentOption/videos

    One thing he talked about for getting nice heel work was a combination of having magnetic balls for his dog to target - one in the middle of his back, one in his armpit (roughly ) and one in his right hand..

    And also doing what he called "backwards heelwork" where the handler walks (or runs) backwards and rewards the dog for being "in the reinforcement zone". And he says - it's a good way to get all your behaviours how you want in terms of position and carriage and targeting - before you and the dog move off in the same direction - as for competition... ie then you can start with nice working style - going forwards - straight away having got all the dodgy stuff with the handler going backwards.

    Cheeky me - asked about the dog going backwards - which is what I've been trying to get but I've got bunny hopping backwards or her butt winding out...

  2. #2
    Join Date
    Jan 2012
    Location
    Melbourne VIC
    Posts
    607

    Default

    I taught my boy to heel backwards by working against a fence line.

    I like the links. They show what is possible, which is great.

  3. #3
    Join Date
    Aug 2009
    Location
    Adelaide
    Posts
    12,583

    Default

    Yes - I could do more with "controlling the environment" would keep her straight. I'd need to find a fence that isn't peed on a lot...

    Maybe a wall.

    Chris uses a lot of perches and targets to get his dogs going where he wants. In the trial performance prep - it's all about getting the reinforcement (treats and toys) off the handler - so quite often - they're under the black bowl things in the video - you see one of the merle dogs jumping onto a black thing... that's where the reward is.

    I hope he posts more videos.

  4. #4
    Join Date
    Jan 2012
    Location
    Geelong, Vic
    Posts
    871

    Default

    He's not exactly revolutionary. As for the magnetic ball the old way was to tuck the ball under your armpit and then pin it to your lapel, then hide it in your shirt to maintain the dogs focus. Magnet on the back can help with a forging dog that refuses to remain in position. I don't like the magnet I have a vest with a ball dropper that uses a cord hidden on the right hand side so the dog cannot see your hand movement and predict the reward.

    Heeling is not a walking exercise, it never has been. Here in Australia we insist on making it a long distance plod where it's a positioning exercise in essence. Teach the position and glue the dog to your leg you can move how you like and the dog will perform perfectly. It's why our dog school puppies find competition style heeling so easy to learn and keeps them engaged.

    IPO has always been about motivation through prey drive. The thing is there is pressure in IPO (well there was.... that's going now compared to what it was) which is why the courage and bitework component is involved. You can't fudge a good bite no matter how you train.

    As for backwards heeling without proper hind end awareness you will not get a good straight line. The dog needs to go one step at a time, do some climbing and awareness exercises first. We do backwards heeling at dog school and some dogs with poor awareness take a bit longer but we take it slowly also to prevent the exercise itself becoming aversive.

    Watch videos by Balabanov, Michael Ellis, Bart Bellon, and other famous IPO trainers.

    As for being a navy seal trainer, working and police dogs are not IPO. Two worlds apart. I attended a week long seminar with corrections Vic, VicPOL and army these dogs are a whole different league. You're not getting a dog to stop dirty biting or refusing to out with more rewards now are you?

    I opened a Sportdog club here in Victoria, we offer IPO style training, drive training, tracking/trailing, scent detection and starting some new sports not available in most other clubs if at all.
    http://i24.photobucket.com/albums/c11/Mali_nut/K9LOGO.jpg

  5. #5
    Join Date
    Aug 2009
    Location
    Adelaide
    Posts
    12,583

    Default

    He's not exactly revolutionary.
    The USA IPO National Champion at their last competition - invited Chris Connnell to do a demo with his new puppy and said this is the future of IPO.

    What Chris said - is that he's always under pressure from the IPO traditionalists to use +P or add something aversive as punishment to stop the dog doing something you don't want.

    What makes him revolutionary is he's decided to train all +R with a bit of Response Cost and "no treat for that". No corrections. But not letting his dog do whatever the hell it likes either (stealing +R).

    No slip chains, no heeling sticks, no ecollar, no scolding...

    And he's still dealing with comments like you have a soft dog and I couldn't do that with my dog I have to use +P...

  6. #6
    Join Date
    Jan 2012
    Location
    Geelong, Vic
    Posts
    871

    Default

    Once again.... he's not revolutionary. He's marketing himself as that though and Susan Garret I take with a grain of salt regarding how she can over inflate things. Clickers and R+ is being used in Germany too FYI where it was born. IPO is just a sport re now compared to its original function. I refuse to breed with dogs that have SCH/IPO in their lines for that reason. There's labradors in the USA with IPO titles as well.

    With some dogs you could get away with minimal P+. Puppy is one thing, see how some nasty adult entires are handled. I think too many people misconstrue that these sports are already about high levels or reinforcement and not using punishment as it can plunge the drive down... then you've ruined it. I think people also mistake the fact that chains, sticks etc are not actually used to hurt the dog for some people but to communicate with the dog what you want as they afford finer adjustment in sensation while you're training.

    The promo videos are always glossy and perfect. NOt everyone has a dog of an obsessively biddable temperament or start with a well raised 8 week old.
    Last edited by Nekhbet; 02-28-2015 at 09:02 AM.
    http://i24.photobucket.com/albums/c11/Mali_nut/K9LOGO.jpg

  7. #7
    Join Date
    Jul 2010
    Location
    Rural Western Australia
    Posts
    2,637

    Default

    I am not even going to try and comment on the training of protection or protection type sports as I know little about them. I do know even on the working dog forums there are often highly charged discussions about the use of various training techniques and tools and even from reading at a distance there seem to many misconceptions. There is a very old school trainer in his seventies I would say who is ex police dog trainer on this forum who has transitioned over to using mainly clicker training and positive renforcement in his GSDs in the last 10 years and really quite likes it.

    When it comes to building drive I have found with my highest drive dog that it would be very hard to plunge that drive down. I train him as a working sheep dog and certainly dont use things like toys and treats for this. However I do let him know he is working well by keeping my mouth shut or using a soft tone of voice to mark what I like and allowing him to keep working (highly reinforcing). I try and read him and the sheep so that I can redirect before he makes a mistake. So yes I have found that training using mainly positive reinforcement is good

    However I have used a long line on him as a pup which can be an aversive, but it also teaches them pretty quick and I have found this to be most effective in teaching the drive and you can then move on. You can show them what you want and at the same time prevent what you dont want from ever happening and it gives you the level of control in which to do this. His drive was already so high that I had to find ways of controlling it and using it to my advantage with sheep. He is a very strong dog and I sometimes need to be pretty firm with him, but there is nothing worse than a dog that you have to keep continuously shouting at to control, as this indicates that somwhere your foundation training has gone wrong. I had to find a way to work with his enthusiasm but develop it where I can trust him to work reasonably independently, but at the same time he acknowledges what I want him to do.

    He is certainly my go to dog when dealing with a particularly dog aggressive ram that I own. He and the ram have come to an understanding.

    Training him in agility is very different where with him I use a combination of treats and toys, but he is the sort of dog that will keep trying regardless, nothing puts him off, he just wants to work, but the use of positive reinforcers certainly make training him easy, however you still have to withold them if they dont get it right, he gets it very quickly and soon is performing what I have taught him with his reinforcers, in the trial ring without them. He loves trialing agility, the sight of a trial ring and his speed doubles LOL. He is totally my type of dog. He is not exactly biddable because he but is highly trainable and loves to work and if you are not fast enough in your thinking he does his own thinking LOL which sometimes works well and sometimes not so much. He is incredibly responsive to me in the ring which is why I cant afford to be slow thinking with him.

    I have another dog who is an abused working dog rescue and would shut down quite easily in agility but he was incredibly responsive to positive reinforcement (mainly my praise which he would die for and treats as he is scared of toys) and this has built a very high level of trust of me in him.
    This same dog when I when I work him on sheep can absolutely take it if I have to tell him off because his instincts to herd seem to over ride the fact that I might be using a harsh tone of voice that would have once had him cowering in agility training. However he is a dog that I had to lay a foundation of trust where he knows that I will be fair and even in agility these days I can tell him if he has done something wrong without him losing his mojo.

    One thing I have noticed among the top agility trainers is that they rarely have dogs that dont perform. They choose their dogs well and they train them using high levels of positive reinforcement and by golly some of the dogs are superfast and accurate. Nice dogs off the trialing field too.

    My current young cattle dog has also been highly responsive to positive renforcement so I generally really like it as a training concept. I generally mix and match a bit how I train. I like to break things down in to small chunks and highly reinforce what I want. I also try and make sure my timing is spot on and that I reinforce exactly what I want. My dogs look at it as a way of life and love working with me. But I will tell them off if I need to and I do withhold reward if they get it wrong but I try and set them up for success as much as possible which keeps this to a minimum.
    Last edited by Kalacreek; 02-28-2015 at 11:47 AM.

  8. #8
    Join Date
    Aug 2009
    Location
    Adelaide
    Posts
    12,583

    Default

    Nekhbet

    Just because they're using +R in Europe - doesn't mean that it's not revolutionary here in Oz or the USA. Even you poo poo it despite the mounting pile of scientific studies that show that +R gets results and fast.

    My best analogy is to say - I'm trying to teach you how to spell a word correctly. I hand you a pen and a piece of paper and every time you make a mistake or sit still for too long - I slap you with a ruler. It doesn't hurt much but it does sting a bit and it certainly wakes you up and makes you hate me just a little bit...

    That's what so many dog owners are doing to their puppies when they try to teach them heel work and recall etc. And most of the instructors and the management of SACA and I've seen the German Shepherd Club in Victoria training at KCC park - really crap at telling the dog when it's doing what they want, and really quick (or slow too) to use physical punishment when it's not doing what they want. The dog has no clue what is required - just that it's being yanked around all the time.

    And this kind of training shut my dog down so fast - I had to find another way - hence my migration to Susan Garrett. I don't have it perfect. Chris says he doesn't have it perfect but I do have a dog I can walk to the park now and I didn't before. She even comes back when I call her - tho we could improve this. Before - I had to trap her or trick her to catch her any time I let her off lead. Including in the back yard.

    I refuse to breed with dogs that have SCH/IPO in their lines for that reason
    At least these dogs have proved they can do something other than just look "pretty" - which is what we get from the conformation lines.

    But even the current police and military dog trainers and drug sniffer dogs and all the zoo animal trainers are moving to +R. It is true that a lot of elephant trainers used pain (+P) in their training and control but a lot of those were also killed by their elephant when it was pushed too far.

    I've seen Spanish horses being trained in Vienna and getting +R. They do use a bit of pressure to get forward movement - because it is a flight animal and most of the reward is in removing pressure... But a horse that is not doing what they want does not get hit or yelled at. They just wait, and then try again, or try something else or give the horse a break from training and try again later.

    I think people also mistake the fact that chains, sticks etc are not actually used to hurt the dog
    The story that Chris used was similar to my "get slapped for making spelling mistakes" story. And he said the dog was performing correctly but there was no joy (drive?) in the way it moved. I think that's what you were trying to say - that too much +P kills a dog's drive. Any +P kills my dog's drive. She's instantly upside down apologising, but she has no idea what she did wrong or what people want her to be doing instead.

    I don't think we're going to agree on this - and that's ok - I like diversity of opinion.

    I've seen so many trainers swap over to all +R with amazing success. Susan was one of them. It is possible to screw it up, but it's also possible to improve the performance of an older dog like mine where I'm just learning. I don't think my dog would be so much fun to be around if I'd persisted with the choke chain. We'd both be much fatter for starters.

  9. #9
    Join Date
    Oct 2010
    Location
    Southern NSW
    Posts
    3,784

    Default

    I love heeling games.That is how i taught Yussie to heel. She loved the me walking backwards and spinning to her and also being sent around cones and coming back in flying and also reward balls and throws.no fancy stuff for me, just my hands and arm-pit holding toys.

    I think as we train more we learn to teach each dog individually and what suits them. we copy bits and pieces from all over the world. I learned the backward heeling at a Turid Rugaas work shop in Finland, from another trainer who was at her place, many years ago.nothing new that is for sure.

    I think though that there are many differnt dogs that have to be taught the way it suits them.....Especially so in newfies, where the drive is not always there, no matter how well you can get other dogs to work in drive. My Yussie was so differnt and I suppose that is why when she died I felt so devvastated..such fun and control and she thought of it all as a game.

    My Father trained dogs for the Police and RAAF....in the days when there was still alot of Aversion, but he was already coming around to the early clicker training, though not a click, just "good" throw treat...the differnce in the dogs was amazing.that was in the eighties, so not new

    And having been to a few Chicken workshops by both Terry Ryan and some others, you cannot ever "cheat" with chickens, it does make your clicker work better. Chickens never work to please you, just for the reward LOL......Makes you realise how you did cheat
    Pets are forever

  10. #10
    Join Date
    Aug 2009
    Location
    Adelaide
    Posts
    12,583

    Default

    the differnce in the dogs was amazing.that was in the eighties, so not new
    The science for it has been around for erm since Bob Bailey started - maybe before... He trained albatrosses to carry messages, one was so set on its trained task that even when it got injured by a sea hawk - it kept going and completed the task. There's no way you'd get that using aversives.

    What I don't understand is the persistence and insistence by club instructors on the old school aversive based methods. So even if you show up and you want the class environment for the distraction proofing, you're not welcome.

    And so many of them scold their students and take the student's dogs and punish those too. They told me my dog would be untrainable till she was over 12 months old. They were wrong. She was untrainable with their methods. Age had nothing to do with it.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •