Aidan

A dog can learn tricks and still be in general disobedient. Sitting for you food, high fiving, sitting on command etc is not an obedient dog. Unfortunately the dog learns to behave for that particular situation and as soon as its done BANG back to normal. Sure, we both agree on that. The difference is that I see no difference between conditioning a behaviour using food or with anything else (leash pops etc) when it comes to actually getting reliable behaviours that aren't just tricks. Just because you don't use punishing consequences like leash pops (and I do sometimes, for the record) doesn't mean you can't condition a dog to become more obedient overall or in the general sense. That "more obedient in the general sense" is just a collection of behaviours, which includes things like listening to you, respect for space and proximity, knowing what you expect, knowing that you have boundaries etc. The person who is clear in their own mind about these things, and how to achieve them, will achieve them no matter what method they use so long as they don't try to defy (knowingly or otherwise) the laws of operant conditioning. The biggest problem with using food is that it becomes an antecedent - part of the collection of what behaviourists call "discriminating stimuli" so when it's not there - "BANG back to normal" as you put it. Intelligent use of food (and other reinforcers) avoids this. It certainly IS an issue for a lot of people, but it needn't be. The same could be said for leash pops, although that antecedent is always going to be there anyway I'm willing to make a bet on that one Same here. Not the first time I've heard it either! What do you think the lady in the video said to me when we started? That dog isn't perfect (and if you pay attention, neither is my timing) but this was her third class. The method isn't important, and it's rare that I would change what I do to suit the dog, the execution is what matters.

If he's running circles then at some point he's going to be on a loose leash. That's when you click and give the treat close to you, then you click again before he can make his way to the end of the leash. Another approach is to set up a clear target, a bowl of food or someone with a toy, as examples. Do these in low distraction environments at first. A little trick I use is to use different walking gear to mean different things. E.g the front attaching harness could be his "do whatever you damn well please" cue, and a martingale collar could be his "we're walking strictly on a loose leash" cue. It's "verbal cue" then "hand signal", over and over until he looks like he's anticipating the hand signal when you give the verbal cue. Then you pause and see what happens. There is the possibility that you have INADVERTENTLY taught him to ignore verbal cues. It happens, no-one means it to happen. If this has happened you could do something like this - wait until he is resting quietly. You go about your business, then suddenly say his name clearly - when he looks up you click and treat. Do this whenever you get the opportunity (within reason!) Then try it in other situations where he is not paying attention to you, but not too engrossed in what he is doing. You just need to find your approach, stop believing that he is drastically different. They are all different, but if he's not deaf he can listen to commands and if he can walk he can walk on a loose leash

I agree with Nekhbet on the making excuses for him thing (not on the "tricks" comment, I take the Skinnerian view of conditioning). He doesn't listen to verbal commands because he needs to learn to listen to them. Some dogs, most dogs in fact, don't really pay much attention to verbal cues and prefer visual cues. But does your dog bark at sounds he hears in the night? Sounds you don't even hear? If so, his hearing is fine and he knows how to tune in (or out) sounds. If you believe that he has some special need that you need to work around, that is what you will do. So problems like "not coming when called" and "pulling on the leash" are avoided because you believe he has special quirks. Of course, you cannot be blamed for taking this view because you haven't had to deal with these problems before and your attempts up until now have been unsuccessful. But you do have to change it if you want to fix these problems! Do you know how to transition from one cue to another? You use the new cue, then immediately use the existing cue. So you would say "Sit", then give the hand signal for "sit" immediately after. Repeat until the verbal cue "sit" begins to PREDICT the hand signal, he anticipates it. At this point you wait for him to sit, if he doesn't sit, give it again and then use the hand-signal. See if you can teach him just one thing using a verbal cue, something he knows well and does reliably when you use a hand signal. Then get cracking on the other things, especially that recall!

So long as he isn't deaf he should be fine. The important thing to look for is - are you getting more of what you clicked?

Are you sure he isn't a cat?

Hi Judy, there is some discussion on the topic here: http://www.dolforums.com.au/index.php?showtopic=179746 Personally, I would take the risk from day one. It depends on the prevalence of communicable disease in your area though, and no doubt on whatever immunity pup already has developed.

It's not random though, that is the point. It just isn't "perfect". You tell me how many people have perfect leash handling? Practical experience tells me that if you tell someone what to click for they get it right enough to reinforce those responses - and that is all that matters, are we seeing more of those responses? People worry about the method too much and take their eye off the behaviour. If it really worries you, do it without the clicker. The problem with that is that there are MANY conditioned reinforcers at play here, the reason I put a clicker in someone's hand is because then they are consciously aware of it and it becomes salient to the dog so they begin to tune out those other conditioned reinforcers. You can if you like. Again, practical experience tells me that you usually get that response regardless and that it isn't important so long as the response that you are MARKING is increasing. It doesn't take 500 clicks to make a difference. There is a continuum from "100% unwanted behaviour under any circumstances" to "100% the behaviour we would prefer under any circumstance". Getting from one extreme to another takes a lot of repetitions, whether they be negative reinforcement through tiny collar pops or positive reinforcement through clicks and treats. For the record, I liked the method you described.

I trained my cats to get somewhere high as soon as the pup became excited, pup had nothing to chase, fun ended promptly. A lot of the escape strategies for cats don't work with smaller breed pups (e.g a small pup can go through a cat door too), but this will because cats can jump very high. A drawback is if you don't like cats on your kitchen benches, dining table etc You also need to teach pup to settle around the cats. There is no sense having a cat on your table with a pup racing around indefinitely below. A short tether or crate can be used to good effect. Have a cat in the room, click and treat the pup every time she is calm. Ignore her when she is not. I'm inclined not to let the excitement build too much though, if that is happening you need a different strategy, have someone hold the cat much further away (or even on a harness and leash). What you are looking for is the pup settling more readily every time she is exposed to the cat, until eventually she does not get excited at all. If pup is never allowed to see the cats as prey, they eventually settle in as part of the family together.

It is extremely rare that I ever charge up the clicker, but if it is necessary it only takes a moment anyway. The reason for the controlled set-ups is that you set the distance so that you're not balancing a clicker and food with a "straining" dog. I rarely encounter clients who cannot manage it, they usually have more trouble with the leash handling. Perfect timing isn't necessary. It gets the job done faster, but the job still gets done. People worry too much about stuff that doesn't matter and forget the stuff that does. When you've clicked 500 times the 10,20, 30 really bad ones don't matter, nor do the 10,20,30 really good ones. It is the 300-400 acceptable ones that the dog learns from.

If he is allowed to pull into the harness I'm afraid you've got an uphill battle. I don't believe that there are dogs out there who can't learn to walk on a loose leash, but I do believe that some people find it difficult to teach and some time with someone who is competent in teaching this skill would benefit you. If you can set up, say 4 or 5 sessions with helpers with dogs over a two week period then do that. Have no other walks during that time, or if you do, go late in the evening. I can't say for sure how much progress you will make on your own, there is definitely a benefit in having someone who has done it before to show you what to do but you will make progress. Find other ways to exercise your dog during this time, get creative. Don't underestimate how much these sessions will tire your dog out. It is impossible not to pass other dogs but once you realise that you are going to HAVE to control the environment for a while if you want a resolution to this problem you will find places to walk that make it easier. In fact, if you can find a place where there are definitely other dogs around, but you can stay far enough away to set your dog up to succeed then that will benefit you. Obedience clubs are usually a good place to do this, fenced off-leash dog parks, that sort of thing. What do you think your threshold would be currently? (Not the point where he notices other dogs, the point where he vocalises?) You only have to do this until the probability of him giving you calm behaviours around other dogs and being able to give you attention is high. Once he has learned to give you attention around other dogs (not when you say his name, just as something he does) more than half the battle is won.

http://www.dragonflyllama.com/%20DOGS/%20Dog1/levels.html Sue says " They aren't aimed at obedience competition, or agility, or tracking, or any other particular sport or job. " - but don't let that fool you.

One of the ways I do it, you're going to need friends with dogs to help you. Not a group all at once, but you want to be able to repeat this with more than one dog over a period of time. You park at one end of a paddock, have your friend at the other end. Find the "threshold", the point where your dog begins to become excited. Stay below that point (it will change as you progress). Explain all this to your helper, if it is a long distance you will need hand signals. Next you need to teach your dog what you want him to do when he sees another dog, in other words, an appropriate way to approach them while on-leash. Have a clicker and lots of easily swallowed treats ready (I use sausage packed food, e.g VIP Chicken cut into tiny cubes). We don't start off by insisting on the finished product, we start off with an approximation. If your dog isn't barking, or isn't straining on the leash, we click. That should be easy to arrange, just make sure you are far enough from the other dog. If he is "bolt upright and watching them" - that is fine. You can click that. He isn't barking and he isn't spinning on the leash. The point is that we are shaping the behaviours that we want, so we can begin with some excitement so long as, on average over a period of time, we are reinforcing responses that are closer and closer to what we want the finished picture to look like. As your dog starts to offer the really nice responses more often, we can stop reinforcing the less desirable responses. <- read that bit again because it is important We're playing with probabilities here. We're not insisting upon anything, we're just stacking the deck in our favour. Very much in our favour. We set up the environment (distance between dogs) so that we're more likely to get the responses we want. Then we reinforce those responses so that they become MORE PROBABLE even when we change the environment a little (move the dogs a little closer). I'm making a point about talking about probabilities here because there is a tendency for people to have unrealistic expectations, or to become disappointed (devastated even!) when they make a little mistake. This is not necessary. Over time the probabilities of the responses that you want become very high if you set up the environment (distance between dogs, reinforce the best responses etc) to your advantage - CONSISTENTLY. So far I've talked about approximations towards a finished product but I haven't given you any clue what that finished product might be! I would suggest walking on a loose leash, and able to give you more attention than the other dog (giving other dogs attention is fine, so long as that doesn't include barking, lunging etc). Definitely reinforce those things. Actual greetings tend to follow from there without too much help, but feel free to write back when you get to that point (might take one session, might take twenty to get there). If you want a dog to walk on a loose leash you need to never, ever let that dog succeed in getting anywhere on a tight leash. Stop, give him a second to make the decision to stop pulling, then back up a step or two. The instant that he turns to you, click then treat. Click again before he makes his way to the end of the leash, and again, and again. You can start fading out the clicks and treats when he really understands what you are clicking for. You can do this with or without your helper, in fact I would start immediately, which again, stacks the deck in your favour. (For a video demonstration, see http://www.positivepetzine.com/loose_leash_youtube ) I hope that makes sense!

It's a two-way street. I'm sure Sabella thinks she has me trained to come over and rub her belly, all she has to do is go to her mat and roll over when I come in the door.

"Learning theory can be summarised as two basic principles and the first one is that if a behaviour is considered rewarding then it gets repeated and the opposite of that is that if a behaviour is unrewarding or not enjoyable then the motivation to repeat that behaviour is reduced and so it does not occur again." "The difficulty is that unless one is looking at the outcome of using punishment in terms of its effectiveness, the temptation is to increase the strength of the punishment rather than to evaluate the approach to training. The way to evaluate the effectiveness of punishment is to see whether the behaviour is repeated. If the behaviour is repeated it demonstrates the dog didn't learn it." -- Nick Branson, Landline interview One thing about learning theory is that it isn't very prescriptive. It doesn't say "do that this way" it says stuff like "did what you just do increase or decrease the behaviour?" There's nothing fancy about that.

The way I look at it is "you get what you reinforce" If you start to get sub-par responses, take a break.

From the transcript: "NICK BRANSON: There's now a great opportunity with the large number of responses that we got to take that back to the Animal Welfare branch of the Department of Agriculture and say, look, there's a really important need here for us to collaborate more closely with working farmers and give them opportunities to gain access to the information that many of us take for granted now." The strategy proposed has been broadly outlined here: http://www.daff.gov.au/animal-plant-health...are/aaws/online

It might be quicker and cheaper to do a phone consult with a veterinary behaviourist who is licensed to practice in NT. They will zero in on the problem much faster and will be able to tell your regular vet which tests to run and how to run them.

I agree with Nekhbet. If you have a friend with a dog who can help you, even better. Have them at a long distance from you, train the leash manners, and gradually bring them in closer. Keep your dog succeeding at keeping a loose leash and not misbehaving.

I won't hesitate to recommend you see another vet. Everything Erny has said has been correct, IMHO. Look into the possibility of brain injury first. In the meantime, crate and rotate. Keep your dogs separated unless they are actively doing something that they know well and you are there to supervise, even then keep a keen eye out.

Agree with everyone else, sounds more like a medical problem than a behaviour problem which makes the instructor's advice not only dangerous but negligent.

Thanks, I enjoyed that!

The obligatory medical checks have already been suggested, if you can see a veterinary behaviourist, even better. Spray bottles, loud noises etc can make matters worse. I know it's hard, but unless you have a very well trained recall or drop, just try to stay calm until you really have to break something up. I say "until" because in the absence of well-intentioned but misguided intervention often all you will see is some ritualised aggression and then it's all over without any blood or real harm. You have a problem when it goes beyond this, one dog ignore the social signalling of the other, or the more submissive dog starts to live with stress, pushed to the fringes, avoiding anything to upset the aggressor. Certainly separate them when you are not there to supervise. Here is an excellent article, heavy reading, but very informative: http://www.vin.com/proceedings/Proceedings...1&O=Generic

Glad to help Yes, you've got it. You can use the same approach when ticking off your Levels exercises, too.

Hi Bub, lucky dog being trained throughout the day like this! There are a few ways to know when to raise the criteria, all involve making sure that you are getting about 80% success. So, do 5 trials of the behaviour you are testing, if she gets 4 out of 5, start making it harder or different. When she can get 4/5 on that, change it again. Another method is 300 Peck: http://positivepetzine.com/300_Peck ... which works great for distance or duration criteria (e.g going around the pole, or long stays)

Hi DaveS, try to be proactive rather than reactive. Ask for what you want rather than responding to what you get. If you ask for what you want and reward it, you will get more of that (instead of what you don't want). A pup with an awesome sit is going to make this much easier for you.